JPRS ID: 9843 LATIN AMERICA REPORT

• Document Type: CREST
• Collection: JPRS
• Document Number (FOIA) /ESDN (CREST): CIA-RDP82-00850R000400070024-4
• Release Decision: RIF
• Original Classification: U
• Document Page Count: 181
• Document Creation Date: November 1, 2016
• Sequence Number: 24
• Content Type: REPORTS

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL t)S~E ONLY
JPRS ~L/ 10109
10 November 1981
- USSR Re ort
p
CYBERNETICS, COMPUTERS AND ~
AUTOMATION TECHNOLOGY
CFOUO 24/81)
- Fgs$ FORElGN BRAADCAST INFORMA~'ION SER:~ICE
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-00850R440400070024-4
NOTE
JPRS publications contain information primarily from foreign
newspapers, periodicals and books, but also from~ news agency
transmissic~ns and broadcasts. Materials from foreign-language
sources are translated; those from English-language sources
are eranscribed or reprinted, with the original phrasing and
other characteristics retained.
Headline;;, editorial reports, and material enclosed in brackets
are supplied by JPRS. Processing indicators such as [Text]
or [Excerpt] in the first line of each item, or following the
last line of a brief, indicate how the original information was
processed. Where no processing indicator is given, the infor-
_ mation was summarized or extracted.
Unfaaiiliar names rendered phonetically or transliterated are
enclosed in parentheses. Words or names preceded by a ques-
tion mark and enclosed in parentheses were not clear in the
original but have been supplied as appropriate in context.
Other unattributed parenthetical notes within the body of an
item originate with the source. Times within items are as
given by source.
' The conten~s of this publication in no way represent the poli-
cies, views or attitudes of the U.S. Government.
COPYRIGHT LAWS AND REGULATIONS GOVERNING OWNERSHIP OF
MATERIALS REPRODUCED HEREIN REQUIRE THAT DISSEMINATION
OF THIS PUBLICATION BE RESTRICTi:D FOR OFFICIAL USE O~1LY.
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFF[CIAL LJSE ONLY
JPRS L/10109
10 November 1981
USSR REPORT
' CYBERNETICS, COMPUTERS AND AUTOMATION TECHNOLOGY
(FOUO 24/81) ~
CO~~ITENTS
HARDWARE ~
Architecture of Computing Systems With Frogrannned Structure 1
Meet :.r.,~ of USSR _ra~io~al Committee of International Association
on Mathematical and Computer Modelling IMACS (AICA1 4
Selection of Military Compu~er Architecture 5
Multiprocessor Computer Systems T~'ith Cormnon Instruction Flow..~....... 8
Abstracts From the Collection ~HYBRID COMPUTERS AND COMPLEXES~....... 13
' SQFTWARE
Para.llel Microprograrrnning Methods 19
Transfer of BAMOS Prograrrmzing System To Control of Operating
System DISPAK 23
Technology of Developiiig Control Computer Complexes Using
YaUZA-6 Programming and Debugging Automation System 27
Technology of Developing Control Computer Complexes Using
YaUZA-6 Programming and Debugging Automatien G~stem 29
Abstracts From the Journal 'PROGRANA9ING', July-~,ugust 1987. 32
Prospects for Systems and Theoretical Prograrmning 35
_ Automated Control System Software Reliability 36
~ - a- (III - USSR - 21.C S&T FOUO]
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407/02109: CIA-RDP82-00854R000400070024-4
- FOR OFFI('IAL USE ONLY
OPTICAL PR~CESSING
Hybrid Optical-17igital Pulsar Signal Processing System Lt0
Measuring Coordinates of Terrain Reference Points and Determining
Shifts of Cloud Formations by M~ans of Heterodyne Optical
Correlator ~5
Acoustooptical Properties of Inforriation Processing Based on
Nonlinear Acoustical Interaction 60
Effect of Elastic Wave Attenuatian on Output Signal of
Acoustooptical Device for Correlation Analysis 69
Effect of Nonlineari-ty of Acoustic Ligilt Modulators on
Correlation Processing of Narrowband Signals 81
APPLICATIONS
Automation of Settlements With Suppliers in Supplies and Salss
Organizations 8C~
Principles of Software of ~Nefteprodukt' 88
Icie rrtification of Programs and ,~ata Files :in USSR Gossnab
` Information System 90
Sanctioned Access to Data in USSR Gossnab Iilformation System 92
Forming and Issuar;ce cf Certifica~ces of Most Impor~tant Lines 94
Complex of Tasks in Determining Mineral Resources Requirements
of Lithuaniax~ SSR National Econo~y .......o......,, 96
Technological Process of Developing Vocabulary f or Ye5 Computer 98
State of Debtor Indebtedness of Republic Consumers to Organizations
~ of Azerbai.~an SSR Gossnab 100
Automation of Process of Personnel Administration 102
Analysis of State Standardized Working Capital and Current Assets . 101~
Development and Introduction of Completely Automated System for
Col~ecting and Processing Data on National Economy's Fuel
Supply 106
Analysis of Trade Turnover, Incomes, Distribu+.,ion Costs and
Profits of. USSR Gossnab Orga.nizations 108
Automated System for Operative Control of Operational Work of a
Railroad (ASUDO-D) 110
- b -
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFIC[AL USE O1VLY
Gor'kovskiy Railroad Computer Center ~.ilfills Plan 117
Improvement of P7:inning TechnoZogy and Int.roduction of Automa.ted
Control Systems for Railroad Transportation 119
PERSONALI~IES
Computing Mathematics, Cybernetics Faculty Dean Tikhonov~s Work
~ Honored 133
PUBLICATIONS
Abstracts From Journal ~FOREIGN RADIOELECTRONICS~, August 1981 l:~l
Table of Contents From Journal ~EI,ECTRONIC MODELLING~,
March-April 1981 143
Table of Contents F'rom Journal 'EI,ECTR~NIC MODELZING', May-June 1981. 11~6
Supervisory Control of Manipulator Robots 149
~ List of Articles cn Computer Technology, Jatiuary-February 1981 155
List of Articles on Computer Technology, Maxch-April 1981 160
List of Articles on Computer Technology, May-June 1981 175
- c -
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
HARDWARE
ARCHITECTURE OF COMPUTING SYSTEMS WITH PROGRAP'Il'~D STRUCTURE
Novosibirsk /lRKHITII{TURA VYCHISLITEL'NYKH SISTEM S PROGRAMMIRUYEMOY STRUKTUROY:
VYCHISLITEL'NYYE SISTEMY ir! Russian No 82~ 198A (signed to press 2 Dec 80) pp 2,
110-112
[Annotation and abstracts o.f articles from book "Architecture of Computing Systems
With Programmed Structure"; editorial boaxd: V. L. Lyatlov, E. V. Yevreinov~ Yu.
S. Zab'yakov, N. G. Zagoruyk~, Yu. G. Kosaxev (editor-in-chief), V. A. Skorobog~.tov~
� and V. G. Khoroshevskiy, USSR Academy of Sciences, Siberian Department, Institute
- of Mathematics~ 800 copies, 112 pages~
[Text~ The results of investiga.tions of architectural f~atures of computing sys-
tems with a programmed structure are published in the book. A. method. of distribut-
in~ tasks in the system and a method of addressing machines are proposed which
- assure effective accomplishment of computations. Methods are developed for
parallel micropro~ramming: ttie property of non~contradic~ion of paxa11e1 micro-
progr~,ms was investigated, the language af simula,tion of parallel micropro~ram
- structures and a method of ~onverting Petri networks are described which reduce the
complexity of analysis of parallel microprograms for correctness. Some results are
presente~i on developments of softwaxe of systems of mini- and micro-computers and. a
pr.ogramming s~stem for logica,l ana structural simulation.
The materidls of the collection can be useflxl to specialists in the area of com-
- puter technalogy, scientific workers and graduate students in the specialty of
- computers and systems.
uDC 681.3zz.o6:681..3~3z3
ON DECENTRALIZED DISTRIBUTION OF TASKS IN UNIFORM CON1~'UTING SYSTEMS WITH PROGRAMMED
STRUCTURE
CAbstract of article by Kerneysv~ V. V., and M~nakhov~ 0. G.~
~
[Text] A method. is proposed for the separation of cor.nected subsystems intended
for the solution of parallel tasks, each of which requi~es for its accomplishment
- subsystems with a set number of ma,chines. The method does not require knowledge
of comglete informa.tion on the state of the computing system and tasks en+,ering
the system. A me�hod. is presented for addressing ma.chines of a subsystem, a method
which assures Offective realization of interactions between ma.chines.
1
FOR OFFICIAL USE JNLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400074024-4
FOR OFFICIAL USE ONLY
vnc 681.325�5
RECOGNIiION OF THE CONSISTENCY OF STATZONARY SUBSTITUI'ION AIG ORITHM5
[Abstract of article by Sergeyev, S. N.~
~Text~ The article examines questions regarding recognition of the consistency of
parallel substitution algorithms. For a subclass of the class of stationaxy sub-
stitution algorithms--Neumann algorithms--necessary axid sufficient conditions of
consistency are formulated. T he condition of reducibility of a stationary to a
Neumann algorithm is formulated. The class of consistent algorithms is investi-
gated and it is shown tha.t it is enclosed in the class of stationaxy algorithms.
vnc 68i.325~5
LANGUAGE FOR MI~ROFROGRA?`~`'~D DESCRIPTION OF UTIIFORM PARALLEL COMPUTER MODELS
~ [Abstract of article by Piskunov~ S. V.~
[Text? A description is preseizted of the principal constructions of a langua.ge
based on concepts of parallel microprogramming and parallel subs~itution algor-
ithms. The language is intended for microprogrammed description of uniform
parallel computer models. It reflects the localness of cell interactions and the
parallelism of data pr_ocessing in equipment. Application of the language for con-
struction of a micropr.~grammed mod.el of a computer with a uniform structure is
demonstrated. ~
UDC 681.322.01
PETRI NETWORK REDUCTION
~A~straat of artiele by Anishev, P. A.~ ~
[Text] A set of conversions (reduction rules~ of Petri networks is proposed.
Those conversions permit curtailing the initial network while preserving the pro-
perties of animation and safety. An algorithm for network reduction is discussed
from the point of view of the sequence in the application of rules~ their complex-
ity, independence, etc. Some general questions of network reducibility are dis-
cussed.
UDC 661.32�32~
. INVESTIGATION OF A MULTIMACI-IINE MINIMAX SYSTEM FOR DEBUGGING OF PARALLEL PROGRAMS
[Abstract of article by Kolosova, Ye. I.~
rText] The author examines the main distinctive features of the flznctioning of a
two-ma,chine debugging program ir.tended for complex debugging and check routine
tesing of parallel programs in a given configuration of a uniform computing sys-
tem.
. 2
FOR OFFICIAL USE bNLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
- UDC 681.3.06
PRINCIPLES OF CONSTRUCTION' OF A SOFTWARE SYSTEM OF ~ARALLEL COMPUTING STRUCrURE5
BASID ON SPECIALIZID MICRCPROCESSORS
[Abstract o~ axticle by Guzik, Y. F., and Piterskiy~ A. I.]
[Text] A new approach is proposed for construction of software for parallel digit-
al integrating structures (using specialized microprocessors~ ba.sed on use of a
high-level language for description of software and hardware and the metY~od of
cross-development using ~ universal computer.
tTDC 681..142.2:681.31.
TIiE R-LYAPAS-BESM-6�PROGRANIMING SY5TEM. REALIZATION OF THE FIR5T LINE
[Abstract of article by Vorob'yev, V. A., Goralik, V. M.~, and Monakhova~ E. A.~
[Text~ The article describes the structure of the R-LYAPAS-BESM-6 progranuning
system, intended for description and realization of a pro~am system with a total
volume of about 120,000 instructions written on the ma.crolevel~ ba,sic and systems
level of the R-LYAPAS instrumental language. A formal static model of the product
of the system is given and the functions of all system units are described. The
- system is constructed by the method of dynamic unwinding and allows the possibil-
ity of further accumulation.
UDC 62-507
LOGICAL METHODS IN MACHINE VISION
[Abstract of article by Levin~ V. I., and Perel'royzen, Ye. Z.]
[Text~ Logical methods of analysis of a three-dimens3.ona1 environMent are
described which represent a special sorting procedure. Used as mathema,tical ap-
paratus are infinite-val:ued logic and.the theory of loglcal determinants.
COPYRIGHP: Institut matematiki SO AN SSSR, 1980
2174~
cso: 1863/265
3
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
- . FOR OFFICIAL USE ONLY
MEETING OF USSR NATIOAIAL COMMITTEE OF INTERNATIONAL ASSOCIATION ON MATHEMATICAL
AND COMPUTER MC~ELLING IMACS {AICA)
_ Kiev ELEKTRONNQYE MODELIROVANIYE in Russian No 3, May-Jun 81 (signed to press
24 Apr 81) pp I06-107
[Excerpts from article by Ts. S. Ia?atiashvili and A. Ye. Stepanov]
[Exczrpts) A meeting of the USSR National Committee of the International Associa-
tion on Mathemati.cal and Computer Modelling IMACS (AICA) in which specialists of
different organizations of Tbilisi, M~oscow, Leningrad, Kiev, Odessa, Tashkent,
Kishinev, Riga, Kuybyshev and other cities of the Soviet Union parkicipat~d, was
held on 26-29 October 1980 in Tbilisi at the Tbilisi Scientific Production Associa-
tion Elva.
Academi~ian of the Georgian SSR Academy of Sciences I. V. Prangishvili, in his re-
port "Highly ProciuctivP Mul~iprocessor Problem-Oriented Caanputer Complexes for Geo-
physical Problems," preser,.ted two models of highly productive multiprocessor compu-
ter complexes witfi rearranged structure for expeditionary and regional geophysical
computer complexes. One model, containing up ta 64 processors and providing a
speed of up to 100 million operations per seconci, is related to configuration of
machines with many data flows and a single instruction flow for operating in the
single-problem mode with regular problem algorithms havir.q natural parallellism.
Another model that provi3es speed up to several tens of millions cf operations per
second is related to a configuration with many data and instruction fl~ws and oper-
_ ates effectively in single- and multi-task modes with less regular problem algo-
rithms subject to parallell~ng.
The high productivity and survivability of the de~eloped multimicroprocessor compu-
ter systems provided as a result of dynamic r~arrangement of the syst~mm structure
according to the requirements of the computing process, parallelling of the calcu-
lation process at the level of problems, branches of vector and scalar operations
and instructions accomplished without access to an operating system, were noted.
COPYRIGHT: Izdatel'stvo "Naukova dumka", Elektronnoye modelirovaniye, 1g81
- 6521
CSO: ~863/249 '
4
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/42/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
,
- . unc 519~59
SELECTION OF MILITARY COMPUTER ARCHITECTURE
Moscow ZARUBEZHNt~YA RADIGELII{TRONIKA in Russian No 8, Aug 81 pp 24-45
[Article by V. A. Vedeshenkov, N. A. Vlasenko~ and A. M. Shevchenko, candidates of
technical sciences]
[Excerpt~ 6. Final Recommendat~ons on Comput~r Family Architecture Se7.ection
T he conducted analysis ma.de it clear tha.t the strong and weak aspects of the three
most completely evaluated computing system architectures are the following [1~ 11].
Interda.ta 8~32. The architecture of thls computing system ha,s the highest total
evaluation for quantitative criteria and gives better results during the running
of test programs. It has good organiza.tion of the interruption system~ which as-
- sures its efficiency in real time. However, the software of this computing system
is very poor, and this has a direct effect on the cost of its life cycle. There
also are certain doubts about the efficienay of organization in this computing sys-
tem of protection of the state of the ma.chine during.processing of interruptions by
the processor.
IBM 5~370 . The ma.in merit of thi s computing system is its power~l softwa.re ~ as a
result of which its architecture ha.s ~ood indica,tor of cost of the life cycle,
especially when the cost of program development is higY~. On the other hand, the
_ interruption system adopted in this computing system makes it difficult to use in
real-time systems. The running of test progra.ms ha.s also shown that the efficiency
of its architecture is lower than that of the Interdata 8~32 and PDP-ll computing
system architectures. The investigations have ma,de it clear tha.t the junior models
of that family do not assure sufficiently hi~h indicators during their use in con-
trol systems presenting low requirements for the computing system parameters.
PDP-11. The 70th model of this family was exa.mined. This computing system has
fairly complete software and also shows high cha.racteri5tics during the running of
test programs. Because of good organiza,tion of the interruption system it is well
suited for woxk in real time. The PDP-11 ha.s definite advantages over other con-
sidered computin~ systems as regards cost of the life cycle. To the merits of this
computing system one can also add the fact tha.t its microprocessor realization
practically exists. A negative aspect of this computing system is a virtua.l address
space limited to 16 bits.
As a result of the work done the commitee for selection of CFA architecture prepared
the following recomnendations:
5 .
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
ruec urrt~tEw u~~ unLY
1. Zt uas acknowledged that the PDP-11 computing system is the most suitable for
construction of a military computer family ~MCF). The IBM 5~370 was xanked second,
and the Interdata 8~32 third.
- 2. The committee concluded unanimously tha.t a single instruction system and a
single architecture must be adopted f:~r the entire military computer family. ~he
- recommendation that one of the three investiga,ted architectures be used is an
alternative.
3. The committee considers it necessary to mo~erate the individua,l limitations
on selection of the computer family axchitecture (CFA). Among such limitations, in
particular, is the digit capacity of the virtual address space, which r~strains
selection of the PDP-11 architecture as the CFA.
The adknowled.ged winner~ the PDP-11 computing system, has the best architecture in
the sense of its application in practice among all the computing s,ystems created
- up to now. It is produced in tens of thousa.nds and used in quite var3ed applica.-
tions. It h.a.s po�lerful software~ which also continues to be developed at tk~e pre-
sent t,ime. For purposes of the conducted investigation it wa.s not important tha,t
the DEC Company had not been designated to produce a military vexsion af ~the PDP-11
family. The purpose of the project was to determine the computing system architec-
ture with which va.rious producers (not necessax~ly prod.ucers of the commercial
var~ant of the system) wo~ld be able to develop computing systems satisfying with
their pa?~ameters the military requirements.
In accorda,nce with the recommendations the Norden Division of United Technologies
has acquired frodn the DEC Compan a license for the production of a military ver-
sion of the PDP-11 family [8, 19~. That compnay ha.s been designa,ted to produce
- military analogs of all models of the PDP-11 familg--from the LSI-11 microcomputer
to the large PDP~.~70 computer. The militar y version of the PDP-11 family has been
given the name AN UYK-~l. It is assumed that small, medium and large computing
systems of thi.s family, corresponding to models LSI-11~ PDP-11~34~ and PDP-11~70 of
the PDP-11 family, will a very large distribution in tactical data processing sys-
tems. The operating conditions of all Norden computing systems satisfy the corre-
sponding requirements of the military standards and ha.ve smaller dimensions and
sma.~.ler ma.ss tr~a.n similar commercial models of the DEC Compa.ny. Thus, the comput-
ing system AN~UYK=-4~1~34-M occupies about one fourth the volume of the PDP-11~34 com-
puting system, and weighs about one third as much. As a result of a more complex
technology of production the AN~UYK-41 computing syste~r~ is much cheaper than the
corresponding models of the PDP-11 family.
BIBLIOGRAPHY
l. Burr, W. E.~ Coleman, A. H., and Smi.th, W. R. In: AFIPS Conference Proceed-
ings, 1977, Val 4~6. AFIPS Press, Montvale, N. J., pp 131-137�
2. Fuller, S. H., Stone, H. S.~ and Burr, W. E. Ibid~ pp 139-146.
3. F~.iller, S. H. ~ et al. Ibid., pp 14~7-160.
4~. Barba.cci, M., et al. Ibid., pp 161-173.
5. Wagner, J.~ et al. Ibid.~ pp 17~-183.
6
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-00850R040400070024-4
FOR OFFICIAL USE ONLY
6. Cornyn, J. J., et al. Ibid., pp 185-199~
7. Coleman, A. H., and Smith, W. R. COMPUTER~ 1977, Vol 10, No 10~ pp 12-15.
8. Burr, W. E., and G ordon, R. Ibid., pp 16-23.
9. F`uller, S. H., and Burr, W. E. Ibid} ~ z4-35�
10 . Wa1sh, L., and T enold, S. Ibid ~ pp ~-63.
11. Burr, W. E., and Smith, W. R. DATAMATION, 1977, No 2~ pp 48-63.
12. Salisbury~ A. B. In: EASCON'76 Records, pp 106A-106D.
- 13. Larson, C. A. SIGNAL, 19?5, Vol 29, No 9, PP ~-9�
~ 14. Popek, G. J.~ and Goldberg, R. P. CACM, 1974, Vol 17, No 7. pp 412-421.
15. Bratman, H., and Court, T. IEEE TRANSA'ETIONS, 1975, Vol C-24~ No j, pp 365-
368.
16. Barba.cci~ M. R., and Siewiorek, C. P. COMPUTER~ 1977, Val 10, No 10, pp 36-4~3�
_ 17. Burr, W. E.~ et al. COMPUTER, 1979, Vol 12~ No 4, pp 11-22.
18. Dietz, W.B., and Szewerenko, L. Ibid,~ pp 26-33�
19. Stone, H. S. Ibid., PP 35--47.
20 . EI,EKTROIIIKA ~ 1969, Vol 17, pp 72-73 ~
21. Smith, W. R. COATPUTER ARCHITECTURE NEWS~ 1975~ No 3~ PP 15-21.
2~. AEROSPACE DAILY~ 1976, Vol 81~ P;o 11, p 72�
23. ELIICTRONIKA, 1975~ Vol 4t3, No 26~ pp 73-76~
24~. ELEKTRONIKA, 1976, Vol ~9~ No 26, pp 66-b8.
25. Tanenbaum! A. S. CACM, 1978~ Vol 21~ No 3, pp 237-2~�
26. Wald, B., and Salisbury, A. C~MP'UTER~ 1977, Vol 10, No 10~ pp 9-11.
27. Brickner~ D. R. COMPUPER DESIGN~ 1979~ No 12, pp 93-99�
28. Stone~ H. S. COMPUTER, 1979~ Vol 12, No 4~~ pp 9-1C.
29~ Radkowski, E. J., and Blake~ R. G. In: IEEE National Aerospace and Elec-
tronics Conference NAECON'?7 1977, PP 38-~�
COPYRIGHT: Izda.tel'stvo "Radio i svyaz
"Zarubezhnaya radioelektronika"~ 1981
2174 -
- cso: 1863/262
7
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
~ FOR OFFICIAL USE ONLY
UDC 381.325
MULTIPROCESSOR COMPUTER SYSTEMS WITH CONIl~ION INSTRUCTION FLOW
Moscow NII~IOGOPROTSESSORNYYE VYCHISLITEL'NYYE SISTEMY S OBSHCHIM POTOKOM KOMAND in
Russian No 19, 1978 pp 2-4, 106-107
[Annotation, table of contents, foreword and abstracts from the collection: .
"Multiprocessor Computer Systems with Common Instruction Flow", Ordena Lenina
Institut problem upravleniya, 500 copies]
[Text] The structures, sets of instructions and methods of realizing multiproce~-
sor computer systems with common instruction flow and problems of modelling asyn-
chronous processes.are considered. Methods are suggested for making calculations
in parallel and the characteristic features of optimizing the algorithms in the
systems under consideration are investigated.
Contents
- Foreword 4
I. L. Medvedev, Pririciples of I~esigning Multa.processor Computer Systems
- with Common Instruction Flow 5
Ye. A. Fishcher.ko, Selection of Sets of Instructyons .for a Multiprocessor-
~ Computer System with Common Instructian Flow 22
A. Ya. Biryukov, N. I. Golovan, I. L. Medvedev, A. S. Nabatov and Ye. A.
Fishchenko, Resolving Fields of Multiproceasor Camputer Systems' 33
A. Ya. Biryukov, Local,Management in Multiprocessor Computer System with
Common Instruction Flow 40
Yu. S. Zatuliveter, Method of Making Group Calculations in Paxallel in a
Multiprocessor Computer System with Common Instruction Flow 46
S. I. Rozenblit, Organizing the Calculating Process in a Multiprocessor
Computer Complex 55
S. Ya. Vilenkin and A. A. K,omarovskiy, The Sliding Method of Solving Or-
dinary Differential Equations in a Multiprocessor Computer System
with Common Instruction Flow 64
- 8
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/42/09: CIA-RDP82-00850R000400070024-4
~'OR OFFICIAL USE ONLY ~
S. I. Uvarov, Calculation of Arithmetic Expressions in a Multipi~cessor
Computer System with Common Instruction Flow 67
I. M. Grishina, Applying the Dynamic Programming Method to Opt:i.mization
of Statistical Data Processing Problems in Multiprocessor Computer
Systems 779
A. G. Fedotov, The Probability Model of A Multiprocessor Computer Complex 85
A. S. Yermakov, Iterative Calculation of Some Parameters of a Multiphase
Queueing System with Interlocks 93
A. S. Yermakov and A. S. Nabatov, Selection of Buffers for the Stochastic
Models of a Multiprocessor Computer System with Common Instruction
_ Flow 99
For~word
The attitude of users toward multiprocessor computer systems (MVS) with common in-
struction flow fluc:tuates from approval uf these systems as the only possible ver-
sion of the digital ~computer of the future to complete rejection of their effect?.ve-
ness. Essentially, systems of this type are problem-oriented. Therefore, their
use in the corresponding fields may actually yield unusually high effectiveness.
On the other hand, atempts to use multiprocessor computer systems in fields and
modes not inherent to them do not yield the proper effect.
Multiprocessor computer systems with common information flow are effective for prob-
lems whose resolving algorithms first met sufficiently good parallelling and second
require either high speed or low cost-productivity ratio.
The range of problems encompassed by the conjunction of these two requirements is
~ rather broad. They include problems of random process filtration, pattern recogni-
tion, solution of equations in partial derivatives, modelling of dynamic object~
and many others, i.e., problems of processing the results of.geophysical investi-
gations, information from satellites, design and so on.
The proposed collection is dpvoted to the most important aspects of development and
application of multiprocessor computer systems. Problems concerning bath the prin-
ciples of multiprocessor computer system design, sets of instructions and control
systems and organization of the computing process, including making calculations
- in parallel anr~ urganization of algorithms that minimize problem-solving time.
Since the same problem can be solved by several methods, a number of articles is
devoted to selection of aigorithms that permit broad parallelling and to organiza-
tion of the sequence of calculations if broad parallelling is impossible.
Problems of developing complexes that include control computer systems as the cen-
tral computer are considered in four articles.
As a whole, the collection will be useful to both engineers w~orking in fields of
multiprocessor computer system design and application and to students and graduate
students studying specific problems of modern electronic machine building.
FOR OFFICrAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400074024-4
C062 OFF~CIAL USE GNLY
UDC 681.325
PRINCIPLES OF DESIGNING N!L'LTIPROCESSOR COMPUTER SYSTEMS WI`~H C(3MMON INSTRUCTI6N FLOW
(Abstract of article by Medvedev, I.. L.]
[Text] The principles of desig;~ and organization of control of multip~ocessor com-
- puter systems with common instruction f1ow, intencled for solving problens having
various types of parallellism, are considered. The effectiveness of these systems
for different applications are analyzed.
- UAC 681.326.3
SELECTING THE SET OF INSTRUCTIONS FGR A MULTIPROCESSOR CON~'UTER SYSTEM WITH COMMbN
INSTRUCTION FLOW
[Abstract of article by Fish~henko, Ye. A.]
[Text~ The requirements on the set of instructions for a multiproc~ssor computer
system with common instruct~on flow are foz~mulated. Sets of instructions for multi-
- processor computer systems oriented toward solvinq problems with various types of
parallellism are described.
~ UDC 681.3.2
'~,ESOLVING FIELLS OF MULTIPROCESSOR COMPUTER SYSTEMS
~ [Abstract of article by Biryukov, A. Ya., Golovan, Ne I., Medvedev, I. L., Nabatov,
A.S. and Fishchenko, Ye. A.]
[Text] The characteristic features of the structure of the resolving field, ensu-
ing from the concept of a highly productive computer system with common instruction
flow that utilizes the structural programming mQthodKare censidered. Examples of
resolving fi~elds of two systems oriented toward solution of ~ifferent classes of
problems--primarily with logic or arithmetic data processing--are discussed.
UDC 681.326
LOCAL MANAGEMENT IN MULTIPROCESSOR COMPUTER SYSTEM WITH ~ONINION INSTRUCTION FIAW
[Abstract of article by Biryukov, A. Ya.]
[Text~ Problems of using local management to increase flexibility, to expand the
classes of effectively resolved problems, to si.mplify central management and to or-
ganize asynchronous~processes in multiprocessor coinputer sys~ems with common in-
struction flow are considered. The principles of designing local management devices
are outlined and examples of their application are discussed.
10
~ FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-00850R400404070024-4
FOR OFFICIAL USE ONLY
UDC 681.326
METHOD OF MAKING GROUP CALCULATIONS IN PARALLEL IN MULTIPROCESSOR COMPUTER SYSTEM
WITH COMMON INSTRIJCTION FLOW
[Abstract of axticle by Zatuliveter, Yu. S.]
[Text] A method is proposed for makinq group calculations in parallel based on
the asyr.chranism of making the calculation and realizing these calaulations in a
_ multiprocessor computer syster~ with common instruction flow. It is shown that a
- uniformly h~.gh processor load is provided for large-dimension problems.
UDC 681.3.06:51
ORGANIZING THE CALCULATING PROCESS IN A MULTIPROCESSOR COMPUTER COMPLEX
[Abstract of article by Rnzenblit, S. I.]
~ [Text~ A method is proposed for dynamic organication of the calculating process in
- a multiprocessor computer compZex based on the use of a hierarchical N-graph and
its matrix of states.
UDC 681.3.057
SLIDING METHOD OF SOLVING ORDINARY DIFFERENTIAL EQUATIONS IN A MULTIPROCESSOR
CONR~UTER SYSTEM WITH COMMON INSTRUCTION FIAW
_ [Abstract of article by Vilenkin, S. Ya., and K,omarovskiy, A. A.]
[Text] A parallel sliding algorithm for solving the ordinary differ~ntial equa-
tion y~l~ = f(x, y) with the initiaL condition y(xp) = yp is considered.
~ UDC 681.3.06
CALCULATION OF ARITHMETIC EXPFtESSIONS IN A MULTIPROCESSOR COMPUTER SYSTEN ~QI~H
COMMON INGTRUCTION FIAW
[Abstract of article by Uvarov, S. I.]
[Text] TFxe pro~lem o.f compiling optimum schedules for calculation of. arithmetic
exgressions in a multiprocessor camputer system is considered or~ the assumption
- that all processors can. perform the same operation simultaneously. A schedule
com~silation algorithm for two processor~ and the expressions that r.antain addition
= and multiplication operators are presented and investi.gated.
UDC 681.3.U6
APPLYING TiiE DYNAMIC PF20GRAMMING METHOD TO tJE~~.'IMIZR.TIQN OF STATISTZ~Ai, D31TA PRACES-
SIN~~ PROBI~EMS IN MULTIPROCESSOR COP~lPUT~R SYSTEMS
~ [Abstract of article by Grishina, I. bi. ]
~
11
FOIt OFFICI~?L USE !)NL'Y
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400074024-4
FOR OFFICIAL USE ONLY
[Text] A metho~d of selecting the time-optimum realization of an algorithm of given
structure in multiprocessor computer systems with comznon instruction flow, based
on the dynamic ,programming principle, is considered.
UDC 681.327
PROBABILITY MGDEL OF A MULTIPROCESSOR COMPUTER COMPLEX
[Abstract of article by Fedotov, A. G.J
[Text] A twr~-phase model of a computer complex with arbitrary number of devices in
both phases and with a buffer of arbitrary length between. phases and interlocking
of the first, phase for filling the buffer is considered. A solution is found for
the problem and the mean values and standard deviation of the length of the line in
both phases of the system and also the maximum capacity of the system are
determined.
UDC 681.327
ITERATIVE C~sLCULATION OF SOME PARAME`TERS OF A MULT2PHASE QUEUEING SYSZEM WITH
INTERLOCKS
[Abstract of article by Yermakov, A. S.] ~
- (Text] An iterative algorithm is presented for calculating the probability of in-
terlocking of individual phases of an open multiphase queueing system with ~.z?ore
than two phas~s. The resu~ts of calculation are compared to the results of itera-
tive modelling. ~
UI)C E81.327
SELECTION OF BUFFERS FOF2 THE STOCHASTIC MODEL OF A MULTIPROCESSOR COMPUTER SYSTEM
WITH COMMON INSTRUCTION FLOW
- [Abstract of article by Yermakov, A. S~ and Nabatov, A. S.]
_ [Text] The order of solving the problems for determining the lengths of the buffer
- exchange zones is presented for a stochastic model of a multiprocessor computer
- system under development. The problem is solved by means of the iterative method.
The results of iterative calculati.on are compared to the precise solution for a
special case.
COPYRIGHT: INSTITUT PROBLEM UPRAVLENIYA 1978 .
6521
CSO: 1863/242
12
FOR OFFICIAL U5E ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400074024-4
FOR OFFICIAL USE ONLY
_ UDC 681.142
RBSTRACTS FROM THE COLI,~..CTION 'HYBRID COMPUTERS AND COMPLEXES'
Kiev GIBRIDNYYE VYCH25LITEL'NYYE MASHINY I KOMPLEKSY, VYPUSK 3: KONTROL' I DIAG-
_ NOSTIKP. VYCHISLITEL'NYKH I UPRAVLYAYUSHCHIKH SISTEM in Russian No 3, 1980 (signed
to press 25 Nov 80) pp 111-114
[Abstracts from the coll,ection: "Hybrid Computers and Complexes, No 3, Control and
Diagnosis of Computer and Control Systems", Institute of Electrodynamics, Electron-
ics and Modelling Sector, Ukrainian SSR Academy of Sciences, 1,000 copies, 114
pagesj .
UDC 681.142
COMPILING FUNCTIONAL TESTS FOR MICRORPRCESSORS
- [Abstrac~ of article by Gulyayev, V. A.]
[Text] Problems of forma~ized compilation of tests for microprocessors are consid-
ered. P, method is propose.d for preparing functional tests, including compilation
of test microprograms and selection of operands and addresses.
- UDC 65.011.56:681.3.06
- SOFTWARE FOR AN AUTOMATED MONITORING DESIGN SYSTEM
[Abstract of article by Shcherbinin, Yu. G.)
' [Text] The software structure of an automated design and programming system of
automated ntonitoring and control systems ASPP ASIQJ is considexed. Main attention
is devoted to means of describing the data in the system and to means of design and
programming automation. The order of program operation in the system and the main
" characteristics of the programs are shown.
UDC 621.317.7
UNIVERSAL INPUT UNIT FOR A MULTICHANNEL IIJFORMATION-MEASURING SYSTEM
[Abstract of article by IQzoroshko, V. A.J
- FOR OFF[CI~L USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFtCIAL USE ONLY
[Text] The input unit of a multichannel information-measuring system is consid-
ered which permits one to use various types of sensors and to carry out preliminary
processing and signal linearization and normalization. The configuration of the
unit is based on the use of a parallel-series structure with multilevel commutat3on.
UDC 681.142
SELECTING A CONTROL COMPLEX MONITORING SYSTEM
[Abstract of article by Burev, S1. A., Sorochinskaya, 0. A. and Chichikanov, I. V.]
[Text] The effectiveness of introducing a hardware and software monitoring system
for a control computer complex is considered.
UDC 681.142
ANALYSIS OF SEQUENTIAL CIRCUITS IN DEDUCTIVE MODELLING OF MALFUNCTIONS
[Abstract of article by Zalchar'yevich, I. A. and Chaprak, L, I.J
- [Text] Problems o� increasing the effectiveness of deductive modelling of logic
circuits with malfunctions are considered. An algorithm is presented for prelimin-
ary analysis of series circuits that permits one to reduce the storage capacity re-
quired to store lists of malfunctions and to simplify operations on the lists.
UDC 681.142
APPLICATZON OF LOGIC CIRCUIT MODELLING IN STATISTICAL MONITORING METHODS
[Abstract of article by Pleshkanovskaya, 0. M. and Chaprak, L. I.]
[Text] The possibility of using logi.c modelling in statistical methods of monitor-
ing the circuits of digital devices is considered.
~ UDC 681.325.6.001.4
MFTHE.MATICAL MODELZING OF COMBINAT20N CIRCUITS WITH REGARD TO POSSIH~LE MALFUNCTIONS
[Abstract of article by Krasnov, V. V. and Krulikovskiy, B. B.I
[Tex~] Problems of minimizing the generalized logic models of logic components are
considered. Algorithms are proposed for formation of optimum models of combination
circuits.
UDC 681.326
CHECKING A PARALLEL BINARY CODE FOR EVENNESS
[Abstract of article by Rybakov, S. V.j
[Text] Known series and parallel me.*.hods of realizing devices for checking a bi-
nary code for evenness are considered. A series-parallel method of realization is
~4
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400074024-4
- FOR OFFICIAL USE ONLY
suggested that permits one to cl-~ange the speed and amount of equipment of the check-
ing device within permissible limits, thus adapting it to the given requirements.
An example of realizing the proposed method is presented. The speed and amount of
equipment of the checking devices realized by parallel, series and series-parallel
methods are compared.
UDC 681.326.74
FINDING THE SHORTEST DISTINGUISHABLE SEQUENCE FOR A FINITE AUTOMATON
_ [Abstract of article by Belichenko, T. P., Belov, Ge I. and Derbunovich, L. V.]
[Text] An algorithm and program are described for finding the shortest distinguish-
able sequence for a finite automaton. It is sug~ested that simple Q-sets be ex-
cluded from A-groups and that the input data be represented in the form of files to
reduce the number of sampling operations and the required computer storage capacity.
UDC 62-50
- METHOD OF CLEAR IDENTIFICATION OF AN AUTOMA'i'ON BY DETERMINING THE NUMBER OF TRANSI-
TION TRACKS
[Abstract of article by Myz', A. N. and Liberg, I. G.]
[Text] A formal method of clear identification of a sequentia.l automaton by deter-
mining the :ninimum number of transition tracks used to form test sequences in the
case of masking of the output realization of the automa;-~n, which is a composition
component and ather components of this composition, is considered.
UDG 381.326.7
USING BOOLEAN DIFFERENCES TO SYNTHESIZE TESTS OF CIRCUITS WITH PSEUDO-PRIMARY
INPUTS
[Abstract of article by Frolova, I. Ye.]
fText] A mo3ification of an algorithm for synthesis of tests by the Boolean dif-
- ference method is sug~ested for covering all malfunctions upor. branching of primary
inputs of circuits with branching coefficient pi > 1. Analy~is of" the structural
redundancy of a logic circuit with pseudo-primary inputs is presented.
UDC 681.31
DETERMINATION OF MALFYJNCTIONS IN A SYSTEM FROM THE RESULTS OF CHECKING THE
PARAMETERS '
(Abstract of article by Dotsenko, B. I.]
. (Text] Diagnosis of a system fram the results of analyzing the measured parameters
is considered. A dir�erential equation is presented for determining the weight
function of the parameter measuring device. An algorithm for determining the mal-
function in the system is found from the condition of minimum average risk.
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FO~t OFFICIAL USE ONLY .
UDC 621.317
EFFECTIVENESS OF SOME EFFICIENCY CHECKING ALGORITHMS
[Abstract of article by Belokon', R. N., Kendel', V. G. and Kuznetsov, A. M.]
[Text] Three effici.ency checking algorithms--optimum, majority and secluential ma-
jority--are aonsidered and investigated. The checking algorithms are compared by
reliab~lity, average risk and hardware realization. The advantages of a sequential
majority efficiency checking algorithm are shown.
UDC 621.396.66
ANALYZING THE RELIABILITY OF RESULTS FROM QI3ANTITATIVE CHECKING OF THE EFFICIEN~Y
OF ARTICLES
[Abstract of article by Belokon', R. N.]
[Text] The problem of determining the reliability of results of quantitative check-
ing of the efficiency of articles D is considered. Analytical expressions are sug-
gested for calculation of indices D. The relationship between indices of the reli-
_ ability of quantitative and qualitative control is s~own.
UDC 62-50
DETERMINING THE PERIOD OF CHECKING THE PARAMETERS OF A MONITORING AND CONTROL
F'ACILITY
[Abstract of article by Ivano~, V. M.]
[Text] The problem of determining the period for checking parameters during tol-
erance monitoring is considered. The derived processes are introduced bX~using
nonlinear transformation of the initial process when fixing the tolerances. The
problem of determining the checking period is solved as one of time quantification
of the derived processes. The properties of one- and ~wo-dimensional and probabil-
ity densities and correlation functions of derived processes are used.
UDC 681.32.019.3
ANALYZING P.LGORITHMIC METHODS OF INCREASING.THE NOISE STABILITY OF LOGIC CONTROL
DEVICES
[Abstract of article by Koloskov, V. A., Koloskova; G. P. and Tipikin, A. P.]
[Text] Algorithmic methods for increasing the noise stability of control devices
are considered. A method is proposed for analyzing the probability of correct so-
lution of the control prc+blem with regard to the self-correcting capability of the
input algorithms. The dPpendence of the probability of correct solution of the
problem by transformed algoritYuns on self-correcting capability, the detecting ca-
pability of checking and the reliability of the units of the device is found. The
effectiveness of the method of increasing noise stability, based on joint use of
periodic and forced returns, is shown. .
16
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
UDC 62-50
USE OF ODD SETS FOR DIAGNOSIS OF CONlPLEX SYSTEMS
[Abstract of article by Vakhid, A., Voronov, V. G., Derbunovich, L. V. and Kardyu-
mov, A. I.]
- [TextJ The use of odd sets in diagnosis of complex systems is described for which
~ precise mathematical methods are unacceptable due tq the large volumes of calcula-
tions. Diagnosis of a system consisting of m units whose aggregate syndrome is de-
scribed by the state of diagnostic signals of each unit, is cnnsidere3.
UDC 681.327:681.326.74.06
ALGURITHMIC METHOD OF CHECKING SEMICONDUCTOR STORAGE DEVICES
[Abstract of article by Belov, G. I.]
[TextJ A method is described for checking constant mal�unctions in semiconductor
storage devices with arbi~rary access. It is shovm that the considered method is
easily realizable by hardware.
UDC 518.12
ADAPTIVE ALGORITHMS FOR CHECKING THE NUMERICAL SOLUTION OF EQUATIONS OF DYNAMICS
[Abstract of article by Latyshev, A. V.]
[Text] Algorithmic checking of breakdowns at each step o� real-time numerical
solution of equations of dynamics is considered. Adaptive versions of checking
algorithms based on extrapolation and interpolation of the process being monitored
~ are proposed.
UDC 62-50:519.14
AUTOMATIC DESIGN OF A TOPOLOGICAL MODEL OF A COMPLEX SYSTEM FOR SOLUTION OF PROB-
LEMS OF DIAGNOSTICS
[Abstrar.t of article by Grundspen'kis, Ya. A, and Tenteris, Ya. K.)
(Text] Computer design of a topological model of a complex system in the form of
an oYganizational flow chart is considered for solving problems of diagno~is of
complex analog systems. A method of forming a set of functional phrases from a
meaningful description of the system, on the basis of which the procedure for fcr-
mation of the arc between two vertices is developed and an algorithm for design of
a topological model realized in the form of a program written in PL/1 language, is
pr~posed. The main principles of operation of the algorithm in two modes are
outlined.
FQR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FnR nFFIC1A1, IISF, ONI.Y
UDC 621.3.019.3
MATHEMATICAL MODEL OF THE RELIABILITY OF A WEAK-CURRENT SLIDING CONTACT
[Abstract of article by Strel'nikov, V. P., Pranik, B. V. and Kostra, M. G.]
[Text] A mathematical model of the reliability of sliding electrical contacts,
designed on.the basis of analyzing the dynamics of the process of variation of
contact resistance, is considered. Variation of the dynamic contact resistance is
approximated by a diffusion Markov process. An experimental-statistical check
showed that the proposed model does not contradict experimental data and has more
effective approximating qualities upon equalization of the statistics of sliding
contact failure compared to other distributions (exponential, Weibull and normal).
UDC 62-501.14
ALGORITHM FOR DETERMINING THE TOLERANCES OF THE PARAMETERS OF INFORMATION DEVtCES
[Abstract of articie by Pampuro, V. 2. and Novitskiy, V. V.]
[Text~ The solution of a problem and an algorithm for determination of the toler-
ances of the determined rational function of the output parameter with incomplete
statistical data on independent variables is outlined. An example of computer so-
lution is presented. Some theorems that determine the point of the highest (low-
est) value of the function of many variables are proved in the appendix, which
permits one to exclude selection of all its boundary values.
UDC 62-50:621.311.22
PREDICTING THE VARIATION OF THE PARAMETERS OF TI~RMAL MONITORING AND CONTROL
EFFECTS OF THE STEAM GENERATORS OF ENERGY UNITS
[Abstract of article by Ivanov, V. M.]
[Zext] The problem of predicting the variation of parameters and the control ef-
fects of a complex object is considered for use in real time by means of control
computer complexes. Metnods of ident~fication and self-adjusting systems are useci
as the basis. The parameters of the model and facility are adjusted by the compen-
sation scheme with a parallel local model.
COPYRIGHT: Izdatel'stvo "Naukova dumka", 1980
6521
CSO: 1863/243
18
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ON?.Y
SOFTWARE
_ PAFALLEL MICROPROGRAPM~SING METHODS
Novosibirsk N~TODY PARALLEL'NOCO MIIQtOPROGRAMMIROVANIYA in Russian 1981 (signed to
press 13 Feb 81j pp 2-4, 179-180
[Annotation, f oreword and table of contents from the book "Parallel Microprc~gram-
ming Methods", by Petr Aleksandrovich Anishev, Svetlana Mikhaylovna Achasova, O1'qa
- Leonidovna Bandman, Sergey Vladislavovich Piskunov and Stan~slav Nikolayevich Ser-
geyev, Izdatel'stvo "Nauka", 4,60~ copies, 181 pages) .
[Text] The book is devoted to theoretical problems of parallel microprogramming
and methods of designing microprogram computer systems and parallel in�ormation
processing structures. Parallel microprogramming is considered as a combination
af algorithmic means of describing parallel calculations with the principles of
modern microprogramming. Methods of representing parallel microprogram algorithms
- and methods of interpreting them by automaton networks and homogeneous machines are
outlinefl in detail. The possibilities of using the considered methods.to design
parallel information processing devices based on a modern component base (micro-
processor sets, programmable logic n:atrices and so on) are shown and practical
algorithms for design of microprogram control are also presented.
The book is intended for scientific workers, graduate students and students special-
izing in the field of computer technology and engineering cybernetics and also for
developers of computers.
Foreword
The results o� research conducted at the Institute of Mathematics, Siberian Depart-
ment, USSR Academy of Sciences, in the field of developing the microstructure of
parallel processing computers are outliried in the book. The investigations were
begun in the mid-1960's under~the supervision of E. V. Yevreinov, who determined
the directions of research and developed the basic theoretical propositions." Dur-
ing those years microelec*_ronics was seemingly being developed at such rates that
production of homogeneous structures consisting of millions of switching and stor-
- age devices in a single production cycle which could then be converted.through pro-
gramming to specialized parallel processing computers, would become a reality
. *
E. V. Yevreinov and Yu. G. IG~sarev, "Odnorodnyye vychislitel'nyye sistemy vysokoy
proizvoditel'nosti" jHighly Productive Homogeneous Computer Systems], Movosibirsk,
Nauka, 1966.
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-40850R040400074024-4
FOR O~FICIAL USE ONLY
within several years. These structures were called computer media aince they were
regarded as spaces filled with 1~hysical objects whose funcLional properties could
be varied externally.
The search for ways to develop computers based on computer media was conducted in
all directions. The technological poss~bilities of producing media were considered,
the optimum functional component bases were determined, the theory of structural
organization was developed and methods of logic synthesis in the media and algo-
rithmic parallelling devices to the level of microoperations were worked out. New
problems were determined and new difficulties arose during the investigations. De-
velopment of microelectronics was not as vigorous as initially assumed, but the suc-
cess achieved determined the solution of a number of problems. 2'hus, the extensive
use of microprogramming and the appearance of microprocessor units and programmable
logic matrices had a very important effect on the development of the idea.of creat-
ing homogeneous structures. Development of theory of parallel processes influenced
the solution of some software problems.
The concept based on a combination of devices for parallel representation of algo-
rithms and methods of microprogram interpretation of them was established during
the investigations. This concept was called parallel microprogramming. Devices
for microprogram description of algorithms in limiting-parallel form, methods of
interpreting this description by automaton networks, methods of asynchronous compo-
sition of algorithms, methods of embedding microprograms in logic matrices and
language for microprogr~m description of parallel processing structures and a sys-
tem for si~:uiating their operation in an ordinary computer were developed on its
basis.
All the foregoing is outlined in six chapters. The contents of the chapters are re-
lated. However, chapters 2, 5 and 6 can be recommended to those interested only in
limit-parallel algorithms, while chapters ~ and 4 can be recommended to those who
are working in the field of synthesis of asynchronous control and parallel processes
ses, omitting from these chapters the few places which concern parallel substitution
- systems. Moreover, chapter 4 can be understood as an outline of the original method
of synthesizing logic function system~ for ~.nclusion in programmable logic matrices
and can be read separately..
The following persons participated in wrzt.ing the book: P. A. Anishev--sections
3.1 and 3.3, S. M. Achasov~-sections 4.1, 9.3 and 4.4, 0. L. Bandman--chapter 1
and sections 3.2-3.5, S. V. Piskunov--chapter 5 and sectioas 6.1 and 6.4, S. N.
Sergeyev--sections 2.2, 2.4, 2.5, 6.2 and 6.3. S. M. Achasova and O. L. Bandman
wrote section 4.2 jointly, O. L. Bandman and S. V. Piskunov wrote section 3.4 .
jointly and S. V. Piskunov and S. N. Sergeyev wrote sections 2.1 and 2.3 jointly.
Besides the enumerated authors, Yu. N. tGornev made an important contribution to
development of the idea and methods.
The authors are deeply grateful to V. G. IQioroshevskiy for support and kind advice,
to I. V. Ilovayskiy and Ya. I. Fet for valuable comments and useful discussions of
the book and also to L. V. Alekseyeva for assistance in organizing the manuscript.
20
FOR OFFICIAL U5E ONLY '
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R004400074024-4
FOR OFFICIAL USE ONLY
Contents Pa9e
Foreword 3
Chapter 1. The Object and Problems df Parallel Microprogramming 5
1.1. ' Modern trends in design of compnter microstructure 5
1.2. Microprogramming as a method of algorithm intezp:etation 8
1.3. Main concepts of parallel microprogramming 10
1.4. Models of parallel calculations 12
1.5. Problems of synthesizing parallel microprogram structures 15
Chapter 2. Parallel Substitution Alc~orithms 16
2.1. Determination of parallel substitution algorithms 16
- 2.2. Main properties af parallel substitution algoz~ithms 23
2.3. Interpretation of parallel substitution algorithms by automaton
networks 30
2.4. Parallel machines 35
2.5. Parallel microprogram structures 41
Chapter 3. Asynchronous Camposition of Parallel MicroproRrams 44
3.1. Parallel graph-scheme of algorithm 44
3.2. A Petri network as an abstract model of asynchronous inter.actions 50
3.3. Checking qraph-schemes for co~rectness 59
3.4. Microproqram description of the cc~mposition of paralle]L substitu-
tion algorithms 66
3.5. Conversion of th~ control microproqram 71
3.6. Automat interpretation of asynchronous paratlel control algorithm 75
Chapter 4. Matrix Representation of Parallel Microprograms and Synthesis
of Microprogram Automatons 83
4.1. Matrix representation of microprogram automatona and problems of
synthesis 83
4.2. The theoretical-structural approach to matrix synthesis problems 91
4.3. Algorithm for reducing microinstructions to abbreviated form 101
4.4. Microinstruction coding algorithm 107
Ghapter 5. Microprogram Models of Homogeneous Computers ?13
- 5.1. Parallel microFrogramaning language 113
5.2. Procedures for design of microprogram models of homogeneous
devices . 12~1
- 5.3. Examples of microprogram models of homogeneous devices 12~
5.4. Overall block diagram of parallel microprogram modelling system 13Q
Chapter 6. Microprogram Parallel Processing Structures 144
6.1. Hom~geneous microprogram processor for solution of linear
algebra problems 144
� 6.2. Homogeneous signal processor 152
6.3. Realization of parallel substitution algorithms in microprocessor
systems 159
6.4. Modelling of calculations in a homogeneous micraprocessor
system 168
21
F4R OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00854R400404070024-4
FOR OFFICIAL USE 4NLY ~
Conclusions 1~2
Bi_bliography 173
Subj2ct Index 177
- CO]PYRIGHT: Izdatel'stvo "Nauka", 1981.
6521
CSO: 1f363/241
22
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAI. USE ONLY
TRANSFER OF BAMOS PROGRAMMING SYSTEM TO ~ONTROL OF OPERATING SYSTEM DISPAK
Moscow PERENOS SISTEMY PROGRAMMIROVANIYA BAMOS FOD UPRAVLENIYE OS DISPAK in Russian
- 1980 (signed to press 25 Mar 80) pp 3-8.
[Excerpt from the report "Tran sf e r of BAMOS Programming System to Control of
Operating System Dispak", by D. Karl, M. A. Kopytov, V. F. Tyurin and E. Yunker,
Computer Center, USSR.Academy of Sciences, 301 copies, 31 pages]
[Excerpt] i. Introduction. Purpose of Paper
Active cooperation between the USSR and the GDR in the fie].d of usi.ng the BESM-6
computer began at the moment when the internal systems so~tware for the BESM-6 had
already been mainly formulated in both countries in the form of independent operat-
ing systems (OS) and programming systems. The operating system DISPAK [expansion un-
known] (the operating systems DUBNA, IPM and ND-70 were also developed in the IISSR)
of the MONITOR SYSTEM DUBNA and BESM-ALGOL programming system achie~red the greatast
distribution in the USSR. The operating system BAMOS (Batch-Proaessing Multilan-
guage Operating System), which (also like the ~perating system DISPAK) was essen-
tially based on the first operating system--the D-68 for~the BESM-6, was developed
in the GDR for the BSM-6.
The MONITOR SYSTEM DUBNA (as of 1969) was taken as the basis in developing the
BAMOS programming system.
Different systems interfaces were offered to users in both countries for developing
their own programs as a result of this development of the systems software of the
BESM-6, which caused spAcific difficulties in exchange of programs between the or-
ganizations o~ both countries.
Investigations were conducted in the USSR to transfer ir~c~iviciual elements c,f the
BAMOS programming system. The ALGOL-GDR and FORTRAN-GDR translators built into the
MONITOR SYSTEM DUBNA were widespread among users in the USSR. The presence of these
translators within the MONITOR SYSTEM DUBNA facilitated adaptation in the USSR of
the programming syst.ems developed in the GDR, but did not completely resolve the
problems of transfer. Therefore, USSR and GDR system programmers agreed on joint in-
vestigations in 1974 to connect the BAMOS programming system to the operating system
DISPAK.
FOR OFFIC[AL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-00850R400404070024-4
FOR OFFICIAL USE ONLY
The main purpose of this investigation was that any program or programming system
~ which was developed for the opera~ing system BAM~JS could be realized without any
modification on the BESM-6 under the control of the operating system DISPAK.
The work was conducted in stages during the period from 1974 through 1979. A num-
ber of USSR organizations utilized the investigation to operate specific program-
ming systems developed in the GDR.
Two preprints were pubZished in the USSR on the BAMOS-DISPAK programming system
[20 and 21] .
This preprint is a unique report on the investigation. The brief structural char-
acteristics of DISPAK and BAMOS are presented in it and special attention is devoted
to interaction between the operating system and programming system.
The characteristics of realizing BAMOS-DISPAK and the results achieved are then
described.
In conclusion the authors consider the given paper among the common problems of
program system tran5fer.
2. Principles of Joining
2.1. Functions and Structure of DISPAK
Multiprogramming is realized in DISPF,K for 16 virtual channels (four for the paper
tape operating system).
The information buffering mode is used extensively in the operating system as input
and output information. This permist one to specifically orqanize effective load-
ing of the central processor of the BESM-6. There may be up to 2~5 problems in ths
input buffers of the operating system and ~ach problem reports information an the
required resources to the operating system through special sections that compxise
the problem certificate. A special planning part of the operating system from the
total line of input problems selects the corresponding problems and transfe~s them
for processing. The planning system algorithm tak~s into account the total volume
_ of physical resources of the BESM-6, the presence of resources required for per-
_ formance of a given task, the priority of the problem, its weight (some functior., of
reading time and the volume of indicated resources) and also the specific mode of
selecting problems for solution assigned by the human operator (actually the ratio
betvreen the numh?r c~f "lc~ng" ~nd "4hort" programs) .
The operating system allocates the resources to the problem which occupied the vir-
- tual channel for processing according to the requirements of the problem. These
resources are allocated statically and are attached to the problem during the entire
time of solving it. one can demonstrate (and reject) only magnetic tapes and disks
dynamically. The capacity'of the internal storage should not exceed 32K words.
There is an apparatus (so-called basic capabilities of the operating system) in the
ciisk version of the operating system which permits one to construct programming sys-
tems consisting of the main and subordinate problems which, occupying the same 16
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOE~ OFFICIAL U~E ONLY
virtual channel~, can exchange resources and mes~ages with ea~h other. Information
can be processed hy means of this apparatus by u$ing inte~::nal storage cagacity
greater th~n 32K words.
The operating svstem DISQAK supports the operation of BESM-6 complexes joined on
- the basbs of a common disk memory. There is the capability of operating with mag-
- ne~ic ta~s installed on different BESM-6 and on message ~xchange betwe~n machines,
ThESe capabilities are essentially sufficient to const~uct systema under. the control
of the operatir~g system DI5PAK which could be used by the aggreyate resou~ces of
several BESM-6 computers.
A set of programming systems operates unaer the c~ntrol of the operating syctem
DISPAK. ;.'he operating system "knows nothiny" about the internal structures of the
programming systems. The operating system only supports a specific set of functions
- (extra codes) required for functioning of the progr~maning systems. The operating
system also makes available specific information security devices to programming
systems. Most programming systems do not now ut~lize the basic cap,abilities of the
- operating system DISPAK.(but these devices can ~perate at the level of fulfilling .
the working program) and operate in the ar~a with capacity of no more than 32K
words. The programming systems are arranged on a special disk (tape), from whence
each problem belonging to a specific programming system loads the corresponding
parts of the given prc~gramming system into its own internal storage. The program-
ming system thus operates on the resources of a specific user problem in the mathe-
matical mode. All the specific functions are realized by means of the extra codes
of the operating system DISP~K. This method of organizing the work of pragrammina
systems (outside operating systems) has botY~ its positive and negative aspects. The
following should be included among the negative aspects.
Each programming system is forced to have its own "instrument" components: a moni-
tor, editor, loader and so on. For this reason the user pxperiences difficulties
if he must use components (for exam~Ole, libraries) of different programming systems.
This disadvantage is smoothed over in real operation by the fact that the multi-
language programming system MONITOR SYSTEM DUBNA is finding ever wider distribution
at most computer centers.
A positive aspect in this organization of ogerating system and programming system
interaction is that there is definite simplir.ity in connection of any new program-
ming system. The programming system should be oriented only toward the set of
extra codes of the operating system DISPAK. This partially explains the fact of
how rapidly the BAN[OS programming systems could be connected to the operating sys-
tem DISPAK.
2.2. Functions and Structure of BAMOS
_ The bench processing mode using multiprogrammi;ng that provides effecti.ve use of a
central processor and internal storage is real:ized in the BAMOa.
The specific goals of developing the BAMOS also included the following:
--connection of different computers (Siemens 4004/26, YeS-1020 and control com-
puters of TPAI type) to the BESM-6;
FOR OFFICI~L USE ONLY ,
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02109: CIA-RDP82-00850R400400070024-4
FOR OFFICIA.L USE ONLY
--connection of nonstandard peripheral equipment to the BESM-6; .
--realize remote batch processing by using government communication lines (the
GDR mail system).
COPYRIGHT: Vychislitel'nyy tsentr Akademii nauk SSSR, 1980
6521
CSO: 1863/247
26
- FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
TECHNOLOGY OF DEVELOPING CONTROL CQMPUTER COMP7~EXES USING YAUZA-6 PROGR.~MNNUNG
AND DEBU~GING AUTOMATION SYSTII~I
Moscow VYCHISLITEL'NYYE SISTEMY in Russian No 2, 1981 (signed to press 16 Mar 81)
p 168
/Table of contents from the collection "Computer Systems",~edited by E.V.
Yevreinov, Izdatel'stvo "Finansy i statistika", 15,000 copies, 168 pages7
~ext7 1. Development and Improvement of Camputer Systems
Yevreinov, E. V., N. N. Yegor.ov and V. D. Ryzhavin, The Administrative
Rayon and the Distribizted Computer System ~
Korneyev, V. V. and V. G. IQzoroshevskiy, Organization of System Manage-
ment in Computer Systems with Programmable Structure 20
Lipayev, V. V., L. A. Serebrovskiy, F. A. Kaganov, B. A. Korepanov,
M. A. Minayev and A. A. Shtrik, Technology for Development of Con- .
trol Computer Compl~xes Using the YaUZA-6 Program and Debugging
Automation System 38
Zhiratkov, V. I., Analyzing the Engagement of Computers in a Distributed ~
_ Computer System Upon Entry of the Problem Flow 57
Mamzelev, I. A., Some Problems of the Model Theory of a Computer Collective 63
Kakulin, N. N. and A. P. Pshenichnikov, The Effect of Information Trans-
mission Rate Through Communication C'hannels on the Productivity of
Distributed Computer Systems 69
Arkhipov, V. D. and B. Ye. Tomilov, Requirements on Development of Parallel
- Data Processing Syste~n5
Yel'kin, Ye. A., Characteristic Features of the Hardware of Distributed
; ParalleZ Data Processing Systems 85
Kim, Ye. x., Using Situational Control Principles in Parallel Data Process-
- ing Systems 92
Bagrinovskaya, G. P. and V. V. Tereshenkova, Design and Analysis of the
Data Base Structure 99
Vasil'yev, V'. F., G. I. Yaxalov, A. P. Pshenichnikov and V. E. Yevreinov,
Prospects for Designing the Information Services of Telephone Net-
works on the Basis of Distributed Computer Systems 114
2. Automation and Mechanization of Accounting and Computer Operations
Polkovskiy, L. M., Problems of Integrated Accounting Information Process-
ing in Agriculture 119
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
- FOR OFFICIAL USE ONLY
Bartel's, Ye. I., Formation of Reference Infor-mation in Complex
Mechanization of Accounting in Agricultur~ T9ith the Use of
Computers 133
_ Dagayeva, M. D. and Ye. I. Samuseva, Experience of Using the SPD-9000
- Automated Data Preparation System in the "Accounting" Subsystem
of the ASU-Pribor 144
Tomlyanovich, A. D., The Possibility of Formalizing the Primary Docu-
ment for Automation of Subsequent Programming 154
Tsigel'nik, M. A., The Possibility of Improving the Organization of
Machine Carrier Preparation Using Multikeyboard Systems 161
COPYRIGHT: Izdatel'stvo "Finansy i statistika", 1981
6521
CSO: 1863/252
~
28
~ FOR OFFICIAL USE ONLY .
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-40850R040400074024-4
. FOR OFF[C[AL USE ONLY
TECHNOLOGY OF DEVELOPING CONTROL COMPUTER COMPLEXES USING Y~U~A-6 PROGRAMMING
AND DEBUGGII~G AUTOMATION SYSTEM
Moscow VYCHISLITEL'NYYE SISTEMY in Russian No 2, 1981 (signed to press 16 Mar 81)
pp 38-56
[Article by V. V. Lipayev, doctor of technical sciences, professor, L. A. Serebrov-
skiy, candidate of technical sciences, F. A. Kaganov, B. A. Korepanov, M. A.
Minayev and A. A. Shtrik, candidate of technical sciences, from the collection
- "Comguter Systems", edited by E. V. Yevreinov, Izdatel'stvo "Finansy i s~atistika",
15,000 copies, 168 pages]
[Excerpts) A significant shift has been noted recently in the approach to software
development (MO) for control computers and specifically for specialized machines
(STsVM [Specialized digital com~uters]). Whereas attention was concentrated mainly
in early developments of technology on programming languages and mean~ of program
translation, complex program de~ign systems, including debugging devices, wer~ sub-
sequently developed. The need to develop complete program design technology, be-
ginning with development of algorithms and ending with embc~diment of the debugged
proyram as a software article, its documentation and accompaniment during circula-
tian,has now become obvious [1, 2j. The characteristic features of the structures
and specifications of STsVM require development of the technology for working out
prc~rams based on cross-systems realized on powerful production (instrument) uni-
vprsal type digital computers.
_ An example of taking this approach toward develogment of control computer complexes
(KUP) may be the YaUZA system [3]. The first version of the YaUZA-1 programming
automation system based on the M-220 production digital computer (1970-1975) mainly
performed functions of program translation and did not solve debugging problems.
The first versions of the YaUZA-6 programming an3 debugging automation system SAPO
based on the BESM-6 digital production computer (1g75-1977) were develaped as
adapted cross-systems that perform a rather wide ranqe ~f complex of problems of
design, debugging, development control, STsVM program documentation and so on.
Modern versians of the YaUZA-6 SAPO (1980) in combination with benches for c~mplex
dynamic debugging provide a complete production cycle of program development for a
wide range of cqntrol and specialized digital computers. The production process
of developing KUP for STsVM can generally be divided into the following phases
(Figure 1):
The first phase is general design of the control system and the algorithms, in--
cluding cietermination of the system problems, selection of control methods, the
29
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
type of specialized digital computers, methods of organizing calculations and the
structure of control program complexes, formulation of general and snecial algo-
- rithms and organization of the STsVM memory;
The second phase is preparation of hardware, which includes adaptation of the
cross-system for a specific STsVM, preparation of the SAPO data base and working
out instructional materials from programming and debugging technology;
The third phase is development of programs, including writing the programs in algo-
rithmic languages~ macrolanguages or autocodes and translation, monitoring and
loading of the programs into the STsVM memory;
The fourth phase is program debugging by determined tests through an instruction
interpreter, obtaining the statistical and tim2 characteristics of programs and
checking the connections of program control and informatibn modulest
The fifth phase is production of machine carriers for program input into STsVM and
documentation of the program complex.
These phases require a rather powerful digital production computer. Moreover, pro-
gram developr.!ent includes phases of final complex dynamic debugging and testinq;
The sixth phase is complex dynamic debugging of the KUP based on the STsVM with use
of simulation benches to model the external conditions of using the STsVM in the
control system;
The seventh phase is half-scale or full-scale tests usinq real control facilities
or si.mulators of them to check the efficiency and characteristics of KUP during
functioning of the ~sy~stem under conditions which were formulated by assignment of
the control system �or development. ~
The latter two phases are usually conducted by using special equipment and are not
considered in this paper. After development and testing of the KUP has been com-
pleted, the complex can be circulated, modernized and corrected during the entire
operating time of the system. To do this, the technology of accompanying the KUP,
which can be based to a significant degree on program design devices, should be
developed.
The first five of the indicated phases of the unified production process for design
of complex program systems using the YaUZA-6 SAPO are described below.
Conclusions
The considered phases of program development using the YaUZA-6 SAPO have unified
engineering, programming and organizational bases. The unity of the enqineering
base includes the fact that the control program complex is being developed and be-
ing debugged autonomously on a single digital production computer--the BESM-6.
Whereas STsVM have devices for autonomo~zs program debugging and correction, the
unified technology requires organization of a channel for the STsVM to communicate
with the digital production computer for operational exchange of IaJP variations
and support of design unity. Dynamic debugging on simula*_ion benches permits the
us~ of other digital production computers as well.
= 30
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-00850R440400070024-4
FOR OFFICIAL USE ONLY
The unity of the programming base is supported by the general design data base
which contains the initial texts of all programs, Yibraries of absolute or load
modules, certificates of translated programs, descriptions of global variables,
debugging assignments and also other data on the KUP and individual programs.
The unity of the organizational base includes monitoring the progress of programs
through the phases of the production process, keeping an account of the results oy
each phase, the expended resources and the number of era:ors of one or another type.
Moreover, the command supervisor accomplishes automatic monitoring of the course
of design with various degrees of detail.
- The standardized technology of transfer of the YaUZA-6 SAPO by users to other or-
- ganizations should especially be determined, in which the following is transferred:
a magnetic tape with a copy of the standard refe~ence version of the SAPO ad-
justed to the standard STsVM;
a copy of the reference magnetic tape containing test and contrnl information;
a control pack with complete control task that checks the operation of the
- subsystems of the YaUZA-6 SAPO;
a model for unsealing the standard output information.
A complete set of the operational documentation formulated as enterprise standards
is also transferred. The complete set of standards (31 standards) contains infor=
mation on the system as a whole and about each subsystem for desc.ribing the SAPO,
programming and debugging control languages and also supervision on preparation and
operation of the system.
- The technology of transferring the YaUZA-6 SAPO permits the user not only to become
familiar with the system but also to assimilate all of its operating modes on the
example of the standard STsVM and then to convert ta adaptation of the system for
a specific STsVM.
The YaUZA-6 SAPO consists of 23 subsystems that realize the enumerated problems.
The SAPO is adapted to different conditions of application (different STsVM, auto-
codes and command characteristics) by the automated method through the adjusting
devices included in the SAPO. The overall program capacity of the SAPO comprises
330,000 of the BESM-6, of which 302,000 are machine-independent and remain un-
changed when the STsVM is replaced. Information modules for adjustment to the
STsVM and the command comprise 16,000 words, while the machine-independent program
procedures occupy approximately 12,000 words. The YaUZA-6 SAPO was adjusted to 26
types of STsVM in 14 organizations during the period from 1976 through 1980.
COPYRIGHT: Izdatel'stvo "Finansy i statistika", 1981
6521
CSO: 1863/252
. 31
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
ABSTRACTS FROM THE JOURNAL 'PROGRAD~IING~, JULY-AUGUST 1981
Moscow PROGRAMNIIROVANIYE in Russian No 4, Jul Aug 81 pp 95-96
[Text]
UDC 519.1
SEMIGROUP MOD~LS OF PROGRAMS
[Abstract of article by Podlovchenko, R. I. , Yerevan]
[Text] Given in this art.icle are the suff icient conditions for a formal model of
programs to have a semantic model equivalent to it; thus the concept of a semigroup
model appears. The problem of equivalence of schemes of programs is considered
for the semigroup model. Bibl. of 4 titles, 2 f igs. and 3 tables.
UDC 681.323
LANGUAGE FOR GRAPH-SCH~MES OF PARALLEL ALGORITHMS AND ITS EXPANSION
(Abstract of article by Aref'yev, A. A.; Korablin, Yu. P. and Kutepov, V. P.]
[Text] Described in this article is a language for graph-schemes of parallel
algorithms and its expansion oriented to: a) effective description of parallel
schemes dynamically adjustable to the needed level of parallelism as a function of
the values of the parameters to'be computed; and b) multilevel hi~rarchical
representation of graph-schemes that simplify the process of developing thein.
Bibl. of 7 titles, 9 figs.
UDC 681.142.2
MODIFICATION OF LR(k)~PARSER FOR PARALLEL SYNTAX ANALYSIS
[Abstract of article by Babichev, A. V.J
[Text] A class of MLR(k)-grauunars, an expansion of the class of LR(k)--grammars,
is introduced. The strategy of syntax analysis of languages produced by MLR(k)-
grammars is described. Solvability of problems of inembership of unambiguous
KC [context-free] grammar in the class of MLR(k) for a given k is shown.
32
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-00850R400404070024-4
FOR OFFICIAL USE ONLY
Methods o� Programming UDC 6$1.142.2
OPTIMIZATIUN Or CODE FOR ARITHIY~~EETIC EXPRESSIONS
[Abstract of article by Stepanov, A. N.]
[TextJ A method of realizing an algorithm for generating optimized postfix notation
of expressions in a push-down list translator is discussed. The structure of the
- translator and instructions to control lists in generating code are described. This
method is oriented to hardware implementation in computers with a high-level
machine language. Bibl. of 5 titles, 1 f ig.
UDC 681.3.06.44 : 62-52
CHARACTER OUTPUT IN PROGRAM SYSTEMS (MINI-FORMATTERS)
[Abstract of article by Belokopytov, Yu. A.; Kaminskiy, L. G.; Klimenko, S. V.;
Lebedev, A. A. and Polovnikov, S. A., Serpukhov]
[TextJ Authors present a method of organizir~g character oui.put in program systems
based on developing a package of subroutines to form the lines to be output.
Advantages of this method are discussed. Bibl. of $ titles.
Computer Software and Systems Programming
~ UDC 518.5-512.831
MATRIX COMPUTATIONS IN MULTIPROCESSOR COMPUTERS WITH COI~ION STREAM OF INSTRUCTIONS
[Abstract of article by Sukhov, Ye. G.]
[Text] Questions of realization of basic matrix algorithms on multiprocessor com-
puters with a common stream of instructions (SIrID architecture) are discussed.
Complexity of algorithms is evaluated in terms of matrix multiplications of the
dimension (W x W). Requirements for matrix computers are formulated and questions
o� organization of calculations are discussed. Bibl. of 5 titles.
'JDC 681.8.06 : 51
USING FUNCTIONALLY FQUIVALENT MODULES IN DEVELOPING PROGRAMS FOR DIGITAL CONTROL
COMPUTERS
[Abstract of article by Bergson, A. and Raud, R., xallin]
[Text] Authors discuss use of functionally equivalent modules (FEM) in the
- language of process control computers. They discuss problems of obtaining FEM
and organization of joint execution of programs in l~nguages of various computers.
Bibl. of 5 titles.
- UDC 681.3.06
SYSTEM OF ANALYTIC COMPUTATIONS IN ALGOL IN 'DUBNA' MONITOR SYSTEM
[Abstract of article by Dem'yanovich, Yu. K., LeningradJ
[Text] Author describes system of analytic computations in ALGOL-GDR ("SAVAG")
which makes it possible to perform differentiation~ identity transformations (fac-
toring out common multiplier, combining 1_ike terms, removing parentheses and others),
- transformations witli trigonometric expressions and other operations. Bibl. of 9
titles.
33
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-00850R440400070024-4
FOR OFFICIAL USE ONLY
UD~ 681.3.06
DATA TRANSIIISSION FACILITIES FOR REAL TIME SUPERVISOR
[Abstract of article by Naumov, V. V. and Petrakov, S. A.]
[Text] Authors discuss data transmission facilities of the applications program
package "Real-Time Supervisor," that supports automatic exchange of data between
peripherals of an arbitrary type, in particular Unified System peripherals and data
queues. Bibl. of 1 title, 1 fig.
_ Complex Systems and Their Software
UDC 681.31 : 681.326 : 681.142
METHODS OF SEARCHING FOR FREE COMPUTER IN DISTRIBUTED COMPUTING SYSTEMS
[Abstract of article by Gavrilov, A. V. and Zhiratkov, V. I., Novosibirsk]
[Text] Authors discuss prc~blem of searching for a free computer to execute a job
that has come in in distributed computing systems (RVS). To solve this problem in
large.distributed computing systems, they suggest using decentralized methods of
search with incomplete information. Results of computer studies of them are given.
Bibl. of 7 titles, 4 tables, 1 fig.
Communications
- UDC 681.142.2+513.83:519.55:518.5
ENUI~IERATION OF PARTIAL GRAPHS WITH A GIVEN CYCLOMATIC NUMBER
[Abstract of article by Mazepa, Ye. Yu.; Silin, I. N. and Fedyun'kin, Ye. D., Dubna]
[Text] A fast algorithm for searching for all partial graphs ~ith a given cyclo-
matic number is suggested based on analysis of topological properties of a graph.
The algorithm can be used to solve problems of optimal programming and for
opti~nization of networks.
Application of the algorithm for optimization of urban power netowrks has yielded a
tangible, economic eff ect. Bibl. of 7 titles, 7 figs.
- ~ UDC 681.142.2
SPALM SYSTEM OF PROGRANIMING AND SIMULATION LANGUAGES
[Abstract of article by Sedol, Ya. Ya., Riga]
[Text] Author discusses SPALM system of programming and simulation languages that
contains languages of various levels and functions~ united by common principles of
construction. SPALM has been implemented on the Unif ied System of Computers.
Bibl. of 9 titles, 6 figs., 2 tables.
COPYRIGHT: Izdatel'stvo "Nauka", "Programmirovaniye", 1981
8545
CSO: 1863/255
34
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-00850R000440070024-4
FOR OFFICIAL USE ONLY
PROSPECTS FOR SYSTEMS AND THEORETICAL PROGRAMMING
Novosibirsk PERSPEY.TIVY SISTEMNOGO I TEORETICHESKOGO PROGRAMMIROVANIYA in Russian
- 1979 (signed to prESS 17 Dec 79) p 3
[Table of contents from the collection: "Prospects for Systems and Theoretical
Programming", edited by Igor'Vasil'yevi~h Pottosin, Computer Center, Siberian De-
partment, uSSR Academy of Sciences, 800 copies]
Contents Page
Samarskiy, A. A., Applied Proqram Packs as a Means of Supporting Complex
Physical Calculations 5
Karpov, V. Ya. and D. A. Koryagin, Problems of Systems P�rogramming Related
to Development of Applied Program.Packs 17
Katkov, V. L., Machine Graphics and Communication with Computers 32
- Narin'yani, A. S., A New Phase ~f Development, Natural Language 42
Kotov, V. Ye., Problems of Developing Parallel Programming 58
Glushkov,~ V. M., A. A. Stogniy, Yu. V. Kapitonova, A. A. Letichevskiy
and I. V. Vel'bitskiy, Some Results of Research of the Institute o�
C~bernetics, Ukrainian SSR Academy of Sciences, on 8ystems Programming 7~
Kaufman, V. Sh., The Projection Approach to Development of Translators 87
Stolyarov, G. K., Evolution of Data Control Systems 104
Yershov, A. P., Some Subjective Remarks on Timely Problems of Programming 113
Tseytin, G. S., Nonmathematical Thinking in Programming 128
_ Tseytin, G. S,, Systems Programming Language ASSOL 133
Prospects of Development in Systems and Theoretical Programming (Survey
of the Seminar) 140
COPYRtGHT: VYCHISLITEL'NYY TSENTR SO AN SSSR
6521
CS~: 1863/241
35
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY ,
UDC 658.012.011.56:681.3.06
AUTQMATED CONTROI, SYSTEM SOFTWARE RELIABILITY
Moscow NADEZHNOST' PROGRAMNII~i0G0 OBESPECHENIYA ASU in Russian 1981 (signed to press
19 Mar 81) pp 2-6, 241
[Annotation, preface and table of contents from book "Automated Control System Saft-
- ware Reliability", by Vladimir Vasil'yevich Lipayev, Energoizdat, 10,00A copies,
241 pagesJ
[Text] Author discusses problem of ensuring reliability of complex software systems
for control from the posirions of mo dern systems engineering, methods of enhancing
reliability of the components and the software system as a whole, and methods for
checking functi.oning. Factors aff ecting reliability are analyzed.
ror engineers and technicians engaged in design and operation of various types of
automated control systems and for students in WZ's and post-graduates studying
corresponding specialties. ~
Preface
The theoretical and practical level of the modern theory of reliability of hardware
is rather high, and not one complex hardware systEm is designed without a simultan-
eous analysis of its future reliability. To ensure a given hardware reliability, a
broad spectrum of inethods and means are used that allow developing highly reliable
complex systeans from relatively unreliable components. These methods and means use
various forms of redundancy to prevent failures and minimize their effect on the
quality of functioning of the systems. A new type of product--complex software sys-
tems for information control and processing sys~ems--has recently entered the area
of interests of the theory and practice of the study of reliability. During opera-
tion of such systems, malfunctions and failures occur that have be.en caused by dis-
tortions of programs and data. These distortions not only occur in connection w3th
anomalies of operati~n of apparatus, but caa also show up during trouble-free opera-
tion of the computer that implements the given software system. The critical impor-
tance of trie systems that control and process information has dictated the need of
studying the causes of software failures and methods of countering them.
The absence of aging and physical de"struction of software has led to the emergence
of the view ~y a number of experts that methods of the existing theory of reliabili-
ty are wholly inapplicable to the study of the reliability characteristics of soft-
ware. However, analysis of the malfunctions and failures during prolonged operation
of complex software systems has allowed finding analogies to hardware breakdowns and
36 .
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONT.Y
failures. A numtaer of concepts, problem statements and methods of analysis of hard-
ware reliability have been successfully applied to the study of.software reliability.
A new branch of the general theory of reliability has begun to be developed: the
theory of software reliability; its sub~ect of study is the serviceability of com-
plex software systems for control and processing of information.
The major factors in ensuring software reliability, just as for hardware reliability,
are the various forms of redundancy: structural, informational and temporal. Redun-
dancy allows monitoring of software functioning processes and data validity, and pre-
venting failures or reducing their effect on system eff iciency. Resources for using
redundancy in a nrogram have always been limited; this le3ds to the need of rational
distribution of resources between various forms of redundancy and various methods
of raising software reliability.
a The lack in the majority of cases of physical destruction and need of repair of pro-
grams has raised sharply the possibility of automatic restoration of programs after
their fa~lures without human participation. There has arisen the problem of quanti-
tativ e research and development of inethods of expeditiously restoring ~rograms and
data that enable reducing the period of restoration and consequences of the failure
to the level of a short-lived malfunction. It has proved possible to distinguish
the phenomenum of malfunction and failure by time af restoration, the tolerable
values of which in turn are a function of the inertia of the subscribers--the
- sources and consumers of the information.
In chapter 1 of this monographz the main problems of studying an enhancing the
reliability of complex software systems are considered; the factors affecting soft-
ware reliability and the features of applying the main concepts and mathematical
~ apparatus of reliability theory to these products are investigated; and the meth4ds
and criteria of evaluating software reliability and prerequisites for determining
the various indicators of reliability arP analyzed.
In chapter 2, the characteristics of disturbances capable of causing malfunctions
and f ailures during functioning of programs are generalized; garbling of source data
coming from the human operator and garbling caused by malfunctions in data transmis-
sion systems are considered. Special attention is paid to investigating the statis-
tical characteristics of errors in software systems and to the methods of
- describing them.
Chapter 3 generalizes the methods of designing reliable software systems. In it,
features of structural design of individual subprograms, data files and software
systems as a whole are considered; methods of checking the correctness of standard
program struc:tures are formalizAd; and the main methods of determinate testing of
programs in the process of debugging thezn and eliminating errors are presented.
Considerable attention has been paid to determining the state of debugging of pro-
grams and the link between indicators of the state of debugging and software func-
- tioning reliability. Here an analysis has been made of the problem of statistica],
checking of software system reliability in the process of debugging it and tests
aimed at determining the main indicators of reliability.
~
= In chapter 4, methods of using redundancy to check functioning of programs and to
expeditiously restore them after malfunctions are presented. Methods of evaluating
- the quality of algorithmic and program guaxding against malfunctions and failures
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
during program operation are given. Optimization of outlays far introduction of
redundancy and debugging of programs to ensure prescribed reliability of their func-
tioning is considered.
Given in the appendix is the general technology of design of reliable software sys-
tems that considers the limitation on resources to be allocated to ensure reliability.
The importance of the problem of Quant3tative study and enhancement of reliability
of software functioning is confirmed by the materials of all-union (41] and inter-
national [76~ 77, 88] conferences and is reflected in the sharp increase in the num-
ber of journal articles in this field [8]. There are no generalizing books on soft-
ware reliability in the domestic literature and the author hopes that this book will
fill this gap to some extent. One would expect that a number of serious generaliz-
ing studies will be devoted to this problem in the near future to create the bases
_ for the theory of software reliabiiity.
Production of software systems is developing rapidly and accordingly the theoretical
base is developing to analyze and synthesize them. Nevertheless, many concepts are
just forming and there is a lack of precise definition~ of a number of new terms
that have appeared. Even such major terms as mathematical and program facilities
[software] [matematicheskoye i prog~ammnoye obespecheniye] cannot always be separa-
ted. The author uses primarily the texms software [programmnoye obespecheniye] and
software systems [kompleksy programm]. Attempts have been made to refine and com-
plete the definition of a number of terms used in analyzing software reliability~
however there is still much work to be done in the field of software terminology.
Reliability is a major indicator of the quality of software; hawever, it is still ~
characterized by a number of functior.al and design criteria of quality, the selection
of which to a considerable extent is a function of its purpose. Being used more and
more extensively to evaluate the constructive characteristics of programs are the
efficiency of use of storage and throughput of the computer., complexity, correctness,
- structurization, mobility, documentability, cognizability and other criteria. A pre-
- cise measuremen't of the actual values of indicators of quality of the progr~m product
is becoming an ever more important problem as a consequence of the increase in soft-
ware production volume, the increase in its complexity, and as a resu~~ of the
- increase in importance of the functions perfonned by software systems in various
fields of the national economy. It is necessary to develop unified methods of
measuring, monitoring and forecasting of program quality indicators as well as of th~
outlays to achieve them and to provide an ob~ective evaluation of the quality of
each program.
The author wishes to express his gratitude to his colleagues who helped him in the
work on the monograph: T. V. Kolganova, A. A. Shtrik, M. A. Minayev~ B. A. Pozin and
, S. A. Blau, and to E. A. Gladkova for keeping him regularly informed or new
publications.
The author wishes to express special thanks to the revieurer~ B. A. Golovkin, doctor
of engineering science, who made a number of valuable comments that led to substan-
tial improvement of the book, and to V. V. Kul~ba, candidate of engineering ~cience,
for thorough scholarly editing of the manuscript.
The Author
38
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
EOR OFFICIAL USE ONLY
Table of Contents Page
Preface 3
Chapter 1. Principles ~f Analysis of Complex Software System Reliabilit~ 7
1-1. Problems of Raising Reliability of Complex Software System Opzration 7
1-2. Crit~ria for Evaluating and Factors Determining Reliability of
Functioning of Complex Software Systems 32
Chapter 2. Disturbances Affecting Reliability of Software System Operation 59
2-1. Garbling of Source Data 59
- 2-2. E~rors in Program Systems 75
Chapter 3. Methods of Designing Reliable Software Systems 93
3-1. Structured Design 9~
3-2. Determinate Testing of Programs in Design 124
3-3. Statistical Checking of Software System Reliability 161
Cnapter 4. Raising Software System Reliability Th~ougti Redundancy 182
- 4-1. Methods of Using redundancy to Raise the Reliability of Software
System Functioning 182
4-2. Analysis of Process of Expeditious Noise Protection of Software
Systems 211
Appendix. Technology of Design of Reliable Software Systems 230
Bibliography 236
COPYRIGHT: Energoizdat, 1981
8545
CSO: 1863/268
~
39
Ft~R OFFICIAL USE ~ON~.Y
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/42/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
OPTICAI, PROCESSING
- UDC 621.393
HYBRID OPTICAL-DIGITAL PULSAR SIGNAL PROCESSING SYSTEM
~ Leningrad RADIOGOLOGRAFIYA I OPTICHESKAYA OBRABOTKA INFORMATSII.V MII~ROVOLNOVOY
- TEKHNIKE in Russian 1980 (signed to press 24 Oct 80) pp 135-140
[Article by N. A. Yesepkina, N. A. Bukharin, Yu. A. Kotov, B. A. Kotov and A. V~
Mikhaylov from the collection "Radio Holography and Optical Information Processing
in Microwave Equipment", edited by corresponding member of USSR Academy of Sciences
L. D. Bakhrakh and candidate of technical sciences .A. P. Kurochkin, Scientific
Council on Problems of Holography, Department of General Physics and Astronomy, USSR
Academy of Sciences, Izdatel'stvo "Nauka", Leningradskoye o~tdeleniye, 2,150 copies,
184 pag~s. Additional sections of this book appeared in the USSR REPORT: CYBER-
NETICS, COMPUTERS AND AUTOMATION TECHI~IOLOGY dated 28 August 1981 (JPRS L/9945) pp
1- 39 ]
[Text]
1. Much attention of radioastronomers has recently been attracted by the investiga-
tion of pulsars, which are poir~t sources emitting wideband signals in the form of
periodic short pulses. Because of scatt2ring of the interstellar medium, the pulse
. length on earth considerably exceeds thsir length near the radiation source and may
comprise tens and even hundreds of milliseconds. The effect of scattering also
leads to the reduction of the amplitude of the signals being received.
Several methods of elimi.nating the effect of scattering in observing pulsars are
- known in which multichannel ele~cronic systems or digital correlators are used [1].
~ These methods are similar to those of complex signal processing with large multipli-
cation of the band by the pulse.length (TOf � 1).
We feel that acoustooptical devices [2], specifically devices that permit correla-
tion processing of wideband signals of long length, are promising to eliminate the
effect of scattering when observing pulsars. These are acoustooptical correlators
with time integration [3] in which special photodetectors (f.or example, devices with
charge coupling) that integrate or store the signal are used. Unlike ordinary
_ acoustooptical correlatoxs [3], in which the correlation function is shaped sequen-
t:ially in time, the correlation function in the devices under consideration is cal-
culated simultaneously for different time shifts of the initial functions by time
integration, i.e., the �ollowing function is formed at the output of this correlator
!~0
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-00850R000404070024-4
FOR OFFICIAL USE ONLY
Z,
~ ~t~~ _ ~ ~ u - t~~ dt
0
where all values of t' lying in the interval from -D/2S to D/2S, where D and S are
the aperture of the acoustic drive of the acoustooptical modulator and the velocity
of the elastic wave in it, T is integration time and u(t) is the received signal.
The possible length of the signals being processed T in these systems, when using
photodetectors based on charge coupling devices (PZS) [4], c~:nprises f.rom units to
hundreds of milliseconds and it may be increased significantly by using an addition-
al computer buffer storage.
The maximum storage time used ia this correlator and consequently the maximum im-
provement of the signal/noise ratio received at the system output, is limited by the
dynamic range of the photodetector.s being used R(in our case the dynamic range of
the PZS [3]) and comprises T~f < R2.
The required time for storing the acoustooptical. modulator is determined by the max-
imum possible pulse length near the radiation source.
The bandwidth of the devices being considered is detern:ined by the acoustooptical
modulator band and can. reach hundreds of inegahertz [5, 6]. .
2. One of the possible schemes of an acoustooptical correlator with time integra-
- tion, which can be used to process pulsar signals, is presented in Figure 1. As can
be seen from the figure, the signal u(t) from the antenna A, which receives the
pulsar radiation, is fed through a power divider DM to two independent receivers
P1 and P2 and after amplifica~ian and conversion to an intemediate frequency , is
fed to the two channels of the correlator that calculates the autocorrelation func-
tion of the input signal. ~ao receivers are required so that the natural noise of
the device N1(t) and N2(t) are not correlated with each other in different channels.
As can be seen from Figure 1, the signal u(t) + N1(t) from P1 modulates the OKG
[laser] emission by means of intensity modulator M. The light is then split by
means of a collimeter consisting of lenses L1 and L2 and impinges on the acousto-
optical modulator AoM, onto the piezotransducer of which the signal u(t) + N2(t)
from P2 impinges. A travelling ultrasonic wave is propagated in this modulator,
which ensures a time shift of the signals required to calculate the correlation
function. Lenses L3 and Lq shape the image of the AOM in the output plane of de-
vice R3, where the linear components of the PZS are located. The undiffracted part
- of the light is filtered in plane R2. The signal at the output of each of the
- linear aomponents of the PZS is proportional to the time integral due to the liqht
intensity impinging on it.
One can show [4] that the following signal is shaped at the output of the considered
device
T .
`~(s>=Alo T-}-o~M f u(t)u(t-.s )dt -k- a (i) ' ' ~1)
L~l
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400074024-4
FOR OFFICIAL USE ONLY
A
. i----~-----~
n . ~M, . i A~ny ~
3BM ~
� ' 0c
. n9 ` n2 ( � ~ - i
. u~t)'' N~ u,~t~+N2 y ~
- - . ~
..i ~
oKr
~ i(~ M n' n3c I
' I ~
I - ~Z P~,aoM~3 . P n4 P3 I
~ -------------J .
Figure 1. Diagram of Acoustooptical Correlator With Time Integration
where Ip is the laser emission intensity, a and A are conatant values, ~ is the
modulation index of light by the acoustooptical modulator, x is the coordinate ~n
plane R1 and s is the speed of light in the sound conductor of the modulator.
The second term in this expression is the desired autocorrelation function and the
first is a constant component. The term a(x/s) contains cross-correlation functions
and reduces the signal/noise ratio at the device output [7]. ,
As can be seen from (1), the autocorrelation function at the output is formed as a
function of coordinate x, i.e., the signal distribution in the linear series compon-
ents of the PZS corresponds to the desired autocorrelation function. These signals
are transferred by the control pulse coming from the computer to the storag~e sec-
- tions (the linear shift registers of the PZS) and are then fed sequentially by a
special interface [8] to the Elektronika-100 computer.
After the necessary processing, the signals are fed to an oscillograph (OSTs), to
an automatic digital printer (ATsPU) and so on. Thus,.the considered device is a
hybrid optical-acoustical processing system in which correlation analysis is accom-
- plished in the optical system. The remaining processing is accomplished in a com-
puter that permits the operating modes of the carrelator to be changed by program,
for example, to print out the results of observing individual siqnals, to operate
in the pulse starage mode.and so on.
3. We developed and investigated a mockup of an acoustooptical correlator which ~
can be used to process pulsar signals. An acoustooptical moculator was also used
as the modulator M. Both modulators o,perated on a frequency of 30 NIIiz and had a
band of 3 MHz.* .
*
Use of modulators [5, 6, 9] operating at higher frequencies permits an increase of
the working band of the device.
42
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
The Elektronika-100 computer. was used in our experiments. The investigation was
conducted in two modes. In the first storage was carried out only in the PZS of
the components and in this case storage comprised T= 10 ms; in the second mode an
addi~ional computer memory was introduced where the results of 10 separate millisec-
ond storage cycles were added.
Figure 2. Signal at One of the Correlator Inputs (a) and Signal at PZS
_ Output (b) at T= 10 ms
Noise signals from two noise generators were fed to the outputs of the correlator,
in addition to identical pulse signals (~F = 260 kHZ) that simulate pulsar signals.
These signals simulated the input noise of the receivers.
A signal 10 ms long at one of the carrelator inputs is shown in Figure 2, a. The
result of processing it, achieved at the linear output of the PZS, is illustrated
in Figure 2, b. Each point on the photograph corresponds to a separate linear com-
ponent of the PZS, while the vertical deflection is proportional to the value of
the signal on it. The distance between adjacent linear components comprised 0.32
microsecond in the time scale in our case. As can be seen from Figure 2, b the
half-width of the autocorrelation function is equal to T~ 3.9 microseconds at cal-
culated valuz of i= 1/~F ~ 3.8 microseconds. Thus, the derived contraction coef-
ficient comprised K= T/T = 2,500.
Similar experiments were conducted with signals T= 1.28 second long, when not only
the linear components of the PZS were used for storage but the memory of the Elek-
tronika-100 computer as well. In this case a contraction coefficient K= 3�105 was
obtained. It is obvio~s from the given results that the signal is confidently sep-
arated from the noise at the correlator output.
1~3
FOR OFFICIAL USE OIVLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
Thus, one can hope that th~ use of acoustooptical devices will make it possible to
simplify the pulsar signal processing system and to improve the parameters of re-
ceiving systems.
The correlation devices considered here can also be used to process interferomet2r
signals, specifically multicomponent radiotelescopes with aperture synthesis [10]
and in this case devices based on spatial-multichannel modulators [9, 11] are of
special interest. These devices are very promising not only to process signals in
_ radioastronomical but in other different radioengineering systems where correlation
processing of wideband signals of great length must be carried out.
BIBLIOGRAPHY
1. Manchester, R. N., "Pulsars," in: "Galakticheskaya i vnegalakticheskaya radio-
astronomiya" [Galactic and Extragalactic Radioastronomy), Moscow, Mir, 1976.
- 2. Cole, T. W., "An Acoustooptical Radio Spectrograph", ASTROPHYSICAL LETTERS,
Vol 15, 1973.
3. Sprague, R. A. and C. L. IG~liogolous, "Time Integration Acoustooptic Correla-
- tdr," APPLIED OPTICS, Vol 15, No 1, 1976.
4. Seken, K. and M. Tflmpset, "Pribory s perenosom zaryada" [Devices With Charge
Transferl, Moscow, Mir, 1978.
5. Chang, I. C., "Acoustooptic Devices and Application," IEEE TRANSACTIQNS ON
SONICS AND ULTRASONICS, SU-23, No 1, 1976.
6. Hecht, D. L., "Multifrequency of Acoustooptic Diffraction," IEEE TRANSACTIONS
ON SONICS AND ULTRASONICS, SV-24, No 1, 1977.
7. Bromley, K. A., "Multichannel Optical Correlation System," OPTICA ACTA, Vol 21,
1974.
8. Yesepkina, A., B. A. Kotov, Yu. A. ICptov et al, "Hybrid Oiatico-digital Systems
for Spectral Analysis," AVTOMETRIYA, No 3, 1978.
9. Chikada, Y., T. Miyaji, N. Kaifu et al, "A High Resolution Acoustooptical Radio
Spectrometer for Millimeter Wave Astronomy," BULLETIN OF THE ASTRONOMICAL SO-
CIETY O~ JAPAIv, Vol 29~ 1977.
10. Fomalont, E. V, and M. K. I~. Raynt, "Interferometry and Aperture Synthesis,"
in: "Galakticheskaya i vnegalakticheskaya radioastronomiya", Moscow, Mir, 1976.
_ 1976.
11. Yesepkina, N. A., V. Yu. Petrun'kin, Ye. T. Aksenov et al, "Multichannel Acous-
tooptical Modulators," ZHTF, Vol 45, 1975.
COPYRIGHT: Izdatel'stvo "Nauka", 1980
6521
CSO: 1863/270
44
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400074024-4
FOR OEFIC[AL YJSE ONLY
. UDC 621.393
MEASURING COORDINATES OF TERRAIN REFERENCE POINTS AND DETERMINING SHIFTS OF
CLOUD FORMATZONS BY MEANS OF HETERODYNE OPTICAL CORRELATOR
Leningrad RADIOGOLOGRAFIYA I OPTTCHESKAYA OBRABOTKA INFORMATSII V MIKROVOLNOVOY
TEKHNIKE in Russian 1980 (signed to press 24 Oct 80) pp 140-155
[Article by A. I. Balabanuv, G. Ye. Korbukov, A. A. Feoktistov and Ye. R. Tsvetov
from the collection "Radio Holography and Optical Information Processing in Micro-
wave Equipment", edi.ted by corresponding member of USSR Academy of Sciences L. D.
Bakhrakh and candidate of technical sciences A. P. Kurochkin, Scientific Council on
Problems of Holography, Departinent of General Physics and Astronomy, USSR Academy of
_ Sciences, Izdatel'stvo "Nauka", Leningradskoye otdeleniye, 2,150 copi.es, 184 pages.
Additional sections of this book appeared in the USSR REPORT: CYBERNETICS, COMPU-
TERS AND AUTOMATION TECHNOLOGY dated 28 August 1981 (JPRS L/9945) pp 1-39]
[Text] Introduction
Correlation processing can be used for cosmic video information in at least two as-
pects: for coordinate tying of images of the earth's surface and other planets ob-
tained at different moments of time and from different jensors and for extracting
meteorologxcal data.
When investigating the chronological changes in cosmic photoimages and when the im-
ages obtained in different spectral bands are compared, it becomes necessary to join
them precisely. Accurate correction of the geometric disto~tions of the images be-
ing joined should be carried out to ensure the possibility of this joining. Data
for this correction can be obtained by correlation detection of the fragments of the
photoimages to be correc~ed and the measurements of the coordinates of terrain ref-
erence points [1]. It is obvious that the presence of the indicated geometric dis-
tortions will have a negative effect on the reliability of recognitiont therefore,
preliminary correction of these distortions is necessary, which is accomplished when
the image is recorded.
_ To gather data on the direction and velocity of wind in different sections of the
globe, space photographs of the cloud cover transmitted from g~osynchronous satel-
lites at intervals of 30 minutes are used more extensively. The required data are
obtairied by recognition and measurement of the motions of individual cloud forma-
tions. The negative factor here that affects the reliability of identification when
using the correlation method may be deformation of thP cloud formations. Extensive
experimental investigations on a sufficient number of imxges must be conducted to
1~5
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFiCIAL USE ONLY
determine the boundaries and conditions of applicability of the correlation method
in this problem, the same as in the previous problem.
The given paper is the initial attempt to conduct these investigations by using an
optical heterodyne corr~lator (OGK) circuit [2]. Its main advantage compared to
the known holographic matched filtration circuit [3] includes the absence of the
need to manufacture holographic filters for reference images along with retaining
the very valuable capabilit~ of spatial filtration of the frequency features which
coherent optics gives. Moreover, the OGK scheme has a potentially higher instrument
accuracy due to the photoelectric correlation image readout system and due to the
_ absence of such error sources as aberration, inaccurate focussing, spatial noise and
nonlinear distortions of the holographic filter and the objective of the second
Fourier transform.
We recall that there are two spatially dispersed input apertures in the OGK scheme:
one for the reference slide h(plane P1) and the other for the signal slide g(plane
P2) (Figure 1). The image of the signal slide 3+1 is superimposed on the image of
_ the reference slide h~ in the plane of diaphragm E(plane P6) by with a hologram
lens GO in the first Fourier plane Pq created by lc:ns L1 (plane P3) upon illumina-
tion of the input slides h and g by a parallel coherent light beam. Images hp and
g+l Were multiplied out and,spatial integration was accomplished by a quadratic pho-
todetector of the FBJ [Photomultiplier]. Since the moving image of the signal slide
g+l has a doppler shift of light frequency proportional to the spatial carrier fre-
quency and the rate of motion of the hologram lens in the Fourier plane, electrical.
oscillations with doppler division frequency whose envelope describes the cross-
correlation function of the input images as the hologram lens moves, occur at the
FEU output. This envelope was observed on the oscillograph screen after additional
electric filtration and detection. The images were processed with suppressed opti-
cal constant of the component for which ~he point shutter 3 was introduced to the
center of the Fourier plane P4 directly in front of the hologram lens. Thus, the
outline images were actually subjected to correlation processing. Moreover, addi-
- tional spatial filters OF could be introduced in the second frequency plane Pg form-
- ed by lens L2 (plane P6) in front of the photocathode of the FEU to optimize the
_ signal/noise ratio at the correlator output. Accordingly, the positions and dimen-
sions of the reference fragment were established by regulation the position and
dimensions of rectangular diaphragm E(which was two crossed micrometric slits).
The fragment of the reference slide established by means of diaphragm E usually
had two squares with side of 1.8 mm in all the measurements.
Investig~tions were conducted on a mockup of the OGK in which a hologram modulator
in the form of a drum 200 mm in diameter with 17 hologram lenses havinq focal dis-
tance of 300 mm fixed on 35-mm Mikrat-900 photographic film, was used. The dnm?
was rotated uniformly at a speed of 1 rpm by an electric motor with friction reduc-
er. Thus, the scanning period of the signal image with respect to the reference
fragment comprised 59.ms through the X axis. An electronic oscillograph of type
51-55, the scanning of which was synchronized by passing the hologram lenses through
the optical axis of the correlator, was used for the measurements. An example of
correlation recognition c~f an image fraqment of the earth's surface is shown in
Figure 2. The illuminated square in the photograph o� Figure 2, a denotes a frag-
ment selected as a reference fragment while the entire photograph denotes the signal
slide. Scanning is along the horizontal.
46
FOR OFFIC[AL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
; P~ Pi P3 . Py Ps p6 p~ pe ,
ZZ J
. I ~
Z~ Z34=f Z z56 Z6A
~n n2 3~; 9-~
h~x'bh,yl 3 ~ h-t ~ i 0~
- ~ I go
~ ~
, . � . . . ~3Y
\
� A ~
~ gl='~g,y) ~ h; ~~y*r
�D~ t~:) h�,Ii I
Oca n
a~ou 1 ,
vaCmomei
Figure 1. Diagram of Optical Heterodyne Correlator
Key:
1. Axis of zero frequency ~
.
;
r^igure 2. Example of Correlation Detection of Image Fragcnent with Coanplex
Structure in OGK: a--input slide (signal); illuminated fragment is
identical to that used as refer~nces b--pattern of Fourier plane of
~ OGK: the flag covers the uninformative bxight central part of the
spectrum; c--signal at OGK output: the signal peak corresponds to
the moment of total coincidence of the fragment noted in (a) with
the reference image
47
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407/02/09: CIA-RDP82-00850R000400470024-4
FOR OFFICIAL USE UNLY
An oscillogram of the correlation signal is shown in Figure 2, c. The peak corre-
sponds to detection of the fragment in the signal image identical to the reference
fragment. The oscillogram of Figure 2 b is a central cross-section of a two-dimen-
sional cross-correlation function of the reference fragment and the signal image
produced by search motion of the signal slide along the Y axis.
l. Investigation Correlation Identification of Fragments of Photographs of the
Earth's Surface ~
Space photographs of sections of the earth's surface (region of the Volga upland)
obtained from Meteor satellites on 1 and 15 September 1976 in the spectral range
_ of 0.6-0.7 micron (Figure 3) were used for experiments in correlation identifica-
tion using the OGK. The indicated images were processed on a computer, at the in-
put of which preliminary correlation of the geometric images was carried out.
. r . sy~
- ~ ~y V''i~?'~r~.,
n rt: "
.
.
~
~
n
Figure 3. Fragments of Space Photographs of Section of Earth's Surface
The image component had the shape of a rectangle with dimensions elX = 50 microns
and ~lY = 200 microns. To i.ncrease contrast, the photographs were subjected to con-
tact printing on contrast photoqraphic paper with subsequent rephotographing on
Mikrat-300 film. Rephotographing occurred with reductions of l:l, 1:2, 1:4 and 1:8.
The dimensions of the image components corresponding to the scales are presented in
_ Table 1.
- Since ~lX is one-fourth the size of dly, the requirements on the accuracy of deter-
mining the motion in the direction of the X axis are considerably higher. For this
reason the accuracy of ineasuring the coordinates of the fragments was determined
along the X axis in conducting the given investigation.
~'he accuracy of determining the coordinates was estimated in the following manner.
The signal slide was moved along the X axis and at the same time the position of
1~8
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE QNLY
Table 1. Dimensions of Image Components at Different Scales of Rephotographing
Dimensions of Imaqe Components, Microns
_ Scale of Rephotography ~lY ~
l:l 50 200
1:2 25 100
1:4 12.5 50
1:8 6.25 25
the maximum correlation peak was no~ed on the oscilloqraph screen. The measurements
- were made with displacements of the reference slide from 0 to 250 microns with spac-
ing of 10 microns (the rate of oscillograph beam scanning was 0.5 ms/cm) and from 0
to 500 microns with spaeing of 10 microns (the scanning rate was 1 ms/cm). The
measurements were repeated three times on the average in each case.
Tw~o situations were investigated: 1-) th~ reference and signal slides were a fragment
of the same image and 2) the reference and signal slides were fray~nents of images
received from s~tellites on different days. To denote the first situation let us
use the tenn "autocorrelation" and to denote the second situation let us use the term
"cross-correlation."
d X3
40 '
30 �
20 ,
~n ~
,
~ SO tU0 150 ~OQ 230
dX~ ntn,w
Figure 4. Dependence of Displacement of Maximum Correlation Peak ~}Ce on
_ Displacement of Signal Slide aX
The dependence of the displacement of the maximian correlation peak ~Xe (in units of
the small divisions of the oscillograph screen) on the displacement of the reference
slide ~X is presented as an example for the case of autocorrelation in Figure 4. The
image scale is 1:1. Similar functions were found for cases of auto- and croas-corre-
lation for all four scales.
The results were processed on a computer and the measurements were approximated by a
straight line [4J with respect to which the values of the mean square error v and
ratio ~lX/Q--a dimensionless value that characterizes measurement accuracy (see Table
2), on the basis of the well-known least squares ariterian.
FOR OFFIC[AL9USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FAR OFFICIAL USE ONLY
b ao~rr
o k r' ~r .-1 ~
0 ro ~ ch ~ ~ O
w
O ~ p~
O S-i o0 to ~ rn
~ V V . . . .
~ O ~ it1 O r~
~ ~ ~ ~
~ \ ~ b
O m
U ~
W r-I �
N
~ > 11
,~.u y ~ k m ~�om~
~ r~ ~ a ri ri o 0
i ~ ~
o s~
FC ~ 1~ ~
V m U tn 61 ~N ~
w U] ~ ...1 .
O O ~ ~-I ~ O
G) ~p~ U ~-I ~-1 r-I rl
~ ~ b
U tT
O
q r-I
~
q U
~ O ~ \ I u~i rMi v~i
a O a ~ ~O N ri
0 VI ,-%I
~tq ~ ~ .
PG rt1 p ~
b~ � ~
O U NN~n
~ i tOMO
b U ~ r1 ttl d'
- tA b N b
0 v1 ~
~ ~ ~
W V II
> O \ I ~NO~o ~
~ ~ b a ~ MNO '
~ ~ ~
w p p
~ w
~ A i~ ~
71 U '~y'
pN 1 CO u1 ~
N y.~ ~ l~ lf1 1~
d) V b
~,p
td
H
N .
~-I r-i N d' ~0
ro
U '-1 ~-i rl rl
~
~
. 50
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OF'FICIAL USE ONLY
The individual values of the mean square errora vary ~,ver a rather wide range.
- Nevertheless the following conclusior.s aan be made. First, no appreciable depen-
dence on the image scale can be detected at all. Second, the mean value of the mean
square error Q for six measurements made at scanning rate of 0.5 ms/cm (5.57 mi-
crons) is approximately one~half the correspondirg value for eight measurements at
scanning rate of 1 ms/cm (11.8 microns). Third, the mean values of the mean square
errors in the case of cross-correlation, although few, are higher than in the case
_ of autocorrelation (these values comprised 6.75 and 4.40 microns at scanning rate
- of 0.5 msjcm and 12.45 and 11.1 microns at scanning rate of 1 ms/cm).
Detailed analysis of the experimental data showed tkiat the accuracy of determining
the displacement of the correlation maximum is limited by the operating in~tabili-
_ ties of the corre lator (specifically, the instabilities of the rotational speed of
the hologram lens, synchronization of hologram lens rotation and scanning of the
oscillograph beam) and by radio noise. It is obvious upon transition from scanning
rate of 0.5 ms/cm to 1 ms/cm that errors related to rounding off the value of the
coordinate of the correlation maximum with accuracy to the small division of the
cathode-ray tube screen begin (one division corresponds to 5 microns on the input
slide at scanning rate of 0.5 ms/cm and to 10 microns at scanning rate of 1 ms/cm).
Due to the fact that scanning was accomplished within the dimensions of the refer-
ence fragment (a square with sides of 1.8 mm), the effect of aberrations of the op-
tical system was probably insignificant.
As can be seen from Table 2, the value of the mean s~uare error is hardly dependent
on the image scale. Consequently, the relative accuracy of ineasuring the coordin-
ates increases with an increase of image scale. For example, the value ~1/T in-
creased fourfold upon transition from a scale of 1:8 to one of 1:2 for the case of
cross-correlation at scanning rate of 0.5 ms/cm.
1~
20 1
~6 \
,z r~ 2
,
~
B i
y 4 - `-\-=S
1~ 1 1~2 1~ 4 f~8
!docu~ma6 (1)
Figure 5. Dependence of Signal/Noise Ftatio on Image Scale
- Key:
- 1. Scale
_ It follows from.consideration of Table 2 that the coordinates of fragments ca,z be
measured with an optical heterodyne correlator with an error considerably less than
the dimensions of the input image resolving component. Of course, this result is
achievable if the natural resolution of the correlator exceeds that in the image.
51
_ FOR OFFIC[AL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFiCIAL USE ONLY
We note that similar results were found earlier with respect to a holographic corre-
lator [3]. An approach was developed in [5] that permits one to give a natural ex-
planation to the indicated results within the framework of already developed theory
of signal separation on a background of additive Gauss noise.
The values of the signal/noise ratio for the cases of auto- and cross-corre].ation
at four values of the scale (the analyzed fragments were selected on the same sec-
tion of the photngraphs) are presented in Figure 5. Broken lines 1 and 2 join
points that characterize the ratio Is~Ish.sr for cases of auto- and cross-correla-
tion, respectively. IS is the value of the correlation maximum and Ish.sr is the
- mean value of noise within the oscillograms. Broken lines 3 and 4 connect points
that characterize the ratio IS/Ish.max for cases of auto- and cross-correlation and
Ish.max is the value of the maximum.noise ejection within the oscillogram. Al1 four
broken lines have minimum values at M 1:1, reach a maximum at M 1:4 and again de-
crease at M 1:8. As follows from the results, th~ optimum value with respect to the
- signal/noise ratio should be regarded as a scale of 1:4. An ihcrease of the scale
results in a decrease of this ratio due.to constriction of the band of spatial sig-
_ nal frequencies effectively used by the correlator. A decrease of the scale to 1:8
also causes a decrease of the signal/noise ratio, this time because of losses of
~ image information content that arise upon r~phctography on this scale.
~ l~/1~
10 ~ ~
5 ' Z .
. '
' , ' ~
� 3
0 f0 20 30 S~ MM Z
Figure 6. Dependence of Signal/Noise Ratio on Area of Fragment of Reference
Slide S
- The dependence of ratios TS/Ish.sr ~1~ and Is~Ish.max ~2) on ~he area of the slide
fragment for the case of cross-correlation, M 1:4, are presented in Figure 6.
Curves 1 and 2 reach a maximum with area of the fragm.ent of the reference slide on
the order of 8 mm2. The value of the signal/noise ratio then begins to drop, which
may be explained by the fact tha~ the effect of aberration and phase errors between
the signal and reference images at the output of the optical system of the corre-
lator, introduced by the Fourier transform lens, the collimator, the holographic
lens and also by phase inhomogeneities of the slide substrates, increases as the
aperture increases. The dependence of the ratio Ish~m~/Ish.maX on the area of the
_ fragment of reference slide (3), where Ish.max is the value of the noise discharge
on the oscillogram that follows as a maximum value, is shown in this same figure.
This ratio hardly differs ~from 1. Therefore, a simple algorithm, according to
which the values of two maximum blips on the oscillogram must be measured and the
52
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
- FOR OFF[CIAL IJSE ONLY
ratio must be det.e:rmined. If both blips are of noise origin, thQ ratio is on the
- ~zder of 1. If tne ratio significantly exceeds 1(on the order of 5-7 in our case),
then the hiyhest ratio can be related to an identified fragment.
It should be e~r;phasized that the cnrrelation function module rather than its square
ia measur~d with the optical heter.odyne correlator, as in the case of a holographic
- correlator. Therefore, the values of the signal/noise ratio equal to 5-7 are equiv-
ale~t ta va~ues of signal/noise ratio equal to 25-49 in the case of a holographic
cerrelator.
The effect of installing an optimizing amplitude space filter (OF) in the second
�requFncy p3ane was studied during thE investigations. 1~ finished filter previous-
ly obtained far several different images was used as the OF; consequently, this
~ filter was not optimtun in the tatal s~nse. Nevertt~.eless it turned out that the
signal/noise .ratio improved by a factor af 1.5-2 when the given filter was in-
sralled, whereas the ratio Ish.max/rsh.max decreased on the average from 1.31 to
1.16. The results ~onfirm the usefulness of using qF spe~ially manufactured for
the given type of images.
Let us dwell on the results oF inv~stigating the dependence of the correlation max-
imum on the angle of rotation of the reference slide with respect to optimum orien-
tation. This problem is primaril~ af interest from the viewpoint of requirements
on the accuracy of installing the fragments to be analyzed by anc~ls.
jc ~~c.max
.
~ 0. 8
0. 6
v. v
~ a.2
3 0 3
Figure 7. Dependence of Correlation Signal (in relative units) on Angle of
Rotation ~ r,f Reference Slide
The angular dependence of the correlation maximtun (in relative units) is shown in
Figure 7 for the case of autocorrelation and image scale of 1:2. The curve reaches
a maximum at angle equal to zero and drops one-half at angles of rotation ~~.5 of
approximately + 2�. It must be noted that the separate curves found in different
scales are characterized by rather large variety of values of angles ~p.5: from
1.5 to 4�. The main reason for this variability consists in the fact that, on the
one hand, individual image fragments differ significantly from the viewpoint of
_ informative parts in them and on the other hand it vras difficult to join the center
of rotation to the center of the fragment being analyzed and as ~ result the frag-
. ment not only rotate~ but is shifted along the Y axis, which leads to a more rapid
53
FOR OFF'ICIAL USE UNLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/42/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
_
t
~
~
- Figure 8. Photograph Taken From Satellite on 22 July 1975 (negative)
.
- ~ ~1~
'~r.: ~
. +
,Fl j '
~
~
f.
.1 . ,
I~~ I
. ~ ~I
Figure 9. Enlarged Pho*~ograph of Four Sections (negative)
51~
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-00850R400404070024-4
FOR OFFICIAL USE ONLY
decrease of the correlation m�zximum compared to the case when this shift is absent. ~
On the whole the values of angles $p.5 correspond to those which should be expected.
In the general case the value of ~p.5 should decrease as the area of the fragment
increases, i.e., the critical orientation of the fragment wi11 increase.
2. Investigating Correlation Identification of Picture Fragments of Cloud
Formations
Three photographs transmi~ted from a geosynchronous satellite on 22 July 1975: one
at 2115, two at 2145 and three at 2245 Gre~nwich time, were used in the experimental
- investigations. The coordinates of the point under the satellite were 0� latitude
and 115� west longitude. The image component had the shape of a square with side
of 50 microns, which correspondsd to a square with side of 3.6 lan on the earth's
surface. The investigations were conducted both in the autocorrelation mode when
sections of image 2 were used as the signal and reference slides and in the cross-
correlation mode when sections of images 1 and 2(at time interval of 30 minu~es)
were used. Moreover, single experiments of images 2 and 3(at interval of 60 min-
utes) were conducted in the cross-correlation mode.
A photograph which is one-fourth of the total image 2 obtained from a geosynchronous
satellite is presented as an example in Figure 8. Four sections (denoted by the
numbers in the figure) were selected in the photographs being analyzed. 2'he frag-
ments were subjected to correlation processing when conducting the given investiga-
tion. Magnified images of the indicated sections with the following types of over-
cast are presented in Figure 9: cirrus 1, cumulus and stratocumulus 2, cumulus 3
and cumulus with closed cells .
As wh~n processing the images of the earth's surface, the sections to be anal~zed
were rephotographed on Mikrat-300 film, as when processing images of the earth's
surface. To select ~he most optimum information recording density, the rephota-
graphing was carried out both with variation of scale ar~d with reduction of it by
a factor of 2, 4, 6 and 8. The functions for the signal to maximtun noise blip ratio
Is~Ish.max for auto- and cross-correlation at five values of scale are presented in
Fic~ure 10. Broken lines 1 and 2.connect the poir~ts that characterize the signal/
noise ratio for the case of auto-correlation of image fraqments belonging to sec-
tions 2 and 1, respectively. Broken lines 3 and 4 correspond to cross-correlation
of fragments of the same image sections (the fragment of the same name had the
shape ~of a square with side of 2 mm). As follows from the results, the highest
values of ~he signal/noise ratio are reached at a scale of 1:4; therefore, subse-
quent i"vestigation was conducted with images of this very scale.
The accuracy of determining the displacements of the picture fragments belonging to
the four select~ed sections was then estimated. The values of the mean square er-
rors v.-uid ratios ~1/cr are presented in Table 3. As can be seen from the table,
the displacements of the cloud formations can also be measured in the case of im-
ages of cloud cover with error much less ~han the resolution. The maximum m~asure-
ment accuracy is determined by the design of tihe correlator and optical and elec-
tronic noise.
The accuracy of ineasurements in the cross-correlation mode was then estimated for
image fragments obtained at time interval of 60 minutes. On the whole the si~nal/
noise ratio decreases one-half in this case while the value of the mean square
55
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-40850R040400074024-4
FY~li ~)F F Ic 1~?1. I~~F' t1Nr
,
8 , .
_ ~ ' / ~
6 ~ / ~
~ 1 , / /~i~-�.~.3 ~
~ 4 / 4 .
_ ~ . . y`''
a .
_ r~~ ~~2 r�~ ~~s ~~6 ~
Macu~ma6 (1)
Figure 10. Dependen~e of Ratio Is/Igh.~ on Image Scale
~
Key:
1. scale
Table 3.
. Cro~s-correlation, Cross-correlation,
Number Autocorrelation interval of 30 min interval of 60 min
- of
Section a, mi.cron ~ ~1/Q a, micron ~1 v Q, micron ~1/Q
1 4.2 2.98 7.3 1.71 3.7 3.4
2 7.2 1.74 4.4 2.84
3 5.8 2.16 7.7 1.62
4 3.8 3.29 4.6 2.72 5.3 2.36
_ error a and ratio ~1/Q is approximately the same as in the case of cross-correla-
tion of i.mage fragments obtained at interval of 30 minutes.
The dependence of the signal/noise ratio on the area of the identified fragment S ~
is presented in Figure 11 for all four sections (image scale of 1:4). Curves 1 on
each of three graphs a-d characterize the ratio I8/ISh.~~ for the case of autocor-
relation and curves 2 characterize the same ratio for cross-cor~elation. Curves 3
characterize the ratio of maximum noise blip to the next noise blip Igh.max/Ish.max�
As one would expect, the ratio IS/Ish.max is higher in the case of autocorrelation
(curves 1) than for cross-correlation (curves 2). It is typical for all graphs .
that the maximum values of curves 1 and 2 occur on areas of identified fragments
less than 10 mm2 wh~.ch generally corresponds to the resulta ob~$ined in experiments
on identification of f ragments of the underlying.aurface.
The values of the areas of identified fragments at which the curves are maximum de-
pend on the information properties of specific fragments and v~ry randomly. These
curves can sometimes have two maximums, which is explained by inclusion of new
cloud fQrmations in the boundaries of a fragment as its area increases.
56
~ FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/42/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
~a
- 12 . ~
.
8 \ -
x k~~x
4 2
x
_~x
~-..z~N � � x 3 x~
D
- ~ b
12 ~
8 �
~ 4 x~`x~ , 2
x..~~~
~ x' 3
! x
0
~v
~2 C
B .
4 ~
x
IF~ ~.-~w X~~F
~
. ~
8 . ~
, x--x ~
~ 4 % "`,c~
i ~ i'~-~-e.-w . M.. w x--.,- ~ .
~ ~ f0 20 30 40 S, mna~
Figure 11. Dependence of Signal/Noise Ratio on Area of identified Fragment S
Finally, i~ should be emphasized that whereas the signal/noise ratio is practically
always higher than 5, the ratio Ish.max~Ish.max remains in the range of 1-1.5
(curves 3). This result permits one to talk about the applicability of a simple
identification algorithm consisting in measuring the amplitudes of the two greatest
surges of the oscillogram and comparison of ~their ratio to a threshold value equal
to 2-3. Since cloud formations generally move along curvilinear trajectories, the
angular dependence of the identification signal must also be taken into account.
The dependence of the cross-correlation maximum ~n the angle of rotation nf the
reference for section 4 is presented in F3.gure 12.
57
FOR OFFIC[AL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
~'t1R tl~'FI('IA1. l~F (1N1.\'
!i
0.5
0. 4 �
0. 3 �
- 0. 2
' 0.1 �
0 S f0 15 ~O, �
Figure 12. Dependence of Correlation Maximum on Qacillograph Screen Is on
= Angle of Rotation.of Reference S1ide ~
The angular dependence of the cloud formation identification signal together with
data on linear motions permits one to follow the variation of the total velocity
vector by finding the maximum correlation signal.
Conclusions
The estimates of the accuracy of ineasuring the displacements and angles of rotation
of the fragments of space photographs of the earth's aurface and cloud forn?ations
obtained in the given paper by correlation comparison with references in an OGK in-
dicate the appli~ability of this method for solving different problems of space
video information processing. It is significant in this case that processing the
images in an OGK does not require manufacture of holographic filters, which permi.ts
compar~tively easy use of it in the optical-electronic operational information pro-
cessing romplex. If the scale of theinput slid~s is properly selected and if the
device is executed with high quality, the error of ineasuring the displacement of
the identified fragment is less than the dimension of resolution in the image.
- The results of #he experiments demonstrate the possi.bility of correlation identifi-
cation of fragments of space photographs of the earth's surface by means of a sim-
ple algorithm consisting in separation of the two hiqhest correlation signals, cal-
culation of their ratio and comparison of it to some threshold value determined
upon study. However, a preliminary rough correction o� the projective geometric
distortions in space photographs is required for successful realization of this
algorithm, which can be accomplished by electronic systems during prelfminary re-
cording. Further investiga~ions must be conducted toward improving the OGK schemes
for given application and accumulation of statistical data on identification of
fragments from a large file of space photographs of dffferent types to refine the
requirements on the image quality and to create a set of optimum filters for dif-
- ferent types of images. ~
~
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
BIBLIOGRAPHY
1. Nad', "Digital Processing of Images Obtafned During Remote Irivestigation of
Natuz�al Resources," in "Raspoznavaniye obrazov pri pomoshchi tsifrovykh vychis-
litel'nykh mashin" [Pattern Identification by Using Digital Com~utersJ, Moscow,
Mir, 1974.
2. Korbukov, G. Ye., V. V. Kulikov and Ye. R. Tsvetov, "The Optical Heterodyne
Method of Correlat~on Processing of Images," in "Golografiya i obrabotka in-
formatsii" [Holography and Information Processing], Leningrad, Nauka, 1976.
3. Kozinchuk, V. A. and A. A. Feoktistov, "Measuring the Coordinates of Fragments
of Space Photographs by the Holographic Correlation Method," TRUDY GOSNITSIPR,
No 5, 1978.
5. IQiudson, D., "Statistika dlya fizikov" [Statistics for Physicistsj, Moscow,
Nlir, 1979.
4. Balabanov, A. I., A. G. Nikolayev and A. A. Feoktistov, "Estimating the Accur-
acy of Measuring the Coozdinates of Fragments of Space Photographs," in "Vtoraya
Vsesoyuznaya shkola po opticheskoy obrabot7ce inforniatsii. Tezisy dokladov"
[Second All-Union School on Optical Information Processing. Report Topics],
Gor'kiy, 1978.
COPYRIGHT: Izdatel'stvo "Nauka", 1980
6521
CSO: 1863/270 ~
59
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407/02109: CIA-RDP82-00854R000400070024-4
FC1R O! ~1('1 AI. t~F: t1N1
UDC 621.393
ACOUSTOOPTICAI, PROPERTIES OF INFORMP,TION PROCESSING BASED 0[~1 NONLINEAR ACOUSTICAL
INTERACTION
Leningrad RADIOGOLOGRAFIYA I OPTICHESKAYA OBRABOTKA INFORMATSII V MIIQtOVOLNOVOY
TEKHNII~ in Russian 1980 (signed to press 24 Oct 80) pp 155-163
[Article by Ye. T. Aksenov, N. A. Yesepkina and A. S. Shcherbakov froa? the collec-
tion "Radio Holography and Optical Information Processinq in Microwave Equipment",
edlted by corresponding member of USSR Academy of Sciences L. D. Bakhrakh and candi-
date of technical sciences A. P. Kurochkin, Scientific Council on Problems of Holog-
raphy, Department of General Physics and Astronomy, USSR Academy of Sciences, Iz-
datel'stvo "Nauka", Leningradskoye otdeleniye, 2,150 copies, 184 pages. Additional
sections of this book appeared in the USSR REPORT: CYBERNETICS, COMPUTERS AND
AUTOMATION TECHNOLOGY dated 28 Auqust 1981 (JPRS L/9945) pp 1-39)
[Text] ~ number of acoustooptical devices for processing optical and radio signals
has recently been developed, for example, on the basis of colinear acoustooptical
interaction, multibackground bragg scattering of light and so on. Along with this
investigations of yet another new class of devices are Y~eing conducted which utilize .
nonlinear interaction of elastic waves. Development of these devices was determined
by the appearance of materials with high coefficient of acoustooptical quality and
high elastic nonlinearity.
The results of experimental investigation of a number of acoustooptical devices
based on colinear nonlinear interaction of volumetric elastic waves are presented
in the article.
1. 'I`t:e operating principle of nonlinear acoustooptical devices (ANU) is based on
scattering of a luminous flux impinging on a sound conductor on an elastic wave
generated by tw~o initial elastic wavea propagated in a medium with elastic nonlin-
earity. Let us consider in more detail the case of colinear interaction of elastic
waves during propagation along an acoustic axis.
As is known, in this case nne of the elastic waves is longitudinal and two others
are transverse. The two waves 1 and 2 excited in the sound conductor generate a
third wave 3 upon interaction if the conditions of synchronism are fulfilled
' 1it1~=~~~ Ki-I-Ka=Ka, (1)
60
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/42/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
where fi and Ki are the frequency and wave veator of the i-th wave. If condition
(1) is fulfilled, eight types of interactions are possible [1] which have permuta-
tional syaunetry in the sense that the two longitudinally selected waves interact
and generate a third wave.
One can show that in the absence of attenuation (ai = 0), the outputs Pi of para-
metrically related elastic waves are determined by elliptical Jacobean functions.
Pi ~z}= Pi ~0) dR=Es, (2a)
p~ (t) = fi Pi I~) k4~nsEz,
(2b)
pa ~x) _ ~i pi k~Rn~Es~ ~2c)
where
kZ = /ipz ~~1 r (BngPi ~~)1sla 1'/,
1zPi(~) ~ E� y` Ps~~Va J'
I' is a parameter of nonlinearity, p is the density of the sound conducting material,
P1(0) and P2(0) are the outputs of the i.nitial waves, S is the cross-section of the
acoustic column and x is the current coordinate. From (2c) at small values of ~x
we find
Ua ~t) = 4 Ki~~Vi ~0) U~ ~0) z~ (3~
where Ui(x) is the amplitude of the i-th wave. If attenuation (ai ~ 0) is taken
into ac,count, coor3inate x in (3) must be replaced by xeg:
ezp (-a~s) - ezP I-~ai a~) tJ (4 )
a.,~= al+,a9_a' .
Relations (3) and (4) permit one to determine the amplitude of the generated elas-
tic wave.
If the initial waves Ui and U2 are pulsed signals whose length is short compared to
the travel time of elastic waves through the sound conductor, then amplitude U3 is
determined by the formula
~ �
Ua = y- Ri/Ca ~ Ui � s) U~ (t) ds, ~5)
i.e., it has an integral of the correlation or convolution type.
It is obvious from relations (3)-(5) that ANU permit one to realize the shift of
two signals, their correlation and convolution, and also the controlled delay of a
radio signal with simultaneous introduction of the received information into an op-
tical system for further processing.
2. Typical for an ANU is the need to d~velop sufficient output Sensity of elastic
waves that, as can be seen from (2) and (5), permits one to increase the efficiency
61
FOR OFFiCIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/42/09: CIA-RDP82-00850R000400070024-4
FOR OF'FICIAL USE ONLY
of the device and also to operate in a wide frequency band which is determined upon
- fulfillment of conditions (1) only by the piezotransducer band. This circumstance
determines a number of characteristics in manufacture of acoustooptical cells for
ANU. ~
It is known that the amplitude and frequency characteristics of acflustoop~tical ~
cells are considerably dependent on the method of attaching the piezotransducer to
the sound conductor. As shown by calculations I2]e the highest working frequency
- bands are realized when the transducer is attached to th~ sound conductor by metal
binding layers. The cells used in the given paper were manufactured by a specially
_ developed technique and the transducers weze attached by the thermocompression me-
thod by copper-indium binding layers [3J.
The process of manufacturing the acoustooptical cells reduced to the following main
operations.
Sound conductors of PbM~oOq and GaP and transducer plates of LiI~03 were initially
machined by optical grindi.ng and polishing methods. In this case the nonparallel-
lism of the transducers and edges of the sound conductors to which the transducers
were attached did not exceed 1 micron. After chemical cleaning in toluene and iso-
propyl alcohol, Iayers of inetals that served as,the binding layer in attachment of
the transducers to the so~d conduaicors and also as one of the transducer electrodes
. were applied to one of the ends of the sound conductor and to one of the sides of
the transducer by the thermal vacuum spraying method. Moreover, in our case sub-
layers of chromium that ensure adhesion of the subsequent metal layers, copper and
then indium, were sprayed on. Spraying the copper improved the conductivity of
the binder layer at high frequencies. The thickness of the sprayed copper and in-
dium layers was such that an approximate weiqht ratio of 60 percent copper and 40
percent indium was provided and in this case the thickness of the indium layer was
approximately equal to one-fourth the wavelength of ultrasound in indium on the
central modulator frequency. The transducer plate was then attached to the sound
conductor by the thermocompression method. It should be noted that the piezotrans-
ducers were attached to opposite ends of the sound conductors si.multaneously. More-
over, the ends of the sound conductors to which the transducers were attached were
not polished to improve adhesion during manufacture of lead molybdate modulators.
The thickness of the piezotransducer was then brought up to the required thickness
- equal to half the elastic wavelength in the transducer material with xegard to the
electromechanical coupling ccefficient. The process of manufacturing the acousto-
optical cell was comoleted by vacuum spraying of an upper electrode (Ti, Cu and In).
The manufactured cell was attached to a apecial support aft~r which the transducer
electrodes were soldered through tl~ie transition contact surface to the contact of a
high-frequency plug.
T~ie conversion losses measured by the echo pulse method comprisefl approximately
3-5 dB per conversion for longitudinal waves and approximately 5-6 dB for shear
waves w~th regard to diffraction and dissiputive losses. T'he working frequency
band of these cells was equal to 30-40 percent wfth electric matching of the imped-
ance ~f the transducer and power supply circuits.
62
- FOR OFFIC[AL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ON~,Y
3. Let us dwell further on the results of experimental investigations that we con-
ducted on an ANU based on acoustical collinear interaction in lead molybdate and gal-
lium phosphide crystals.
a. Lead molyk,date crystal mixer. Lead molybdate has the hiqhest elastic nonlin-
earity f~nr longitudinal acoustic waves propagated alonq a fourth-order axis in one
direction. The variance diagram for degenerate colinear interaction of longitud-
inal waves is presented in Figure 1. The laws of canservation of energy and pulse
- (1) are identically fulfilled in this case and consequentlx mixing will occur for
- any pairs of initial wave frequencies. For the selected configuration ~f interac-
tion the parameter of nonlinearity is equal to [1].
Ir~=~3+ Caa 1-17.5. -
The sound conductor had dimensions of 8 X 9 X 23 mn with longest length along the
[OO1J axis. The piezotransducer operated on frequencies from 80 to 350 MHz and en-
sured acoustic output density up to 5 W/em2 (the dimensions of the upper electrode
were 1.2 X 5 mm).
f
_ V~
f~fZ
~
~ I
I
I
. f~ ~
- I ~ ~
` fZ - - ' i i
I
i I !
f~-f - ( 1 ~
I I I ~
/f~ /(Z /(Z /f~ l(~+l~Z l~
= Figure 1. Variance Diagram for Degenerate Collinear Nonlinear Interaction
~ of Longitudinal Elastic Waves
Mixing was investigated on the follow=^g frequency pairs fl and f2: 80 and 320,
115 and 320, 150 and 324 and 320 and 320 MHz in the continuous mode uaing modulation
of one of the input signals with frequency meander of 1 kHz. A photodetector re- ,
corded the light intensity i3 scattered on the generated wave with frequency f3.
The dependence of intensity I3 on the length of interaction of initial waves x is
_ pre~ented in Figure 2 for different levels of input power P. 'The contributions of
sequential scattering, the nonlinearity o� the piezotransducer and spurious liar-
monics of SHF' generators which did not exceed several percent were subtracted wh~n
plutting the experimental curves. The extreme nature of the curvea was determined
by the presence of elas~ic wave attenuation.
FOR OFF[C(A~ USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
o ~ ~ .
1.0 P,~Pr~>p m pr>P~~ .
t r a . . . + + .
o.e P,' , . ,
. . .
' I
0.6 ' ~
� ~ i
n P~
0. 4 pf
.
Pnr
0. Z � ~ . ' , P,
H
~ 0
~ b d
.
1.0 P~ ~ P~~r~ P~rr? p~r~p~n~P~ n
0.8 ~ P;
r ~
0.6 ' Pi
Ptr , . . . Pn
0. 4 > >
. �
� ~ � P ~n
0.2 ' ~ � ~'p m
. � . t
/1
2 4 6 9 ~0 ~Z ~4 2 4 6` 8 ~0 12 14
X , MM
Figure 2. Dependence of Scattered Light Intensity I3 on Lenqth of Interaction
X of Initial Waves for Different Levels of Input Pow~er P: fl =
= 320 MHz, a--f2 = 150 MHz and fl + f2 = f3 = 470 MHzi b--f2 = 115
MHz and f3 = fl + f2 = 435 MHzj c--f2 = 80 MHz and f3 = fl - f2 =
= 240 MHLj d--f2 = 320 MHz and fg = fl + f2 = 640 MHz
For quantitative analysis of the mixer parameters, let us select far example the
case of mixi.ng of frequencies fl = 320 MHz, f~ = 115 MHz and f3 = fl + f2 = 435 MHz.
In the case when acoustic output of initial waves comprises P1 = P2 ~ 1C10 mW, we
find fron, (3) and (4) P3 : 5 mW, i.e., the relative efficiency of the acoustic mixer
is equal to P3/P1 + p2 ~ 2.5 percent. The effectiveness of light scattering on the
acoustic wave f3 comprises approximately 1 percent, which is in good agreement with
the experi.mental analysis. As can be seen from Figure 2, effective mixing is ob-
served over a wide frequency band which is limited only by the piezotransducer band-
_ pass. In this case the effectiveness of mixing incrQases with an increase of input
signal power and frequency, whereas there is an optimum length of initial wave in-
teraction due to attenuation.
b. A gallium phosphide cr~stal correlator. Nondegenerate interaction of acoustic
waves propagated in opposite directions along the [lllJ axis of a GaP crystal (class
43 m) was used to develop'this 3evice. The variance diagram for this case is pre-
sented in Figure 3 and the laws of conservation (1) have ~he form
~i,-ls,=1B~~. KL,'Xs,=x,~~. (6)
A shear wave S3 propagated in the same direction witih the initial longitudir.,al wave
~ is generated as a result of interaction of the longitudinal Ll and shear S2 waves.
r
~4
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407/02/09: CIA-RDP82-00850R000400470024-4
FOR OFFICIAL USE ONLY
The ratio between the initial wave frequencies follows from (6) with different
polarization waves participating in the interaction
_ '2VL ~ (7)
fl yi _ yg ~
and the qenerated v~ave frequency is also deteYmined
~t ='11 vL + vB~ ~ f~ 1 v~ ~ vB~ I ~ 8~
2Vy Vy-~T
8
Relation (5) with regard to the velocity sign has the form
� ~ '
- U, ~e~ rxlx,ve ~ U, (t~~ v, (t~ ati; � ~ (9 )
.
where
~
_ ,
2s to ~ ' -
~ = t y- - y-B . s -�a (t et) -}-�~s::
. a . .
xp is the coordinate of the point for tapping the s~.gnal with a laser beam. Conse-
quent].y (9) is a correlation integral with parameters: a=(VL + Vg)/2VL is the
time scale factor, Ot =-xp/Vg is the time shift fac~or and S=(VL - Vg)/2VL is
the amplitude scale factor.
, f
V~ ~1. -
VS ~
VS f~
. f3 ~ 1
. ~ i
~ i. ~ .
_ . f ~ ' ~ .
~ ~ ~ .
~ I .
-K KE 0 If~ I(~ K
Figure 3. Variance Diagram for Nondegeneratc~ Collinear Nonlinear Interaction
of Elastic waves ~
Experimental investigation of the operation of the correlator was conducted at in-
itial signal frequencies fLl ~ 440 MHz and fg3 ~ 90 MHz with generated wave
65
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2047102109: CIA-RDP82-00850R400404070024-4
F'nR O~FICIAL USE ONLY
frequency of fg3 ~ 350 MHz. In this case the scale factors are equal to a= 0.79,
R= 0.21 and I~t) = 5.7 microseconds (xp = 2.15 cm).
Initial wave pulses 'rl = 2.0 microseconds and T2 = 3.0 microseconds were propagated
counter ~o each other at v~elocities VL = 6.45�105 cm/s and Vg = 3.76�105 cm/s in a
- GaP crystal 55 mm long. The acoustic pulse lengths were approximately identical.
The acoustic output densities created by the piezotransducers reached 5 W/cm2 with
dimension of the upper electrode of 1.2 X 6 mm.
An optical diagram of light scattering on the generated wave is presented in Figure
4. The bragg mode of light scattering was realized since the diffraction parameter
Q exceeded the value Q= 10. The point of impingement of the light beam on the
crystal was selected so that the acoustic pulse with frequency f2 intersected it
first. After this pulse f2 and pulse fl delayed by tzad = 11 microseconds, inter-
act to the left of the optical beam and generate an acoustic pulse with frequency
f3 which is propagated to the right at velocity Vg and intersects the optical beam.
" The oscillograms of the input pulses and the correlation signal are presented in
L^igure 5. The length of the correlation signal is equal to T~ 2 microseconds by
t~:e level of half power. The maximum signal is recorded at moment of time t~ 16
microseconds with respect to the moment the input pulse of shear wave with fre-
quency f2 is started. The effectiveness of light scattering on the generated wave
comprised 10'2 to 10-3. The values of the nonlinearity parameter T, whose analyti-
cal expression for the used configuration of interaction has the following form,
were analyzed on this basis
3Cas6~.~~
r-i+
~ii 2Cia 4C~t ~
A value of I 1' I~ 6 was found for I'.
f
f~=440M1'~ ~ f =90M/'q ' .
~
~VL VS V J L.
. ~
Z
Figure 4. Optical Diagram of Light Scattering on Wave Generated in ANU
Cell Based on Gallium Phosphide: F--photodetectorj 1--impinging
light; 2--to amplifier
4. The given consideration of acoustooptical devices based on callinear~nonlinear
direction of acoustic wares indicates the possibility of developing these devices,
which have the special main characteristics:
66 .
FOR OFFICIAIL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-00850R400404070024-4
FOR OFFICIAL USE ONLY
the radio signals ar~ processed in real time over a wide frequency band de-
termined only by the piezotransducer band when phase synchronism conditions are
fulfilled;
the efficiency of these devices is no lesa than that of most devices now used
and increases with an increase of working frequencies, which is very pramising from
the viewpoi.nt of developing devices that operate at hypersonic frequencies;
, t=L~ .
~ ~ taa8 = 9~Mkc ' - ZMKC ~ . . . ;
_ . . I ;
. U �i
a� .
~y, 3n~Kc
fZ=90M1'~`~.
. .
,
, U; ,
;i~,~2MKC .
f~=44t7IGlrt~ .
, . .
. ;
~~'1
i, f
/ 5^-e
i, _
:
- 3 ~ ';L
~ a ,
~
~ ~L'8, 7'~r.;
~ Y ~ 1
~a�` .z
_ , , ; ~~r'." ~ t~..
r, y
r~`
~ ~
- , t =~.16 MKC . ~ ;
~ .
_ . -t~- - _
Figure 5. Oscillograms of Input Pulses (a, b) and Correlation Signa7. (c)
the use of nonlinear acoustical interaction does not @eteriorate the dynamic
range of optical information processing systems since the dynamic range of acoustic
processes themselves, determin~d by very weak thermal noise in a solAd, comprise
approximately 90 dB [ 1 ] ;
- 67
FOR OFFICiAL U~E ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
the gener�ated acoustic signal is read by an unspZit light beam or a beam
focussed at the point of tapping; consequently, the aperture nf the first lens in
the optic.31 sysfem of the ANU is small and is not dependent on the dimensions of
the souzid conductor, where~s ardinary acoustooptical devices require a collimating
system with lenses whose dimensions exceed those of the sound conductort
the use of an ANU makes it possible to simplify the arrangement of the pro-
cessing system for achieving convolution or correlation of signals canpared to
acoustooptical devices oi~dinarily used;
- universal materials now used, for example, such as lead molybdate, gallium
phosphite, paratellurite and so on, can be investigated for development of ANU.
The indicated characteristics permit one to cor.clude that acoustooptical devices
based on collinear nonlinear acoustic int~raction will find their place in a number
of radio signal processing devices.
BIBLIOGRAPHY
1. Bridoux, E., J, M. R,ouvaen, C. Bruneel et al, "Correlation and Convolution of
Bulk Waves in Nonpiezoelectric Solids,"JOURNAL OF APP~IED PHYSICS, Vol 46, No 6,
1975.
2. Aksenov, Ye. T., N, A. Bukharin, V. A. Grigor'yev et al, "The Frequency Ch�rac-
teristics of Acoustoop~:ic Modulators," ~HTF, Vol 46, No 10, 1976.
3. Aksenov, Ye. T., V. A. Grigor'yev, N. A. Yesepkina et al, "High-Frequency Ultra-
sonic Bragg Light Modulators," ZHTF, Vol 42, No 11, 1972.
COPYRIGHT: Izdatel'stvo "Nauka:", 1980
- 6521
- CSO: 1863/270
68
FOR OFFiCIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICiAL USE ONLY
UDC 621.393
EFFECT OF ELASTIC WAVE ATTENUATION ON OUTPUT SIGNAL OF ACOUSTOOPTICAL DEVICE FOR .
CORRELATION ANALYSIS
Leningrad RADIOGOLOGRAFIYA I OPTICHESKAYA OBRABOTKA INFORMATSII V MIKROVOLNOVOY
TEICHI~III~ in Russian 1980 (signed to press 24 Oct 80) pp 163-174
[Article by S. v. Kulakov from the collection "Itadio Holography and Optical Informa-
tion Processing in Microwave Equipment", edited by corresponding member of USSR
Academy of Sciences L. D. Bakhrakh and candidate of technical sciences A. P. Kuroch-
kin, Scientific Council on Problems of Holography, Department of General Physics and
- Astronomy, USSR Academy of Sciences, Izdatel'stvo "Nauka", Leningradskoye otdeleniye,
2,150 copies, 184 pages. Additional sections of this book appeared in the USSR RE-
PORT: CYBERNETICS, COMPUTERS AND AUTOMATION TECHIJOLOGY dated 28 August 1981 (JPRS
L%9945) pp 1-39)
~ (Text) Introduction
The concept of functional space L2E [1, 2] is used extensively in the theory of
analog device�~ for processing continuous causal signals with limited energy. Its
application specifically permits one to select the criterion of comparison (the
measure of proximity) of signals shaped by real devices (real signals) and the sig-
nals of ideal devices (ideal signals).
This criterion can be used to analyze linear distortions of signals in acoustoopti-
_ cal devices. They may include distortions detezmined by attenuation of high-fre-
quency elastic waves in an acoustooptical interaction medium at low levels of input
signals.
Thes~ distortions of signals in an acoustic light modulator (AMS) and of output sig-
nals of correlation analysis acoustooptic devices (AOUKA) are considered in this
paper.
Let us limit ourselves to the case of relatively narrowband signals, practically
complete infozmation about which is contained in their complex envelope.
Let us use as a measure of proximity of real and ideal signals normalized Euclidean
metrics [3] deternuned by the relation ~
~B '~Cn~-C~~ ~1~
- ~sroaa NC~xII-I-pC~~ ~
69
FOR OFF[CiAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
F'OR OFFICIAL USE ONLY
where ( ~ ~ ~ is a s i g n of the no rm and Cid = Cid (x, y) and C= C (x, y). are the '
moduli of the complex envelopes of ideal and real signals. We note that the value
~E/~Em~ corresponds to the normalized mean square error.
The Signal in the AMS
~ It has been accepted to characterize attenuation of elastic waves in an acoustoopti-
cal interaction medium by frequency-dependent coefficient a(w). It is specifically
taken into account in the electroacoustical transfer coefficient of the AMS intro-
duced in [4, 5]. This coefficient characterizes the frequency properties of the AMS
and establishes the clear relationship between the spectrum of the input electric
signal and the instantaneous spectrLUn of the spatial frequencies of an elastic wave
packet (space-time signal) in the modulator aperture.
Relations that determine the signal shape must be found to estimate the distortions
in the AMS.
~ � z
. Z
I L .
_ ~z
~ ~
ses (tJ
Figure 1. Acoustic Light Modulator: I--plane light wave; L--length of aper-
t~~re along direction of elastic wave propagation= 1--piezotrans-
ducer; 2--light-sound conductor; 3--acoustic absorber
Let us assume that the AMS aperture (Figure 1) is sufficiently large (greater than
the length of the elastic wave packet). The space-time aignal in this aperture,
corresponding to the input electric signal s(t), can then be represented in the fol-
lowing form:
r(z) s(vt. Z) = Z~Z) ~ S(~i) 1~ ~'~r~ Z) e~W' ~�~'`~dw� (2 ~
-w
where K(wZ, z) is the electroacoustical transmission coefficient of the AMS [4, 5],
~(~Z) is the signal spectrum s(t) expressed in spatial frequencies wZ, v is the
speed of propagation of elastic waves, z is a spatial coordinate
r~z~ = r i 1t -~r~ ~ Z~ ~t~
l ~ at s~vt and = ~o D~o 3rid - ~ < Wo - ~wo, ( 5 )
where c~~ is the mean bandpass frequency of the AMS usually coincident with the
signal carrier frequency s(t) and 2~w~ is the width of the AMS bandpass.
Moreover, the signal delay time in the piezotransducer of the AMS is negligible due
to its smallness while we subsequently omit the multiplier r(z).
The following analytical signal will then correspond to signal (4)
wo+emo ~o
3 - (~t-~)
s' (ut~ z)=~~� ~ e-�~'"~�S` e' d~,, (6)
. ~,-e~,
which can also be represented in the following form
/f�-� (.t-r) ~7 )
i~ ~U~~ Z~ = l~ ~U~~ 1~ 9~ ,
where
+e~u, -J9
G~ (ut, a) S e-6(We fl)`S`' (~a._ g) e ~ d4 (8)
_e,,,
is the complex envelope. In this case ,
G` (ut, z) =C (ut, a) elp~�~, (9)
The results of numerous investigations of materials which find application for de-
velopment of various acous~ooptical devices [5-8] can be used for practical calcula-
tions of the signal shape in the AMS. These experimental data indicate that the
71
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
the the quadratic dependence of the attenuation coefficient on frequency is most
frequently counterec~, which is tygical for most crystals and glass over a wide fre-
quency and ter.?perature range. Therefore~ let us limit ourselves in this paper to
consideration of the function a(m) = T2c~2, where r2 is the dimensional proportion-
ality constant.
CN
a ' ~ ~ivmas
ao=0 .
0. 25 �
D. 5 .
~ 0.5 0. 7 ~
3 2 1 0 1.. 2,, 9.
m ~ m m r?~ � ?n , m Z
. ~ p- a
Z~ y cio=o .
' ' 0. 7
. 0.25
. n , ,
~ i
2 1' D 1 2 3
-m -m m , m , m
L
. ~ ~ � - ~ 3 .
' -2n ,
8 sin 6~,ot ; a--modulus of
Figure 2. Signal at AMS Aperture at e~ ~ co~ ~ar.
complex enveloaes b--phase characteristic _ _
Let us introduce the following AMS parameters:
_ a- 52/w~ is the relative detuning, a0 -~WO~WO is the relative detuning
corresponding to the boundar_y frequencies of the AMS bandpass, R= z/L is a dimension-
less spatial coordinate; t' = tv/L is dimensionless time= aQ = I'2~2 is attentuation
of the harmonic oscillation of the carrier frequency on length L, ~L T~ L/v is the
travel time of the elastic wave packet through the AMS aperture and m= 2~f~T is the
information capacity~of the AMS, greater than or equal to the information capacity
of the signal at the input.
With regard to the parameters, let us write (8) in the following manner:
� +ao xmo ~l~ ~
~t~~~) - nu� ~ e~~(1-0~~5~ (u) e f ~ ~da.
-ao
72
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400074424-4
FOR OFFiCIAL USE ONLY
Let us make the calculations according ta the relation
cNc~~. a~=~ c~~, Ko,~,. c~~~
determi.ning the mcdulus and independent variable.
Let us use the signal in the AMS in which attenuation is absent as the ideal siqnal.
Let us compare to the value of CN id (t'~R) by using the metrics determined in (1).
In this case let us fix mament of time ti > tu, where tu is the length of the input
electric signal (the moment of feeding the electric signal to the electroacoustic
transducer of the modulator is taken as t= 0)
It is obvious that the distortions at the end of the a~;erture (at high values of t')
are more significant. Consequently, being given the permissible value of the error
and analyzing the possibility of using the AMS, one should fix those values of ti at
which the signal will be close to the end of the AMS aperture.
Let us represent (1) in the following form for calculations:
4E ~~t~ ~ C ~~I p) -CN ~t~~ Y~~1dt'1,~~ ~12~
. 1~~ Nex { ~
l ~
p (ti) / h L
Emaz ~ ~(~Cz P) d~ I' ~ f CN ~>~As)~cos[~ot(p Bt+et81~Q]ds.
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407/02/09: CIA-RDP82-00850R000400470024-4
FOR OFFICIAL USE ONLY
Thus, (14) and (15) are nonlinear determined models of an acoustooptic system that
performs correlation processing of narrowband signals during Raman-Nat diffraction.
With Bragg diffraction (1) is written in the following form:
E ~ ![m-~'-cP�~t-�,)-6~],l
e~ (t, a~) = p. rect tza) e~Weat cos 0.5 A1-}-. sin�0.5A ie p'~ ~ X
� ~ ; ~ (16)
~~~~~~e-`~ ~ ~
[cos O.SA~ sin 0.5A,eJ Q " .
Having perfozmed simple transfoxmations, we find
+m
il_(t)=ZB9 ~ rect(:a)sinAisinA~cos[~ot(P-Q)-9i-}rA~SignQ~dz~. ~17)
_o,
Expression (17) is a nonlinear determined model of an acoustooptic system that car-
ries out correlation processing of narrowband signals during Bragg diffraction.
Relations (~5)-(17) take into account the nonlin~arity of acoustic light modulators
and permit one to make numerica~ calculations of the shapes of output signals of
~ acoustooptic correlation devices and also the amplitude characteristics of para-
metric quadupoles based on them.
BIBLIOGRAPHY
1. Kulakov, S. V., O. D. Moskalets and B. P. Razzhivin, "Generalized Block Diagram
of Acous~ooptic Correlation Processing Device," in "Akustoopticheskiye met~dy
obrabotki informatsii" [Acoustooptic Methods of Information Processing], Lenin-
_ grad, Nauka, 1978.
2. Kulakov, S. V., "Akustoopticheskiye ustroystva spektral'nogo i korrelatsion-
nogo analiza dignalov" [Acoustoopti~ Devices for Spectral and Correlation Analy-
sis of Signals], Leningrad, Nau}ca, 1978.
3. Fikhtengol'ts, G. M., "Kurs differentsial'nogo i integral'nogo ischisleniya�'
(Course of Differential and Integral Calculus~, Va]. 2, Moscow, Nauka, 1966.
- COPYRIGHT: Izdatel'stvo "Nauka", 1980
6521
CSO: 1863/270
FAR OFFICIA~, USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400074424-4
FOR OFFICtAL USE ONLY
APPLICATIONS
_ unc 658.7/.8.003�3:657.~71.7~r.o11.56:681.32
AU'1'~MATION OF SETTLEMENTS WITH SUPPLIERS IN SUPPLIES AND SALES ORGANIZATIONS
Moscow MATERIAL'NO TIICHNICHESKOYE SNABZI~NIYE~ SERIYA 4s PRIMENENIYE MATEMATI-
CHESKIKH METODOV, VYGHISLITEL'NOY TIICHNII{I I ORGTII{HNII{I V MATERIAL'NO TII{HNI-
CHESKOM SNABZHEDTII (NAUCI-INO TIICHTIICI~9CIY REFERATIVNYY SBORNII{) in Russian No 3~
_ Mar 8o pp 5-6
[Article by G. P. Bogdanova~ senior engineer, computer center of the ~Tolgo-Vyat-
skoye Main T erritorial Administration, USSR Gossna,b~
[Text] In the computer center of the Volgo-Vyatskoye Main.Territorial Administra--
tion of the USSR Gossnab the problem "Settlements With Suppliers" ha,s been worked
out and introduced into the Association "Volgovyatma,shelektrosna.bsbyt."
The solution of ~:.'ze problem is ba.sed on informa,tion about the actua,lly arriving
goods~ the bulk of which is created during recording of the movement of prod.uction
and packing, and on informa,tion arriving frum the bank about payments.
Invoices of suppliers which ha.ve arrived in the association accounting office
through the bank or by ma,il are grouped according to the commodity ciivisions flznc-
tioning in the association. Workers of the cofnmodity divisions verify tha.t the
d.a.ta of the supplier's documents ma.tch all the conditions of delivery, accept
invoices of the supplier and write up arriving orders. In accordar.ce with the
da,ta on the good.s which ha.ve actually arrived, the incoming order is printed on
"ZQyemtron-393" ma,chines witM perforator accessories in the electronic billing
ma,chine unit. ^lhe prepared punched. tapes are transmitted daily for processing on
a "Minsk-32" computer.
The accounting off9.ce transmits the verified and accepted invoices of suppliers
to the computer center.
- To make the settlements ar.id print outgoing sta.tements the daily mass of invoices
monitored and formulated on magnetic tape is used~ ::~,s are the daily ma.ss of the
arrival of goods by invoice items, the starting balances at the beginning of the
month and the referet~~e inform~.tion "Chara,cteristics of Supplier Ente~rises."
The starting ba.lance at t~he moment the ta,sk was introd.uced is entered f`rom the
6-SN order register. The daily and monthly records of settlements with sc:ppliers
are made in a cross-section of each wa,rehouse section and commodity division. The
- results of the settlements ma,de with suppliers permit ha.ving data, each day from
86
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
E'OR OEFICIAL USE ONLY
the beginning of the month in total against the staxting balance of account 60
"Settlements With Suppliers and Contrac-tors" (debit and credit); debit account 60
in accordance with the credit and credit account 60 in accordance with the debit
of the accounts; credit the balance at the end of the month for account 60, credit
for unbilled deliveries and debit for "goods in transit."
After the processing nf data with a special complex of progr~ms the informa.tion
about the arrival of goods and invoices is grouped in autput sta~tementss the in-
voice register, the sta.tement of the arrival of goods, the statement of settlements
with suppliers for warehouse deliveries, the sta.tement of "goods in transit" and
unbil].ed deliveries.
Computer solution of the problem contributed to increa,se of the effectiveness of
monitoring settlements with suppliers for ea,ch invoice, the adoption of effective
measures for the timely arrival of "goods in transit" and "unbilled deliveries" and
the timely and reliable obtaining of accounting data assurirxg the synthetic and
analytical recording of se~tleinents with suppliers both for the association as a
whole and for subdivisions.
COPYRIGHfs Tsentral'nyy na.uchno-issledovatel'skiy institut informatsii i tekhniko-
ekonomicheskikh issledovaniy po ma.terial'no-tekhnicheskomu snabzheniyu
- Gossnabe. SSSR, 1980.
2174
= cso: 1863/26~
87
FOR OFF[CI~+L U~E ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FCR OFFICIAL USE ONLY
unc ~658.oi2.ou..56:ooz)::[F,a.7/.s~665.7i]
PRINCIPZES OF SOFTWARE OF 'NEFTEPRODUKT'
Moscow MATERIAL'NO-TII~HrtICHE~.KOYE SNABZI~TIYE~ SERIYA Ws PRIMENENIYE MATEMATI-
CHESKIl{H METODOV, VYCHISLITEL'NOl' TII{HNIKI I ORGTII{HNIKI V MATERIAL'NO TII{HNI-
CI~5KOM SNABZI~NII (NAUCHIdO TII{~TICI~SKIY REF'ERATIVNYY SBORNII{) in Russian No 3,
Ma.r 80 pp 12-13
[Article by A. Fishel�, section chief of "Nefteprod.ukt" Centra]. A3U Design
- Bureau]
[Text] T o solve problems ba.sed on the YeS-1o33 computer~ in the Cen;ral ASU Design
Bureau of "Nefteprodukt" a second line of a package of applied progr~.ms has been
creatPd~ one assuring the processing of sets of data on direct-acce:,s equipment
(PPP "RIP-2-15C") under the control of a YeS opera,ting system.
The RIP is characte::~:` ;ed by a traditional approa,ch to da.ta processing. An advan-
- ta~e of the method ci discrete sets consists in more rapid results than with an
integrated approach (data banks~, and a shortcoming is the fact tha,t the creation
of different se:~ which include different iaforma,tion on the sa.me ob~ects and make
use of information fz~om different sources can lead to incompatibility and mutual
non-equival~nce of the data.
The RIP uses a magnetic disk store for basic storage and da.ta processing and mag-
netic tapes for lon~-term storage.
The methods of processing sets assure rapidity~ precision and completeness of ~iata
reirieval, reliability~ simplicity ~nd convenience of opera.tion and the ordering of
information in the absence of c,~xeiul external sortings.
T he RIP represei:ts a complex of programa~ a language of user-RIP interacti~n and
= RIP da,ta sets.
T he programs permit formulating, correeting and apening up consecutive-index sets,
looking up and reading RIP information, etc. They use the entire RIP and OS axsen-
al for�optimum performance of input-output operations and the discovery, analysis
and loca.tion of possible errorg.
The langua.ge of t~aer-RIP interactioii is a set of concepts~ agreements a.1d rules
which germit receiving and renewing informati.on of data sets~ looking iip and issu-
;ng the necessa.ry information and analyzing the arising situa.tions.
88
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R004400074024-4
FOR OFCICIAL USE ONLY
All the RIP sets are of the consecutive-index type frith blocked recardings of taken-
dm~n data. The b~sic principle of the logical arganizatioai of record.ings is the
field structure of the recording, in which the recording consists of fields singled
out by the user forr storage of data of a specific type. The apeed of retri~val ~nd
reading of ~ata is achieved because of the absence of regio~s of redunclancy in the
set storages, the presence of infrnrmation in Which ~ea~r: slo~ws datn the reading.
Information is corrected, not in the data. set storages, tnit in the RIP set carrect-
or~ xhich permits realizing invariability and independence of the o~ganization of
sets from the correctian methods. The RIP set catalog c~fiains certain inform~tion
about the disposition of sets, the results of successive carrectian, etc.
Address for inquiriess 109390, Moscoir~ Ul Ax~tyukhinoy~ 6'B", TsKB ASU "Neftepro-
dukt" . ~
COPYRIGHP: Tsentral'nyy na.uchno-issledovatel'skiy institut informa.tsii i tekhniko-
ekonomicheskikh issledovani~ po ma.terial'no-tekhnicheskomu snabzheniyu
Gossnaba SSSR, 1980.
2174
cso: is63/z64
89
FOR OFFICIAL USE ONLY
~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
UDC 681. 5].8~ 6,58. 7/.8. 008
IDENTIFICATION OF PROGRAMS AND DATA FILES IN TJSSR GOSSNAB INFORMATION SYSTEM
Moscow MATERIAL'NO TII{HNICHESKCIYE SNABZI~NIYE, SERIYA 4: PRIMENEriIYE MATEMATI-
CHESKIl{H METODOV~ VYCHISLITII~'NOY TII~HNII{I I ORGTII{HNII{I V MATERIAL'NO TII{HNI-
CHESK~M SNABZHENII (NAUCHNO TII~HNI~HESKIY REF'ERATIVNYY SBORNIIC) in Russian No 6,
Jun 80 gp 1-2
[ArticZe by A. I. Pronchenkov~ deputy section chief, Main Computer Center, USSR
G ossnab]
[T~xt] The USSR Gossna.b Information Systes (fixst line) ha,s been developed and
introd.uced in the USSR Gossna.b Main Computer Center. ~
In ~:arking out the informa,tion system, interna.7. atandax~ds were introduced into a
- minicomputer for the identification of softwaxe, data, files and the stand,axd-refer-
ence complex, and also some special standa~x~l.s connected with distinctive features
of the technical ba,se of the information system.
The programs and data.files axe identified by serial number. The identifier ha,s a
length of 8 bytes and determines:
l. Tha.t the set belongs to the class of programs or ddat files, and aZsn tha,t it
~~~longs to the informa,tion system. .
1.1~ For programs relating to the information system, the first four symbols must
be "ISOR." For programs not relating to the information system the identifier is
established. specially in each specific ca,se and is entered in the system as permit-
ted.
1.2. For data files relating to the ~nformat ion system, the first four symhols
must be "ISOR." For data files not relating to the informat ion system, the iden-
_ tifier is established. specially in each speaifi~; ,as~ and is entered in the system
as permitted.
2. Tha,t the set belongs to some specific task.
2.1. Symbols 5 and 6 in the identifiers of programs and data files indicate that
they belong to a certain task in the information system and ca,n vary in the range
of o0-99.
- 3~ The specific number of the da,ta f31e or program.
90
~ FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400074024-4
~ FOR OFF7CIAL USE ONLY
3.1. Symbols 7 and 8 af the identifYer determine the number of the program o~r data
file tirithin a specific task. The identifier of dictia~arles also contains 8 sym-
bols and has the folloKing forms SIA~V No, Where No is the number of the dictiaa-
ary, xhich varies in the range of o0-99� There also is a service file of the rior-
ma.tive-reference . complex, xhich is called '~IASSARY" and co~tain~ inforn~a,tian abaut
all the dictia~.aries activated in the system.
There also is a nwnber of standards introduced due to s~truct'ural features of the
m;n;computers and the absence of an operating system in them. Thus, on a disk a
place is allott.ed. for stor~ge of the data and ordinal number of the current da.y in
the year (the date is entered each t,ime the system is Yoaded)~ for stozRge of the
disk number and the date of the last reorganizatian of c]ata on it~ an~d for storage
of the number of the minicomputer complex.
The introduction of a single identificatian in the information system permits put-
~ ting in order the a.xzangement of the data. files a.nd progiams on disks of informa-
tion complexes, writing a numb~r of programs xhiah simpli~ data file maintenance
and facilita.tes the writing of applied programs.
Address for inquirles: Moscow, Promyshlennyry proyezd~ d. 3, Main Computer Center,
USSR Gossnab.
COPYRIG}1T: Tsentral'nyy nauchno-issledovatel'skiy institut informatsii i tekh~iko-
1 ekonomiche~ki~ch issledovaniy po material'no-tekhnicheskomu snabzheniyu
Gos~aba SSSR, 1980.
2].74~
CSOs 1863/264
91 .
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00854R000400470024-4
FOR OFFICIAL USF ON1.Y
Unc 681. ~8:658.7/.8.008(47 + 57)
SANG�rIONED ACCESS TO DATA IN USSR GOSSNAB INFQRMATION SYSTEM
Moscow MATERIAL'NO TII{HNICHESKG'YE SNABZI~NIYE, SERIYA 4: PRIMENEtiIYE MAT,EMATI-
CHESKIKH METODOV~ VYCHISLITEL'NOY TII{HNIKI I ORGTII{HNIl{I V MATERIAI,'NO TII{Ii1~TI-
CHESf{OM SNABZI~III (NAUCHNO TII{HNICHESKIY REFERATIVNYY SBORNIK ) in Russian No 6~
Jun 80 pp 4-5
[Article by A. I~ Pronchenkov, depu-ty section chief, Main Computer Center, USSR
G ossnab]
[Text~ The introduction of sanctioned access to data in the USSR Gossna,b Informa-
tion System was stipulated by the need to limit the circle of users of the system.
Sanctioned access permits putting in order the obtaining of inquiries on various
problems and protecting information stored in the system.
On the system level, access to data is protected by the use of user identifiers and
passwords. For each user of the system an especially selected responsible person
establishes a personal identifier and an original password. During ea,ch reference
to the system the ~~ser must present himself with his iden~ifier and password., which
are then verified in the list of system users and, if +h~~ wre nct grsUon~ th~re ar
do not correspond with one another~ the attempt to enter the system is denied.
Lach user is given the right t o change his password as desired at any time. The
~.dentifier is changed only by the especially selected person.
On the data ievel~ protection is accomplished in the following manner~ T he manage-
ment esta,blishes a list of problems accessible to each user, and within each task
the degree of competence of the user is determined~ that is, the working conditions
of the user for the given task (data reading~ data renewal, change of programs for
the task, etc) . 1'he sys~:em gives the user the possibility of accesa only ~o tasks
- determined to be permitted for him a,nd of working on a selected task only under the
conditions permitted him for tha,t task. The system is entered from a terminal in
an interactive mode.
- Sanctioned access assures three levels of access to data and programs of the infor-
ma,tion systems the chief dispatcher level~ the level of the central apparatus user
and the special level.
The chief dispatcher level permits ha.ving access to a11 informa,tion in the system
only in t,he data reading mode, except information of the special level, T he level
92
. ~ FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-OQ850R000400Q70024-4
~FOR OFFICIAL USE ONI~Y
of the central apparatus user permits ha.ving access to reading and recording of
informa,tion on tasks permitted to each specific user. The special level permits
having access to individual information only to the user himself wl.th additiona,l
passwords for a specific task.
The introduction of sanetioned access will permit excluding unqua,lified access to
the system, putting information renewal in order and limiting the circle of users
for each specific task, which is necessaxy in col~ective-use systems.
Address for inquiriess Moscow~ Prom;~rshlennyy proyezd., d. 3~ Ma,in Computer Center,
USSR G ossnab.
COPYRIGHTs Tsentral'nyy na.uchno-issledovatel'skiy institut informa.tsii i tekhniko-
ekonomicheskikh issledovaniy po materia~.'no-t~khnicheskomu snabzheniyu
G ossna.ba. SSSR, 1980.
2~74
cs~: 1863/z64
93
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-40854R000400070424-4
FOR OFFICIAL U5E ONLS'
unc [69.003:681.32.06]
FORMING AND I5SUANCE OF CERTIFICATES OF MOST IMPORTANT LINES
Moscow MATERIAL'NO TII{FINICHESKOYE SNABZI~NIYE, SERIYA 4~: PRIN~NIIJYYE MATEMATI-
CHESKIl{H M~'rODOV, VYCHISLITEL'NOY TII{HNII{I I ORGTII{HI1IKI V MATERIAL'NO-TIICHNI-
CHESKOM SNABZHENII (NAUCHNO-TII{HNICHESICIY RE~ERATIVNYY SBORNII{) in Russian No 6,
Jun 8o pp 6-7
[Article by I. B. Ayngorn, deputy section chief, and senior engineer Yu. N. Anurov~
Main Computer Center, USSR Gossnab~
[Text~ The task "Forming and issuance of certificates of the most important lines"
has been develope3 and introd.uced into industrial operat~.on within the framework of
the firsl; l~ne of the USSR Gossnab In~orma.tion System. The task was accomplished
on a minicomputer and is intended to implement monitoring during�the introduction
of capacities into operation during the course of construction and equipping of the
most important lines of enterprises for mineral fertilizer prod.uction. The task is
solved , onthly.
T he complex of minicomputer programs permits monitoring up to ~00 lines for 30 in-
dieators. The indica,tors include the introd.uced ca.pa,cities far each line~ the
volumes of ca,pital investment~ and construction:and installation work~ the numbers
of workers, the delivery of raw ma,terials and supplies, the delivery and installa-
tion of equipment and reinforced-concrete and metallic structures.
The voiume inc3.icator., deadline for putting ~.n operation and the actually introduced
volume are kept tracI~ of for each capa.c3ty. An annual plan~ a plan for the report
period. and ~alfilment of the plan from the beginning of the year in an increasing
total are prepared for the rema.ining indica.tors.
Provision has been ma.de for two kinds of data input: automa,tic, from ma,gnetic tape
obtained as a r.esult of solution of the problem of deliveries of inetalwork on~.the
YeS comp�.ater~ and manL.al, from documents and displays on video terminals. The com-
plex permits issuing documents in the form of reports from a video terminal on a
screen or printer.
There is the ~ossibility of obtaining the certifica,te of a specific li.ne on all the
mineral fertilizer lines in succession, starting from any one of them~ and also
,summary data in a cross-section of contractor~ customer, territorial organ or for
all mineral fertilizer lines. Two days were iec}uired for tha.t work when it was
- done manually, and after automa,tion of the issuance of certificates not mor~ tha,n
94
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R004400070024-4
, . FOR OFFICIAL USE ONLY
1 minute was required. This made it possibl.e to issue certiflcates ozi mineral
fertilizer li}~es monthly, Khere~a,s when it was done manually they Kere issued quar-
terly. ~
Address for inquiriesi Moscow, Orankov per 5, 1~1ain Ca~puter Center, USSR Gossnab.
COPYRIGHPi Tsentral'~yy nauchno-issledova.tel'skiy iristitut informatsii i tekhniko-
ekonomicheskikh issledavaniy po material'no-t~khnichesgomu�.snab$heniyu
G ossnaba SSSR, 1980.
z174 ~
cso: z863/z64 ~
95
~ FOR OFFICIAL USE ONLY "
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-40854R000400070424-4
FOR OFFICIAL USE ONLY
~ unC 681. 32. 06: [6,s8.711. z+ 339 � 6] (474. 5)
CON1~.'LEX OF TASKS IN DEPERMINING MINERAL RESOURCES REQUIREMENTS OF I,ITHUANIAN SSR.
NATIONAL ECONOMY
Moscow MATERIAL'NO TII{HNICHESKOYE SNABZI~NIYE, SERIYA 4~ PRIMEfiTEPiIYE MATEMATI-
CHESKIKH M~,'rODOV~ VYCHISLIT~'NOY TII{HNIKI I ORGTII{HI~1If~I V MATERIAL'NO TII{Hr1I-
CI~{OM ~ SNABZHENI2 (NAUCI~10 TII{HIdICHESKIy REFERATIVNYY SBORNIl{) in Russian No 6,
Jun 80 pp 14-15
[Article by K. A. Lyaugminas, deputy chiei, Republic Coir~puter Center, and V. V.
Stoshkus, section chief, ~epublic Computer Center~
[Text] The Republic Computer Center of the Lithuanian SSR Gossna,b ha.s developed
_ and introduced a complex of tasks in determining the requirements of the nati~onal ~
economy of the republic for ma.terial resuurces on a group and specified products
~ list. The group products list includes the production, the balances and plans for
the distribution of which axe prepared by the USSR G ossnab. The complex of tasks
is oriented towaxd the existing organizationa.l structure of the Lithuanian SSR
G ossnab with consideration of interaction with the seconc~ level o~ the Lithuanian
_ SSR Gossnab ASU (soyuzgla,vsnabsbyts and soyuzgla,vkomplekts) and with related. z~.ys-
_ tems of the republic. The informational interactions of those systems are accom-
plished on the basis of a standa,rdized system of documentation of ma,terial and
technica,]. supply with use of t~he production coding system a,dopted in the soyuz-
7.avsnabsbyts and soyuzglavkomplekts.
The reliability of formation of input dacuments is monitored in two sta,~esi when
the documents axe transferred to the users and whsn the data are processed on the
computer. In the first stage the specializod ad,ministrations before order asaign-
ment inform the users about the assignment order~ and in tha,t case special atten-
tion is given to the reliability of the codes. Applica,tions from users axe �
received by special delivery after caxefLil checking of the filling of all codes
by them.
In the second sta,ge counting and logical monitaring of the dat~, are accomplished
by a programmed method in the Republic Computer Center.
In the process of~~solving the complex of tasks five types of machina reports axe
formed: "Swnrnary executive ba,lance" (Ml) with a summing up of the results for the
_ ~n.d holder and group of production, "Record of production requirements" (M2~ with
a summing up of the results for the ~nd'�holder and group of prod.uction, "5umma,ry
_ report of production requirements" (M3~, "Report of ord.ers for delivery of speci-
fied production (in a cross-section of the ba.ses" (M4}, with a summing up of the
96
FdR OFF1C[AL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
� FOR OFFICIAL USE ONLY
results on t,ype, variety, brand and size and group oF productiaai, and "3ummary .
rep~ort of orders for delivery of specified productim" ~+15), irith summing up of the
results on type, variety, bra~nd and size and ~p of prodsctia~
1~lachine reparts are used by workers of specialized administratiaais for xork ~th
users and substantia~iari of the needs of the rep�sblic .for mater3.al resaurces.
The soft~are xa.s created on the b~sis of a padsage of applied programs preP~red ir!
"[lkrglavs~nabs~.stema" Praiuctio~ Technicx~l Associati~ far tt~ "Minslt-32" computer.
- Introduction of the complex of ta.sks pezmitted coasiderahly improving the quality-~
and reliability ois order assignment. The actual saving i'Yom introduct3a~ of the
t,a~ amounted to 162,400 rubles per year.
Address for inquiries: 232006~ Vilnius~ ul Partizanu~ 57, Republic Computer Cen-
ter Lithuanian SSR ~ossnab.
COPYRIGI~s TsentraZ'nyy nauchno-isslndovatel'skiy institut informatsii i tekhuiko-
- ekonomicheskikh issledovaniy po materi3].'no-tsk~r~nicheskomu snabzhe~i.yu
G ossnaba SS5R ~ 1980 .
2174
CSO: 1863~264 .
- 97
' FOR OFFIC[AL USE ONLY . .
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407/02/09: CIA-RDP82-00850R000400470024-4
Ft)R O~l~IC1A1. 11~M: ON1.1'
vnc o25.4~3. 06: [681. 3. 06: 5~.9.6821
TECHNOLOGICAL PROCESS OF DEVEIAPING VOCABULARY FOR YES COMPUTER
Moscow MATERIAL'NO TII{HIJICHESKOYE SNA~ZI~NIYE~ SERIYA 4: PRINI~3~IENIYE MATEMA-
TICHESKII{H METODOV~ VYCHI~'LITII,'NOY TF~CHNIKI I ORGT~KHNIKI V MATERIAL'NO TII{HNI- ~
CHESKOM SNABZI~IJII (NAUCHNO TIICHIIICHESKIY REFERATIVNYY SBORNIl{) in Russian No 12~
Dec 80 pp 11-13 _
[Article by V. S. Birger, group lea.der, "Les" ASU section, M. V, Sheatyakov,
computer center deputy chief, and V. Ya. Shmidtman, head~ NSI (not flirther iden-
tified) section head]
- [Text] In the computer center of the Krasnoyarsk Main Tez^ritorial Aci~ainistration
of the USSR G assnab, in work on the creation of a voca,bula.ry for the Ye5 computer
common to the entire system, a variant of text curtailment and punching durin data
input into the computer was proposed. Processing languages r~ritten in.the PI,~1
language ha.vP been prepaxed for implementation of the proposed method.
- The newly developed tachnological process ha.s the following appeaxance: �
1. During punching of the production voca.bulary in abbreviated form the punched
� data include the items and GOST's, the initial cod.es of the standarci size groups
and the increase of codes, and within the groups, a li'st of standard sizes. In
addition, the ordinal numbers of the ca,rds in the pack are punched~ starting with
_ ~ ity.
2. Each pack is processed in the'input flow of programmed input, which generates
lines of the vocabula.ry~ computing successively the prod.uction codes by means of
, the given increase and forming complete texts from the recurrent part and standa,rd
sizes.
The formed recorciings axe brought out into the next set in the volume of externa,7. .
memoryandpr�inted out with decima.l. numbers which are read off from the atart of
the set
Cha,nges can b~ introduced in the set in the volume of external memory by means of
a vocabulary editing program. The pgr.o~am~CLSFRDP realizes such ~netions as
removal of recox~cLings according to their decimal numbers, insertion or removal of
discrete symbols of the textual paxt, replacement of some symbols by others~ cor-
rection of production codes and also the printing out of recordings with their
ordinal numbers. ~
98
- FOR OFFICIAL U5E ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPR~VED F~R RELEASE: 2007/02/09: CIA-RDP82-00850ROOQ400070024-4
FOR OFFICIAL USE ONLY
3. The adjusted and c:orr~cted vocabulary is put in order in accordance with the
growth of the production codes by means of a programmed system.
4. The program CLSFRPC for monitoring and printing out codes permits printirlg out
completely or pa.rtially the vocabulary, arganized as a successive set in cemplete
form or shortiened. (only the production code).
- 5. The IF~ISAM system program trans~ers the finished voca,bulary into a volume of
direct access, equipping it with successive index organiza,tion.
6. The voca,bulax correction program in a volume of direct access GL5FRDP. perndts
the removal, replacement or addition of recardings to the vocabulax~ organiz~d as
a successive-index set. In that ca,se a11 the correction is braught out to the
= printer.
7. The printing of the voca,bula.ry out in the form of a classifier with addition
of a voca.bulary of graup names of products is done according to the program
CLSFRPR. The printing-out intervals are given by an ord.er in the input flow.
The applica.tion of an improved technologi~,l proce~s to create large voca,bula.ries
will permit reducing the labor-intensiveness of work on the preparation of texts
- of the vocabulary and punching by three fourths and facilitate the selection of
optimal condi�tions for the correction and maxia,gement of vocabularies.
Address for inquiries: 660025, Krasnoyaxsk, pr Krasnoyaxskiy rabochiy~ 14~, Com-
puter Center, Krasnoyarsk Ma.in Territorial Administration, USSR G ossnab.
COPYRIGH'i': Tsentral'nyy nauchno-issledovatel'3kiy institut informatsii i tekhniko-
ekonomicheskikh issledovaniy po ma.terial'no-tekhnicheskomu snabzheniyu
Gossnaba. SsS~i, 1980.
2174
- cso: 1863/264 .
99
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00854R400404070024-4
F'UR UFFiC'141. UtiN. UN1.1'
uDC 68~.3.06:[657.432.5:658.7/.8.000.06]
STATE OF DEBTOR INDEF3TIDNESS OF REPUBLIC CONSUMERS TO ORGANIZATIONS OF -
AZERBAIJAN SSR GOSSNAB
Moscow MATERIAL'NO TEKHNICHESKCYE SNABZF~NIYE~ SERZYA 4: PRIME'NFNIYE MATEMATI-
CHESKIKH METODOV, VYCHISLITII,'NOY TII{HNIl{I I ORGTIICHIdII{I V MATERIAL'NO TII{HDTI-
- ~r~S'ICOM SNABZI~NII (NAUCHNO-TII{HNICHESKIY REFERATIVNYY SBORNIK) in Rusaian No l,
Jan 81 pp 2-3 .
[Article by F. A. Gasanov, section hea.d, Republic Computer Center., Azerbaijan SSR
Gossnab, and A. M. Guseynov, senior engineer~
[Text] The task "Recording of deb~cor indebtedness of enterprises of the republic
to organiza.tions of Azerba,ijan SSR Gossnab" is being solved on a"Minsk-32" com-
put~r in the Republic Computer Center of the Azerbai jan 5SR Gossnab. Ix~ the
machine report received twice a month da.ta are presented on the debtor indebted-
ness of enterprises of the republic as a whole to G ossnab~ including to organiza,-
tions subordina,te to it.
To effectively provide the management of Gossna.b nf the republic with brief infor-
ma,tion about tr~e debtor ird ebtedness of ministries (departments~ in 1980 the Repub-
lic Computer Center of the Azerba.ijan SSR Gossr.a.b developed and introduced a draft
of the task "Formulation of effective reference information on the state of debtor
'.ndebtedness of consumers of the republic to organiza,tions of the Azerbaijan SSR
G ossnab" with the use of minicomputers.
The input informa.~ion of the task is the ma.chine report "Report on the recording
of debtor indebtedness," obtained. on a"Minsk-32" computer. In addition~ the
directory of the ministries (departments) is used as norma.tive-reference informa.-
tion in solving the.task.,
The output informa.tion is "Inquiry about the state of debtor indebtedness of con-
sun?srs of the republic to organizations of the Azerbai jan SSR Gossna,b." The in-
quiry is printed on an alphanumeric printer and the conclusion of the inquirer on
a display screen.
- The task is solved by means of a program working in five modess data input, print
out of the inquiry on an alphanumeric printer, print out of the inquirer of minis-
tries on an alphaxiumeric printer an~ rea,dout of the inquirer on a displa~ screen.
Data are entered in the memory of a minicomputer from the display keyboard, and in
_ that ca.se programmed monitoring of the entered da.ta is accomplished (arithmetic
correlation between data of a line is used for monitoring).
100
FOR OF'FICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPR~VED F~R RELEASE: 2007/02/09: CIA-RDP82-00850ROOQ400070024-4
PUR OFFICIAL USF: ONI.I'
Data on the final line are entered last. Then the minicomputers vertic~lly monitor
all the introduced lines. In case of error the data are entered repeatedly. i~Then
ea.ch line is entered the correctness of the code set of the ministry is verified.
The intrcduction of the task provides the management of Gossnab of the rc~public
with aggr~gated.informa.tion on the state of debtor indebtedness on the whole to
G ossnab~ including to organizations subordinate to it, and pravides the possibility
of adopting effective mea,aures for its regulation.
C~YRIGI~P: Tsentral'n3ry nauchno-issledovatel'skiy institut informatsii i tekhniko-
ekonomicheskikh issledovaniy po material'no-tekhnicheskomu sna.bzheniyu
Gossnaba. SSSR, 1981.
~ ~7~ .
cso: 1863/z64
ioi .
_ FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R004400070024-4
i~lllt c11~b'1('l11 l'`h' t111 \
t1DC 658.3.o12s681.3.06
AUTOMATION (7F PROCE5S QF PERSONNII, ADMINISTRATION
Moscow MATERIAL'NO-TII{HNICHESKOYE 5NABZI~NIYF, SERIYA 4s PRINIQiENIYE MATEMATI-
CHESKIKH 1KET(7DOV, VYCHISLITEL'NOY TII{HNIKI I ORGTII{HNIKI V MATERIAL'NO-TII{HI~R-
CHESKOM SNABZI~NII (NAUCI~IIdO TEKHNICHESKIY REFERATIVNYY SBORNIl~) in Russian No l~
Jan 81 pp 4-~
[Article by I. A. Yashayev, group head, and D. 3. Petrosyan~ ~enior ma.thematician,
Republic Computer Center, Azerbaijan SSR G ossnabj
[Text] In the Republic Computer Center of the AzerYa.i jan SSR Gossn,3.b a complex of
tasks "Automa,ted record-keeping and selection of personnel in the G ossnab of the
republic (the ma.in territorial ad,ministra,tion)" ha,s been developed.
' The complex af tasks is being planned on the basis of the YeS-1022 computer with
the use of the algorithmic languages PL~1 and COBO~L.
The complex of tasks is intended for the centralized collection of information
about workers of the territorial organ at the computer center, the forrmation and
issuance upon xequest of inquiries on poaitions and work~rs satisfying definite
requirements= the conducting of an analysis of the organiza,tion and position atruc-
_ ture of the a.dministra,tion and the conducting of an analysis of the composition and
movement of personnel.
- The analysis is made on the ba.sis of features determined in a survey.
The ana.lysis of tasks is planned with consideration of the following requirementss
standardiza,tion of the report doczunents on personne~i the cr8ation of a personnel
informa,tian retrieval system; the applica,tion of union-wrlde classifiers in the
formation of a voca,bulary fundi standardiza.tion of machine reports formed as a
resul~t of solution of the task.
'1'o solve the complex of task report machine-oriented documents are introduced~
"Worker's report ca.rd~" "Form for making cha,nges in worker's reported character-
istics" and "Form for chaxige in workers' positions~" as the available report '
documents are not adapted to computer processing. In the development of the in-
formation retrieva.l syatem proviaion ha,s been ma,de for the possibility of ad~ust-
ing to the varying structure of the machine-oriented started documents.
To realize the algori~hm for obtaining the outpu~ machine reports, provision has
been ma,de for the grouping of all the starting information into three data basesi
. 102
FOR OFFICIAL USE ONLY ~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4
FOR OFFICIAL USE ONLY
, i
the ba.sic:, iiicludl~i6 thc~ organization and position structure of the administratioii
and information about individual workers on the of~`i.cial list; the "Archive," con-
taining the necsssary informa.tion about workers excluded ~om the official list
(in that ca.se the "Archive" is subdivided into short-term and long-term) and in-
quiries necessa.ry to obta,in the output machine reports (continuous and one-time
inquiries).
The content of specific output ma,chine reports is determined by the content of the
corresponding inquiries. The autput ma,chine reports are used by personnel sub-
divisions for the coirrpila.tion of sta,tistica,l. reports and inquiries about personnel,
and also the analysis and a.dopting of administrative decisions on personnel ques-
tions. The principles laid down in the planning of the complex of tasks permits
using it also in other territorial organs of the USSR Gossnab.
COPYRIGH'i's Tsentral'nyy na.uchno-issle~lovatel'skiy institut info~matsii i tekhniko-
ekonomicheskikh issledovaniy po ma,terial'no-tekhnicheskomu snabzheniyu
G ossnaba SSSR, 1981.
2174
cso: i~63/z64
103
FOR OFFIC[AL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-40854R000400070424-4
FUR OFFICI:~1. l'SE UNLl'
unc 68i.3:658.153.o1z.7
ANALYSIS OF STATE STANDARDIZED ~GJORKING CAPITAL AND CURRENT ASSETS
Moscow MI~TERIAL'NO TEKHNICHESKOYE SNABZI~NIYE~ SERIYA 4s PRIMENEriI~'E MATEMATI-
CHESKIKH MFTODOV, VYCHISLITEL'NOY TII{HNIl{I I ORGTII{HNII{I V MATE~IAL'NA T'II{HNI-
CHESKOM SNABZHENZI (NAUCHNO-TII{Hr1ICHE~{IY REFERATNNYY SBORNII{~ ~.n Russian No l,
Jan 81 pp 16-17
[Article by V. I. Smelyanskiy~ senior engineer, Main Computer Centery USSR G ossnab]
[Text] The task of ana.lysis of standardized workin~ ca.pital an~ standa,rdized cur-
rent assets has been developed and introduced into the Ma.in Computer Center of the
USSR Gossna,b. The task consists in the composition of a complex for analysis of
the indica.tors of financial activity of the subsystem of planning and financial
activity of the USSR G ossnab ASU. Data are processed a.nd the output documents are
issued on a Ye5-10~V0 computer.
The ma,in software of the ta.sk is a data ba,se which assures the input, storage~ re-
newal, checking and correction of dictionaries and files.
Tasks of analysis are solved on the basis of accounting re~ords, which are entered
- quarterly in the dQ.ta ba.se of the subsystem of planning and financial activity of
the USSR Gossnab ASU.
T he ma.in indicators according to which the standardized working ca.pital is analyzed
are: equipment (Soviet and imported)~ materials structures and low-valued objects,
accounts receivable, uneredited assets and above-quota uncredited assets.
A questionnaire is prepared with consideration of deviations of ~he actual ba,lances
from the established standard of self-owned assets.
The result of solution of t:~e problem is formula,ted in a summary repoz�i on USSR
G ossnab organizations in a cross-section of the main territorial adminis+ratior.+~,
the soyuzglavsnabsbyts and soyuzglavkomplekts, the gossnabs of union republics and
other organiza,tions. The report is prepared from the results for a quaxter~ a
half-year, 9 months and a year.
Introduction of the task permitted providin~ workers of the planning and financial
administration with reliable and timely informa,tion about the financial situa,tion
of the branch and made it possible to free workers from uncreative maxiual data
processing operations and use the freed time for more profound analysis of process-
es of ma.terial and technical supply.
104
FOR OFFIC[AL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPR~VED F~R RELEASE: 2007/02/09: CIA-RDP82-00850ROOQ400070024-4
FOR OFFICIAL U~E ONLY
Address for inquirles: Moscm~r, Proa~shlennyy prayemd, d 3, I~fain Ca~puter Center,
USSR Go~saab.
C~YRIGRTs Tsentral'nyy nauc~o-issledavat~l'siciy institut inf~~atsii i tekt~ik0~
_ eko~canicheskikh issledavaniy po material'no-tel~Yna3cY~esgo~ai snab~eniyu
G ossnaba SSSR, 1981. ~ ~
2174
CSOs 1863/264 ~
io5
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R040404070424-4
. FOR OFFICIAL USF ONLY
- unc 681. ~8s [658.7/.8.oi2s 66z.6]
DEVEI,OPl~NT AND INTRODUCTION OF C~ETEI,Y AUTOMATED 5YSTII~I FOR COZZECTING .AND
~ PROCE5SING DATA ~N NATIONAL ECONOMY'S FUEZ SUPPZY ~
- Moscow MATERIAL'NO TIICHNICHESKOYE SNABZI~NIYE, SERIYA 4: PRIMENENIYE MATEMATI-
CHESKIKH METODOV, ti'YCHISLITEL'NOY TEKHNII{I I ORGTII{HI~1IKI V MATERIAL'NO TII{HNI-
CHESKOM SNABZHEDiI~ (NAUCHI~TO TII{HI~TICHESKIY REFERATIVNYY SBORNIK) in Rusaian No 4,
Apr 81 pp 5-6
[Article by I. N. Kryuchkov, group head, Main Computer Center, USSR Gossna,b~
[Text] Work is being done in the U5SR Gossnab or~ the creation of a single state-
wide automa,ted information system for fuel supply. In 1979 the first lin~ of the
system was~turned over to industrial operation by the Ma,in Computer Center of the
USSR Gossnab. The "Fuel" Informa,tion System ("~opliva" IS~ is oriented towa,rd the
accomplishment of flznetions of operative monitoring and analysis of the state of
~.iel deliver3.es to the na.tion.31 economy and effective traaking of ftitel stocks in
industry. in it provision ha.s been made for automat~~. processing of in~ut and out-
- put data~ the preparation of controlling documents for dectsion-making and an in-
teractive mode of man-computer communica.tion.
In the structure of th.e "Fuel" Information aystem th~ leading pla,ce has been allot-
ted to a dis+.ributed data basc which performs ~.inctions of data, accumulatio:,, storm
~,e, renewal and retrieval necessaryr for the solution of tasks in directory servic-
ing of the administrative appa.r~,tus~ prediction of ftiiel supply proceases for many
da,ys, operative a.nalysis and regulatiun of the flinctions of suppliers in assur~n~
the flil.filment of f1xe1 delivery ~lans. Realiza,tion of conceptions of the data barik
under the conditions af a sfst~m f~r ~ontrol of the flzel supply of the nationa,l
economy proved gossible only in conditions of the distributi~n of both the infor-
ma,tion itself and ~f its processing on hierarchic levels of a,dministration.
- In the "F~zel" Informa.tion System practica,l verification of the principle of dis-
_ tributed da:ta processing has proceeded on the level of the control apparatus of the
USSR Gossna.b. At the Central Information Point of the USSR Gossna,b N1a,in Computer
Center, on the basis of minicouiputers a computer group ha,s been formed which is
intended for performaxice of op~ra,tions of collection of primary information in the
system. Questionnaires are issued to the administrative appaxatus by the computer
- group loca,ted. in the sectior~ for monitoring flzel deliveries to the nationa,l econ-
omy. The interconnected work of those groups assured the development and intro-
duction of 13 tasks of the first lino of the "Fuel" Informa,tion Sy~tem~ The compu~
ter group network alrea.dy embraces Soyuzglavugol', Soyuzglavneft' and RSFSR
106
FOR OFFICIAL USE QNLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407142/09: CIA-RDP82-00854R000440070024-4
FOR OFFICIAL US~: ONLY
Goskomr.efteprodukt. Also pla,nned is the organiza,tion of automa.ted col.lection of
~ operative da.ta over communication channels from oil supplier organiza,tions of the
~xnion republics and uglesbyts~ and exchange of informa.tion between ma,chines with
the fuel and energy administrations. Introduction of the sacond line of the "Ptizel''
Informa,tion System is to be accomplished in 1980-1983 with an a.nnual sa.ving of at
least 100,000.rubles.
Address for inquiries: Moscow, Orlikov per. 5, Ma,in Computer Center, USSR Gossnab.
CQPYRIGIi~s Tsentral'nyy nauchno-issledovatel'skiy institut informatsii i tekhniko-
ekonomicheskikh issledovaniy po ma,teria.l.'no-tekhnicheskomu snabzheniyu
G ossnaba SSSR, 1981.
2174~
cso: 1863/z64~
~07 .
. FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R004400070024-4
FOR OFFICIAL USE ONLY
- unc 681.32.06:[658.155.o1z.7s658.7/.8]
ANALYSIS OF TRADE TURNOVER~ INCOMES, DI5TRIBUTION COSTS AND PROFITS OF USSR GOSSNAB
- ORGANIZATIONS
MoscAw MATERIAL'NO TECHNICHESKOYE SNABZI~NIYE, SERIYA 4s PRIMENENIYE MATEMATI-
CHESKIKH METODOV, VYCHISLITEL'NOY TII{HNIKI I ORGTII{HI1IKI V MATERIAL'NO-TII{HNI-
_ CHESKOM SNABZHENII (NAUCHNO-TEKHNICHESKIY REFERATIVNYY SBORNIK) in Russian No 4,
Apr 81 pp 13-14~
~ [A~rticle by V. I. Smelyanskiy, group head, Ma,in Computer Center, USSR Gossnab]
[Text] A ta.sk of ana,lysis of trade turnover~ incomes, distribution costs and
profits of USSR Gossna,b organiza.tions ha.s been developed and introduced in the Main
Computer Center of the USSH Gossnab within the framework of the subsystem of plan-
ning and financial act~.vity of the USSR G ossnab ASU. The task is included in the ~
complex of tasks in analyzing indica.tors of financial activity. The computations
~ are ma.de on a YeS-1040 computer.
The starting information for solving analytical tasks is taken from the data. ba,se
of the subsystem of planning and financial activity of the USSR Gossna,b. The da,ta
base, in turn, is formed from quarterly accounting reports.
The ma,in indica.tors of trade turnover in the system of computor ca,lculations azes
wholesale sa.les (total); sa,les of goods ex warehouse; sa,les of goods through .
storeU; sa,Ies of goc~d.s directly from prcxiucer tA consumer with and wri.thout parti-
~ipation in the computations; release of g~cxis writhin the system ex warehouse and
direct frorn producer t~ consumer with participation in the computations; sales and
release of materials~ work and services.
- The ar.alysis is made on the ba,sis of the following income indica,torss from sales
_ of goods ex warehouse, through stores, directly from prod.ucer to co~lsumer with and
without participation in calcula,tions; incomes f~om release of goods through the
system; income~ from other sales; total income.
The data of tasks of analysis of trade turnover and incomes of USSR Gossnab organi-
zations are used in solving the task of analyzing costs and profit.
The ma,in indica,tors of costs and prof`lt are: expenges of warehouses, ba,ses and
stores; transportation expenses; expenditures on delivery and shipment of goods;
warehouse expenses~ non-productive expenses; expenditures of organizations subor-
dinate to the USSR Gossnab; profit from sa,les.
108
- FOR OF~'ICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407/02/09: CIA-RDP82-00850R000400470024-4
FOR OFFICIAL USE ONLY
,
- The result of solving the task is formula,ted in a awnmary~ report an costs and
profit for organiza.tions of the USSR Gossnab in a cross-section of the ma,in terri-
torial a.dministrations, soyuzgla,vsna,bsbyts, soyuzgla,vkomplekts and gossnabs of the
union republics and other organiza,tions. The report is issued 4 times a_yea~.on
the results for. a qua,rter, a half-year, 9 months and a yeax.
Introduction of the ta,sk under consideration has permitted greatly reducing la.bor
expenses and impraving the quality of analysis of indicators of trade~turnovers~
distributian costs, incomes.and profit.
Address for inquiriess Moscow, Prnmyshlennyy proyezd 3~ Main Computer Center, USSR
~ o5snab.
' COP:IRIGI~i Tsentral'nyy nauchno-issledovatel'skiy institut informa.tsii i tekhniko-
ekonomicheskikh issledovaniy po material'no-tektinicheskomu snaY~zhen.iyu
G ossnaba SSSR, 1981. ~
2174 .
_ csos 1863/z64
~
- 109 . ~
~OR ~FFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R004400070024-4
~ FOR OFFICIAL USE ONLY
~ unc: .65s.oiz.oll.~:656.2
AUTOMATID SY5TEM FOR OPERATIVE CONTROL OF OPERATIONAL WORK OF A RAILROAD (ASUDO-D~
Moscow ZHEI,EZNODOROZHNYY TRANSPORT, SERIYA VYCHISISTEL'NAYA TII{HNTKA= NAUCHNO-TII{H-
1JICHESKIY REFERATIVNYY SBORNII{ in Russian No 2, Jul-Dec 80 (signed to press 10 Sep
80) PP 1-32
[Booklet entitled "Avtomatizirovannaya sistema, operativnogo upravleniya, eksplua,~
tatsionnoy rabotoy dorogi (ASUDQ-D)" (Automa,t~ed System for Opera.tive Control of
. Operational Work of a Railroa.d. ~ASUDO-D])] by A. L. Pisa,rev, chi.ef of the PKTB
ASUZHI'i Candidate of Technical Sciences R. A. Sakov, division head.; G. L. Shele-
nyakov, division head; D. A. Sosnov, A. F. Feoktistov, A. S. Kuz'min, and Ye. I.
Skla,d.chikov, chief project engineers] .
[Exr.erpts] Ed.ict of the Mi,nis~ryr. o~' Ra3.3.roads No 30Ts dated 10 Ma,y 1978, entitled
"On improvement of the technology of the transportation process and increase of the
efficiency and quality of work of the railroads" anel "Complex program for the
development and increase of the efficiency of au~toma.ted control systems in railroad
transportation; provision is made for the creation in 1980-1982 in the main direc-
- tions (west-east, above all) of automated systems for operative control of the
operational work of railroa.ds--systems n.gw in principle, based on informa,tion :
arriving from pri.mary sources~ which permit rea.l.izing not only a.referral mode but
also the solution of problems in planning and predi eting work indica,tors, the op-
1iza,tion of var3ous processes and interconnection with the ASU of line enter-
prises and the Ministry ot' Railroads. As a result of pre-planning developments by
the Planning, Design and� Technologica,l Bureau (PKTB~ of ASUZht, the VNI:LZht (All-
- Union Red Banner af Labor Scientific Research.Institute of Railroad Transport~,the
- computer centers of the Belorusskaya,, Oktya'~r'skaya~ Yugo-Zapadnaya, Yuzhno-Ural'-
skaya, Sverdlovskaya, Gor'kovskaya and Vostochno-Sibirskaya railroa.d.s, the tech- .
nical task for such an ASUDO-I1 sy~stem ha.s been develo~ed and approved. tiy the Min-
istry of Railroads (No a-2657, dated 23 January 19'!9 .
In the process of pre-planning investi~ation a ba.sic list of functiona,l problems
to be solved was determined, and also a list of the starting d,a,ta for their reali-
za,tion, the information was systema,tized and informa,tional messages wera ma.de up
assure the solution of 90 percent of all the tasks of the systemi missing 3nforma.-
tion is entered by means of special messages. The main distinctive feature of all
the messages is tha,t they were developed on the ba.s3s of unified requirements for
their structure in order to assure genera,lity of the technology of work with them
on all levels and processing in computer centers by means of unified standard pro-
gra,ms. In the creation of the system the requirement of a single input of infor-
mation was observed.
~ 110
F(~R OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407142/09: CIA-RDP82-00854R000440070024-4
FOR OFFICIAL USF, ONLY
The system being develvped consists of two parts. The general system paxt of
ASUDO-D envisages the collection of necessa.ry informa.tion, its input into the com-
puter, monitoring and correction, the forma.tion of general and specialized data
bank files, and the formation and issuance of results to users by stancl~.rd means.
The practical part of the system implements procedures fvr the solution of func~
- tiona,l problems; in tha,t ca.se it is assumed tha,t a11 the information (both vaxi-
able and normative-reference) is present in the computer in a form prepaxed for the
given problem.
Division of the system into the general syste~a and practica,l parts per~mits greatly
x~ducing labor expenditures on pro~rram development, as input-output procedures are
- worked up once (analysis shows tha,t the working up of those procedures for an
individua.l problem occupies.40-45 percent of the total la,bor expenditures) and
maIces possible parallel elaboration of both parts, which lea.ds to a curtailment of
. the total times required for creation of the system.
The actual system for control of the operational work on the ra,ilroad level is
limited in its possibilities by facilities for the manual collection and processing
of prima.ry informa,tion, which do not provide the urgency, reliability and cotap~ete-
- ness of da,ta on the entire variety of opera.tiona,l events on the monitored sections
of ra.ilroa.ds which are necessa.ry for the development o~ both parts.
The limitations are reduced ba,sically to the following:
--the periodicity of the obtaining of basic data for control of the transpcrt pro-
cess (freight car pools, rolling stock transmission over link-up points, loading,
unloading, tra3:n position) is limited in operationa.l reporting to days and in dis-
patcher reports and summaries to half-days;
--foreca,sting of ~zture work and possible difficulties, especiaLly for 2-3 or more
- days, in the manual variant is practically impossible due to complexity in the col-
iection of the information needed for tha,t and the need to use as a basis in cal-
culations a laxge number of mutually influencing factors in short period.s;
--the impossibility of accomplishing in the ead.sting control system effective mon- .
- itoring of the state and disloca.tions of ind3.~rldual units of the rolling stock
_ (refrigerator units, self-contained refrigerator ca.rs. transporters, etc);
--the presence~ during tk:e monitoring of observance of the established technology
and standards of operational work~ of la,bor-intensive work with manual selection
of individua.l daia, which on the wrhole does not give a complete characteriza.tion
of the situa,tion. This applies, for example, to monitoring violations of the plan
for formation and shipping by directions~ the provision of shunting yaxds with com-
plete lists by telegram of a11 trains.arriving for break-up~ to monitoring of their
quality, etc.
The ma.in purpose of the development and introduction of the ASUDO-D is the elimin-
ation of those limitations; the assurance on tha.t basis of the necessaxy effective-
ness and completeness of data cha.racterizing the operational situation on the rail-
road; reduction of the time required for estimating the production situa,tions by
operations workers and help in decision making by them; reduction fo the time for
bringing the results of a decision to executives; efficient and operative monitor-
ing of execution.
111 ~
FOR OF~ICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-04850R000400070024-4
FOR OFFICIAL USE ONLY
With consideration of distinctive fea.tures of the transportation process the
ASUDO-D musts
-~process and present to managers of vaxious ranks a laxge qua,.ntity of accounting
and report data as compared with the standaxd and planned tasks=
--organize d.a,ta chaxacterizing the general operating situation on la.rge sections
with consideration of the interconnection and dynamic interaction of controlled
- and monitored objects;
--as~ure the necessary effectiveness and depth of primary information, the forecast-
ing of expected results of work for the purpose of timely predic~ion of possible
failures and tracking disloca.tions of separate ca.tegories of trains and types of
rolling sta~c.
T o bring the system close to local conditions and assure infb~-mation interaction
of organs of c~ntrol on various levels, provisions axe ma.de, on the one hand, for
a differentiated approach to the selection of tasks, the form and content of the
prima,ry documents for each railroa.d by starting fr~m its distinetive features and,
on the other ha.nd~ standardiza.tion of the ba,sic documents used for the excha.~ge of
3ata between subc]ivisions which are participating in the proce~s of decision.
Those requirements determined the set of standard complexes of tasks which must be
solved on the common information base with the use of the unified ~eneral system
facilities ASUDO-D. The list of standard complexes~ the sequence of their devel-
opment and organiza.tion and the developers were determined by ~he coordina.i;ion plan
o.f development of the system approved by the Ministry of Railroads.
All complexes of tasks are interconnected technologically and with respect to in-
fo.rma,tion through the general system part of the ASUDO-D--a unified complex of
hardware and a set of s�tanclaxd program and informa,tion modules which accomplish
the collection of information and ma.intenance of a da,ta b~.nk on the transportation
process.
.
T'Ze system opens up fairly broa,d. possibilit~ies for improvement of the control of
~.ailroad operations, permitting railroad and section managemer:t to obtain a com-
plete idea of the operating situa,tion on monitored sections at moments close to
_ real time. To describe tha,t situa,tion to the user~ tYie following analytica,l data
are presenteds
--shift of the f`reight ca,r flows--total exchaxige of the railroad and sections,
transfer of trains and freight caxs over specific link-up points with the neces-
sa,ry breakdown of the pool by structure and dest~.nations;
--the presence, disloca.tion and state of locomotives of the oper~,ting pool by types
of traction and series on the sections and railroa,d, including turnaround sec-
tions,~addition depots and monitored stations;
~ --the presence of locomotive cre~as for freight movement on turnaround sections and
- turnaround points;
- --loading and unloa,ding for the ra.ilr.aad~ sections, monitored stations and enter-
prises by types of freight and. freight caxs;
On border railroa,ds, in addit~.on, similar informa.tion is issued about freight
cars with export, import or transit foreign trade freight~ and stations near ports
and border:stations are included in the monitored ob3ects.
112 �
FOR OFFICIAL LISE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R004400070024-4
~FOR (1FF1('IAI. t'SF: ONt.ti'
- The possibilities of forecasting and operative planning of flzture work in the very
near f~.iture are being especially expanaed. Here one can distinguish:
--fore~a.sting tra.in situations and operational evenis--the feeding of trains and
locomotives to monitored railroad stat~ons, train fortnation an.d readiness to
ship from shunting yards trains with the attachment o~ 7.ocomotives to them to
_ organize train work on ra,ilroa,d sections;
--foreca.sting the arrival of goods at monitored unloading and reload.ing points to
orga.nize rhythmicity ~f the shipping work of ports, laxge freight statio~s and
consignees;
--foreca,sting fulfil;nent of the ma.in indica.tors of operating work--loading, unload-
ing, transfer of trains and fre~.ght caxs, the locomotive requirements, the trans-
portation of local freight for various periods as the bas3s for daily-shi.ft plan-
ning of the railroa,d.' s work;
--issuance of detailed informa.tion to operations workers about the train situatic~n
on rail sections for 3-6-hour periods~ including tnformation about as~umed diffi-
culties in operational work and recommenda.tions on their elimination.
The system provides nonitoring of the disloca,tion .and statP of ind.ividual. categor-
ies of trains and separate units of rolling stock and coi:tainers, detez~mi.ning the
location of ea~h monit.ored unit4 its state~ the type of ~eight bei.ng transported,
the time of shipment and arrival and the tra.cing of monitorEd. points, the ~xecution
_ of freight and technical operations and quallta~ive~ inc933.cators of use (the time of
finding en route and during loading opera.tions, the route speedl.
A number of pra.etica.l tasks of the system permit, monitoring the observance of tech-
nologica,l discipline and adopting effective measures to eliminate faund violations~
contributing to normal conditions of oper~.tional works
--execution of the plan far the formation flf stations;
--assurance of the formation of fl~ll-weight and complete trains;
--observance of the established procedure for direction of trair.s in a circuit;
--performance of regulatory tasks with eli~r.inatioii of counterflows of freight caxs
of the same kind,
The ASUDO-D assures the issuance to aperations workers of stations, sections and
rail administrat~ons of a number of technologica.l documents for each train. A11
results of computations are combined by the referral system, which accomplishes the
necessa.ry communica.tions with the user. Its work consists in prepaxation fbr the
- railroaci management of the ba,sic services (movement, locomotive, fre~.ght, etc) and
the railroad sections of the corresponding information, reports, farecasts, etc.
- In the first stage within~the fYamework of the referral system summaxy r~ports are
prepared in regulated conciitions on movements and transport work, analytica.l rPp~rts
for ra.il administrations and sections and report~ for transmission to the Main Com-
puter Center of the Mini:stry of Railroa.ds. In the ser;ond stage an interactive mode
of fluictioning of the referral system is rea,lized, wj th the possibility of issuin,~
the results of solu~ions of practical tasks of the ASUDO-D on displa,ys at the
request of railroa.d workers.
113
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-40854R000400070424-4
FOR OFFICIAL USE ONLY
The ASUDO-D Haxdware Complex
The flow of da,~ta on reception and tr^~snussion in +he ASUDO-D system amounts to:
from ~,000 to 1 million signs~day at the stations; 3�103 signs~da in the TCh
[expansion unknown~ and up to 4�105 in the NOD [expansion unknown~; up to ~�105
signs,aay in mutual exchange b~tween railroad computer centers. The total input
into a ra3lroad cor~puter cent~r per day in messages amounts to: flill-scale sheet-
- telegrams, up to 3�106 signs; information about train movements for s~lect~d sta--
= tions, up to 6�105 signs; inf~rma,tion from the NOD~ up to 10~10~ signs; informa,tion
- about locomotives, up to 5~10 signs. Information is transmitted from line sub-
~iivisions over communica,tion cha,nnels (Fig 3) directly to the computer. For tha.t
purpose in the first sta.ge of development of the system data tranemission multi-
plexors (N~PD) are used in the following manner:
. --the N~'D-1 fGr the co~lection and transmission of informa,tion ovsr telegra.ph com-
munication channels and organization of referral mode through AP-64. When appro-
- priate modifica.tions are ma,de, telegraphic commutat~,ble communica,tion channels can
be connectea to the MPD-lA;
--the MPD-1 ~YeS-840.1) for the colle~tion and transmission of data over telegraph
separat�ed a.nd commutable communication channelst organiza.tion of a referral ma3e
through the AP--62 and AP-64~, connection with the Yew-1010 computer over separate
. telephor..e communica,tion channels in an asyrrchronous mode in accordance with
exhange algoxithm AP-70 (AP-1);
_ --the :_~L-1 (YeS-84~10) for reception and transmission of dat~, over telegraph and
telephone corurutable and sepaxated communlca,ti~n channels with use of subscriber
points FKG-T50, TAP-2, TAP-3, AP-62, A-64~, etc.
- 1 2
Figure 3. Structural diagram of the ASUDO-D
communication system
~ eu � arc ara
1 Automa,ti~ telephone office
- ~ 2 Automa,tic telephone exchanga
(,qeq H0,[( QBU cdu 3-- Railroa,d compu~cer cent e.r
_ ~ I 4~ Station computer center
i 1 S-- Ma.in computer cent~r
~d'U ATC 5 Users
- 7-- Stations with aver~ge processing volume
- 9 8-- Sm,,ll stations, locomotive depots
- H~Q aa ao,q 9-- Railroad section chief
~
CBq `
~ ~
~
/ I ~ ~
- ATC /~ony~o a- ATA
menu
- Cm~yqr~u c~ Maneie ~
~ op6eeMOr I ~no
aM muB ~
= Le~padamn�u ~bie ver,o
~ ~._----1
~ 11~.
' ~OR OFFICIAL ~JSE OPdLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R004400070024-4
FOR OFFIC7AL USE ONLY
~ m~ b)
.s .s .
- M ' y~( . H H
2 Z
y9Hl~Q yydM,Q y9Nl~q ~ ByNM~
~ -
fc- ~ozz 4 ~c- iarz E~ -~~s 4 Ec-so~s
(fC-IOJJ) - RKK - . ~'fC-f0 (EC-1QB0J . ~X (fC-1060J
~ ~6 6 ~
y9Np9/1~ 9yNM/I S?9NM0 y~INM//
HM 5 N N HM
Ml1,Q MnQ NJHO
nPoM~~~
oP
III ~II
~
' I(OMM~/IlIQ7770p S~NJ(L A~OMCH//16l C![C/IlC O
MN ~CCI71 b .~~1 y~
/1 ON~N/7/M C/(CTCHb/ ~
a Figure 4. Struct~re of the ASUDO D computer complex.
a- First stage 5-� ~agnetic ~ape store
b- Secorid stage 6- Magnetic tape sto~re comtrol device�
- 1- Magnetic disk stare 7- Comau~nicatiaa~ commutatorr
. 2-.Magnetic disk store ca~trol device 8- Communicatia~ processor (SM-2~
3- EC = YeS 9- System subscribers
4- Cha,nnel-to-channel adapter 10 - System subscr~bers (computer netWark)
11 - Data. transmission multiplexos
_ Ynformation transmissi~ in the first stage betKeen railroad computer ceaters in
_ the presence of variaus data transmissian ~ltiplexors can be done aver duplex
(frn.ir-wire) telegraph courmunicati~ channels at 50 bits~s o~r Hith the use of an
intermediate machine carrler--punched ta.pe or magnetic tape. Subsequently the.
' �railroad computer centers must be equipped frith data transmissiom multiplexors YeS-
~ 8410 of the s~.~s. type and cmmmunicatiaai proc:essors.
� For Kark of the system in the first sta~e it is suffici.ent to have a tKO-machine
complPx based on the YeS-1022 (YeS-1033) computer, constructed on the level of the
magnetic disk or magnetic tape store, crr a"channel-to-channel" adapter (or a data
transmis~on multipleaor to issue co~tr~l si~nals). Thp main memary of each com=~
puter must be at least ~7.2 Kbytes, and the externa~. memory 6 YeS-5052 m~gnetic ta.pe
- stores or 4 YeS-5061 ma~netic disk stores. Figure 4a gresents a structural diagram
of an ASUDO D co~puter complex of the first stage.
>~7
The system must be developed thraugh increase of the computatioma7.: capadtg of the
complex by a transitiam to nex YeS computers of model YeS-104~ (YeS-1060) or
115
= FOR OFF[CiAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R004400070024-4
MUR OFF7CIA[. USE ONLY
replacement of low-ca,pa,city processors by more powerflxl ones and enlaxgement of the
ma.in and external memories. The next sta,~e in the development of the haxdware com-
_ plex and organiza.tion of exchange betwesn ma,chines is the introduction of comimini-
as.tion processors based on the SM-2 mini-computer (Figure 4b presents a structural
diagra.m of the computer complex of tha.t stage) and message commutators. The use of
communica.tion procdssors will permit creating a uniform computer network and red.i~-
tributing its resources.
Ths prepaxa.t,ion and trar.smission of informa.+.,ion from line subdivisi~rs in the first
~ stage must be done with the use of existing haxdwa,re (a T-63 teletype and Akkord-
1200, TA-o00, TAP-2 and TAP-3 subscriber points in a punched tape-punched tape
(or ma~etic tape-ma.~netic tape) mode. In propor-tion to the equipment of the ra,il-
road computer center with YeS-&I-10 multiplexors and communica,tion processors the
transition must be ma,de to direct data input; in the first stage over telegraph
communication cha,nnels (YeS-84~01 and Ye5-8410 MPD~ and la.ger over telEVhone chan-
nels (YeS-8410 MPD and a communica,tions processor .
For the organiza.tion of an intera,ctive working mode it is advisable to use AP-62
and AP-64~ subscriber points and . YeS-7920 group informa.tion display equipment.
In each specific ca.se equipment must be selected by starting from the specific con-
ditions. Thus, in the presence of communica,tion processors based on the SM-2 for �
the organization of interaction it is advisa,ble to use SM video terminal subcom-
plexeG. At large f`reight stations and shunting yards the collection, preparation~
, processing and transmission of da.ta must be done on the be,sis of mini-computers of
the type of YeS-1010~ YeS-1011 and SM-2. At stations which reclassify less than
30 t.rains per d.ay in the first sta,ge of introducti~n of the system T-63 teletypes
, with FKG-T50 equipment and TAP-2 and TAP-3 subscriber points will be used; in the
_ following sta.ge of development of the system at those stations for automation of
- the collection, preparation and transrnission of information it is necessaxy to
introduce intellectual termina,ls and micro-computers of the types "Iskra-126",
"Iskra-226", SM-1800~ TAP ~{h, etc. Depending on the volumes of transmitted infor-
mation, the depots, sectional stations and link-up points are equipped with tele-
types ancl TAP-2 or TAP-3 subscriber points.
CUPYRIGHTs T sent~�al'nyy nauchno-issledovatel'skiy institut informa.tsii~ tekhniko-
ekonomicheski.kh issledowaniy i propagandy zheleznodorozhnogo trans-
- parta (TsNITTEI MPS)~ 1980
_ 2174~
CSO: 1863/261
116
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R004400070024-4
FOR OFF'IC'IA1. 115F: ONI.I'
GOR'KOVSKIY RAILROAD COMPUtER (~1TER FUI~FTLIS PLAN
Moscow ZI~EZNO~OROZHNYY TR~SPORT. SERIYA VYCHISI~ITEL'NAYA TFXFII~TIKAs NAUCHI~O TFXH-
~~~g{Ty ~TIVNyy ggpgNII~ in ~tussian No l, Jan Dec 81 (signed to press~ 3 Feb
81~ pp 1-2
[Article by Ya. M. Lembrikov, deputy ch3.ef, computer center of the Gor'kovskaya
Railroad~
[Text~ Stand3.ng up in honor of the coming 26th CPSU Congress~ the collective of
the computer cent~r of the Gor'kovskaya.Railroad has assumed enhanced socialist ob-
liga,tians in which are envisaged ~Zrther increase in the efficiency of use of com-
puter capa.c3.ties and, in particular~ of the YeS-1o33 computer, the use of TAP-3S
equipment for transmission of informa.tion over communication channels from magnetic
ta.pe, accelera.tion of the start-up of the Yudino Automated Shunting Yard Coa~trol
- System, improvement of orga.nization of the comp~uta.tional process aud a flxrther
sa,ving of punched ta.pe, paper, red tape and other materials.
The a.dministration and the party~ trade-union and Komsomol organizations in the
course of a socialist competiti.on have devoted seriaus attention to increasin8 the
reliability and the ra.tes of organization of third-generation computers~ to e~cpan-
sion of the use of inedium-speed data transmission equiPment and to inyrease of the
rates of develapments and improvement of their quality.
For 9 months of 1980 the plan tasks on problems to be soZved and the voZumes of
infc~rma.~ion to be processed ha.ve been over-fulfilledr and la.bor Pr�ductivity ha.s
a,mounted to 106 percent of the planned. The loading ~f "[1rar~-14" com~pute~ hae
amounted to 20.2 hours and of "YeS-1.o33" computers to 15.8 haurs per daY
average and ha.s exceeded the planned; the cost per hour eT ent f~rtthe "Y sbio33"�
reduced by 23 percent for "[lran-14 ma.chines and by 39 P f~ the
In 9 months 0.4 percent of the planned Wage flind was saved. The program.
development and introduction of tasks Has flil.filled.
Since September 1980 the wo~3sers of the traffic and locomative service ~nd~the man-
- agement of sections and depots have have obtained by computer data on dislocationa
and monitoring the sta.te of main-line diesel engines servicing the.Cherusti-
Druzhinino section. In that task~ as in some others, use is made of apparatus
transcription of the results of Woric of an integrated aomplex for processing the
machinist routes on "Ural-14" ar~d "Ye9-1o33" computers.
� The YeS-1010 computer put in operation at the G orkiy shunting yaxcl and. the Zyanga-
sovo station to automate operations in processing train informatia~ in real ti.me
117
FOR OFFICIAL USE ~!?NLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-04850R000400070024-4
FOR OFFICIAL USE ONLY
are being successfully ogerated. At the present time those ma,chines axe connected
by a direct telegraph cha.nnel over which flil.l-sca,le sheet telegra.ms axe exchanged
between the ma,chines automa,tically. A similar system is also being prepaxed for
� introduction at the Yudino station.
In 9 months of 1980 the computer center personnel introd.uced 16 ra.tiona.lizers'
suggestions which permitted obtaining a sa.ving of 7200 ruhles. Considerah~le work
was done on the creation of a system for technical servicing of the Ye5-~o33 com-
puter~ based on the use of non-autonomous tests, which permits accomplishing main-
tenance, the diagnosis of faults of the input-output equipment, checking the work
of equipment after it is installed or modernizing it under the control of the Ye5
OS operating system.
The personnel of the computer center are applying every effort to honorably ~lfill
_ the socialist pled~es in honor of the coming 26th CPSU Congress.
COPYRIGPT: Tsentral'nyy na.uchno-issledovatel'skiy institut informatsii, tekhniko-
ekonomicheskikh issledovaniy i propagandy zheleznodorozhnogo tr~.nsporta
(TsrTIITEI I~S), 19~1
2174
cso: 1863/z61
118
FOR OFF[CiAL USE ONLY ~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407/42/09: CIA-RDP82-40850R000400470024-4
FOR OFF'ICIAL USF ONI.Y
- ~ ~
vnc 658.oiz.ou.56:656.z
IMPRpVEI~N'P OF PLANNING TECHNOLOGY AND INTRODIICTION OF AUT~IATED CONTROL
S�STEMS FOR RAILROAD TRANSPORTATI(7N
Moscow ZI~LEZNODOROZHNYY TRANSPORT. SERIYA VYC}QSLITEL'NAYA TII~HI~TIICAs IICSPRESS-
INFORMATSIYA in Russian No 2~ May-Aug 81 (signed to press 29 Jun 81~ pp 1-56
- [Booklet entitled "Sovershenstvov~niye tekhnologii proyektirovaniya. i vned~eniya
ASUZhfr" ("Improvement of the Technology of Planning and Introduction of ASUZhT"~ .
by A. P. Pisarev~ B. M. Filimonov~ N. Ye. Tarasov~ et al~
[Excerpts] ~
Softwaxe and Improvement of the Technology of Its Devslopment
pages 4-9
The main distinctive feature of third-generation computers is their assurance of
c3eveloped opera.ting systems which perform a large set of fluictions of automa.tion .
of the computational process and opti~mization of the use of computer resources,
and also the presence of translators from algorithmic languages.
Besides the YeS OS standa.rd software the system includes the following companents
developed by organizatirns of the USSR Ministry of Rai7~roads~(MPS)s
--a translator from the algorithlaic language SPCK -2 [1]=
~ --a universal set of procedures wh3.ch accomplish f~netions of the basic YeS
utilities,
--a program for correction of symbolic libraries with a Ye5-?906 display complex
and triggering of tasks for execution in a package mode; ~
--programs fer the processing of statistical infbr~,tion accumulated ~by a systems
monitor routine;
--a program for analysis of accaunting information of a JOB ca.rd; ~
--a program far separation of a systems conclusion, one which issues at the end of
a listing a line of sepaxating symbols and the naaae of the,uaer=
--ma,crodetermina.tion of t.he structural programming in t~he Assembler ].~ngixage.
Althaugh at the present time the YeS OS operating~system version 4.1~ third edition~
with an executive program ~n the MVT mode~ has been selected as the standard for
- NII'S subdivisions, it has turned out tha,t that system alrea~y does not satisfy the
119 .
FOR OFFICIAL USE ONLY . ~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-40854R000400070424-4
N'OR 4FFICIAL USF. ONLY
~ requirements of the Railroa,d Computer Center and machines of the second Ye5 series
(YeS-1035, YeS-1045, etc) . 7'he ASUZhT [Avtoma,tizirovannaya, sistema, upravleniya
zheleznodoroshnym transportom--Automated Control System for Railroa.d Transporta-
tion] PKTB [Proyektno-konstruktorskoye tekhnologi.cY~esY.oye byuro--Planning and besign
Technological Bureau~ considers it necessa.iy to accomplish a transition in the very
neax flxture to work with OS 6.1. 3~ whi ch in contrast . with OS 4.1. 3 preserit s the
users with completely new softwaxe which offers:
--the possibilit~ of time sha.ring, which permits crea,ting, translating, editing
and debugging prog-rams in an intera.ctive mode;
--the possibility of working in a multiprogram mode with a vaxiable number o~ tasks
jointly using virtua.l memory;
--means of regen~ration for the Ye5-1035 and YeS-1o33 computers;
--means of grouping YeS computers.
The YeS computer softwa,re also has availa,ble laxge sets of standard. means of real-
izing various flznctions. They include input-output generators, data ba,se control
systems and pacIcages of applied progra,ms to solve varic~us standard. problems.
The results of investigations conducted by the ASUZhT PKTB show that for various
_ reasons certain standard means caxinot be used ta so?v~ various ASUZHf problems.
For example, the "Q{A" 3a.ta ba,se control system can be used to process laxge data
files with hierarchic and network structures of-:data not requiri~g frequent renew-
al, and cannot be used for tasks of operational control, whereas the "INES-2" data
base control system can be used to cr~ate information and reference systems~ but
in tha,t ca,se it is required to additiona].ly develop file maintenance facilitiss.
Therefore together with the wide use of standard YeS computer softwaxe a need ha.s
arisen to develop our own means for the realiza.tion of various systems, Thus,
in the development of a system of opera.tional control of operating road work
(ASUDO-D) [2] a need was revealed for the additiona.l development of systems for
da,ta gathering and preliminaxy processing, the servicing of files, control of the
comput~,tional process, and special softwaxe for the descrip�tion of information and
ference systems (a data description language and a ca.lculation description lan-
- gua,ge) . In the process of development of an information and reference system for
the ASU of the Railroa,d Economy a system was created for the processing and issu-
ance of output forms. In that ca,se it should be noted that the modules develoged
for certain systems in a number of ca,ses are becoming universa.l and can be applied
in many developments. Thus the file maintenance system has alrea.dy found applica-
tion in the system of integrated processing of a road record. and in the DISKOR
road level system; it can also be used in many other prablems . The system for the
processing and issuance of output forms~ which ha,s been formulated as a package of
applied, programs and is used by many organizations of the MI'S and ~ other departments ,
as a means of the technologica,l engineer for obtainin~ successf~.il materials, has
also become widespread in use. Very great acknowledgment ha,s been obtained by the
system for data gathering and preliminaxy processing~ which has been officially
approved as the ba,sic system for remote data processing in the ASUZhT and is open
for further development.
The subsequent development of the ASUZhT within the framework of the organiza,tion
of the computer network requires the creation of da,ta exchange between ma,chines
120
FOR~ OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPR~VED F~R RELEASE: 2007/02/09: CIA-RDP82-00850ROOQ400070024-4
FOR OFFICIAL USF: ONLl'
for the purgose of assuring its efficiency and reliability for use on different
levels and the exclusion of inechanical data carriers~ which leads in the final
_ account to the combining into a single whole of all railroa.d computer centers of
the ra.ilroad network and to the creation of a so-c.~,lled distributed computer net-
work. This will permit not only solving the problem of data exchange but also
using available computer ca,pacities very completely and uniformly.
The problem of creation of such a network at the present time requires selection of
a universal and flexible algorithm for exchange between computer complexes on the
level of control of the channels for tranamission of data to the data transmission
system, for known algorithms for exchange bet*aeen ma,chines (exchange protocols~
developed within the framework of protocol X.25 upon the recommendations of the
International Consultative Committee for Telephony and Telegraphy such as ANSI,
~ SDLC, ~IDLC, etc, have neither softwaxe nor hardware ca,pable ~af realizing them. In
the PKTB ASUZhT a detailed analysis ha,s been ma,de of all existing exchange algor-
_ ithms capable of becomin~.universa.l and not limiting the development of the net-
work, as a result of which an exchange algorithm of the type of AP-70 has been
- selected, which ha.s been technica.lly realized in all types of YeS computer da,ta
_ transmission multiplexors and ha.s a number of advanta.ges important and necessaxy
for the organiza.tion of a computer network:
--the possibility of organizing conditions of "rivalry";
--the presence of program support in all OS versions under control of the V`rAM and
TSAM method.s of access;
--the use of asynchronous exchange cond.itions with the KOI -7 exchange code;and
rates of d.a,ta. exchange of 100-1200 bits~second over telephone and telegraph chan-
nels of communica.tion.
The adopted algoritnm for intermachine exchange of type AP-70 was realized in the
PKTB ASUZhT during organiza,tion of the interaction of the Ye5-1030 compu~er and the
MPD-lA ~Ye5-8400) with the Ye5-1010 computer [3]. Additionally for the realization
of exchange the ma.crocommands R~ADTV, READTB and WRITETL were developed; they have
been included in the BTAM pr.ogram resources. Distinctive features of the realiza-
tion of that algorithm for intermachine exrhange are:
--the introd.uction into the algorithm of priority timing for succe~sful expansion
of conflict situations during simultaneous initializa.tion of a call f`rom both
sides (3 s for the YeS-1030 and 20 s for the. YeS-1010~; .
--the establishment of communica.tion, accomplished by means of a set of service
synbols identical to the symbol "Who is there?", presented in the form of the
combination "NT-NT KP-KT" (NT--start of te~; KP--end of transmission; KT--end
of text~ in a~cordance with the KOI-7 cod.e table~;
--confirriation of reception, accomplished by the combinations "YES-YES~CP KT" or
"NO-NO KP~CT" (YES--symbol of a positive reply; NO--symbol of a negative reply~.
When a negative reply has been obtained, transmission of the da.ta block is rep~at-
ed three times;
- --conclusion of the communica.tion, accomplished by means of transmission of the
combination "YES-YES~{P~T" .
Results of the work show that the~~applica,tion of the given algorithm for inter-
ma,chine exchange is very realiza,ble in the ASUZYiP from both the ha,rd.ware and the
121
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPR~VED F~R RELEASE: 2007/02/09: CIA-RDP82-00850ROOQ400070024-4
FOR OFFICIAL USE ONLY
softwaxe points of view on its different levels--"Main Computer Center-Railroad
Computer Center", "Ra.ilroad Computer Cer_ter-Railroad Computer Center"~ Railroa,d
Computer Center-ASU SS", etc.
The creation of the computer network in the ~.iture will require dPtailed considera-
tion of problems in organization of the computational process,and protection a,gainst
unautk~orized access~ and also problems of the technology of system nlanning.
Contemporary requirements for software presuppose a high technica,l level of the
organization of its planning. On the agenda stands the question of transforming a
program into an article of the industrial type, satisfying certain standards, and
readily corrected and ma.intained. In tha,t direction the PKTB ASUZhT ha.s been doing
some work, including;
--study of the lead.ing technology of planning on the ba.sis of avai.lable translated
and 5oviet literature~ and also experience of other organizations;
--analysis of labor productivity during the performance of planning work by pro-
grammers, and the revelation of reserves and directions of applica,tion of the
leading technology;
_ --the development of planning standa,rds;
--teaching of advanced programming technology and the organiza,tion of planning;
--study of experience in the introduction of lea.d.ing methods of planning and the
development of corresponding methodica.l ma.terials for the Railroa.d Computer
Center.
One ~lement of advanced technology is the introduction of standards of mod.ular pro-
gramming. Such requirements for the construction of a progra,m complex as the pre-
- sence of a single executive routine with a common working field for all other pro-
grams, a standa.rd input and output for each module ,and limitation on the mod.ule
length are justified and are already used by the leading programmers. The follow-
ing stage was the introduction of structura,l programming. Modulax and structural
programs~are creating the basis for a new approach to the organization of planning
--so-called planning from the top down [4]. This method requires, primaxily on the
part of the rnana,ger_, through understanding of the problem and much preparatory work
- for direct programming and debugging (breakdown of the task into parts forming the
standard structures, the plan for testing and debugging, solution of the quest~.on
of the possible deviations from an ideal circuit, etc).
Planning from the top down must be introduced into both the hardware and software
sections. Substantial su~port for the introduction of that me-thod is the HIPO
technology, which gives rules for documentation of the program and assumes in tha,t
case use o.f the method of planning from the top down.
The enumerated method.s of programming and planning are not the only ones. At the
present time it is important'tm study already existing experience in their use and
to make it a~eneral achievement.
' Of no little importance in increasing the productivity of computers and developers
is the system of organization of .debugging wo~3c. Used as the basic system in the
PKT~ ASUZhT is the package me~~hod of processing ta,sks in a multiprogramming mode.
122
~ FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPR~VED F~R RELEASE: 2007/02/09: CIA-RDP82-00850ROOQ400070024-4
~ FOR ()FFICIAL LISE ONI.Y
Work hae. been done recently on the seaxch for improved program debugging modes.
The systems "Interactive remote introduction of ta.sks" a.nd 'Time shaxing system"
have been tested; "Correction of staxting texts and triggering of ta.sks for execu-
tion with the XeS-7906 display complex" has been operated experimentally. The
results of that work showed that at a computer installation with a small nwnber of
direct-access equipment and controls for them~ a small main memory of the computer,
small speed of the processor and, wha.t is the most important, when. a�ie. encounters
unstable work of the computer, the applica.tion of ineans of remote debugging does
not bring perceptible effect. ~
Of the above-enumerated systems, "Time sharing system" has proven to be the most
effective, but for its applica.tion it i.s necessa.ry that ~he computer have a main
memory of at least 1 Mbyte.
The ASUZhT soft-aare is subdivided into general-system and applied; the applied
sof~~ware is based c~n the genexal-system, which performs general fluictions--input,
ou~put, da.ta bank maintenance, intermachine informa.tion exchange, etc. The applied
performs fhnctions of ASUZhT ~nctional tasks. In the development of these two
parts of the softwaxe it is necessary to use improved methods of programming--the
HIPO technology and modulax and structural programm3.ng. Such an approa.ch to devel-
opmer~t will permit consideza.~ly reducing the time required for the development and
- introduction of a system, on the one hand, and will permit standardizing the compu-
tational process in railroad computer centers, on the other. An example of that is
the development of the ASUDO-D system.
ASUZhT Hardware and Distinctive Features of Its Planning
- Pages 28-32~ 3!~-39
The planning of the hardware of the first ASU line was done by each railroa,d indi-
vidually, as the har3ware products list wa.s limited, the systems developed on the
different railroads were poorly connected with respect to information, the ~Znc-
tioning of tasks was accomplished in a package mode, and the exchange of informa,-
tion between the ra.ilroad computer center and the information points, as a rule,
did not provide for remot~ processing.
The ASUZh[r second line required a different approa.ch to hardware planning, f~ the
ba.sis here wa.s problems to be solved in real time~ an extensive haxdware products
list requires the development of standaxd general-system softwaxe, and remate pro-
cessing is the basic regime of information exchange. The close informational inter-
connection between the procesGing complexes of different levels required. solution
of the problem of interma.chine excha,nge, and on the scale of the network, the cre-
ation of a branch computer network. Another important factor which influences the
approa.ch to haxdware planning is the requirement of maximum coaip-atib~lity of the
process of primary document preparation with the abtained information on ma.chine
carriers. ~
AI1 this taken together required a coiaplex approach to the pla~nning of the techni-
ca1 ba'se--.from the information saurces to the processing complexes. On the other.
hand, ta assure the effective use of haxdware it is necessaxy to take into account
distinctive features of ea.ch object, tha.t is~ to assume their definite individual-
ism! which can be solved by standardization of plaxuiing decisicros for d3fferent
123 .
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPR~VED F~R RELEASE: 2007/02/09: CIA-RDP82-00850ROOQ400070024-4
. FOR OF~'ICIAL 11SF: ONLY
~levels and as a fl~nction of the assumed volumes of information. Such a?i approa,ch
permits reducing tA a minimum labor expenditures in the planning of both hardware
and softwaxe and creating unity of technology and technical policy in the branch.
Soviet i.ndustry and the industry of the CEMA member-c~untrles ha.s organized and is
no~r issuing an enormous axsenal of computer harciwaxe, starting from powerflxl com-
puters of the Ye5-1060 type to micro-computers and information registers. One of
_ the main tasks in the construction nf an automated control system for railroa.d
transportation is growth of the ca,pa,city of resourcc:s for the automa,tion of the
gathering~ preparation, prima.ry processing and transmission of informa.tion with the
_ use of microprocessor technology. It will permit increasing the number of users of
the system, contructing a more reliable model of train position, maacima,lly automa.t-
ing data processing in the field, increase the efficiency of use of communica.tion
= chan~els and reduce their total number. One of the variants of the structural
- scheme of an A~UZhT ha.rdware complex is presented on Figure 2. It is assumed tha,t
the ASUZ~P will ha.ve a four-1eve1 system (the Main Computer Center of the Ministry
of Railroads~ the Railroad Computer Center, junction or station computer centers
- and the ASU of Llne Subdivisions, such as sectiona,l, intermediate and freight ~ta-
tiAns, link-up points~ locomoti.ve and freight cax depots~ etc~.
Hardware Complex of the Main Computer Center and the Railroad Computer Center
To solve the complex of problems determined by the technical ~,ssignment on the
second line of development of the ASUZhT it 3.s necessary to gather informa,tion fr.om
more than 300 subscribers. The volume of arriving informa,tion from a single sub-
scriber amounts to from a thousand (path distances~ structural controls, ete) to a
_ million signs per da,y (shunting sta.tions, large locomot~.ve depots, freight stations).
- Even today many railroad computer centers axe processing a flow of informa,tion tha,t
exceeds several million signs per day. By 1990 tha,t flow will amaunt to from 10 to
30 million signs per day, and some railr.oa.d computer centers will surpass tha,t
level. At such an input the computer ca,pacity for the reception~ process3:ng and
issuance of information must amount to from 0.6 to 2 million operations per second.
Input-output information must be produced over commutatable and separated tele~raph
~ i telephone channels of communica,tion with the use in the first sta,ge of an TTPD-1
da,ta transmission multiplexor (YeS-84~10) and a YeS-1022 (YeS-1033) camputer. The
informa.tion will be processed as a fluiction of its volume on a YeS-1035~ YeS-10~5 or
YeS-1060 computer. Fresented in Table 1 is the make-up of type YeS ~computers in
railroa,d computer centers as proposed by the P'KTB ASUZhfr as a flinetion of the vol-
umes of input informa,tion to be processed.
Table 1 Make-up of Computers in Railroa,d Computer Centers
Volume of input data, T}rne of YeS Computer
million signs per day 1022 1033 (1035) 1~4~5 1060
Upto10 2 z
= Up to 20 2 z
uP to 30 2 2
_ Over 30 2 2
_ 121~
- FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPR~VED F~R RELEASE: 2007/02/09: CIA-RDP82-00850ROOQ400070024-4
N'OR OMF'It'IAI. 1 :~I~: (1N1,1'
3 reu ~ .
rP~ rpz 3 r~~ fp1
~03 Reu2 aeu ~~3
rP3 ro3 ro3~
asu Qeu aeu
CBU ~V ~ 98 fpT
rpr rP2 rp~ rz
t'AABNbIN Obl'~, uE11TP O,OPO~(H~lM dbi~.yENT~
,
. eU asu 2 -
0
- ex fc.a~v ex ~raa+u 4
~ ~ tt~~ 5
.
: ;
; ~
0 t : ~
. u~'c ~n,~ aw-aan y~il - ~'C ~
c~u- trua c~ . cc~aa rc+~a
1
r~n~JC ~r~�~so ,404 2
au! ~ r,~ ~ rl,
~ c'ev.~te~.ra! ''�u 7
Figure 2. 5tructural dia~am r~f the ASUZYiP hardxare co,plex.
1-- Main Computer Center 9-- System subscribers
2-- Railroad computer center 10 Data transmission multiplexor
3 Gr-~. 7.1 Modem
4-- Computer compl.ex of YeS computers 12 Shunting statiou with capacity
5-- EC = YeS of 50 trains per day .
6-- Station computer center 13 Link-up point
7-- Junction computer center 14 Freight stations ~
8-- SM = Sma11 Computer 15 Separate railro~ds
_ 125
- FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPR~VED F~R RELEASE: 2007/02/09: CIA-RDP82-00850ROOQ400070024-4
i~~llt c11~ 1~ Ic'1 ~11. 1~~I? ~?N1.~
. _
~9~oeo~ cT~u~ouu~N yyaCINC~~A ct~uUkp
~bly. uEbTp bbly. I~EHTP ~~'~~~�a'�'"" z
7 6 ~ . T.~
~ BL~ ~ 8u 2 c,~~cn .
. 7 snrt fc eaas ,~.x er*
e~ Ecia~ ~ar ~c~aqo
~ fk~21['~Fd Cll~? /7 b Ca11 y?~CT~ Alll~l ~t~w~i
PN n~
.M~n.b � .
a~r~uuva,a~,a~9 ~aw aa~? CTb10lOdOM . IIYHNT
~ ~ 13 ar� ~uny
a Nme~ cucme,uer a a a~~uar ~~~ciati PN
BT - /5 - 20 BT - e - >6 . M ~ ~x ~f
~ T 63 - 20 - 40 T-6J -/0 - 20
Pl: - 5 - /5 P.H - S - !0 ~~~QyA
/R~.91A!/� 6 - /2 /A~SM~O?- 4 � - 6 !A,lM/~
fPy306~E CTAN Ni4 oTaEAEtu~E 101o~N (NOa~tS ~
~ ~ 8 71fn�3C 'lt~l
~ .w, . . ~p .
. ~
BT ~ ,d,9iFA~
- fCQG105 ~ ,nn.,nwr
wec~r wiw~r. ' Q7'*A
_ ~r airr~nrr? .
N ~ T-B~
AOKOMOTNBHbIE N BAfO l~4ACTKOdbIE N' fIP0~1. ~Pl~08i~ CTAN~�~"N
~ Nl(~ dEIlO,DTK,~MC, ~~iquu~ ~ ~'u' a18' .
1 ~ ~ 7 r-e~ ~2
TAO;~C ~
N ~ � pivsae~
~
� MAAbIE ~TAUi(~1
_ . ON T�63
Figure 2. (Continued) .
16 Locomotive and freight ca,r depots 20 Punched tape
17 Sectional and industrial stations 21 Communicazion cha;nnel
18 Capaci~.y of Z5 freight ca,rs per day 22 Good.s office
19 Sma,ll stations 23 Data register
126
FOR OFFIC[AL USE ONLY .
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02109: CIA-RDP82-00850R400440070024-4
N'(1R OI~FI('IAI. lltil~; nNl.l'
The total memory ca.pa.city for a.railroa.d computer center must a.mount to 200-400
Mbytes, of which the capa.city of the store~ of airect access (the magnetic disks)
' is 100-300 Mbytes; the capa.city of the main memury is 3-6 Mbytes. Virtual memory
must be ~ased in the real time mode. The computers axe grauped on the basis o~'~
common magnetic disk store and "channel-channel" aci,apters. The interactive x~rking
mod.e is constructed on the ba.sis of AP-62 and -AP64 ~ sat~b~ points and YeS-7920.11
graup information displays. The following stage in the develapment of the Y~axdware
co~�plex and the organiza.tion of the ~Lini~stry of Railroads computer netwcrYis is the
introduction of connected processors and message commuta,tors based on SM-2 mini-
computers.
Junction Computer Genter and Station Computer Center Hardware Complexes
The main requirements presented for haxdware complexes for station and junction
computer centers are:
--the possibility of work in real time and a developed system of remo{ce processing;
--the possibility of performing ~the fluictions of monitoring and control of techno-
logica.l processes;
--the possibility of increasing computational ca.pacity.
These requirements are satisfied very completely by the Soviet SM-2 compuier and by
~he YeS-1010 and YeS-1011 of Hungarian origin.
- In the Eleventh Five-Year Pla.n it is proposed to conduct an experimental opera.tion
of the SM-2 computer in the ASU SS system. A two-machine speciali~ed co~trol com-
puter complex (SUBK No 17 and 18j ha,s been ordered for that purpose; it has the
following parameters:
--a main memory capa.city of 2j6 x 2 Kbytes;
--a magnetic disk storage capacity of 60 x 2 Mbytes;
--a magnetic ta.pe storage ca.pacity of 30 x 2 Mbytes;
--up to 48 connected commutatable telegraph channels;
--up to 22 sets of video terminal systems equipped with DZM-180 alphanumeric print-
ers or magnetic tape cassette stores, ~ri.th a~Yj~~m distancs between the pair
of telephones of 16 km;
--up to 4 sets of video terminal systems connected at a distance of over 16 km:
All the comiaunica.tiion modules are connected to th~ computer through a multiplex
interface splitter, working in automatic and semi-automa.tic conditions of switch-
ing communica.tion cha.nnels from one computer to another, and the computers them-
- selves are grouped into a single computer complex on the level of the processors.
The productivity of the computer complex based on the SM-2 is increased by increa~-
ing the number of connected. computers or prucessors. The YeS-1011 computer is a .
la.ter model of the YeS-1010 computer; it has a greater speed~ an exp~nded memory
(to 1 Mbyte) and six permanent magnetic disk sto~res with a ca.pacity of 2.5 Mbytes
each; in addition, the YeS-1011 is capable of grouping tHO computers by organizing
a common memory field on the ba.sis of the main memory and magnetic disk stores.
127
- FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-40854R000400070424-4
FOR nFFIC1.41, t'fil�: nNi l' ~
Users of.junetion or station computer centers more than 16 km distant (for the
SM-2) or 5 km clistant (for the YeS-1011) must be connected with the use of micro-
computers or intellectua.l subscriber points of ~he type of TAP~Ch or SM-180n.
Ha.rdware Complex of Line Subdivisions
Of the entire set of micro-comF,uters, subscriber points Knd data registers devel-
oped and organized by industry for purposes of the automation of the gathering~
prepaxa,tion, primaxy �processing.and transmission of informa.tion the ~requirements
of the Ministry of Railroads axe sa,tisfied most conapletely in the present sta,ge
by the micro-computers SM~1S00, TAP Kh and "Isk.r_a,-5~" (or "Iskra-555~~)~ the data
registers R17501b, Ri7801, R164~2, R161h01 and RI2~01 and the subscriber points
TAP-ZS~ F3{G-1001, ~cG T5o and T-63. This haxd~rare can be installed at sectional, ~
freight and intermediate st2.tions, etc. The quantity of Ysardware installed and
_ its structure depend on the volume of work of the given subdivision. When the
volumes of informa.tion to be processed are considerable, arrangement of the ha,rd-
ware with sepaxation of the contro7. ~.inctions is possible--thus, for exa.mple, for
a sectiona,l station with a large volume of ~reight work it is a,dvisa,ble to intro-
duce an ASU for a good.s office, an ASU for a technical office and to install a
mod.el SM-1800 central mic.ro-computer into which the resulting information arrives
= from the t~o A~U's and which accomplishes flanetions of communica,tion with the
station~ junction and railroa.d computer centers. Subscribers of sectional,
f~eight,and intermediate s~;ations, link-up points and others axe equipped with
video terminal systems, data registers and tel~~ypes-and are connected ta that
micro-cor,tputer over physica.l lines. Additionally connected by an object communi-
catic;n device are systems for infarmation readaut from rolling stock,and axl.e.
counters, and the ASU of the shunting station is also connected; it ha,s gravity
yard automa,tic centraliza.tion with monitoring of breakup~ similax to wha,t h~,s been
done at the Shkirutava station, for automa,tic removal of information about the
_ actual breakup of a train. Several variants of hardware complexes.are su~gested
for the creation of an ASU of lineax subdivisionss
a) installation of a YeS-1800 coir~puter in groupings of three types: .
-two SM-1800 computers connected by a connection module (US SM), each of which ha,s
a ma,in memory with a ca.pacity of 6~I- Kbytes, two floppy disk stores (NG~'ID), three
v3deo terminals (VTA), three model D3M-180 alphanumeric printers (ATal'[l~9 two
modules for connection with a modem (MSM) and five modules for connection with a
telet;~pe (MST) .
--one SM-18U0 computer with a main memory ca,pacity af 64~Kbytes, one NGNID, 2 MSM,
~.~P4ST, two VTA, one DZM-180 and four object communica,tion devices (USO)~
--two SM-1800 computers~ each of which ha.s a main memory ca.pa,city of 64~ Kbytes, one
MSM~ one NGNm~ one VTA, one DZM-180 and 8 USO;
b) installation of a TAP Kh, which is an intellectua,l terminal constructed on the
basis of a microprocessor and ha,s a ma.in memory,with a capacity of 32 Kbytes~
- two NGNID, one VTA and 1 MSM;
c) installation of RI2401, R164~a1, R16?~02, R17501 and R17801, the principal tech-
nical chaxa,cteristics of which are presented in Table 2.
128
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-04850R000400070024-4
~ FOR OFFICIAI. USE ONLY
Table 2 Principal Charac~eristics of Data Registers
Chara.cteristic Data Register Model _ �
2401 6401 6402 7501 7801
Data inputs
from keyboard num num -al~num al~num al~num
from punched caxds yes yes yes yes no
Indica,tion 16-digit 16-digit - num CRT
Printer no MP-16 '~{onsul" DARO '7{ansul"
al~num
Transmission speed over
communica.tions channel,
sigYis~s 200 200 200 1200- 1200-
960o z~wo
Magnetic tape ca.ssette store no no no yes yes
Control compu~er computer computer program program
~Computing and Data Processing Computer Network .
_ T.he computing and data processing computer network in railroa,d transportation must
be crea.ted on three main -levels--the flinctional, commutation of inessages and com-
mutation of channels. In the given stage of the creation of an ASUZhT the last-
mentioned is the most developed.
The crea.ti rn of a computer network in the ~7.1 +rolume of its flanctioning is an ex-
tremely difficult and costly task, but it is greatly simplified if the network is
created with only the fluictions most necessary for the Ministry of Railroads~ that
- is, with well-developed. informational and less developed computatioma,l fluictions.
This network must be created simultaneously also as a depa,rtmental autonomous net-
work and as a component part of the statewide automated system for collection and
processing of informa.tion for accounting, plaiuLtng and cantrol in the na.tional
economy with consideration of a11 the effective normative documents~ including the
international (documents of the GEMA~ the International Consultative Committee of
- Telegraphy and Telephony, etc).
Such a network must ha.ve the following properties:
--computers in the network correspond automatically with one another;
--the work of the network is assured by a distributed aperating system which
belongs to a family of network opera.ting systems~ being an expa,nsion of local
opera.ting systems without change of the 3atter. This permi.ts using one and the
same program interfaces of tasks in woxking in net~rork conditions and during
- local work;
--regulation of the axchitecture and characteristics of the netxork is establishPd
by a hierarcY~y of protocols xhich deteraiine the procedures of computer interactian
129 .
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407142/09: CIA-RDP82-00854R000440070024-4
MUR ON'FI(7A1. USF ON[,Y
on vaxious levels~ starting with the physical interface of~communica,tion cha.n-
nels and ending with the level of information exchange between remote informa-
tion processes and remote tasks;
_ --the main paxt of the network resources (processor, communica.tion, etc) is ex~
_ pended An the accomplishment of dispatcheriza,tion, routing of inessages, diagnosis,
ne~work measurement~ adaptation to vaxying conditions of network flinctioning, etc.
The generation of that part of the resources ~about .~O,.~ercent) in the ASUZt~ ca.n
be possible only when relatively inexpensive computer hardwaxe (mini- and micro-
computers) is used.
The SM-2 grouping recommended by the PKTB ASUZhfr (specialized control computer
complexes Nos 17 and 18) is designed for work under network conditions~ and so
even at the present time it is possible to create experimental computer networks
based on unifica,t3on of the ASU SS on a certain polygon of the railroa,d network.
T he created experimental computer network in the genera:l ca,se can be cla.ssified
according to three groups of signs--the structural, ~Znctiona,l and purpose. The
main structural signs of a network axe the topology~ the method of dispatcheriza.-
tion of work to the network computer center (centralized, decentralized or com-
bined)~ and the make-up of the computers (uniform or non-uniform). Developments
made in the PKTB ASUZtiP ha,ve shown tha.t an experimental network created dn the
basis of SM-2 computers must have a radial-circulax topology with a centralized
type of dispatcherization, which greatly simplifies conducting the experiment.
In the first stage it is advisa,ble to include only SM-2 computers (but not more ~
than 30 groups), as the realiza.tion of communica.tion and interaction between the
processes of processing in a remote computer is very complex far non-opera,ting
works, that is, networks consisting of computers with different structural and
functiona,l chaxacteristics (word length, operating systems~ data organization,
i.nterruptions, etc), and is simplified for networks constructed on the basis of
computers of the same type. ~
~-�~ording to their ~.inctional purpose computer networks are divided into the infor-
~r~,tional, computational and computing and data processing. A computer network com-
bining station computer centers (ASU SS) ought to prima,rily ha,ve an informationa,l

- character. It is created to obtain the possibility of more effieient movement of
information for great distances by means of a complex system of exchange protocols,
which musts
--determine the physical connections with communica,tion channe~s;
--include agreements relating to the semantic content of the control informationi
- --describe algorithms for exchange between various elements of the network=
--assure the detection and correction of errors;
--describe methods of routing messages;
--control buffer pools and message lines;
--assure reliability of informa.tion exchange.
. 130
. FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R004400070024-4
~oR 4~~ic~Aa. us~ oN~.~~
Protocols must be quasi-independent, tha,t is, each of them must regula.te independ-
ent exchange between network elements only on its awn level, but at the sa.me time
- the realization of the protocol of ea.ch level must be based on protocols of lower
levels. The lower levels of protocols, ca.lled the physical interface, must in
accordance with the effective standards characterizing the physical connection of
the comnunication channels~ correspond to the requirements of the International
Consultative Committee of Telegraphy and Telephony.
The communica.tions equipment which assures.a. low level of protocols in the SM-2
include the following data transmission equipment:
--a duplex register--a mod.ule which assures parallel da.ta excha,nge between com-
_ puters up to SO m apart;
--an intrasystem communications module~ which assures consistent da.ta excY~ange by
ca.ble at a distance of up to 3 km; �
--an APD-MA.[data transmission device]--appa.ratus which assures communica.tion over
standard telephone communication channe].s;
--an APD-NR'P--apparatus for multipoint connection of equipment or computers at a
- distance of up to 16 km apart;
--a mod.ule for connection with telegra,ph channels;
--sets of coupling modules with C1~ C2~ C3 and C4 cauplings.
In creating a Ministry of Railroads computer network it is necessary to combirne
computers of three different cla.sses--maxi-, mini- and micro- into a single system
which performs a11 the set of algorithmic acts necessazy within the fra.mework of
- the ASUZYir .
We classify as "maxi-computers" all YeS computers (except YeS-1010 and YeS-1011);
as "mini-computers", YeS-1010 and YeS-1011 computers and all SM computers except
- SM-1M and SM-1800. For the class "micro-camputers" in the Ministry of Railroads
network~ RPT~ SM-1800, SM-1M, vTS-561oo and TAP~{h ha.ve been specified.
- ~ Mzxi-computers are used in power~zl fluictional systems of the ASUDO-D type as
information processors workin~; under the control of a communicatian processor. In
the railroa.d computer center several information processors are ba,sic, and the r~st
duplica.te them. The principe.l questions for the given class of co~puter axe ques-
t ions of~the linkage and use of a local operating system by creating flxnctianal
- superstructures (monitors).
- Mini-computers axe used in medium-sized flinctional systems (of the ASU SS type)
as information processors working under the control of the railroad computer cen-
- ter and communica.tion processors, and also for ~the creation of inessage commuta.tion
centers. It is most a.dvisable to rea].ize the communication processers on two-pro-
cessor computers of the SM-2 type.
It is a.dvisa.ble to use micro-computers ir~ sma]1 systems (subsystems) as communic~.-
ti on or terminal processors.
In the creation of an experimental computer network it is necessary to take into
consideration tha.t at the present time in our country a single methodolog,y for
131
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R004400070024-4
FOR OFF'I('IAi, US1~: ON1.1'
the planning of computer networks ha.s not yet been created, and so a methodology
of planning the ASUZh(r computer network represents a fairly laxge and independent
problem.
BIBLIOGRAPHY
1. VYCFIISLITEL'NAYA TFKHNIKA (T sNIITEI MPS), 1978, No 2(4~6).
2. VYCHISLITEL'NAYA TII{HNIKA (TsrTIITEI MPS), 1980, 5cientific-Technical Coliection
_ of Abstracts, No 2.
3. Idem, No 1.
4. VYCFIISLITEL'NAYA TEKHNIKA (TstTIITEI MPS), 1981, Scientific Technical Collection
of Abstracts, No 1.
COPYRIGHr: Tsentral'nyy na,uchno-issledovatel'skiy institut informatsii, tekhniko-
ekonomicheskikh issledovaniy i propa,gandy zheleznod.orozhnogo trans-
porta (TsNII'~EI MPS)~ 1981
2174
- cso: 1863/260
132
FOR OFFIC[AL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407/02/09: CIA-RDP82-00850R000400470024-4
FOR OFFICIAL USE ONLY
PERSONALITIES
COMPUTING MATHEMATICS, CYBERNETICS FACULTY DEAN TIKI~ONOV'S WpRK HONORED
Moscow VESTNIK MOSKOVSKOGO UNIVERSITETA: VYCHISLITEL'NAYA MATEMATIKA I KIBERNETIKA
in Russian No 3, Jul-Sep 81 pp 3-8
[Article: Celebration of the 75th Birthday of A.N. Tikhonov"]
[Text] One of the greatest scientists of our time, Academtcian Andrey Nikolayevich
Tikhonov, was born on 30 October 1906 in the city of Gzhatsk in Smolenskaya Oblast
(now the city of Gagarin).
- ln 1922 Andrey Tikhonov finished secondary school as an external student and
entered the mathematics department of the physics and mathematics faculty o~
Moscow University and in 1927 he completed the university and entered as a gradu-
ate student the Moscow University Institute of Mathematics.
A.N. Tikhonov has been a professor at Moscow University since 1936 and in 1937 he
was elected a corresponding member of the USSR Academy of Sciences. In 1953 A.N.
Tikhonov was given the title of Hero of Socialist Labor and was awarded the first-
degree USSR State Prize. In 1966 A.N. ~ikhonov was elected a full member of the
USSR Academy o� Sciences. He was award2d the Lenin Prize in the same year. In
1976 A.N. Tikhonov became the winner of still another USSR State Pr�ize.
At the present time Academician A.N. Tikhonov is the dean of the faculty of com-
puting mathematics and cybernetics at Moscow University and heads the department
of computing mathematics in this faculty. The faculty was created in 1970 with
the very active and decisive participation of Andrey Nikolayevich. Andrey Niko-
layevich has been the unquestioned director of the faculty since the instant of its
creation. The development of the faculty of nomputing mathematics and cybernetics
over the last 1J years and the great achievements of the faculty's team in the area
of training a broad range of specialists and of constantly improving the efficiency
of scientific research have been constantly associated with the work of Andrey
Nikolayevich, who has directed the team's efforts at the solution of timely prob-
lems.
At the same time A.N. Tikhonov is the director o~ the USSR Academy of Sciences
Institute of App7.ied Mathematics imeni M.V. Keldysh.
Andrey Nikolayevich was the initiator o� the creation in universities of the Soviet
- Union of faculties o~ computing and applied mathematics, where specialiats are
trained not only in the area of studying mathematical models of various natural
FOR OFFICI~CL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-04850R000400470024-4
FOR OFFICIAL U~E ONLY
science prob~ems, but also in the area o~ camputing co~pl.ex software, automated
- control and planning systems, and systetqs ~or pxocessing the xesults o~ complicated
scientific experiments.
~ x
. ~ ~w~,~
. A ':A. ~
r Y :
4n
~i qr~'y
} ~9
i y
k 4~~.
> ~'.e
I
- ~ iti. J ' ;
p ~ :1~ ~
_ ~
-
~ ~ ~~~~;j+ll I
, ;
a,
, -d '
Andrey Nikolayevich Tikhonov
In beginning to characterize the creative scienti~ic work of A.N. Tikhonov we
immedi.at.ely note the quite extensive r~,~nge of his scientific achievements. The
fundamental pioneering results in many areas o~ modern mathematics and its appli-
catians (topology and functional analysis, theory o~ differential equations,

~ 7-3~.
FOR OFFIC[AL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407/42/09: CIA-RDP82-00854R000440070024-4
~'OR OFFICIAL USF O1VLY
computing mathematics and mathematical physics, the const~uction and study o~
mathematical models of variaus natural science pxoblems) are those of A.N.
Tikhonov. A.N. Tikhonov has made an enormous contribution to the creation of new
scientific trends such as methods o~ solving incorrect problems, the automation
of scientific research and the development of sybtems for the automated processing
of the results of complicated scientific experiments. To tttis must be added the
fact that the creative scientific work of A.N. Tikhonov represents a distinct and
rather rare example o~ a combination o~ first-class achievements in the most ab-
stract areas of "pure" mathematics and the profound investtgation of appli+ed prob-
lems directly related to the requirements of practical work.
The earliest period of the creative scienti.fic work of A.N. Tikhonov was devoted
to topology and itinctional analysis. As an 18-year-old youth Andrey Tikhonov ob-
tained his own first results--a proof of the fact that any regular topological
space with a denumerable base is a normal space (and therefore also metrizable).
T~aQ years after this the young scientist published the results which brought him
wor~dwide fame and which placed him among the outstanding topologists of our time.
_ First Andrey Nikolayevich formulated a definition of the topological product of
- any set of bicompact spaces. In mathematics finding the correct definition rather
often plays a decisive role in the entire theory. Tikhonov's definition of. the
product of topological spaces plays just such a decisive role in topology.
On the basis of this definition A.N. Tikhonov proved that the product in his sense
of any set of bicompact topological spaces is also a bicompact topalogical space.
"Tikhonov topology," based on this definition, has firmly entered the arsenal of
fundamental concepts of modern mathematics. The topological work of Andrey Niko-
layevich is the foundation for all modern topology, for the theory of topological
groups, for the theory of dynamic programming and for a number of divisions of
functional analysis.
Then A.N. Tikhonov worked on the general theory of differential equations and on
_ studying problems of geophysics and electrodynamics which are important froffi the
application standpoint.
~ Prominent Topologist A.N. Tikhonov became a world-famous specialist in the general
theory of differential equations and in topical problems of geophysics and electro-
- dynamics. .
The first studies of A.N. Tikhonov on the theory o~' partial differential equations
were the result of his studying a number of topical problems in geophysics, in
particular, the problem of reconstructing the historical climate o~ the earth.
In studying these problems A.N. Tikhonov ~ormulated and solved a number of funda-
mental general mathematlcal problems.
A striking example of the solution of such a fundamental general mathematics problem
is A.N. Tikhonov's study (1935), which has now become a classic, in which he found
in a specific sense the definitive conditions guaranteeing ~he existence and unique-
ness of a solution to Cauchy's problem ~or the heat equation.
- These conditions, which consist oP the requirements for continuity o~ initial
- function u~(x) and o~ the existence of a limit
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407/42/09: CIA-RDP82-00854R000440070024-4
F'nR OFNIC'IAI. IISH: ONI.Y
� 1 im [uo (x) . e-~fxl~~ = p~
~z~~~
aze now customa,rily called Tikhonov~s conditions.
The ideas set down in this stud}r by A.N. Tikhonov were developed in subsequent
years in studies by a number of ~oviet and foreign mathematicians (0. Vider,
O.A. Ladyzhenskaya, O.A. Oleynik, I.M. Gel'fand and G.Ye. Shilov, etc.). Then ~
Andrey Nikolayevich formulated and studied the inverse heat problem. He proved
the fundamental theorem that the solution, u(x, t) , of the heat equation in region
x> 0, < t< t~ , is unambiguously determined from the assigned valup of
u(x, t~) = u~(x) , if only a single derivative of this solution is limited in terms
of coordinate.
Belonging to the same period oP work is A.N. Tikhonov's fundamental study on the
congruencz of regions for which the first boundary problem for the heat equation
and the Dirichlet problems for Laplace and Helmholtz equatiuns are solvable in the
classical sense. Having determined the fundamental region as that ~or which the
individual problem is solvable in the classical sense, A.N. Tikhonov proved the

_ following three statements: 1) any bounded region fundamental for the heat equa-
tion is a fundamental region also for the Laplace equation, 2) any region funda-
mental for the equation ~u - Au = 0 with some a> 0 is a fundamental region
for the equation ~u - Au = 0 with any 0, and 3) any region fundamental ~or
the equation ~u - au = 0 with any a> a~ is a~undamental region for the heat
equation, too.
. These studies by A.N. Tikhonov were continued in la.ter years ~y a number o~ mathe- ~
- maticians: In 1949 O.A. Oleynik and G. Tautts proved that the region fundamental
for the Laplace equation is also fundamental for a general second-order elliptic
function with sufficiently differentiable coefficients; in 1959 V.A. T1'in proved
- that a cylindrical region is the fundamental one for second-order hyperbolic and
p:~rabolic equations permitting the Fourier mer.hod any time when the cross section
ot this cylindrical region is the fundamental region for a Laplace equation.
These studies by A.N. Tikhonov were continued subsequently by V.R. Nosov, S.M.
Ponomarev, A.A. Novruzov, etc.
Affiliated with A.N. Tikhonov's cycle of studies on partial differential equations
is also his doctoral dissertation, defended in 1936, in which the concept of a
functional equation of the Volterra type was introduced and a study was made of the
conditions for the applicability of the method of successive approximations oP
Picard and of the Carchy-Lipschitz method of polygonal approximations for solving
this equation. A number o~ problems for the heat equation were discussed as appli-
cations, in parti~u~.ar the problem o.f the cooling o~ a solid with emission from
its sur~ace ~ollowing the Ste~an-Boltzuqann 1aw. These results of A.N. Tikhouov
were utilized considerably by V.G. Fesenkov in investigating the pr~perti~s of the
surface of the moon.
Then A.N. Tikhonov constructed the strict mathematical, theory o~ a thermocouple,
studied the in~luence o~ radioactive decay on the temperature o~ the earthFs crust,
136
FOR OFFICIAL USE ONLY ~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPROVED FOR RELEASE: 2407/42/09: CIA-RDP82-00854R000440070024-4
FnR OFFiC'IAL USE nN1,Y
~I~~v~+lu~iuil l.l?.~ lli~ury ul. ul~~�l ruuktbu~Clc tqeCltodB ut aluayL~~; llir. c:~{rile'~ ~rusl ~u~~l
mantle (and, in particulax, methods o~ prospecting) and provided ttteoretic~l
~ustification for using the natural e7.ectromagnetic field o~ the earth kor ob-
taining a complete electrical cross section oi' the earth's crust.
Let us emphasize thac in studying these problems a fundamental role was played by
the problem, developed by A.N. Tikhonov, of the stability of inverse problems.
For the purpose of proving the uniqueness of the range o~ inverse probiems con-
sidered, A.N. Tikhonov formulated and solved the problem of the possibility o~
determinins coefficient q(z) in equation u" -H ~q(z)u = 0 Wfith the condition
u(~) = 0 from the assigned function
f ~ _ -u -~0 ~ . - - .
u (0 , 1l)
He proved the uniqueness of the definition of this coefficient, q(z) , in a class
of com~;lex values of spectral parameter a under the assumption of the piecewise
analyticity of q(z) .
~ Now after the lapse of a number of decades it is impossible not to note that in
this cycle of studies by A.N. Tikhonov fundamental mathematical results were ob-
tained for the first time relating to the problem of reconstructing a linear dif-
ferential operator from the properties of its spectrum. These results preceded
the familiar studies af I.M. Gel'fand, M.G. Kreyn, B.M. Levitan and V.A. Marchenko.
~ A.N. Tikhonov completed a cycle of studies on electrodynamics in con~unction with
A.A. Samarskiy. Mention should be made especially of t~ro studies, in the first of
which the general method of constructing a Green's function for a sqstem of
Maxwell equations in a cylindrical region with a random cross section was validated,
and in the second the principle of a l.imiting peak value was formulated and stu-
died, i.e., the question of de~ining a unique solt~ti,on, u(x).,.to the Helmholtz
equation in an unbounded region as the limit as t-} ~ of the solution, u(x, t) ,
of the corresponding Cauchy problem for the wave equation.
These studies by A.N. Tikhonov and A.A. Samarskiy were continued in the studies of
A.G. Sveshnikov, O.A. Ladyzhenskaya, V.M. Babich and other mathematicians.
A.N. Tikhonov was also involved in studies of the mathematical theory of chemical
processes, in particular, studies completed in 1945-46 devoted to the dynamics of
sorption. The timeliness of' these studies consists in the fact that they camprise
the theoretical basis ~~r the design oP various decontamination ~acilities, whose
importance has been increasing steadily in connection with the problem of environ-
mental protection. The mathematical model o~ this rang~ o~ problems has resulted
- in nonlinear systems of parrial dif~erential equat~ons. Andrey Nikolayevich has
found precise ana~ytical solutions for same of these problems and for other prob-
lems which are r_ot sub~ect to an analytical solution he has used the ffiethod oP
combining numerical and as}rmptotic methods. At the present time this cycle of
studies is being continued success~ull}r bg A.V. Lukshin.
FOR OFFIf:IAL USE ONLY ~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400070024-4

APPR~VED F~R RELEASE: 2007/02/09: CIA-RDP82-00850ROOQ400070024-4
FY~a ~~FMirt,~~. ti~h: ntvt.~~
A.N. TikhQnov~s grea~ tu~x'it ~ies in the ~act ~h~t h~ wa~ the ~ix~t tQ ~ox'utiu~a~e
and begin ~o study in 1948 the question af tfie b@haViqx o~ sR7,u~i,an~ Q~ ~y~sxeaqs o~
ordinary di#~exent`_a1, equations wi~h a pertuxhati;on wi~h ~h~ highe~t derivative.
For sys~tems o# the type
_ _ _ _ .
dy` = ft ~t, y, z). (i = 1, 2, n),
dt .
' d~ = F j~t~ y~ z) ~l = 1, 2, m),
I '
in which u are perturbations satisfying the cond~tion u k u and are such
that as u~+ 0 there exists a limit to the ratio u ~_(u)~ul(u) ~ Andrey
Nikolayevich provided a general formulation for the Cauehy~ pr~blem and established .
= the criteria upon the ~ulfillment of which solutions of the initial system tend
- to the solution of the singular system as perturbations tend to zero. .
These studies by A.N. Tikhonov, which formed the bases of a scientific trend under
heated development at the present time--the theory of singular,perturbations--have
been continued by his students and other mathematicians, among whom should be men-
- tioned primarily A.B. Vasil'yeva and V.F. Butuzov, V.M. Voloso~r, Ye.g. Mishcheako
and N.Kh. Rozov, S.A. Lomov and many others. �
A.N. Tikhonov has also made an enormous contribution to the development of modern
computing mathematics and mathematical phy~.:ics. Finite difference methods repre-
sent some of the most effective methods of ;~olving boundary problems for complicated
systems of differential equations. A.N. Tikhonov in cooperation with A.A. Samarskiy
created the theory of homogeneous difference systems designed for solving classes
of problems definable by assigning only the type of differential equation and
- boundary coniitions. They also formulated and proved the principle.of the conser- .
vatism of random homogeneous difference systems as the necessary condition for con-
vergence of the system in a class of discontinuous coefficients. The theory of
h