JPRS ID: 8544 TRANSLATIONS ON USSR SCIENCES AND TECHNOLOGY PHYSICAL SCIENCES AND TECHNOLOGY

• Document Type: CREST
• Collection: JPRS
• Document Number (FOIA) /ESDN (CREST): CIA-RDP82-00850R000100060055-4
• Release Decision: RIF
• Original Classification: U
• Document Page Count: 77
• Document Creation Date: November 1, 2016
• Sequence Number: 55
• Content Type: REPORTS

APPROVE~ FOR RELEASE: 2007/02/09: CIA-R~P82-00850R000'100060055-4
~ . ~ ,

26 JUNE i979 CFOUO 36179~ ~ i OF i
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR OFFICIAL USE ONLY
JpItS L/8544 -
. 26 J~e is~9
s
~ .
TRAP~SLI~TIONS ON USSR ~CIENCE AND T~CNNOLQGY
PHYSICAL SCIENCES AND T~CHNOLOGY
(FOUO 36/i9)
_ .
U. S. JOINT PUBLICATIONS RESEARCH SERVICE
r
FOR OFFICIAL USE ONLY
~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
NOT~
JPRS publicaeions conCain information primarily from foretgn
newspapere, pertodicals and booke, buC aleo from n~we egency ~
erattemieaions ~nd broadcasts. Materials from foreign-language -
sources are translgted; those from English-language sources
are transcribed or reprinted~ with the original phrasing and
other characteristics retained.
Headlines, editorial reporta, and material encloaed in brackeCs
are supplied by JPRS. Processing indicarors such as [TexCJ
or [ExcerpCJ in Che first line of each item, or following Che
lasC line of g brief, indicaCe how the original informaCion was
processed. Where no proceasing indicaCor is given, the infor-
mation was aummarized or exCracted.
Unfamiliar names rendered phonetically or trensliterated are
enclose3 in parentheaes. Worda or names preccded by a ques-
tion mark and encloaed in parentheses were not clear in the
original buC have been aupplied as appropriate in contexC.
Other unattributed parenthetical notes within the body of an
item originate with the source. Times within iCems~are as
given by source.
The contents of this publication in no wgy repreaent the poli-
cies, views or attitudes of the U.S. Government.
I
COPYRIGHT LAWS AND REGULATIONS GO~VERNING OWIvERSHIP OF
MATERIALS REPRODUCED HEREIN REQUIRE THAT DISSEMINr1TI0N
OF THIS PUBLICATION BE RESTRICTED FOR OFFICIAL USE ONLY.
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
~
NOTI~E
Eff~ctive 2 July 1979 tihis reporti will bo discon~inued i.n
i~s present form~ Material.s now published in this report
will be combingd witih the abstract se~tes and wfll be published
under tha existing abstract rsport subiects lfstied below.
A new cover design wfll. also be initiated at tihis time.
USSR REPORT: Biomedical and 3ehavioral Sciences ~
USSR REPORT: Chemistry
USSR REPORT: Cybernetias, Compu4:ers and Automation
Te~hnology
USSR REPORT: El.ectronics and Elec~rical Engineerinq
USSR REPORT: Engineering and Equipment
USSR REPORT: Ma~erials Science and MetiaZlurgy
USSR REPORT: .Physfcs and Mathematics
USSR REPORT: Geophysics, Astronomy and Space
If you receive your JPRS publications through NTIS, you ~
may wish to contact them concerning your sur~~3cripti~n.
If you receive your JPRS publications t~`~rough a distribution
control center, please contact them dire^tly concerning this
chanc;e.
S
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
.
' ~
. FOR Ok'F'ICZAL USE Ot~LY
~ ~ '
- JPR5 L/8544
- 26 June 1.979 ~
;
~ TRANSLAT~ONS ON USSR SCIENCE AND TECNNULOGY
PHYS,ICAL SCIENCES AND.TECHNOLOGY -
~FOUO 3fi/79 )
' CONTENTS , PAGE
CY'BERNETI~S, COMPU"_'6RS AND AUTOMATION TECtiNOLOGY
~
5ocio-Philosophical Problems of Man-Machine Systema: -
Artificial Intelligence as a Complex Scientific and
Technical Problem
(E.V. Popov; VOPROSY FILOSOFII, No 4, 1979) 1
~
~ Virtual Operation System for Small and Medium Computers
- (Ya. Soko1, V. Navratil; PROGItA1~AiIROVANIYE, Nov/Dec 78). 5~ -
GEOPHYSICS, ASTRONOMY AND SPACE
Orbits of Communications Satellites
(G.M. Chernavskiy, V.A. Bartenev; ORBITY SPUTNIKOV
SVYAZI, 1979) 17 :
PEiYSICS
Laser Spectroscopy in Nuclear Physics
(V.S. Letokhov; VESTNIK AKADEMII NAUK SSSR, No 4, 1979). 45
PUBLICATIONS
Systemology and Linguistic Aspects bf Cybernetics
(G.P. Mel'nikov; ~ISTEMOLOGIYA I YAZIKOVYKH ASPEKTY
KIBERNETIKI, 1978) 60
Analytical Probabilistic Models of Electronic Computer =
Functioning
(G.T. Artamonov, O.M. Brekhov; ANALITICHESKIYS -
VEROYATNOSTNYYE MODELI FUNKT~IONIROVANIYA EVM, 1978) 63 �
-a- [III -USSR-23S &T FOUO)
FOR OFFICIAL USE OYLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
~
FOR OFFICIAL U3E ONLY
_ CONTENTS (Continued) p~~c
Economics And Organizatinn of Dara Processing Sygtema
(I.S. Zinger, et al.; EKONOMIKO-OItGANIZATSIONNYYC
OSNOVY S02DANIYA SISTEM OBRABOTKI DANNYKH, 19~8) 66
List of SovieC ArCiclea Dealing With Compoeite Materials
(GOSUDARSTVENNYY KOMITET SOVETA MINISTROV SSSR PO
NAUKE I TEKHNIKE. AKADEMIYA NAUK SSSR. SIGNAL'NAYA
INFORMATSIYA. KOMPOZITSIONNYYL MATERIALY, No 6, 1918). 68
- b -
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR OF'FICIAL t78E ONLY
CYBERNETIC3, COMPUTER3 AND AUTOMATION TECHNOLOC~Y
~ 50CI0-P;!ILOSOPHICAL PROBLEM3 OF MAN-MACHINE SY3TEM8s ARTiFICIAL INTEL-
LIGENCE AS A COMPLEX SCIENTIFIC AND TECHNICAL PROB'uEM
.
Moscow VOPROSY FILOSOFII in Rusaian No 4, 1979 pp 76-~8
(Article by E. V. Popov: "Systems of interactfon of man and computer in
natural language")
(Text~ The concept of "artificial intelligence" was introduced by
specialists in computer technology and programming. It uaually implxea an
- engineering discipline intended for the development of computers of
programa capable of action which would be called intellectual if it were -
done by humans (1). This de~inition focusea attention only on tihe
_ inclination of artificial intelligence (AI) to solve non-numerical prob-
lems of a particular type on the cmnputer which, until recently, were
included in the sphere of hum~n activities. it is not claimed that
programa or devices which solve these problema poasess intelligence in the
human sense. The relevancy of the very term "artfficial intelligence", of
course, could be criticized, but this would hardly be profitable since the
term is so popular.
MOfit specialists (including the author) relate artificial intelligence to
te~:hnical disciplines (2) and feel that the search for a theory of
- a?.tificial intelligence makes no more sense than a search for, let ue say,
a theory of civil engineering. Instead of a unified general theory of
artificial intelligence, there is a series of theoretical disciplines
which are employed in AI. Among these are linguistica, psychology,
mathematic logic, theory of computations, theory o� algorithms, theory of
ir.formation structures, theory of graphs, theory of heuristic inves-
tigation, etc. The basic applied functions of AI include automatic
problem-solving, "comprehension" and synthesis of texts, translation fro!n
one language to another, proof of theorema, recognitfon of visual models
and speech, representation and storage of knowledge, development of
roboi:s, and so forth.
Many of the difficulties with Ai are r~flected in the problem of developing
a system of interaction of man with computer in natural language (i~L) .
~ This allowed several experts to consider thefr problems identical. The
process of interaction in natural language includes the following prob-
lems: "camprehension" and synthesis of texts, automatic decision re-
1
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02149: CIA-RDP82-44850R000100064455-4
, FOR OFFYCIAL USE ONLY
trieval, recognition and syntihesis of speech, representia~ion and atorage
of knowledge.
No sector of the national economy exists where computer technology is not
used. But day-to-day and masa utilization is hindered by the fact that
interaction with computer may now be done only by programmers, the demand
for which is constantly growing. Consequently, it is becoming evident that -
the need Eor problem-solving of human interaction with computer is not only
due to the quantita~ivA increase in programmers, but alao on the plane of
interaction with co~i~rater in natural language, which has several ad-
vantagea over formalized programming languages. Fi.rst o� all, because of
the practical lack of a training staqe, it is poasible to conduct
preliminary training. Secondly, there is a rise in efficiency and
convenience of human fnteraction with computer. Finally, the number of
errors permitted by man in such interaction drops sharply.
Tt?e development oE systems realizinq interaction of man using natural
language with the computer should, we feel, be based on principles of
universality, development and interdisciplinary penetration. Univer-
sality implies: a) univers~lity of choice of inett~od of representati.on of
knowledqe (data bank) which will permit expansion (during evolution of a
system) of the class af phenomena representable in it without change in -
processing method; b) universality of algorithms and prc~grams (software)
to reveal the possibility of expanding functions of the system not throuqh
remodelling, but through adjustment of software; c) independence of
softwar~ and data banks to per?:iit expansion and modification of models of -
the environment without changing programming equipment.
The principle of development anticipates fuZfillment of systems with
natural language in several stages due to the current impossibility of
solving the problem in its total scale. As such, these systems should be
built on a modular principle, according to w!-~ich it is possible to alter
both the sequence of request of individual modules of the system, their
number and their function without changing the entire system. As a matter
of fact, this means the development of complicated versions of a system
!retaininy its continufty) by successive lifting of constrafnts imposed on
the interaction process. The process of selection of these constraints is
now couplec] with problems, on one hand, of specifying constraints
- acceptable co system users and the determination of their consequences (in
other wcrds, there is no guarantee that interaction will be at all possible
witr, a c:;o~zn constraint) ; r~nd, on the other hand, with problems oE
determ~n:ng (pcior to system design) the adequacy of the media used in its
design to solve the problem of interaction in a selected su6set of natural
languages. In practice, attempts are not even made to solve this problem,
and the first constraints encountered are accepted. Thus, for example, the
author knows of several teams (working without lfnguists) who have
attempted to create natural language systems based only on semantics,
neglecting morphology and syntax of Russian. The failure of these
attempts, apparent to linguists, is unclear to prograrr,mers.
2
FOR OFFICIAL US~ ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02149: CIA-RDP82-44850R000100064455-4
!'OR OFFICIAL U3E ONLY _
The principle of interdisciplinary penetration anticipates the need for
solving the problem oF interection oE close and constan~ contact of
specialists o� various fields, primarily linguistiag and programmers. it
seems inadmiasible to us that~there is the trend of specifying the limita
to which linquistis are involved in the program and then programmers in the
proceas of solving the prob~em of interaction. Unfortunately, this trend
has Eound almost universal private acceptance.
i would also like to touch briefly on a description of the s~atus and
outlook for development of the POET system (program of processing economic
texts). The particular choice oE this system is not accidental. The author
is unaware oE any other currently operatinq system which could be related,
- with complete justification, to natural language systems= this does not,
of course, mean that the POET syetem solves all problema of interaction,
but it has been worked out in atrict correspondence with the above
principles, permitting it to evolve while continually increasing its
possibilities. The POET syatem was developed by a team led by the author
with the aid of co-workers of the department of structural and applied
linguistics of Moscow State University, chaired by Prof. V. A. Zvegintsev.
In the first stage of the process of interaction, the following constraints
were introduced: language of interaction is commercial economic Russian _
process; interaction is carried out by individual non-connected, simple
interrogative sentences--free of inversions, e~lipsES or anaphorisms;
internal representation allows only the static (not dynamic) world without
connections between events (dynamic, that is, a change of data base, is
dor.~ without consent of the POET system); the data base has a rigidly fixed
format restricting topics of interaction; the system can not', fn the event
of misunderstanding of an interrogation, direct the user to periphrase it.
The system response time to a 15-wor.d interrogation is 30 seconds fn the
third-generation model YeS-1050 computer.
in the second stage, to be completed in 1979, the process of interactfon
will be characterized by the following features: interaction is carried on
in dialogue mode consisting of several interrelated sentences conteining
inversions, subordinate sentences, elliptical and anaphoristic phrases; in
the event of misunderstanding of an interrogation (due to user error,
ambfguity of interrogation or system constraints), the systems ceports to `
the ~ser on the causes of failure in concept-terms and directs the user
(using queatcions or statements) to change the initial interrogation;
internal representation permits the expression of casual relationships
between events.
3
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02149: CIA-RDP82-44850R000100064455-4
FOR OFFICIAL U3E ONLY
i
The proc~s~ of designing gyetems wfth natural languages is now, we feel, .
retaining the follmwing di�ficultiea: the amall number and discreteness of
groups engaged in its development~ underes~imation of the problems o�
interdisciplinary cooperation. Many linguists �eel that problems are
theirs alone, while programmers feel that they can get along without '
- linguists in general; bath �orget about psychologiats, philosphers and
members of certain other apecializations, without which the solution to
the problem of interaction, in its full scale, is impossible; and finally,
the lack of basic research in the field of representation of knowledge,
understanding o� semantics and structures of a connected text.
COPYRIGHT: Izdatel'stvo "Pravda", "Voprosy filosofii", 1979
8617
CS0:1870
4 .
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02149: CIA-RDP82-44850R000100064455-4
, , .
FOR OFFICIAL U9E ONLY '
CYBEltNETICS~ COMPUTERS AND AUTOMATION TECHNOLOGY
i
~ VIRTUAL OPERATION SY3TEM FOR SNWLL AND ~IEDIUM COMPUTER3
Moscow PROGRAMMiROVANIYE in Russiah No 6, Nov/Dec 78 pp 50-59 manuacript ,
received 23 Feb 77
(Article by Sokol, Ya. and Navratil, V.~
(TextJ This article examt oes pl~ns for a virtual operation system designed `
for small and medium general purpose computers. Let us first touch upon
the basic requirements imposed on an operatfonal system.
Eaah operational system (OS; should support the efficient use of machirte �
resources under various operating conditions: in batch processing, remote
data processing and in limited operation during time-sharinq.
~i'he system should guarantee reliability in the event of hardware mal- ~
functions or programming errors. This requirement is important, es- ;
pecially in data bank applications. !
Considering the labor-intensiveness of systems design, long service life ;
must be supported. It is desirable to anticipate the system's design and
evolution.
It is also very important to provide simple use and maintenance of the
system, often the deciding factor for technical workers and users (pro-
grammers and operators); this has a substantive effect on system ac-
ceptance by the user. If the system can be kept rather sfmple and logically
, consistent, the number of documents required by the user will drop
subsl:antially.
Specia2 attention should be given to compatibility of data and programs
between devices with different hardware (memory capacity, different
peripheral devices, etc.), between separate versions and modifications of
a proposed system which will undoubtedly appear during its service life and ~
which should not cause changes in the user's programs. One condition for
attaining this compatibility is to secure a system interface which would be s
accurately defined and guaranteed for the future at the data and program
level in initial symbolic form. The interface shoulr] anticipate in-
dependence of programs from types of i/0 devices and provide retained
efficiency in the event of further expansion of system functions.
5
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02149: CIA-RDP82-44850R000100064455-4
, FOR OFFICIAL USE ONY,Y � -
The proposed system ensures complete transferability of sequential data
filea from the most common third-generation operating systems and contains
programs for converting index-sequenced data Files. Programs written in
universal programming languages can be transfered in aymbolic form without
change even if they were �irs~ intended for i/0 devices of other form.
Programs wri~ten in Assembler m~.~st be translated. If they are independent
of the form of translation, marking and length o� system array, control
units and macroinstruction execution blocks, they can be transfered
unchanged from unified system disk-operating systems. Caxry-over of
translated programs can only be done in exceptional cases: if no references
to the operational system are contained. -
An important aspect of the project is the labor-intensiveness of rea-
lization, testing, putting into operation and maintenance of the system
within its entire service life. Much attention fs also given to the
problem of incorporating new I/0 devices into the system. This important
practical question has not been satisfactorily resolved in all current
systems, this involving great losses.
The general concept of proposed system is based on the sequential uae of
principle of the virtual nature of inemory not only for the user, but for the
system per se, in centralization of all significant functions of the system
and in dynamic distribution of resources.
Terminology r~nd concepts of operational systems of the Unified System of
Computers are wide~.y used in this article.
2. Systems Resources
The basic task of the operating system nucleus is to control the use of ~
systems resources. Systems resources in this system are the following
devices ana hardware: central processing unit and main storage, virtual
address space, storage region and mag disk working memory, I/0 devices,
program library, data files, and op system programs.
Storage regions and working disk memory are specified dynamically as
needed in units of 64 Kbytes. The entire volume of working memory on mag '
disk is combined for all applications and for all users in a single systems
retrieval file which can be placed on several mag disks. The system
specifies segments of working memory onZy if needed. This achieves
su~stay~tiaZ savings of disk memory and one shortcoming of the current
system is eliminated that has often hindered the use of multxprogramming,
especially in the small computer. ,
Place in the storage region for created (output) files is reserved and is
specified at the moment a file is opened. In the case of changing files,
the earlier file is specified for the availability of one user. As a
result, the danger of deadlocks could appear, but it is rather theoretical
in the given case.
6
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02149: CIA-RDP82-44850R000100064455-4
FOR OFFICIAL USE ONLY~
It is completely different with other devices which are spec~ Fi,:~l only for
_ one user. All I/0 devices, except floppy disks and operator consoles, are
task-specified for exclusive use for all execution time. in some cases (for
example, an alphanumeric printer), the above strategy often involves a
- reduction in productivity of the entire syatem, since a computer prr~vided
with a single alphanumeric printer can execute~ only one task, though
neither the central processing unit nor the alphanumeric printer is
overburdened. The proposed system an~~icipates work with vixtual peri-
pheral devices using the principle of balanced rate of data transmiasion by
means of floppy disk (SFOOL). Virtual I/0 devices, whose number is not
limited, ma~ be simultaneously shr.;ed by any number of users. The
operating principle of virtual peripheral devices will be described below.
Sharing of the central processing unit (CPU) by sevural independent user
tasks and several subtasks of a single assignment or system is a regular
phenomenon and will be described below (see Section 5, S1).
3. Control of Main Memory
Sharing of main memory by several paral,.lel executed tasks is done in the
propc~sed system by the mechanism of virtual addressinq supported by ,
computer devices.
Virtual address space implf~s an abstract concept in this system, denoting
a set of numbers by means of whicb program objects can be identified
(inatructions, data). Each parallel executec] task has its own addreas
space which can be dynamically raised almosti to total c..apacity of 16
' Mbytes. In practice this means that in devices, translation of virtual
adresses is controlled for each parallel ~xecuted task using its own -
segmented array; replacement of segmentQd .arrays in transition from one
task to another is done by altering the coRt,ents of the system reqister of
~he segmented array.
Program objects denoted by virtual addresses may be maintained: constantly
in the phase library (intrinsic programs), temporarily in disk storage
region (changing gortions of programs and working memory).,
~ The form of virtual address space reproduced by portions of the library or _
disk storage reqion is optionally continuous. In the general case, one
part of the virtual dddress space can be arranged in the lit~rary, while
anothe~ part---in disk storage region. An important consequence of this
concentration is that the shape of virtual address space can be reproduced
dynamically. This means that parts of virtual space are appropriated by
the disk mem~ry region only at the instant of appropriation of the
corresgonding virtual address to objects which can be altered during the -
computation. Virtual space appropriated for objects of continuously
- changing nature (e.g., multiple input programs) is placed only in the phase
library, while space whose address at a given moment is not occupied by any
- object remains unpositioned.
7
FOR OFFICIAL USE ONLY i -
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
~
. F'O~t OFF'ICIAi. U8~ ONLY
~
Objecre ere rranefered from the librery or disk etorage region, if
necessery, into reai memory, generally using en ordinery page-by-pege
- mechanl,~m, ~xcept f~r eh~ reoid~nt ~rtf.c~n ~f the gup~rvigor which ig
cerried intio rea~ m~?!~ory ae Che ~tert of ~ystem operetion (using the IPL ~
pr~gra,~) and gQVergl sp~ciai cases oE pha0e entry.
i. Vic~u~i Addrea:~ ~pace (VAS)
VA3 of each uger tagk consists of two basic parts~ sha~~e6 virtuai spac~
S (SV8) end private virrual space (PV8).
Sv~ ig space common to all ueer tasks= corresponding positiona of segmented
arrey~ of aii executed taeke indicate identiical page arrays. SV8 ig found
in the lower pnr~ion of each virtual space, starting with the zero
(virtual) ~ddrea~. The upper limit of 3V8 is detecmined with ays~em
generation~ Since SV3 does not contain any ob~ecte which woulc+ have to be
temporarily entered during computation into the disk storage regton, it~
shepe will alwayg eppear only in phese library. 8VB consiste of the
following parts:
i~ ace nf identicai Addresses (SIA)
~ This contains the nucleus af the control program and correspcmding systema
arrays. SIA constantly appropriates identical real addresses and its
shape eppears in reai memory as a result of execution of the IPL progrem. #
RQa: memory, carried to objects stored in SIA ia not subjected to pege-by-- _
pege distrfbution of inemoryt thus it is not necessary to reproduce the
shape af SiA in the disk storage region. The actual volume of 3IA can be
selected in generetion of the syetem. if the selected voluem of SIA is
qreater than that required by resident portions of the supervisor, other
unoccupied portions can be used for program operation in real memory mode.
Maximum S2A volume is, in this case, defined by the capacity of main memory
used by the computec.
~
Separate Virtual Proqram Space (SVPS)
tlere are kept the remafning parts of thE supervisor, LIOCS tnodules, task
input control program, etc. The shape of tuis portion of virtual space is
agein on2y in the phase libracy. Separage pages are transfe.ced into
,
- r+orki>>g ;nemory directly from the phase 14bcary. 3ince all progcams
inte:~ded ror SVPS are multiple input, :hey can not be cewritten in disk
storage.
C
}
Separ~te Virtual Working Space (SVWS)
This is specified dynamically accordinq to parts of the program executed in
SVP5 foc storage of tempcrary workinq data. Objects stored in SVWS exist
- only in reserved workinq memory which, ducing use, is excluded from page--
by-paqe distribut.ion (R,iocked). After replacement of blxking of reserved
- working memory, objects are considered destroyed. There is no need to
reproduce the shape in disk memory for SVWS. Data of a long-term nature are
retained usinq SVPS programs in regions reserved i.~ SVS.
8
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
F'OR OFFtC~AG U8~ ONLY
Prive~e virtuai spac~ nE all uaer taaks srarte with the virruai eddrees
dQEined by the upper limie of VAB. it ie an eddresr to whieh ere tuned eii
u~er programe, tiranglatorg Qnd other aystemg programs. ~rom this limit,
virtuei apace~ of differenti ueers are eeperatied, i.e., Begmentied arreye of
individual 8V8 indiaate varioug page erraye. Described organizetie~n
permite in ail eimultaneousiy exee~ted uBer taskg tihe use of programs
written in identicai addresses. It is thus noti neaesa~ry to pre-determine
in which user speces to gtart up a speai~ic program and when the program ia
~ntgred i,nto memory, site-by-glte alignment is noti necessery.
At the etert of solution of a tiaek, the usere hav~ evailable e part of SV8
(i.e., writiten correspon9ing page errays or reserved spece of disk
storage) eccordinq to requirementa defined in task input contirol in-
atructions. The volume of SV8 can he dynemieally expendeQ durinq
computetion to a maximum oE 16 Mbyte for eech user. Arrangement of ail SVS
is primarily done in the phase library. After program entry into working
memory, the ahape of unchenged pageg remeina in the phaBe library, wt~ereag
altered paqee of the program and data are pieced into the digk atorage
region as r~quiced. Organization of segmented erraye end page arreys
, p~rmite two or more 3V8 to be combined in this syatem. in such casee, the
stepa of tasks being executed in these spaces use identicel programe. A ~
, condition for separation of SVS is the multiple input nature of programs
and their entry into all SVS in identicel addresses.
2. Pheae Library and Disk Storaae 3pace
The phase library reproducee the long-term shape of aeparate and all
private virtual spaces= in the diak atorage cegion is repro8uced the
ten~porary st~ape of these pacts of private virtual spacea which were changed
, during computation and can not (or should not) be in working meuiory at the
time.
The basic image unit of VAS in :he library or disk storage space is the ~
segment. The 8VS shape in the phase library is always continuous in terms
of segments (i.e. programs are edited in SVS by segments), just like the F
shape of SVS in disk storage space. in observing agreements for forming
user pxogram phases described in section 5 S1, the ahape of SVS in the phaee '
library is also continuoug in terms of segments, permitting the con-
centcetion of all carries between thp phase libracy end working memory
excl~ss~ively with the page-by-page mechanism and eliminating the ordinary
funceion of phase entry.
Physicdl segments in phase library are s~ecified during editing and are
carried to the virtual spacea. Segments in the disk storage region are
carried to SVS seqments at the instant of demand for specification of a
part of SVS, i.e. at the start of execution of a~task or in sequence, with
dynamic expaneion of SVS during execution.
9
FOR OFPICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR OFFiCIAG t18B ONLY
Real Memo~ (RM) '
Reol memory consists of tiwo partig. The firsr part, stiartiing wirh tihe
zero reai eddree~, correapondg to part of 8IA which i~ con~tantly xaupie9
by the resident portian of the syetem. mhe 8econd part (unoccupied) forma _
the entitiy of physical pages which are carried to pages of aeparate and
private virtiual spaces as needed. The subprogram of ~he controi program
which realizes page - by - paqe mechanisms does periodic mo~~itoring of
optionai pre~ence in working memory of pages, releases the eppropriate
physicdi pages for furtiher utilization an8 transmits aitered page8 to tihe
reserved portion nf disk storage.
4. Direct input-Output 8ystiem
,
The greatest hindrence oE the multiprogramming mode is the conetirainti
on the number of peripheral dgvicea, primarily printerg and readere for
punch card equipment. Thus the proposed system includea media for
operation with virtual peripheral devicea, the numbec of which is not
conatrained and which can handle sev~ral teska. Virtual devices operated
by the following principle. ~or each virtuai device and for each user, the
system forms a epeaial working file in diek memory in common storege. Each .
- 1/0 demand in the virtual device is executed in a special systeme
subprogram which performs data exchange. in the event of peripheral input
devices, the deta block is read from the disk and the appcopriate amount of
data blx ks is tcansmitted tio the user region. For an output device, user
datid i~ transfered to the systems buffer or, if it is filled, data are
entered into the working space on the disk. Carry from disk to ~isk xcurs
with the aid of the pag~-by-page mechanism. The actual physical carry of
data from the input device to the disk (or from disk to the output device)
is done by an independent system~ subtask totally asynchronously, but in
" such a way that data carry of one task (task group) takes place jointly.
Thus, simultaneous direction of the resulte of two tasks toward virtual
' printers is ensured, although in physicel form, both results are se-
quentially printed by one printer. Even one task can operate with several
virtual printers.
Victual devices can be used for readers (punchcard and punched tape
_ perforators), for printers, diskettes and remote data processing devices.
i;~direct i/0 systems programs simultaneously eliminate distinctions of
inafvidual peripheral devices, so that all virtual input devices, from a
pre;ra~~ point of view, operate as a reader from punched cards and virtual
output aevices operate as a perforator of punched cards or printinq device.
Th,us data processing in an unusual information carrier is ensured (punched
tape, diskette) or ~nusual peripheral device (such as a remote prxessor)
without changes in the program, even in p:ograms written iR univers~l
pragramming Ianquages which do not contain this prxessing.
10
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR OFFICIAL U8E ONGY
~he operatior conrroi~ th~ wock of Byetems rasks Eor phyeicei carry of data
berween diBk and pRripherai device uaing eimpie instructions. in data at
tihe output tihis concerne start, stop or repeat. Bomewhati more complicated
is input. inputi date ehouid be pcepared before ~tiarting tihe tiasks -
(magnetic tapea etc. are ser up), which is eA~ily detected in deta of the
inputi fiow of aasign~n~nti~, i.e., data contiaining taek controi operetiore.
The propo~ed sy~tem, Eurthermore, supports tihe virtiuel prxesefng of input
data filea, !or example, from punched tape or other unusual inform~tiion
carrier. in the exemples eited, the operator supporre the starti of the ~
systems tesk for phy~ical carry of data and, in addition to otiher
pernmetera, definee the identifier of the input file. Task eontrol
operatora degigned to process the file refer to the i~entifier.
5e P_rogram
Sy�r
s~tem~
1. Supervisor. The Bupervisor of the proposed gystem coneiats of
a comparatively smaii nucleus conetantly storer] in working memory and a
large number of more or less independent sub~progrnmms tranefered to working
memory as needed by the page-by-page mechanism. The reaident portion of the
supervisor is located in virtual space in the SIA region and contains
interrupt signal prxesaing, subprograme, some of the most basic SVS
subprograms, programa for magnetic disk readout error correction and,
fin811y, programs necessary for operation of the page-by-page mechanism.
in addition to the aforementioned 8ubprograms, the reei8ent pcrtiion
contains the corresponding arrays.
The remaining programs of the supervisoc are located in SVP3 and are edited
in turn in a aingle succession, so that any two subprograms from SVPS can
operate as needed in an identical time segment.
in editing parts of the supervisor in SVPS, the editor forms a special list
of all SVPS programs whose carry to working memory ia done by the IPL
progr~n. Positions of this list contain SVPS phase names for each symbolic
name, virtual address of the phase start in SVPS and address on the disk of
the corresponding eegment in the phase library. Swmnona of any part of
SVPS is thus reduced to translation of the symbolic name to the cor-
responding virtual address and transmission of control to this address.
in combined operation with the task input control system, the supervisor
can control the operation of up to five parallel executed independent user
tasks; in each task the user can form and simultaneously execute up to 99
dependent subtasks.
input of new phases duri~g computation of results is done by two different
metho~s in the system. The first metho8 requires ~bservation of the
following agreements during phase Formation:
11
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
,
: ~ ~
FOR OFF'ICIAL USE ONLY
phases are edited to the ~virtual) addrea~ through which taeke wiil be
writitien during execution (i.e., in entry o~ ph~ees no site alignmont is
dr~ne). This condition is noti difficuit tio observe, eince eaoh ueer taak
has its own virtual aQdr~ss apaae in the ~ystem~ ~
the initiial phase address is located at rhe segment boundary, i.e., rwo
different phases one task stiep are never edited into one segment. IN
editing a large number of phae~s, gaps are formed in the virtual epace. 8ut
in reality they only cauee a loss of a certain number of virrual addresses
which are consequently not phyaically depicted in either real memory or in
disk storage. When the above agreements are obaerved, summons of a phase
from the phase library is reduced to retrieval of the address of the phase
start on the disk in the phase library and formation of the appropria~e
_ position~ of segment and p~ge arrays of PVB. Phyeical carry of required
phase pages to real memory is done by tihe page-by-page mechanism in the same =
way as in the case of programs belong to 8VP3. When agreements are not
observed, i.e. if phase site ~lignmentis required during en}ry or the phase
fs not situated at ttre segment boundary, it should be read in the ordinary
manner or its shape should be reproduced in disk storage. In the l~tter
case system efficiency is reduced substantially.
2. Remaining Systems Programs
All other parts of the system are connected to the control program and the
" entire system via a conventional, accurately-3efined interface, primarily
in symbolic form. Reference8 to the control system are always controlled
using macroinstruc~ions. Information contafned in control arrays is also
defined using macroinstructions or empty systems sections.
Concatenation of systems components is much simpler than in current
systems. A basic condition for simplification is strict observance of all
systems agreements. Basic systems agreements are expressed with the aid of
standard media (subprograms, macroinstructions, etc.), whose use is :
_ obligatory for all systems components= LIOCS for systems filee, program
for readout and interpretation of control operators, program for com-
munication with the operator and printout of errors, convention and
macroinstructions of subprogram summons, formation of standard ~page
headings.
The system is also provided with the followinq standard media for
compilers:
readoct and entry of library elements,
LIOC for translated programs,
dynamic signalization of translated program ecrors.
12 ~
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
t~
. t
- FOR OFFICIAL UBE ONLY
;
6. inputi/Out ut 8vstem `
One of ~he be~ia purposes in planning a systi~m ie to raise the flexibility ~
of inputi/outiput and improve ita elgioiency in a virtuel medium. The ~
physiral level of inpur/outiput in the proposed systie:n ie considered an '
exception. in easee of need, the uaer ehouid supporr one og the following
conditionss
a) pither etert the program in reai memory mode= ~
b) or must fix paqes containing tihe CCW ch~in and ail regions of datia '
himself, and transfer the data eddreea in CCW to reai himseif (this method
wi11 be used in cases whece the progrem generetea the CCW chein). o) or the
physical i/0 program muat alwaye work with the virtuai device.
The standard principie of i/0 programming ie the logical level (LiOC)
~including aystems programs). Principles of operation of virtual memory `
permit constan! reaidence of executing modules of LIOC to be realixed in ~
_ virtuai memory. mhey are carried into real memory as needed. Sinae the
executing modules are mulriple-inputi, they can be shared by several tiasks,
which involves substential economy of apace.
The module or a type of actual c~ncdtenated peripheral device can be
considered simultaneously. Parame~erg of file deacription related to the ;
specific device model do not have to be coneidered, like the parameters 3
which determine only module properties.
The result of translation of the DTF description is an array containing t
CCB, pariuneter valuea, storage region for registers and module atorage ~
place of information necessary to the user, e.g. the address and length of ;
a block just prxessed, a disk addrese, error display, etc. Working ~
regions, place for CCW chains, IDAL, disk address field are speciEied
during opening operation (OPEN). Necessary information is simultaneously
fixed and translated to real. it is useful for the program not to contain
a descriptian of buffers, whose sutomatic apecification and fixing occurs
during opening via the macroinstruction GETtiIB. Thus there ie an econany
of the number of fixed pagea and space in the phase library, since the phase
_ texts do not contain buffers.
If there is automatic specification of file buffers, further parameters
such as blx k length can be dynamically altered when interactian with the
file begins. in the proposed syatem it is possible to take the j
correspo~aing information either from DLBL/TLBL operators or from marker
F1. Specification of the place for output files on the disk is aiso a
function of the macroinstructfon OPEN. Prxessing of diak markers ,
guarantees the transferability of successive files from regular third-
generatfon systems.
13
FOR OFFICIAL U~E ONLY
,
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR OFFICIAL USE ONLY
~he Eormati of index-eequenced filee differs from current systeme in tihat
there is blocking of cylinder indexes o� both primary da~a and data in tihe
overFlow region. When the file is formed, the usor can determine the
limita of reserved unxcupied place in each ddta blxk. in the case of
overElow of tihe block itis vnlume is divided roughly in hal~ and a new bloak ~
is formed with a reserve of uno~~upied apace in the overflow region. The ~
proposed format, subjected in advance to eimulation and tieats, also
supporta addressing of blocks by meana of their aigk addresg ~macro- :
instiructiions READ ID), whioh is especially importiant in datia bank
applications. in order to atandardize data handling the parameter IOREG
wa~ introduced for a file with directi acceas (DTEDA) and some il-
logicalities were eliminated.
Working filea (TYPEFLE-WORK) are generally appropriated by virtiua~. memory -
as an informatiion carrier instead of diek memory. For this case the syatem
contains a special module which apecifiea the required volume of virtual :
memory and aimulates in it the behavic~ of the disk module. This metihod
also economizes and makea better use of disk apace, acclerates work of
translators and simlifiea maintenance.
The proposed concept of virtual memory also supporte the possibility of
retaining some portion of virtual user space in a certain f11e which h.e
specifies on disk memory. Thus it is a very efficient method of long-term ;
storage of data for whose addressing a more c4mplex method has to be used
(lists, arrays, matrices, etc.). The program inquirea for expansion of its
virtual space by macroinatruction GETVIS and itaelf determinea tihe
symbolic device and identifier of the file in which the system will
reproduce or whence will be selected samples of pages of the particular
~
space.
Declarative and executive macroinstructions of LIOC on the symbolfc level
do retain the basic features from previous systems, but some parameters of
file description are not taken into consideration. All methods of aystem
expansion are planned not to disturb program compatibility.
The proposed concept greatly simplifies inclusion of new types of I/0
devices into the system: until now this was considered a complex problem
which required intervention into a large number of system components. In
tt.e prop~sed system this only concerns programs of physical and logical
ievels, but always in accurately defined places. �
7. L:brary
All librari2s of the proposed system are ordinary, standard organization
disk Eiles. A list is place first: its positions are sorted and blocked.
To accelerate retrieval, each list blx k has a key. Positions whose
cateloquing fs done by the key are placed at the end oE the list= total
ordering and physical exclusion of dumped items is done during re-
organization of the library, always associated with copying to increase
system reliability. Some text is stored in constant length blxks (each
text position is connected in sequential blocks). All libraries can be
syatemic and private.
14
_ FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR 0~'FICYAL I18}~ ONLY
The ph~se library oacupie~ a partiicular piaoe in the syatiem. Phase ti~xts
ore tr~ns~ered from the phase library or using tihe pdge-by-pags meohaniem,
or by ordinery methode (LOAD, F~TCN) wi~h poseible eite alignment. in
addition to ordinary phases, the phase libracy contains deciaratiive phASee
_ of us~r divieione, the gystems library contiaine the li~t of ahared apace.
The lengtih of bix ks of ali remaining libraries is identiaai, and tiheir ~
poaition may be of different type. in one library file may be stored
modules, maaroinsrructiione, initial texta and prxedures= tihe type of
poaition is defined by rhe identiifier prefix. Aocess to librariee is done
exciusively on rhe logic levei, and its equpment ia accessible to the user
who cen form hia own libraries.
A general maintenance program for aupport of libraries performs catelo-
guing, dumping, corcectiion, punching and printout of individuel poaitions
of entire groups. inalusion of correction punchcarde depende on their
nwnbering in selected card columne. Uaing maintenence programs, it is
poasible even to write initial texte from parts of other texts in arbitirary
aequence. Yn cataloguing macroinetructiions, careful syntactical mo-
nitoring and preliminary prxeseing into a form which greatly acceleratea
generation is done. NonetheleaE even packaged macroinstructions can be
corrected with au~ciliary programs without preliminary downloading. The
maintenance program used to copy and reorganize libraries makea a carry of
whole libraries, sublibraries and groups of programs between magnetic
disks or a magnetic disk and magnetic tape, namely from one or more input
librarfes to one output library. During this carry there ia ordecing and
compression of text.
This solution is based on operating experience during which it ha~ been
found necessary to save at least one reserve copy of each library. In the
proposed system one should use sequential distribution of libraries into
systems and private libraries. Systems libraries are changed only in the
event of a new editing of the system= a package with private library is
always converted fnto a resecve copy a~ter some time has elapsed and is
copied and reorganized into a new working copy. This type of organization
is always safe and convenient for comp~ter maintenance.
8. Task input and Planning Orders
Tasks are introduced into the system via one or more de~~ices designated by
the operatar as systems inputa. if there are several systems inputs or
there :s a virtual device, the system operates in the multiprogramminq
mode. A systems input is a punched card reader, a maqnetic tape or disk,
a virtual ~]evice--an arbitrary device with p.3per medium, diskette or
terminal.
15
FOR OFFICIAL USE ONLY -
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
~
FOR OFFICIAL tISE ONLY
The taek planning unit ia tihe ordor formed by one or mor~ ~asks which
must be startied in a specific eequence using iden~ical sygtiems resources. ~
The order is searted by the BATCH 3.nstiructiion, which can be considered
a partitiion description. Parame~~rs df ~hi~ ins~ructiott describe '
' ~h~ composition of peripheral devices and ~he requirementi o� the task ,
Eor working memory. Deacriptions of varioua compositiions of poriphe-
ral d~vicQS are stored in ~he phase library: ~he 6A7'CH instruction con- .
t~ine 3denti�iers o� the corresponding declara~ive phase. This phase ~
s
berv~s the order dispatr,her to pra-dQtermine whe~her or no~ it has the .
required services available and whether or not it ahould write out iti
LUB array a~ the gtart of the order ,(st~ndard reserving array). The "
user can reproduce, according to his needs, any number of declarative
phases. ~eripheral devicea can be designed uging all methods permittied
in the insrruction ASSGN (i.e. by means of address, clasa, liat or ~
package notation). All tihese devices are reinforced a~ s~artup for ~
private use, except for maqnetiic disks which are only appropriated.
Reinforcement occurs in the disks according to �i.les only when the i
macroinstructiion OPEN occurs.
The order dispatcher has an order list prepared for startup, and starts
orders up according tio priority and unoccupied resources. Since prio-
riry and the composition of required resources are properties of the
order, and not invariably of the partitions set up, it makes sense to
have a constant designation of partitions (BG, Fu). In connection with
the operator, orders are also identified by means of their own identi-
fiers and not by m~ans of partition numbers. Durinq order processing,
thc input position of orders is supplemented by data to define time ex-
penditures and peripheral activity; after the order execution is com-
pleted, a job account is formed.
i
COPYRIGHT: Izdatel'stvo "Nauka", "Programmirovaniye", )978 ~
8617
CS0:1870
16
~OR OFF?CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
~ , , _
_ ~
. FOR OFFICIAL U8E ONLY
;
;
GEOPMt3IC5, ASTRONOMX AND BPAC~
UDC 621.396.946
ORBIT5 OF COr4tUNTCATTONS SATELLITES ~
Moacow ORBITY SPUTNIKOV SVYAZT in Rusaian 1979 s3gned to prese 25 Aug 78
pp 120-131, 185-197, 215-223
[Seceions 4.3, 6,2, 6.3 and 7.3 from the book by G.M. Chernavskiy and ~
V.A. Bartenev, Svyaz~ Publishers, 2,800 copiea, 240 pages]
(TexC~ 4.3. The Orbital StrucCures of Communications Satellites
One of Che basic requirementa placed on satell~fie communicaCions is 24-hour
service and communications continui~y.
The overali number of saee113~es in a aystem for servicing ~ territory, while
providing continuity and 24-hour communicationa service ia determined by the
duration of a posefble communicaCions seasion (t~) through each eatellite of
the syatem and depends on the size and poaition of this territory with re-
spec~ Co the satellite path on the surface of the earth.
The duration of a possible communications aeasion is conditioned by the radio
visibility time of the serviced territory as well as the possibilitq of pro-
viding a apecified power flux density aC the surface of the earth. As was
noted above (see ~3.5), the radio visibility time for "Molniya" type orbits
is basically determined by the geographic positi~ning of the communications
stations, which approximate the territory, by the Greenwich longitude of the
ascending orbital node, as we11 as by the argumenC of the perigee.
The procedures set forth in 54.2 and 3.5 allow for the determination of the
radio visibility time and aervice area from a satellite having any ascending
node longitude and thereby eatablish the optimal, in the aense of a maximum
radio visibility time and specified service area, ascending node longitude
~3 �
The duration of a cammunications session t~ (7~`~)~obtained for the chosen
optimal ascending node longitude permits the determination of the minimum
number of satellites in the syatem:
T, l
n~~ Lt~(%;1 J Y (4.30)
where E indi~ates the integer port~on.
17
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
_ , ; , ,
' FOR OFFICIAL U5E ONLY '
' The gelected numbax of sa~elli~es can provide �or 24-hour communica~ions
over the terriCory under cons~.derg~ion. In th3s case, a11 of the saeellites
ehould have iden~ical pa~he on the aurface of ~he earrh (i.P., iden~ical .
- orbiCa]. parameCers and Greenwich longiCudes of Che ascending nodes) and �ol-
low one afCer the other at equal time in~ervals
.et~T~ /n, (4.31)
The quan~ity Qp also determinea the time during which Che satellire should
provide communicaCions for the specified territory. It 3s obvious ~haC
~p e~ r~ (a~). Depending on the na~ure of the varia~i.on in Che aervice time ~
due Co rhe Greenw3ch longi~ude of the ascending node, and ehe size of the
difference (a~), ~t of the long3tude of the orbital ascending nodea of ~
~he saCelliCes in ehe system can vary within different 1imi~s, which are
greaCer the greater the service time available C~ (a3), ~t of the longitiude -
of the orbital ascending nodes of the satellites 3n the system can vary _
within differen~ 1imi~s, which are greater the greater t~e serv3ce time
availab7.e t~ (a3). The values of the boundaries for ~he permiseib'le range
of ascending nodal longiCudes are determined by work3ng from the aesu~ance
of communications continuity in the system.
A saCellite in n"Molniya~' type orbit makes two revolutions every 24 hours
wiCh ascending nodal longi~udes spaced 180� apart. ~
If the orbital structure of the system of satellitea 3n these orbits is
soughCby working from the optimization of ita characteristics during oper-
aCion over one revolution, then the ascending node longitudes are equal to
a3 and A*3 + 180� respectively. Tn this case, we will call the revolution
with Che ascending node longitude which is opt3mal for the service area the
main revoluCion, and the revolution following 3t, the trailing orbit.
To determine the possibiliCy of using the satellite in Che trailing orbit,
the area is found wh;ich can be serviced during the time AC by the satellite
in the trail3ng orbit. In order to obtain as great an area as poasible, Che
sCarC of ~he communications sessi~n is optimized, beginning with Che minimal
possible start. The minimal possible start o~ a communications session is
found ~uet as for the case of the main orbit, by working from the operational
condi~ions of ~he on-board equipment of the satellite, the time At' Co pre-
- pare it for operation and the positioning of the control points, if the satel-
lite preparation for the start of a communications session is accomplished
via commands from the ground. Tf at the point in time t(the orbital time for
the entry of ~he control point into the sate113te zone of radio visibiliCy)
an instruction is fed to prepare the satellite for service, then the possible
minimum time #or the sta�rC of a conanunications session wi11 amount to t+ At'. _
By now defining the radio visibility zone (or the service zone) for the dura-
tion ~t sequentially for the points in time of the start of a communications
session, beginning with C+ At' having a certain time spacing, we obtain the
optimal start of a communications session in the trailing orbiC.
18
FOR OFFICIAL USE ONLY
~ z.s~:;:. _ : ; _
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
~'Olt 0~'FICIAL USE ONLY
The apecif ic fea~urea o~' a~~Mo1n~.ya~~ Cype orbi~ and tihe geographic arrange-
menC of the ~arrirory of ~he USSR pe~rm3t the utilizaCion of the main and
~railing orbiCs for communica~ions.-
As an example, we shall con~ider a ey~e~tiexq cdnaisting of four eatelli.Ces with
orbital pararnetierss hp~ 500 km; T~ tl 11 h 51 45 sec; i~ 63�; w d 285�;
a*~ = 68� east longitiude [e.1.].
For this ey~tem, communicationa time Q~ is ai.x houra. The southern boundary
of ~he 24-hour service area of the sate~7.3,~e sys~em in the main orbit is
shown in Figure 4.6a, while the dashed line showe'the service area in Che
trailing orb3t.
As can be seen frotn ~'igure 4.6a, during the main orbit the Cerritiory of ~he
USSR ie provided with comple~e radio service. The si.x hour session of the
trailing orbit provides for service to onlp a.porCitin of the USSR. By com-
paring rhe area~ serviced in the main and trailing revolutions, we note that
parC of the conmiunications statioris located in the cambined service area of
the main and trailing orbitis can emplop the satelli~es during both revolu-
eions. The stationa located in the shaded portion of the surface areae can
use the sa~elliCes during only one of Che revoluC3ons~
An example of a. 24-hour servic~ system using kwo satellites in orbit~ wt3h
Che indicated parameters, with Ehe exception of 7~~, is ahown in Figure 4.6b.
Here, for the ground stations which are located north of ~he boundary of the
service area of the satellites in the main and trailing orbita, 24-i~our com-
munications is provided by two sate113tes, ~fhere each of ~hem operates for
six hours in a revolution. The satellites ahould follow one another at a
spacing of T~/2, and in this case, the stations operate alternately, first
for 12 hours through rhe sa~ellites during ~heir flight in the mait~ orbit,
and ~hen for 12 hours during their flight in the trailing orbit.
The longitudes of the ascending no3es of the main and *railing orbits, in
order to assu~e the greateat surface area,should be arranged symmetrically
with respect to the service area. Tn the exanzple given here, Che longitudea
are 150� e.l. ~nd 30� w.l.
To assure continuous service for a spec3fied territory, all the satellites of
the s~stem should have identical paths, and consequently, identical Greenwich
longitudes of the ascending orbital nodes. Then, in order to guarantee the
requisite timewise shift for the passage of the equatorial plane, At, the
orbiral planes should be separated in absolute space under nominal conditions
by the amount of Aii = 2n/n or w3At = et2 when using only the main orbit, and
by the amount of DS2 = n/n when using both revolutions.
Such a separation of the orbital planes in absolute apace is assured through
thE appropriate selection of the launch times for placing Che satellites in
orbit in the process of setting up the system.
19 ~
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR OFFICIAL USE ONLY
-zo o~0 4n eo 8~ iooi~o iao ieo e0-ieo i4�-~zo ioo eo ~0-40 -so
,
eo eo
40 � ~ ~ 40 ~
.
20 20
i
0 Q
/
-40 -20
~ 40 40 60 e0100120140180190 18014014010080 gp 40 a� '
a) ~a~
~
-20 0 20 40 80 80100 t20140180180_180140120100 80 80 40 -20
~pe �
80 - S ~.I 80
40 -r- --i- 40
, ' ~
~0 0 20 40 80 80100120140180180 180140120100 80 80 40 o ~
el (b) ~
Figure 4.6. The 24-hour service area during Che main and the trail-
ing orbits for a system of four (a) and two (b) arti-
ficial earth satellites. ~
To deCermine the method of setting up a system of satellites which assure _
continuous service, it is convenient to introduce an idealized sysrem. We
shall undersCand such a system to be the orbital strucCure of satellites
traveling in orbits, the ascending nodal longitudes of which are separated
in absolute space in a uniform manner in a range of 0-360�, having a constant
rate of precession, while the remaining parameters (inclination, alCitude and
argumenC of perigee, Greenwich longitude of the ascending node) coincide and
remain constant during the exisCence of the satellites. By virtue of the
assumptions made concerning Che equality of the rates of precession of the
l,ongitudes of the ascending orbital nodes of a11 of the sysCem saCellites,
and concerning ~heir constancy, as well as concerning the unchanging nature
~f the remaining orbital parameters, including the Greenwich longitudes of
the ascending nodes, the draconic orbital periods of the satellites will be '
stable and constant over the entire time of existence.
In such an orbital structure, the satellites pass through the same points of
the orbit in equal time intervals (perigee, ascending node, apogee, etc.),
20
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
.FOR OFFZCIAL USE ONLY
r~ 1Y
i.e., synchronic3ty of sa~el~.iCe mo~ion and constian~ cond~.Cions for the con- '
duc~ o~ ~ha commun~ca~~ons sesa~ons tihrough each sa~e1~.~.~e are asaured, while
Che orbi~sl ascending node long~~udea in ~beo].u~e apace aYways rema~.n un~-
formly d3,s.~ributed ovex the range of 0--360�. The introduction of an ideal-
- ized sysCem permits a si.mpl~.fication of ~~tre process of se~Cing up and re-
�ining an acCual sys~em, s3nce the me~hod of set~ing up aceual. syateme in
this case cons:~s~s in assur3ng that an actuaY syaCem is maximally close tio
Ghe idealized one over Che considered period of exisCence of the eaCellites.
'We shall der~ve design formulas for the de~~rm3na~3on of ~he launch time of
Che sa~el7.i~es when se~ting up a system consisting of n satellites. We
shall make uae of ~he fac~ tha~ rhe longitude of the orbi~al ascending node
aC the momen~ oE insertion of Che satellires, ag, coinc:Ldes with the opCitnal
value of ~3, while the or,bi~a1 period during the inserC3on providea �or equal
satellite paths. Tn the system under consideration, balliatic continuity of
the possible communicaC~.ons sess3ona can be asaured if the sate113tes will
sequentially intersect the equa~tor at the ascend3ng node at time inCervals
of 2T~/n, while ~he duraCion of the conununicarions sessiona Chrough each
satellite wi11 be no less than 2T~/n. If the satelliCes ~re inaerted in
orbit from one launch site during a 24-hour day in accordance with an un-
changing plan and the time of ~ravel from the momenr of launch ro arrival
at the ascending node (~ta) remains constant for all satellites, then, ob-
viously, to creaCe an orbital ~tructure for satellites with the conditio~e
assumed above, the difference 3n the launch time of the satellites should -
be:
~ t~,. =`lTQ/ri. (4.32) .
After the satellites are in orbit, the longitudea of the orbital ascending
nodes of Che sys~em satellites in absoluCe apace will prove to be spaced
apart by the angle:
(4.33)
~S2 = 52~-52 i, ~ .
where � .
rAe where .
SZ2 = Sp'~' A,e'I' (il 3~tcr s-r ~~a-~Y~ ;
, s~, = so+~.. ; W3(t~T oce-s~+ s~et~T;
t~ ~ _ ~~T ~ + ot~T ; .
~ (4.34)
SZ2 is the longitude of the ascending node of the last satellite at Che moment
it arrives at the ascending node; i21 is the longitude of the ascending node
of the leading satellite, Caking into account the precession of iCs orbital
plane at Che moment of passing through the ascending node of the next 8ate1-
l.ite; Sp is the sidereal tine at Greenwich midnight of the launch date;
21
FOR OFFICIAI~ USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
;
FOR OFFICLAL USE ONLY
~ t 1 gnd t~~ 2~re tihe Moecow leunch timee gor th~ leading and lAet earei-
- 1~fCeA raspecCively. ;
By BubstiCut3ng rhe expreesione for tt1 and R2 in ~4.33)~ C~king (4.32) inCo
account, w@ finds -
~A~ (~,~a-.p~ ~._.r~..p_,
(4.35)
.
Coneid~ring ~ha gact thgt Che eCable orbitial period T~ti (the draconic orbi- -
t~i p~riod eorresponding to the etabie path) ie defined ~eking (3.1) into
a~:counC~ winc ~he ~cpreesiont
TQ a T~* = s;~(c~~3 - ~2) , (4.36) ' .
we g~nd Char
~S3 ~1rc/n. (4.37)
It followg from (4.37) that the launch time for each succeesive eatellite of ~:Y
the eyetem, consisting of n eaCelliCes, ehould be choeen from the condition
that the longitudes of the aecending node of its orbit and the arbit of the
].eading sateilite, taking the precession rate into account, are spaced an
angle of 2n/n apart. Aa cgn be seen, in this formulation, to determine the
launch rime oE the next eatelliCe miniiml information is required concerning
the orbit of the preceding one, and specifically, only the value of the
longitude of its ascending node at the moment of the impending paseage of
the equator by the satellite being launched, taking into account the rate of
preceagion of the orbital planee.
Since the planes of the orbits of an idealized system have the same and con-
stant rate of preceasion, the angle between the two adjacent orbital plttnes,
which ie obCained ic~ designing the system~ will be preserved throughout the
entire time of functioning of the system. Md since the orbital period of
ali satellites ia equal to T~, then there will also be preserved a conetant
interval in the tima of equatorial passage by the satellites. This specific
~eature musC be considered in setting up systems, when the setup ie not com-
pleted within one 24 hour period, but ie epaced over a longer period of Cime.
Another approach is possible to the determination of the satellite launch
time when establishing an orbiCal atructure for eatellites where ag ~ 1~3;
Tg = Tn ~ T~t, if only the only the data and time o� the launch of the
previoua satellite are known. In this case, for the launch date of the next
satellite, it is necessary to find the time the preceding eatellite passes
Chrough the equator in the main orbital revolutions
f.~~-tr = tcTu_~? ~t, (N--1) (Tu - Tc/2) 24.r., f4.38) ~4.38)
hz
22
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
, ,
t
e
. !
POR OPFICIAL U88 ONtY ~
,
wh~rp tCt ~~..L~ i~ the launch tim~ og the preCeding (i - 1)rh eaC~liite=
'Ce ar~ Aolar days; N is the number of rhe rnain orbi.tal r~volutioa of rhe ~
pr~~eding aataii~te on the launch dare of the naxt eaCelliCe; K is gn inte-
ger, daeerminad lrom the aquations ~
.
- x ~ ~ ~-~~,r~~~~ ~.et~ tN (Ta - Tc1~)_~ i-
.
, � i
indientee rha integer potition). i
The launch time of the n~xr eateliite te determinad from Che expreeeions ~
~~t t~~~_~? et, o~ tz ~ ` `
ct nt~-t? ~ dtit ~4.39)
~r~->> ~~ra - ~r~i~). ;
IC c~n be ehovc? ttuit aith Chi~ method o~ eel.ecting the launch timee, the -
dil~erence in the valuee o! the longitudee of the eecending nodes in abeo-
luCe space, taking preceaeion into account, aiii aiso amount to 2n/n.
Un a certain date, the sidereai time at Greenwich midnight of vhich ~e equai
to 30(i-1)~ let the (i - 1)th eatellite of the syerem be leunched at the
poine in time t~r i~). The eecending node longirude of ite orbit during
the liret p~eeage of tihe ~quacar viil bet
~~_s~: go~~_~~ ,ua~tc~~~-~? + ~t, -3h~ ~4.40)
L~t the next sat~llite be launched after the preceding one has pasaed the
ascending nede in the N-th main revoluCion, the longitude of the ascending
node of i~s orbit aC the poinC in time of equatoriei pasaage iss ~
- So, �i- t~3 (ta ~ ,t~ ~3y~ ~l~ ~ (4.41)
EtY
while the longitude of the ascending node of the orbit of the (i - 1)th
ob~ect at this point in time is
~~-t = So~~-t) ~r w3~ccr(1-t~ 0~~ -3~1) -4~
1 ~ f2(Ta(N-i ) +.~tnJ. ~ ~ (4.42)
then bf2 - So~ --Sai_~t 1' ~~~~a~ ~cr~~p�t1)
~'2~'I'o(~I-'~ 1-1-~tcr~ ~
or taking (4.38), (4.32) and (4.363 into account, we find:
Ofl-cc~ --~oii-i~ n([Y-i)+ Z=-wa(N-i) -T=� (4.43)
n Z
23 -
POR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR OFFICIAL U8~ ONLX ~
'Chg m~gn O~e~nwl,nh C~.me nC midnigllC on rhe ~0uncl? d~Ca of eh~ i.-~11 ~AC~11iCe
c~n be expreeeed in ~erme o~ Che me~n Greenwtch ~im~ at midnight of the .
launch dae~ of the (i - 1)~h aatellire ue3ng Che formalat
So~ � S~t~-~~ wa~Te T3) ~
- ~ ~ (4.44)
3ubatiCuCing (4.44) in (4.43), and coneidetiing the facr Chat T3w g� 2n, we
Eind ~Sl ~ 2n/n, BomeC~hin~ ~hich was eeeerCed above.
The re~ult obtained h~re aCtesCs to rhe complere equival@nce og the golutions
Eor the selection og tihe launch time of communica~ione eateliites in a
"Molniya" type orbit in the two formuiatione degcrihed abnve. :
- The following conclusion follows from ali Chat hae been said aboves if in en
ideglized eystem of n eaCel].ites in "Molniya" orbirs, the ascending orbital
nodea are spaced an angie of 2n/n apart, rhen the timee when the sateliites
pasa the equator are spaced 2T~/n apart regardleee of the time interval uaed
in ee~~ing up the sysCem. In this caee, Che orbi~ai periods of all ChQ eatel-
7.i~es of th~ syet~m should coincide and be equai to Che stable figure defined .
from formula (3.1).
All of the satellites of the system indicaCed here have coincident paths of
travcl o~ the poin~ under the satelliCe on the surface of the earth and identi-
cal. val,u~s og the Creenwich longitude o! the ascending node, equel to ~?B.
If the satellite orbital insertion longitude aB doea not agree with Che
optimal longitude of the ascending node aa~ then the sarellites must be ,
launched with a lead correction with respect to the orbital period ~TB so
that following the completion of N revolutione and the correction of the
orbirnL period with the establishing of the stable orbit, the Greenwich
longitudes of the ascending nodes of their orbits are equal to 7?3. In this
case, a constant angle of 2n/n is preserved tetween the orbital planea if
the difference in the time of equitorial passage by adjacent satellites is
equal to 2T~/n. We shall prove this, noting beforehand that the time of
equatorial passage by satellites traveling in an orbit with a stable paCh
(Tn s TcC) hsia a consCant departure in the direction of an earlier time.
In fact, the time a satellite passes through the N-Ch node can be repre-
sented in Che fonu
t~h � (N-i)~'f�-- 2` ~ 4' [l+(-1)N)-29~ix: ~4 45)
.
where K is an integer, determined from the equality: '
r~ r~
t~~+(N-I)~To- ~ 4 (i+(--1)~~
~ ~r C ~4 ~
(E indicates the integer part, while the time is measured in hours).
Z4
FOR OPFICIAL USE ONLY ~
_ , _ _ _
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
_ .
~
FOR O~FICIAL U8B ONLY :~s
F~
~
~
Wa eha11 reprasant the dxaconic orbieai pa~iod in the form ~
t.
'Tp ~ m~~ ~i- DT,,
(4.46)
i,
A~Cer subet~tuCing !~4.46) in (4.45)~ we obrain
E~N e~ t~ ~~(N.-. rT~ pT~ ~ 1.r J
+~-t~+(--~)`~~+24aK. ~ t4.47)
4 ~
In the caee where 6Tg ! 0~ i.e.~ When T~ � rct=
~
~?H = ~ ~ (IV--t ) ~T~t - ~1 -T-~�-(1 ~ (-1)Nj +2~1~ src, (4.48)
As can be seen from (4.48)~ the time a eatellits paeses the equator moving
in an orbit wieh a sCable paeh, there ie a d~viation in the dire~tion of an
earl3er tiime (T~t < T~/2) by tihe amount of ~C * T~t - T~/2 per revolution.
When ATg ~ 0, there occurs an additional dev~ia,`.~ton in the rime of equaroriai
paesage, equal to ~Tg per revolution.
We sha11 find the relationehip between the additional dieplacement of the :
satellite in the�time of equatorial paesage and the deparCure of the longi-
rude of the node relative to the (~re~nwich eyeCem.
~
We shall deeignate the difference in the Greenaich longitudes of the N-th ~
and the firet of the ascending aodes as eaa, Thena ~
tll~, d (N-1) OT. (~a - t2) - ~-I ~ + (--1)Nj stgn (~T.~. ~
From thie ~?e f ind s ~
(N-i) AT. = ~T~, -~-~(1+(-11Nlsigm IDT,~. ~4.49)
Substituting (4.49) in (4.47), we obtain a formula for the determination of
the ahiFC in the time of passage of the N-th node relative to the firsts
~6~N r: ~T~ _ ~ ~ ~ .
+ Z" (l+(-I)HJ sign (AT.) ; 4` ~I+(-1)Nj-~24~tx. (4.50) -
As can be seen from (4.50), besides the conetanC timewise shift in equatorial
paseage, related to the difference in the draconic period from half of a solar
day, chere occurs an additional ehift equal to aJ1aT~t/n, which depende only
on the amount of departure of the gscending node longitude aaa. Since this
shift does not depend on the number of the revolution, then in a aqstem of ~
n satellites, the longitudes of the ascending nodes of Which on a particular
25
- FOR OFF:CIAI. USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR O1~FICTAL UaE ONLY
d
daee f~llowin~ ~h~ compleCion o� Ch~ ~.nserC~.on proc~~e aa~ume value~ of
X~~, Che eame C~m~wiee ~t~i.EC ~.n equa~orial paeaage wili occur foti each ~aCel-
liC~. Thie ehife wiil have no effece on tihe intervai ati which tihe eat~1].ires
~ollow onQ nnoCher, i.Q., ie wil]. remain Che enme gs it waa for Che cn~e af
:InsexCion ~.nto a e~able peri.od and nt the ineerCion lungitude, which ie ~qunl
to rh~ oprimal longitude. 13ut for Chie caea, it was shown above thaC the
orbi~gl p].anes r~main spaced 2tt/n apare ig tihe time when the eatelliCes paes
the ~quaror ralativ~ to each other ig 2T~/n regardless of rhe length o� time
in th~ process of e~titing up the syeCem. Th~ assertion formulated above is
thug proved. If we now equate tihe longitude o~ the aecending node of the
orbits of an idealized sy~Cem Co Che longiCude of ehe ascending node of ac-
Cual eystema, then all of Che conclueione derived here can be employed in
the design of aceual satell.~te communications ayetema. The deaign ~f a eya-
eem cgn sCarC gti any launch d~te for the fireC sarellire into one of the
pl~nes witih an arbitrary longiCude nl. Moecow launch time is degined in
accordance wieh (4.34) Erom the formula
Ri-se-~~ ~p~~-4~3y, (4.Si)
tet t = ~e3
At the seare di the N-th revolution, the longitude of the ascending node of
[his satelliee orbit will amount to
_ n~cN~Ln,+siT~tcN--~~. c4.sz~
If tlie i-th eatellite is to be launched on the date correaponding to the
N-th revolution of the first satellite, then in accordance ~rith that pr~e-
aenCed above, it is sufficient to provide for its inserrion into the plane
with n longitude of the ascending node determined fYOm the formula:
` = ~~t+S2T~t~hT-1) 2 fn ~u'f�1 (i-l)} . (4.53)
o-st
The last term characterizes the ahift in the orbital plane of Che i-th
satellite relative to the plane of the first, and {}p_2n aignifiea the
resulC of referencing the angle to the 0--2n range. Then the launch time _
for the i-th satellite will be defined by the expression:
f?~-So~ -~e - ~t,~ -F3~t, ~4.54)
tn ~ - ~3
wheze Spi is the sidereal time aC Greenwich midnight of the launch date.
Expression (4.54) can be transfora?ed, by switching from the use of (4.44) `
from Spi to the aidereal tin?e on the launch date for the first saCellite
and employing (4.53), to the form: . , � .
C _ ( S2o-So-~n JL~ ~;-~4'~' ('t-~~ -f-
. cT I i �3 r " (4.55)
+ rT~, Z` ) (N-1) ~~p .
~
26 -
~ FOR O~~ICIAL U5E ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR OFFICiAL USE ONLY
The resu].Cing axpresei~n coincides with (4.39); {}CD means eh~C tihe Yeeult
- regerenced to Mo~cow C~.me. We wi11 noCe ~haC (4.5S) provides #or Che
deCerminarion of the launch ~~ne when complering ~he eyetem.
b.2. The Determinat~.on of the 3ervice Areae and the Radio Interferen~e
T~king into AceounC the Errors in the Orienee~ion of the On-Boerd
Transmirting Mtenna
The method of calculating the eervice area of a erationary satellite treared
in the preceding section does not take ~nCo eccoune the errore of tihe orien-
tation of Che on-board transmitting anrenna. In ac~uai facti, such errore
aiways exisC, and it ie quite underetandable tha~ they ehouid be coneidered
when deCermining the characCerie~ics of a cc~mmunicatione sysCem.
If the on-board antenna is rigidly mounCed in the satellite, than the pre-
cieion of ite orientaeion is compleCely determinpd by the precision in the
orientaCion of the sateilite.
For an antenna having a drive which can be oriented independenCly of the
satellite towarda the service area, the precision in ite orientaCion is
deCermined by the characterietics of the antenna guidanc~ eyetem i~self.
We ehall introduce an orbiCal ayetem of coordinatea: SXYZ. We ehall place
the origin of the coordinaCea in the center of mase of the eateliike, di-
rect the SX axis along the radius vector towards the center of the earth,
and direct the SZ axis along the traneversal in the direction of motion of
the eatellite. The SY axis completes the syatem. Let the satellite in its
travel in the orbit be oriented with one of the axes tovarde the center of ~
the earth, wiCh a precision characterized by rotations With respect to the
SZ axis through an angle Nk~p (roll angle), With reapect to the SY axis
through an angle Nt~p (pitch angle) and with respect to the SX axie through -
an angle Np~p (yaw angle), and let the on-board antenna be rigidly coupled ~
to the eatellite~ while the axis of its directional pattern is direcCed
towarda a point C on the earth's surface having geographic coordinates of
and 7?~. The directian af its axis in thia case can be determined by the
angles ~ and n.
The angle ~ te the angle between the plane Y~ 0(Che plane of the orbit~
and rhe plane passing through the SZ axie and the point C. The angle n ie
the angle betWeen the plane 2~ 0 and a straighC line passing through the
points S and C.
Because of errora in the orientaCion of the satellite, the axis of the on-
board antenna in ehe general case describes a compiex figure, which can be
approximated by a pyramid With the vertex at the point S and ~ith angles
's
i
27
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
,
~OR OFFIGIAL U3E ONLY
~r tihe verCex o� Nti gnd Nk (F~.gure 6.6). The~~ angles dat~rmine ~he devia-
Cions of ~he aneenna axls wi~h respect ~o piCch and tioll reepeceively, where
the angle N~ characterixes the shift of ~he on-bonrd an~enna axis in Che
north-easC direcCion, while the angle Nr applies Co the east-wes~ direction
r~~nCiv~ Co tihe direction 5C. W~ sha11 ~eeume ~hae Ch~ ang~.e Nk ie poeitive
when the antenna ax3e ia incl3ned to Che norrh from tihe S2C plan~, while the ,
~ngle N~ ie positiv~ whan the d3r~ction of the an~enna axie is inclined tio
~he east from the SYC p1~ng. The values Nk and Nt, depending on tihe anglea
n and the precision in th~ orientation of Che NC, Nk and Np axes of th~
serelliee can ba der~.ved from tihe foilowing expreasions
NT ~ Nr,o ~ ~
arctgr sin r~ co~ g- sln r~ cos g(1- cos N p,e) -
L ~ +1 . � (6.23)
cos cos
aln atn N p.o~ ,
~ ~ C.~S C~4 ~ 1
NK = NK,o ~ E (6.23)
-etctg r atn ~ cos ri - sln I- cos N p,o) -
~ ~ I y
L COS ~ COS 1~
atn ~ cos g stn Np,o 1
� � � cos E cws ~ ~
In the case of the abaence of eatellite orientation errors aith reapecC Co
the yaw channel ~Np~~ Q 0), the errors in the orientation of the eatellite
are converted directly to orientation errore of the on~board antenna.
~xpreasions (6.23) for several estimates can be aimplified by means of re-
plgCing the trigonometric functions of small anglea, x, by their ~rgumenCst
e~n x~ x, ana cos x~ 1- x2/2.
5uch a substitution can be made even for the angles ~ and n, since their
values will noe exceed 8� in Che ma~ority of casea. Then, taking What has
been said into account, it follows from (6.23) that: .
Nr = Nr.o -f'~N p.o ~ NK = NK.o -{�t~N o.o . (6.24)
For nntennas which have a drive, the precision in the orientation of the on-
board Cransmirting axis is specified directly by the valuea Nt, Nk and Np,
which are determined by the characteristics of the system. When the axis
of Che antenna is moved within the limits of the angles Nt and Nk there
always exists for each angle a~`ij, which defines the positiorr of the poirtts
Ni~, one worst case position of the anCenna axis, defined by the conditions: _
sign NK ~-sign (cos a;~); sign Nt ~sign (sin a~~) .
(6.25)
28
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
ROk Ub'H'tCtAL U~J13 1)NLY
~ ~ We sha11 now coneider a certain point N~,~ (gee
~N`~ I P~gure 6.2). Th~ gain og rhe on-board anrenna
~ i.e the direction toararda the po~,nt N~,~ ie de-
~ f ined ~n Cerme of Che angl~as 0~~ , a~,~ and a k,
: i,f ir is aeeumed tihaC th~ ~x~~ of eh~ ~nti~nne ie
_Nr 'N, d~recred towarde rhe point C. rg or3enCation
fNN errore N~ and N, Are bYOUghe into coneidera~ion~
a~ g~ , then the poesib~e direction of the antanna axis
can be echematicaily bounded by a rectangle hav-
Q ing sides i.n anguiar meaeure of 2Nti and 2Nk ~n
~ eome plane perpend3cular to the line SC (Figure
. ' 6.7). Then rhe aegiee a~~ and Ai ~ which gre
, .N,,, depicred echematicaliy in Figure ~.7~ wiil cor-'
respond to the nominal oY3entation og the anteana
LM ~ I;~~. 4 ..~.s.. axie~ towarde the poinC C. The aorst caee point
Figure 6.7. On ~he de- ~or the direction af the antenna axis when de-
termination of the woret te~~nieg the gain in the directinn to the point
caee or3entation of the N~~ ~11 be the point of the rectangle abtained
antenna axie. in accordanca with (6.25). To determine ~t~e
gain of the on-board entenna in the direction
towards the point N~~, taking into account rhe
orientation errors Nr and Nk, it is necessary to determine the angies
and a"~~, correspnnding to the aorst case poeition of the anCenna axis.
The expressione for the determination of ~i~ and a"i~ have the forao
. _
COS 9~~ ~
= CoS 9~~ I- tg N* tg 9iflstn a~~~ - tg N~tQ A i~Ico~ aii~
Y I+ta~Nt+ta=Nk '
~
cos a,~ _
tg Aq cos a~~ +tg NK
a ~
Y Itg 6~~~stn aj1l'}' tB NrN +(tQ 9i1) cos a~tI +~Q N~ 1'
sin a,~ = .
tgA~~stna~~ +tgN,
Y(~B e~l~sin a~1I + tQ N,)' +(ta 9U~cos aii~'~ ~g N�~' �
cs.~s~
(6.26)
The angles Ai~ and a
i~ which enter into the expressions of (6.26) are de-
termined from (6.12) and (6.13). The condition which determines the
boundary of the service area in the direction aith an azimuth Ai from Che
center of the orientation, ahich is similar to Che condition (6.18) of 46.1,
can be ~?ritten:
29
FOR OFFICIAI. USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
, Y.
i
I~
-
.
FOR OFFICZAL U9E ONLY
~~~~"~ty~ai~~I Gg, ' (6.27)
, ~ ~ _
To da~erroine a~,~, it is nacesea~y to coneider Che prenieion in the orien- .
CaCion of tihe an~enna as regarde yaw (roraeion about ehe line SC)t
Np � Np.p '
and the poseibie etructural design angle for Che turning of the antenna on
iCe axis, ~
The necessiry of etructur811y turning the antenna can be due to either Che
configuraCion of the eervice area with reepect to the position of Che saeel-
liCe~ or Cha naceeeity of setiafying several limitations. The quaneity a~~ ~
is determined from the relationahipt
(6.~8)
aii :r (a~~ ao?)Ypsign(sin(r~,~ ~K)rc~s(,u,~ ~ ICi,"~)
, . '
_ artien npN ~ sln (ai~- a~) I>sin Np (6.29)
and u sin a~t ~ fE'~.?~:~) ~
when np~~ ~ sin (ui~ - aKl I~sii~ ~p '
sign(cos ~,~)~sign(cos(~c,~ -ti~)~.
The eupplemental term ~f+$iqtt(sin(xi~ ry~)cns(ic,, T�! ~ yields a reduction or
an increase in the angle a i~ by the amounC of the precieion in the orientation
w3th reapece Co the qaw channel in such a manner that the antenna gain is the
loweet. Sxpreseion (6.29) meane that ig `rithin a prec3sion o~ Np the minor
semi-axie of the directional pattern coincides with Che point Ni , then the
angle a i~ ie to be taken as equal to either 0 or 180�, so that t~e antenna
gain is the lo~eaC. All the re~aining conditions~ as, for example, Che con-
diCion of assuring an elevation of no leea than the permiasible Y~, ahould
be mee. We will note that if it proves poseible to determine the gain of
the antenna analytically as a function of Ai~ and a i~, then condition (6.27)
is to be reduced to the form:
IG(~~~. I~~) " (~j,p(~t;~ 'ii;)i '
Where D~~ and Yi~ are found from (6.17). Y
The method remains justified for a8ymmetrical directional patterna. In this
case, it is sufficient to change (6.2) in an appropriate manner, retaining =
_ ~he correspondence of the small aperture of the directional pattern to the i
value of the angle a~ 0.
Along with the problem of determining the reliable reception area (the service '
area), there is aleo subatantial interest in the problem of fir~ding the radio
interference region produced by ground services, especially when using high ~
poaer eatellitea in the comaunicatio~~s system, for example~ those at?ich '
30
FOR OFFICIAI. USE ONLY
;
, ~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR OFFZCIAL U3E ONLY -
prov~,de for ehe recep~ion og a Celevision eignal directly at ehe home an-
tennas oP usere~
~he permiesible interfering eignal level is regulaCed by the no~nns of ~he
VAKR [Worldt,rl,de Admin~.e~rg~ive Rndio Commun3ca~3ons Conference on SpQCe
Communications~ [23). Thie levei depende on the bro~dcaet frequency and is
a funcC3on of Che angle to Che local horizon at which Che interfering signel
arrives at the point ~nder coneideration.
Up Co a cerrain value of the elevation angle Y, a conetiant power flux deneity
~.a perm~.rted, and thereafter, with an increase in Y, the permieaibl.e value
o� Che powex flux density increases by virtue of the spatial selection of
Che 3neerfering aignal and Che received usaful aignal, radiated by another
ground etarion. Thus, for exampie~ in accoxdance w~,th VAKR e~amdards [23~,
ehe tn~erfer~ng power llux dens3Cy of ~he intierfering eignai at Che euxface
of Che earCh at a frequency og 300 MHz should eat3efy the folloaing condi-
~i.ons t
- 129 dBW/m2 at Y~ 20�;
- 129 dBW/m2 + 0.4(Y - 20�) at 20� < Y~ 60�;
- 113 dBW/m2 at Y> 60�.
The methoda presented in ~6.1 and 6.2 can be employed ro calculate the radio
interference area, Caking ehe following into account. Since when calculating
the radio inCerference zone it is necessary to piace an upper limit on Che
1eve1 of the interfering signal, the preciaion in the orientation witt~ re-
spect to pitch, roll and yaw is to be chosen so that the antenna gain in the
direction towards the point under conaideration ie the higheat. For Chis
reaeon, when the following conditions are met:
I sin A,~ sin ai~~ ~ ~ sin NT (
y ~td
~sin 9,~ cos ai~~ sin N~ ~
The following expressions are to be uaed for the determination of ~i~ and
a i~ ' cos 9,~ = cos 6,~ X ,
x 1+Ig Nttg A tl~stn a~l~ " ~B NK~B ~i! (cos Q~I~ ~
Y ~ +ta~ Nt +tQ~ N ~ ~ ' '
cos a,~ _
~ ~B eit cos ai~ -!g N� ~
V (lg9~J~sina~~~+tgNt)!-h (lg~qlcosc~~~+tgN�)! ~ ,
sin a,~ _
~ tgAi~ sIn ai~-lgNt ~
Y(tg B~~IsIn a~~l +tg NT)~ +(Rg 9~~Icos ai~~ +tg N~)~ �
(6.30)
;
F
31
FOR OPFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
~
FOR OFF'ICtAL USE ONLY
Zf only the Eollowing condir~.on ~.s meti, ,
.
~ sin 9 i~ s{n a~~ ~ sln N r
Then Cha values of and a"~~ must be found �rom ~he Eormulae:
. I~t NK t~,9i1_ ,
COSA11,~'COSA~~ t+tg~ f~tg
N ~
(6.31)
sin a i~
sign (cos au ) ~ sign (cos ai~
And ~inally, when the following condi.Cion ~.g mets
~ sin 9 u cos a,~ ~ sin N K ~
~ r
The following relationships should be employed tio compute A~~ and a3~ s
cos 6 i~ ~ cos 8 ~~'t N T t A u ,
l~htg~Nr+tg'NK ~ C I~
. r Cv~JL, , . ~ .
C.~s di~) ~Oj .
si n sina =si n sinai . ~
B ~ U) B ~ I) .
6.3. The Derermination of the Requisite DirecCional Pa~tern and Setting
Angles for the On-Board Antenna
To aelect the preliminary characCeristica of the direckional pattern of the
on-board anCennas of a stationary satell~.te and determine their setting angles
on the satellite, it is convenient to represent the surface of the globe
visible from Che stationary satellite in apherical coordinates, related to
the satellite.
T ^ ~ ` ' ~ � ~ ' ~ ' ' ~ We shall 3ntroduce a spherical eystem
~ ~ of coordinates, and place the origin
p ' of the syatem in the center of mass of
~ the satellite S. We shall pass the
- plane SOP through the point S and
; N the rotational axis of the earth (Fig- -
; o p ure 6.8). We shall define the posi-
: ~ tion of any point N on the surface
' of the earth in Che system of coordi-
: natea introduced here in terma of the
; S~ angle a and S. The angle a represents
, � the angle between the plane SOP and Che
~ r
Ms plane ~passed through the points S and
N perpendicular to the equatorial
plane. The angle a is positive for -
Pigure 6.8. The spherical ~ystem points on the earth~s aurface, which
of satellite coordi- are located to the east of the meridian
nates. below the saCellite and negative for
32
FOR OFFICIAL USE ONLY
ta, , - _ , ~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
.
. x
~ ~
FOR OFFICIAL U3E ONLY �
, poin~e locg~ed Co ~he wes~ n~ ~he merid~,an beloW Che eaCellite. ~he angle
, R 38 the angle beCween Che direc~ion to ehe poin~ N on tihe earCh~s eutifnce
~nd Ch~ equa~orial plane. Then eha enCire surface of ~he globe vis~ble �rom
~ e~a~iona~ry eaCelliCe can be representied in a coordina~e grid of a and ~
in tihe fo~.lo~rl,ng manner. ~For ~he coordinaCes a and R of any poin~ on
~he ear~h's sur�'aca, a and ~ can be de~ermined 3n accordance with ~igure 6.8
f rom the f ormulae t ~ sin ,
~~arotg d ,
a~ aresin R
cos
s~n e~_ ~ (6.33)
where .
d~'?~ r+ R' cos~ q,-Rr cos ~ cos 01~; 0~ ~ 1~--~';
~ ia the geocenCric latitude of the poin~ N at the earth~e aurface. ~
Beceuse of the simplified model for the determina~ion of the on-board anCenna
chaYaceerietics, one can adopt the geographic lati~ude in place of the geo-
cenCric and the average value of the radiue vector of the point on the earth's
surgace in place of ita true value. In ~hie case, the error in the coordi-
natea a and S will noe exceed 1.5~. Ueing formulae (6.33), a11 the meridane
and parallels can be represented in the coordinates a and Tn a eimilar
mgnner, lines of equal ranges or elevation anglea can be draWn on the coordi-
naCe grid, which take the form of circles concenCr3c to the circle of Che
ou~er outline of the earth. The radiua in angular meaeure of circlea of
equal ranges and elevation angles can be determined in accordance wiCh Figure
6.9 �raa the relationshipa:
8= aresin ( d sin ; . (6.34)
- deYr~+R'-Rrcoscp; ya90�--~-b.
In this case, a po3nt with a latitude ~ is choaen at the meridian of the
_ point beneath the satellite. Then, the fo1lo~,ring expreseion can be used Co
find the rela~ionahip between a and s, which correaponds to equai range d
and elevation angle y:
cos a a cos 8/cos ~ (6. 35)
or, considering the small eize of the angles a, S aud d, one can t~ttrite
approxfmately: ~
as+~~~8~. (6.36) -
e
~
Expression (6.36) provides an error of no more than 1~ in the determinetion ~
of the angles a and S. The coordinate grid of the visible portion of the
globe plotted in this manner ie depicted in Pigure 6.10.
;
~
33 t
;
FOR OFFICIAL USE ONLY ~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
, :
~ ~
~ .T, ~z".
~ FOR OFFZCIAL USE ONLY ;
. _ ~
~
~ i
i
The 1~,neg of ~he angu~.ar di.men-
. e3oned grid o� Che spher~,cal
~ sys~em of coo~d~.na~es ~eferenced
Co ~he ea~el~.iCes are indicated ~
y R with tihe le~~ere D and E in ~
~ e ~ Figure 6.10, along the horizon~al :
' and verCical reapecCively. The � ~
~ parallels are represenCed by the
,
. B lines and the merid3ans by tihe
' , ' C linea. The linea of equal
ranges and eleva~ion angles are ;t.
Figure 6.9. On .the construction~of deeignated by the let~er A. 1-
lines of equal ranges zti 8hould be emphasized ~hati tihe ;
and elevatiion angles. L;nes of Che requisite gaina co- ~
3ncide wiCh th~ linea of equal
ranges and eleva~ions. ~
An terr3tary of the earth~e surface visible from the satellite can be mapped ~`j
in Che a and S coordina~e sysCem. Thus, the territory of Che Sovie~ Union, ~
viaible from a sCationary saCellite located a~ 100� east longi~ude is depicted i
in Figut~e 6.11. 3
, ,i
~A -5 -~1 -3 -4 -1 0 I 4 3 4
� ;
p~ A.NN A g ~ ~
D r � . r. 6 $
, e~
g 80' ~8~ 0_ 8 f
55` ` 3~ ~
~1Z \ i
A A SO' ~30~1 �1 ~ � t
~ y "j3~ . .a?, ~ ~ ~ ;
n
~ j
C 40' 3 5 ~ ~ 1.
g 1
e - g / - - \
_ _ ;
3K� � _ 1?0 - , ;
4g 4
6 - 5 , ; ~
_ 30� ~g - 5 1 1
~ S ~ ~
~ . ?5' 3e 8
0 ' .
' ~
i
. ~
-50 -4U -30 -40 -10 0 10 ~0 30 a � ~
Figure 6.10. The earth in a spherical system og coord~nates referenced ,
to the satellite.
i
Now let it be necessary for the servicing of a certain area to determine the '
approximate characteristics of ~he directional pat~ern.
,
34
FQR OFFICIAL USE ONLY
.a. a.. . . ' ' _ ~ _ ' a,. . _ . s . ,`;..a~:' _ _ . . i~~~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
J.. ~ . . . . . ~ .
l i' �
FOR OFFICIAL USE ONLY ~
The aperCure de~ined wiCh respec~ eo a and ~ corresponds to ~he specified
' Cerri~ory ~o be aerviced and referenced ~o the coord~natea a and S. Thus,
a solid ang~.e aper~ure w3Ch respec~ to a of 3� and wi~h respecC Co S of
~..5� corresponds to Che por~ion oE rhe ~erri~ory of ~he USSR lying be~ween
85� eas~ long3~ude and 115� eae~ longiCude for a sa~elli~e ati 100� east
longi~ude.
-7 -9 -6 -4 -3, -2 -1 0 I 4 3 4 a,
: ~u
H - 6
7 4 7
40
8 3 8
6 6
2
60 60 70 80 90 100 IIO 120 130 A�
Figure 6.11. The Cerritory of the USSR in a spherical sysCem of
coordinates referenced to a satellite (7?* = 100� easC
longitude).
The result3ng value for the solid angle aperture should correspond to the
aperture of the directional pattern of the on-board transmitting anCenna.
The gain in the aperture is defined with respect to Che reception area at
the maximum reception range and ~o the corresponding minimal elevation angle
at the reception point. In this case, it fs determir~ed by the range and the
elevation corresponding to ~he point in the territory with the greatest value
of the angle S. In this case, the aperture of the directional pattern must
be increased by the amount of the orientation error.
Simultaneously with this, the coordinates of the point on the earCh's surface
_ towards which the electrical axis of the directional pattern of the on-board
antenna and the setting angle of the antenna in the satellite should be
directed, can be determined in a first approximation. For example, if the
system SXYZ is chosen as the satellite system of coordinates, where the
SX axis is direcCed towards the center of the earth, the SZ axis is directed
_ along the binormal t~o the orbit and the SY axis compleCes the system making
it rectilinear, Chen the antenna mounCing can be specified with three angles:
the two angles ~ and n for the corresponding center of orientation.and by the
angle a k, the angle of antenna rotation with respect to the direcrional pat-
tern. Thus, for our example, the axis of the on-board antenna directional
pattern should be a~tned at the point ~rTth coordinates of 57� north la~i- ~
tude, 11 = 100� east longitude, while the on-board antenna should be mounted
35
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR 0~~'ICtAt U5L ONLY
gn thae the axie o� th~ di.recCionel pattern ie 7� 4S~ and r1 � 0(in thie
cae~~ a� 0). Yn this inetance~ i~ ie nor gt ~11 ne~~~~~~y eh~e Che eBYV~c~d
area be arr~nged eymmeCrically aith reapect Co Che m~r3dian beneath the a~Cei-
lie~. In this caee, ehe cenCer of Che orientation 3e choeen in a simii~r
fashion, whil~ the direcCion of Che elecCricgi axi~ of ehe on-b~atid direc-
Cionai paCCarn ie gseured by the nppropriate mounting of the antenna with
respecC ro Che axee of the g~CelliCe.
When iC ie necesgary to coneider limitations on the radigted eignal powQr
(in accordance with the rc~quirements atipulated by varioue accords) fo~ a
cert~in CerriCory~ deCermined by the coordinates ~ and a, 1imiCatione on the
power of rt~e on-bo~rd Cran~nitting anCenna cae be drawn in figuree auch e9
delinegt~d in Figur~ 6.11 using eyp~ A curves for the correeponding valuee
of ~ nnd a.
In order eo show the ndvantage of the repregent~tion given here Eor Che portion
of the earth~s surface vieible from a~atellite, lines of observaCion ~nglee
frnm a satellite located gti 100� east longieude are shown in Pigure 6.12
(lines o� an angular dimeneioned grid of rhe spherical ~yatem of eatellit~ _
coordinatea), for the semi-conical pro~ection of the visible portion of the
northern hemisphere. Although Chis representation yields the same functions
as thoec shown in Figure 6.12, iC is a more complex matter to ud: thie repre-
eentaCion. io io ao ~o ~0 9o iio iao iso no i~o ~ao
-~o o zo ~o eo eo ~oo ~ao ~.o ~eo ieo -ieo -i~o
s~ � , ~
~ ~
, � `
70 ~ . ` ~
.9 , ~ t ~
~ ~
60 ~4~ - ~ - - ' ~ ~
~.9 1 ` ; , ot 4+ 5,
50 - .
5.9~ ~ ~4 1 '
10 � 5,~� - - ~
q~g ' ,
1.4 ~ ~ - ~ -
30 ---3,9- ,
3A~ , ~ ,
10 ~0 30 ,0 50 60 70 80 00 100 It0 1~0 190 140
Pigure 6.12. Lines of observation angles from a satellite
~ (A* n 100� east longitude).
The direcCional pattern obtained usin~; the described approach should be made
more precise taking into account the possible variants of the design using
ehe numcrical methods of analysis und calculation described in Che preceding
sections of this chapter. In this case, the prablem of more precisely speci-
fying the antenna setting angles ~ and n or the position of the center of
orientation and ,the antenna rotation angle a k is also so'Lved. In the general
case, when selecting Che center of orientation, the area of the reliable
36
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
_ r
i
FOR O~FICIAL USE ONLY ~
s
rac~pCion zona servee ae the opCime~ity criCer~,on. Tn thie caee~ rhera
~1wQye ~xie~e a nwnber of limiCation~ og ~he it~equel~~y tiyp~ on ~he p~r~-
met~rs or rhei~r funcrione. 8ecause of Che impogeibil~.~y of deriving any
kind o~ analytica]. getimates, the solution ie accomp~.iehed numericaily.
Aesuming~ for examp3.e~ Chet of the three independenr parametere Which de-
~ermine the patt~rn, t~to are constane, one can deriv~ the area of the
s~rviae zone ae a gunc~ion og the Chird taking the limiCatione inCo accounr.
~hig function definea the eimultgneously permiesible range of change in th~
ehird pnremeter~ for example~ a k. By varying Che v~lues of the coordinates
of th~ center of orientation in the appropriate faehion (the values of the -
two se~ting anglee for the on-boaYd gntennQ), one can obtain the permieeible
vglues o~ a~ for any combinations o~ ~ and n. ~
1.3. xhe CoordinaCion of the Angular Spacinge Between Communicatione ~a
SaCelli~es
The extremQly saturated r~dia band in the 1-10 GHz frequency range is em-
pioypd �or contemporary sa~eliite communicationa [1-3~. The use of rhie
band is due to the minimal influence on signal passnge via earth to apace
and epace Co earCh aignals of rhe interference generated by natural sources
o~ ~arth and extraterrestrial origin as well ae induetrial noise. M~re-
~ over, this band is maximally well mnstered and rnther inten~ively utilized
by traditional communicatione serviceg.
Under these conditions, the problem of crosstalk interference of not only
the different communicati~ns services, but also within satellite communi-
cnCions is extremely acute. To assure normal operation of Che coannunicationa
equipment which functions in one band or anothex, limitaCion~ are placed on
rhe technical parameters, the layout, and the operational sequence of the
sAtellite coromunications components, including the eatelliCes themaelves.
Tl~e queation of crosstalk is an urgent one not only for syatems which -
employ satellites in atationnry orbitg, because of the limited capabili-
ties oE an equatorinl orbit with respect to the number of satellites which
can be placed in it, but nlso for systems ~rhich use satellites in high
elliptical orbite.
The visible tra~ectories of satellitea against the celeatial aphere of
ground starions are shown in Figures 7.7 and 7.8 for "Molniya" type orbits
in a topocentric system of coordinates, i.e., in coordinates of azimuth
(A) and elevation angle (Y). The direction of satellite travel aga~nst
the celesrial. sphere of thz Cracking starion is shown with the arrow on
each curve. The times in hours, read out from the poinC in time the sate?-
lites pass through the orbital perigee, are indicated wiCh thc dots on the
curves.
~7
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
;
FOR OFFICIAL USE ONLY
Considering Ch~ EacC ~ha~ ehe satellites in Cha sye~mn fo~,low each oCher
along Che gama paChs, ~hey tri17, Aave ~.den~~ca1 Cracks ~gainst the celestial
spher~ for ~h~a poinr undar coneideration~ Therelore~ ae can be seen from
~he figuraa, Chgre is aiwaye a situaCion in which the angular spacings be-
tween tihe sa~ellite~ following one anoCher are sma7.1. For exampla~ such a
confi.gura~ion beCween rising and se~~ing satellites at tha momenC Cha ~round
etation ewi~ches over ~rom one sateilite to the o~her can lead Co the fact
thaC the signal oF one saCe113~a will serve ae ~ntergerence for the signals
Gf the other if borh eaeelli~ee fa11 w3~hin the directiona]. pattern of the
ground antenna.
~
Y'
7 8
e ~ o ~
b
W s27.~ ~
80
7 9 ~
_ ~
6060 60 70 d0 00 A!
~igure 7.7. Visible motion tra~ectories of a satellitQ for _
Moscow (�N = 56�; aN - ap ~-30�; i~ 62.8';
hp = 540 km).
To estimate the angular aeparation between satellites traveling in a highly
ellipeical orbit, we shall make use of an inerCial system of coordinates,
OXYZ wiCh the origin in the center of masa of the earth, the OX axis directed
towarde the ascending orbital node of the i-th satellite, Che OZ axis coinci- -
d~nt with the earth's axis of rotation, and the OY axis compleCing the system
making iC recCilinear.
In the system of coordinates under consideration, the angular spacing d
beCween the i nnd i+l-th satellites is defined as ,
(a~ xv ) (x~+t - xx ) + f Y~ - Yx ) (Yi+t - YN ) + (=i ~ =N ~ ~zl~ 1 1' zN ~
v~ arccos (7.31)
RiR~}~
where xi(i+l)~ yi(i+l)~ Zi(i+l) are the coordinates of the setting and rising
satellites of the system respectively; xp, yN and zN are the coordin~tes of
Che ground station;
Ri. i+i a (Xi �+i?-- xN (Y~ c~+~~ -Yx (z~ p+~~ -zN (7.32)
38
' FOR OFFICIAL USE ONLY
=I
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
� , .
. _ _ - _ . _ - . , _
NOR OFFICIA~, US~ ONLY
F
7
u=
' Yr e WMeo~
ee
~ ~0.
~
eo s e -
~
46 60 76 100 A,� R:
;
;
Figure 7.8. The visibie travei tire~ectoriea of a satellite goY
Novoeibirsk.
- (~N ~ SS�; aN - aa � 15�; i~ 62.8�; hp ~ 540 km).
The coordinates xN, yN and zp ~an be determined from the folloWing formulas
~ith a precieion sufticient for tAie probiems
, XN ~ R C~S ~N COS ~L~ i YN ~ R cos ~N stn l~N ;
ZN ! R sin ~N ; -
aH ~+~3 ~t, -et. c~.33)
where 7~ai i8 the Greenwich longiCude of the aecending orbital noda of Che
i-th sate113Ce; ti is Che travei t3me of Che i-th eatellite from orbital
perigee; ~t3i is the travel time from orbital perigee to the equator; =
To
~t�i= 2~ (E,~-esinE,i);
(7.34)
~Y,=~, ~n_m,~~
E~ i��2 arctg Y~+~i tg .
1'he coordinates of the i-th satellite, in accordance ~rith (1.41) and (1.43),
can be determined from the relationahips:
x, ~ a[(cos E, - e, ) cos -
'~1- e,'sin E, sin c~, j;
Y~ � e I(cos E; e~ ) sin cu~'+ (7.35)
j~l - e~~ sin E~ cos ca, ~cos i~ ;
z~ a a[(cos E, - e~ ) sin ca,
+ t- e~~ sin E, cos u~, J sin i~ . '
The following relationships can be employed to determine the coordinatee of ~
Che (i + 1)-th satellite in the OXYZ system of coordinateas
Xi+~ - x;+~ ~os en,+~ +y~+~ S~n en~+~ t
Yi+~ x~+~ sin Oni+~ +yi+i cos ~it~+i ; (7.36) _
Zi+i ~ Zi+r -
39
;
FOR OFFICIAL USE ONLY i
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
z,,r _ _ _ . _ . _ _ ..,uw
v.
_
~ ~ ; ~ -
~ ~'OR OFFICIAL U3~ ONLY '
whare ~Sl~,.~l is the angle bet~reen tt~8 eecending nodes of rha orbits of rhe i
and (i ~h 1)-tih s~tei'~3~es; x~~,l~ y~.~~ gnd ~~,~1 ar~ ~ha coordinaees detierm~ned
_ fxom the expreseionet x~*~ ~ g~ (cos Ei+i -ei.~~)cos cai*~ -
Y1=e~ sin E~+~ stn ~~*~1 ~ (7.37)
yi+~ � e�cos Ei+i-ei+~)sin ~ai~~ ~ _
~ ~'1-
~+,sin E~+t cos cai+~]cos li~.t ;
~i+~ " e ( ~cos E~+~ - e~+~ ) sin wi+~ +
+ 1~1
~~~sln E~+~cos c~i+~ ~sin i~~~� -
33nce the coordinaCes of the i and the (i + 1)~th satelliees in expresaion
(7.31) should apply ro the same point in time, the value og Li.~~, in (7.37),
corresponding to from (7.35), must be deCermined from Kepler s equation
in the gorms
~
~
E~+t-- e sin Ei+~ = ~ � ~ . (Ei - e~ sin E~ ) - .
~'i+t -
_ .
d A~+~
- ~
8~1+t m3
npNUeM where
e sin E~ ) ~ d � (7.38) _
� 3
~ - ~
By using the resulting expreasions (7.31)~(7.38), one can determine the
angle d for a particular ground communications station as a funcCion of the
eccentric anomoly of the i-th satellite. The calculations should be per-
formed beginning with the values of Ei+l which correspond to the statit of
the communications session through the (i + 1)-th satellite. In this case,
it is necessary to check the conditions for the location of the point being
considered in the radio visibility zone of both satellites, where these con- .
ditions can be urcitten in the following form in terms of the coordinaCes of
the saCellites
xY(xi - xN~ ~F YN~YI - Y~t) zv~=i - ztv~
~ RRi ~ sin~~o;
x x x,+ N - N c~.39)
rt~ - rr Yx(Yi,.~ --Y~1 + I(z sin yo.
RRI+~ ~i;
To [ake into account the displacement of the routes with respect Co longi- .
tude, aJ~3, when determining Che angle v it ia necessary to introduce the #
term Aa3/c~ with a sign corresponding to the decrease in the angle ar for the
station being considered into the right side of equation (7.38). The calcu-
lation of the angle o from (7.31) ~rith a particular atep with reapect to
Ei permits the determination of its minimal value as a function of Afli+l and -
the orbital parameters. Thus, the minimal value of the angle ~ is shown in
Figure 7.9 for a ground station located in the region of Moscow as a function
of the altitude of the orbital perigee and the angular separation of perigee
40 ~
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
, ~ , _ . _ _ .r~
:r
FOR OFPiCIAL U36 ONLY
from the node for Che caee of AA~,.~1 ~ 45�. The curve wae obtainad ueing _
expreaeione (7.31)-~7.38) ~or a six hour communica~~oes seeeion aher8~tfl8~
perraise3bie diepigcement of the path is QA~ 5�.
�~NNHO Ae can be seen from PigUre 7.9, the
angle amin ~s WeBkly dependent on
- ,;~.o rhe orbital perigee aleitude of bath
te ' ? � rhe i-Ch (setting) and the (i + 1)-th
~ (rieing) eatnilitee. Within a pre- -
cision of 30~~ one can consi.der Che -
~ � � � angie omin gor the range of aititudea
8 indicaeed in Figure 7.9 to be inde-
pendent of the altitude of ehe perigee
. ~ ~eo, , of the orbit.
~
A coneiderable inEluence is exerted
p on the size of the angle am~n by the
, 4eo aso s~o r~o Afli,NM argument of the orbital p~rigee of
both the eetting and ehe rising eatel-
Figure 7.9. The angular separation litea. The nature of the change in
be~Ween saCellites as the angle am~ ie 311uaCraCed by
a function of the Figure 7.10, from ahich it can be seen
perigee altitude: that the angle omin ie a linear func-
tion of the argumente of Che parigee
hp i~,l ! 480 km; '.of the eetting (i-th) eaCellite.
� hp i+l � 800 lan. Relationehips s'imilar to thoae ahown
in Figure 7.10 make it posaible to
plot lines of equal anglea for amin in cnordinates of the argument of the
perigee of the rising satellite (wi+l) and the argument of the perigee of
the setting satellite (wi). Such lines of equal angles of omin gre shoan in
Figure ~.11 by way of example for ~fli+l ' 45� and Afli+l ` 90� ~?ere the
satellites operate in the main orbital revolution for the Moecov region.
Using lines of equal angles for omin. values of the perigee argumenCs can
be choaen which aseure an angle ctm~ no smaller than a specified value for
a particular point. Thus, for a point located in the Moscow region, an
angular separation of no less than 5� between the satellites following each ~
other is assured when ~ni+l a 90�, if orbits with two values of the perigee ~
argumenta are empioped in the epstem, for ezample, 280 and 267'. ~
~
Because of the fact that to service the northern hemisphere, etationary
satellites can be employed along with satellites in highly elliptical orbits, -
there is unquestionable interest in determining the angular separation be-
tween these types of satellites. It folloWS from the analysis of the flight j
path of a satellite in a highly elliptical orbiC (see, for example, Pigure ~
2.6), that the minimum value of the angle between the direction to a ~
~
~
~
~ 41 ~
FOR OFPICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR OFFICIAL USE ONLY
- ,.M~~~ stationary sarellite aad a eateliite
- in a highly elliptica~, orbi.t from a -
g~ound etation ~ili occur at the
' 1e ;,,,oa srarC and ehe conclueion of a com-
a~o municaeions eeseion through the setel-
lite in the eilipticai orbit. in
a~ this case, the eize of the angle be-
a ewean the eatellites depends sub-
~ atantially on the loaaeion of the
~o fround etiarion. The case ahere Che
stat3onary ,~teilite, the aareiz~.te in
~o r~~ r~e aez ase ~w� ehe e113ptical orbit and the ground _
atation are poeitioned in the eame
Figure 7.10. om~ ae a function of ineridian plane with the maximum
the argument of the diseance bet~een the point and the
perigee of a rieing ~9uaCoriel plane, i.e., at the boun-
satellite in the main dary of tihe radio viaib3lity zone o�
orbital revolution the BCationary satelliee will corres-
for the region of pond Co the minimal value of the angle
Moacow (~S2 ~ 45�; ~Figure 7.12).
hp ~ 640 lan). The minimum angle made between the
- ground atation and the direction to
the staCionary satellite and the satellite in a highly ellipeical orbit can
be determined taking what has been said above into account as fol~oas.� For
rhe point in time under consideration for the start or concluaion of a com-
munications session through the satellite in the highly elliptical orbit,
the true anomaly A can be determined using (1.15) and (1.16), and thia meane,
also the value of the argument of the satellite latitude u. The corresponding
� values of the orbital radius and the geocentric laCitude of the satellite are:
e~~ � s aresin sin u sin i).
t+ecoso ~ ~ (7.40)
We shall use a aphere of radius R as a model of the earth. In this case,
_ the geocentric latitude of the earth station coineides aith the geodeaic
latitude. Taking into account the fact that the stationary sate113te, the
satellite in the highly elliptical orbit and the ground staCion are arranged
in Che same meridian plane, expressions can be written for the angular coef- ~
ficienCs of the direction of the ground station to the satellites:
g
rs(n~-Rstn~p,~ R~in~Y
K~ �rcos~-Rcos~x ' ~ � - ro-Rcos~N ~ (7.41)
where rp is the radius of the stationary orbit.
;
42
FOR OFFICIAL USE ONLY
;
:~t _ _ _ _ _ _
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
_
FOR OFF~CIAL US~ nNLY
~ ~Mfll" dul~
.L:~_.. -
NN~ ~ . ~ ~ Np~'~O~ j ~
488 9 ~ ~9A
10 8 ~
~A4 0 264 .
8 ~
498 476 8
IO ,
'174 ~~d
18 ~S
~70
?~4 4~6 4tl2 2A6 u.e 470 27~ Z98 29~ 4g~ w.~
p ~g~ 0~ ~b~
Figure 7.11. Linee of equal value~ of cr~in in the main
orbitel revolution for a point in Che region
of Mosca~t:
A~ s 4S�~
b) en = 90�.
'r`e~~T~ , r x~s~a i~~ , . , ~i
. . ~ : ~ ~ r r i: ~'-~'r~' f '1
r'
~ ~ . n
. . . ~
, .1 p~ y:"~, f ~ ,
J'@#~ rr - i .
~ _
S~
` ' ~
. ~a ..2 , k~~ ~.~~i ~ ' ~
.R . . ' ~g~~ 7.,-c ~
~ ~'~~J .a r.~^4~
~~Ya:r .
r : ~ I
? ~
i-
. �
. e ,
7
t F~y E~.~
iA~r ~i~
~j il �
~
~ `a ; ` f : S
ak+ y?` ~I i ..:C.r 'd~~~R ~
~~;r~ '
ni
:;e_ .
Figure 7.12..On the determination of the mutual visibility
betWeen satellite~ in highly elliptical and
stationary orbita
Kcy: 1. Equatorial plane.
43
FOR O~FICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
. _ -
FOR OFFfCIAL U5F ONLY
Ar the boundary o� th~ r~d3o vieibilitq ~oa~ o~ th~ ~~aeiona~cy eareilire,
Che gaodeeic lae~~ude of ~he ground a~aCion ~given the cond3~3on of a epher-
~~ai egrth) 3es
4N ~ arccos R- . ~7.42)
Then the value of the aegle betw~en tihe direct3one to the eatellite~~ taking
(7.36) into account, can be deCermined from Che ~cpreeeions
Q ~ erct~ r~e~tn~-R~atn (~-~N)-reRitn~N (7.43)
~a rrecos~- ~co~ (~-~pH -re co~~H
The raeulCs of an analysie of the angular separetion b~t~?~en the direcCione
eo thg sate113tee, baeed on (7.40)-~'Y.43), are given in Table 7.4 ahere the
angular epacing beeaeen the eatelli,tee ie presenCed gs a guncCion of the
seare and conclueion e~nee for co~mnunications through the eateilite in the
ellipeicai orbit and as a function of eha angular eepara~ion of perigee from
the node.
Tabl~
Bpe~~ Nnia~ ee~Ne~ eu~a roNw e~~Mea a~~r
tl, (Q) lor acorte~l ~ (or oe?Nn~~
1.3 I 20 I!.S ( 9.0 !.0 (!.b IOA 1 b
270 36'00' {6'15' S3'ZO' S8'10' b8�10' b3'10' 46'15' 36'00'
280 ~5'20' S~�~5' 60'35' 6~1'35' 49�50' ~3'b0' 36'OS' 26'00' -
Z8b ~9'S0~ 58b0~ 83~40~ 66'20~ ~S'30~ 39~20' 31'25~ 20'30~ ~
290 b3'30 61 ~0 65 45 61'30 ~1'05 34 ~0' 26'45 d5'b5
Key: l. Value of the perigee argument,
2. Time of the etart of a communicariona aeesion (from perigee); _
3. Time of the end of a communications seseion (from perigee).
COPYRiCHT: Izda~el's~vo "Svqaz~," 1978 -
8225
C50~ 8144/1322
44
FOR OFI~ICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
> ,
:
f
. . ~ . . ~ . ~ . . . #
' FOR OFFIC~AL U3~ ONLY
t
i
PHY9IC8 ~
,
e ~
~
' !
i
UDC 621.183.4 `
'LA3ER SPBCTROSCOPY IN NUCLBAR Pt~tY9IC8 ' ~
~
~
Moecoa VESTNIK AKADEMII NAUtt 888R in Rueaian No. 4, 1979 pp 38-48 ~
~Article by Doctor of phy~icai and riathen~acical science~ V.3. Letokhov] F
LTexe] Coneiderabl~ progree8 hag been achieved durirg the paat saveral yeare
in development of inethode of coherent light generation aad fre uenc coneroi.
4 Y
It hae become poesible to change ta gyetematic inveetigatfon of rather fine
effecte of ~he resonance interaction of coherent light afeh atoms and mole-
culea. Thie field of reeearch is ueualiy called lager epectroecopy. The
deve~oped methode of laeer epectroscopy are glready begineing to be applied
in many fields of ecience and tachnology; in quantum me~rology (quantwn
frequency ~nd length etandards), in chemisrry (laeer eeparation of 38otopee r
and production of pure materiale), fe analytical technology (laeer apecero- '
meters gnd detectore of email trace numbere of atoma and aalecules),
in geophyeice (monitoring polluti~g impuri~iee in ehe atmoephere and the
eearch for fuela by accompanying gasee), in biology (investigation of photo-
eyntheais and the molecular mechanism of vision) and so on. The field of
nuclear phyaics reeearch in Which methods of laeer specrroocopy also open
up nea experimental opportunitiee ie coneidered ia ~he given article.
Ae ie known, many characteriatice of the nucleus the number of protone ;
and neutrone, epin, Che quadrupole moment of the nucleue and the related -
ehape, mean radiue and sxcitation of the nucleue and also ite velocity and :
orientation ~Table) are clearly manifeeted in the fine etructural detaile
of the outer electron ehell of an atom, that is, in the optical epectrum of
the atom.
It ie thie that permits the use of laeer epectroscopy methode in nuclear
phyeics investigatioae in t~?hich operations aith nuclei having the required
characteristics are conducted by the effect of coherentliqht on the electron
cloud eurroundieg the nucleus. Of course, euch methods of effect of laeer
emiasion on atoms which Would permit selective excitation and ionization of
each atom aith a nucleue of eelected variety, ~roduction of an obeerved eig-
nal from each individual atom, Which aould permit oee to extract each selected
45
FOR OFFICIAL USS ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR OFFICIAL U9E ONLY
arnm from the mixture snd eo on muet be developed for thie. The pragreer -
achieved durinp the paet few yeare fn methoda o~ laeer apectroecopy made
- it poeeib'le to's;+ive tihe ennumerated prob~eme and to open up naw proepecte
in nuniear phyeice ti~eearch.
Manifesrarion of Nuclear Chargcterietics in the Opeicai
Spectrum of the Atom
Manfgeetaeion of Ch~racrerietics
Characterietice of Nucleue of Nucleug in OpCfcal Spactrum
of the Atom
Type of nucleus (charge 2) Wave lenqth of abeorpeion lfnes
Ieotope composition (no. of neutrons) Igotope er~ife
8pin gnd magnetic moment Superfine etructure ~STS)
Quadrup~le moment SupeYfine etructuYe
Shap~ and mean radiue Superfine etructure ,
Excitation . Superfine atructure
Orientarion Population of magnetic eublevels
Velocity Doppier ehift of abeorption linee
Laeer Detection of Single Atome
Considerable attention has recently been devoCed to developing methoda of
laser detection of superemall, or as they say~ trace amounte of material.i
The eseentially detectable limit ie one atom eince it etill carries com-
plete apectral information about ite oan structure. Therefore, one of the
main purposes of laser epectroscopy is to develop methode of detecting
eingle atoms. The method of laser excitation of reaonance fluorescence,
Which makes it poaeible to find the maximum number of photona scettered by
one atom, and ~he meChod of eelective step ionization of atom~ by laeer
emiseion which eaeentially permita transformetion of each atom into an ion,
are moet promising. Both methodg were recently implemented aucceeefully2
et the Institute of Spectroscopy of the USSR Academy of Sciencee.
The problem of detecting aingle atoma can be divided into three succeeeive
atag~s: (1) accumulation of the element and production of free atome;
(2) tranaport of atoma to the detection zone; ~3) detection of the atom.
Solution of all theae probleme is very important for ueing methods of laeer
apectroecopy in nuclear phyeice investigations. Succeas hes not been
achieved in Working out the third problem: a eelective aignal (in the form
of photons or ions) significantly exceeding the noise level can be produced
from a single atom interecting resonantly with one or eeveral laeer beama
46
FOR OFFICIAL USE ONLY :
. . s. _ . ,
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
~
~'OR t1Ft~'TCtAt, USC~ ANLY ~
~o thar th~ face of trensie of ghe arom through rhe iaAer lfghr ~i~1d can
b~ religbly establiehed. ~'herefore, 1et ua initiglly ~oneider methode of
detec~ing s~ngi~ atoms and then let ue diecuse th~ firet and eecond prob~ems
wfCh reepec~ eo nuci~ar phy~ic~ inv~~eig~Cion~~ '
Th~ fluoreecence method. Th~ characreri~tic ~ea~ur~ of the meChod of rer-
onanc~ fiuoreecenc~ ie that the eame arom can interact eevera~ timea with
the laeer emiesion, reemirting photone of ehe eam~ frequency ae the exc~k-
ing photone in ali directione. if the 1~eer emiaeion inteneity exceede tihe
reeonance tr~neftion eaturation inteneity, rhe populations of the ground
and excited levele become idlentical. In thie caee ehe aeom reemire Nmax ~
sT/2 ~ pon~ photone ahere ~6 pont ~e the time of eponraneoua decay of Ch@
excited etate to the ground ~tate, dUring time T of intersection of the
light beam.
~
~ As ~n examp~e, 1et us estimate th~ maximum number of photone epontaneously
rer;nitCed by a sodium atom. The mean velocity of the thetmal motion of
Na(vr) aeoms at 50oC compriseg 5~104 cm/s. The decey time of ~he ffree
excited etate of IVa~Gpon~ ie 1.6~10'~ e. Therefore~ a eodium atom is
capable of reemitting N ~ 250 photone on a peth of h~ 0.4 am~correapon-
ding to experimental con~~tione~. One can gather Nd~t � 20 photone on the
cathode of a photomultiplier wfth a value of the eolid angl~ of the ecatrered
emieaion sampling (.fL) of 1 sr. The beat FEU ~photomultiplier~ have quantwn
efficiency of 7'~ = 0.1-0.2 at aavelength of A~ 589 im. Thus~ one can ex- -
pec~ the appearance of 2-4 phoCoelectrone fro~ the FEU phorocathode from
' each atom pesaing through the 188er beem.
Thie experiment aas carrfed out at the Inetitute of Spectroecopy of the
USSR Academy of Sciencea by using a CN dye laeer whose frequency wae tuned to
the D2 resanance line of a Na atom. The fluoregcence si.gnal Waa recorded
- by two FEU and a two-channel recording syet~m operating in the coincident
mode Wae used to separate it. With laser emiseion inteneity providing
_ abaorption eaturation, the atom re�mitted a number of photona aufficient
for at leaet one single-electron pulae to be forn~ed at the output of each
F~U during the time the atom wae located in the be~m. During this time~ the
appearance of pulgea at the FEU output Was regarded as e coincident event.
Thie re~cording acheme permitted, firat, a coneiderable reduction of the
phonon effect and, second, m~de it posgible to establieh the effect of
intereection vf the laser beam by the atom. The ratio of the number of
phonon photons impinging on the FEU cathode to the numb~r of reeonance-
aca~~ered photons of taser emission contained in the cuvette compriaed
10~ for a specially designed cuvette in which the atom beam and laser
beam intersected. A total of 10'2 phonon photons impinged on the FEU
cathode during the tranait of the atom through the laser beam under the
experimental conditione.
47
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
;
FOR OFFICiAL USE ONLY
The depandenca of ~h~ Na aeomic fiux, measured in thie experimen~, through
Ch@ r~cording range on the beam source CempetiaCure ie presented in Figure l.
The minimum eign~l ie d8tectied at furnace temperature of 43oC. In thie ca~e
a glux o~ l0 arome pe~ second 3s detecCed, which correspc?nds to the mean
aeomic deneity o~ 10' in the recording range~ Thie maximum deCection was
relgted to the ~eevitable egfact of phonon emi,esion participating in formation
of the pulaee in the coincidence circuiti.
_
io�~
~ '
. ~o, ~ ~0�1 ~ ~ , ~ .
Qy
~ T _
~ ~ .
~1~~10 ~ 10 ~ ~
~ ~ ,
t+~ + 10'4 ~ ! .
10 ,
? ~
~ 10'd
� 40 60 80 70 80 00 T,�Q
Figure 1. Experimental Dependence of Sodium:�.Atom Flux
Through Laser Beam on Furnace Temperature
Measured by the CW Laser Fluoreacence
Excitation Method.
Y'
- KQ ~ 1. Atom flux s-1 2. Atoms in recording zone
1'he fluorescence meChod of detecting sic~gle atoms ia based on cyclic stim-
ulated excitation, and epontanEOUS reemission of photona. By using exist- ~
ing lasers, thie method can be realized for the time being only for alkali
and alkali earth elementa. The cyclicity of the proceas is easily inter- ~
rupted for moat the remaining, eapecially complex atoms having metasCable
atates near the ground atate, since only the atom drops to the metastable
atate. In these casea the maximum detection is i.ncreased ~to 102-104 atoms,
which incidentally is quite acceptable for some problems of laser phyaics
diacussed below. However, the photoionization method ia more efficient and
universal.
,
48
FOR OFFICIAL USE ONLY ~
~ ;
~,T : - ;nd_ : - ; . - _ ~ , -
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
FOR OF~TCIAL USE ONLY
_ The photoionizaCion method. Thia approach to detection of single atoma ie
based on the idea of their aelective step photoionization, propoaed by the
author in 1969 both for isotope aeparation and for detection of aComz, If
detection ia carried outi in a vacuum (the moaC inCereating case trom the
viewpoint of maximum apectral resolution), laser emisaion pulaea with �re-
quency of approximately 50 kHz (aC velocity of 5~10~? cm/s and interaction
path of 1 cm) muat be tranamitCed for ionization of each atam that has
entered the laser beama. Effective ionization of moeC atoms can be achieved
only iF the atom is ionized through the overlying (Ridberg) atatea with the
mean output of retuned laser emisaion accesaible under laboratory condiCion~
(on the order of 1 W). This method of aelective ionization was propoaed and
inveatigated at the Inatitute of Spectroecopy of the USSR Academy of Sciencea.3
In theae investigationa, the non-resonance procesa of photoionizaCion of an
atom in~ transition from the intermediate state to the continuum is replaced
by a process of resonance exciCation of an atom from Che same state to the
overlying state near the ionization boundary wiCh subsequent ionizaCion by
a pulse of the electric field. The effectivenesa of ionization ia close to
unity in thia process. Since excitation, aC all the subaequent atatea is
resonance, comparatively low energy density of laser pulsea (10'4 to 10'6 J/cm2),
completely achievable by means of exiating dye lasera, is required for satura-
Cion of all tranaitions.
For illustration let us preaent the latest results on detection of ainglE ~
ytterbium atoms.~ A simplified diagram of the energy levels of an ytterbium
atom and the Cypical diagram of the quantum Cransitions uaed in excitatian
of the Ridberg state are presented in Figure 2. A three-sCep excitation.
scheme using three pulsed dye lasers excited by a single pulsed nitrogen
laser was used. By readjusting the emission wavelength of the third laser
in the range of 5,950-5,770 one can transfer the Yb atoms to P-ataCes
with the main quantum number of n= 14-20. The atoms are then easily ionized
by an electric field pulse with intensity of 10-15 kV/cm.
The configuration of the atomic beam, laser beams and Che direction of motion
of the formed ions is shown in Figure 3. The laser beams intersected the
atomic beam between two electrodes to which an electric field pulse was fed.
The ions resulting from selective ionization were extracted through a slit
in one of the electxodes and were recorded a second time by an electron
multiplier tube. The experimental conditions made it possible to extract
practially all the ions from the interelectrode gap.
When the laser pulse energy exceeded the saturation energy of each transition,
the atoms were uniformly distributed along the ground, intermediate and final
states according to their statistical weight. For example, 5/12 of all atoms
located in the excitation- zone converted to the final 173P~2-state. Record-
ing of single atoms with quantum efficiency of approximately 50 percent was
achieved under these very conditions.
~
49
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
a
,
FOit OF~ICtAL US~ ONLY
~~CY~~ ONNII~TY~~' TpNI1MTY~Z' ~eg
iso ip~ i~? as~ apo apo o~�~
00177
e~- i$;` io= $ �
`
e-�- 8..,
e---
~ooo0 8_' e-- ~0 6
y " _
~ ~
~ ,
4 -
00 000
6~-- -
. ~ 3
~O A`
e..- ~ ~
to 000 ~ .
i
0 0 ~
Eigure 2. Diagram of Ytterbium Atom Levels and Quantum
Trangitione Ueed for Three-3Cep nccitati'on.of
an Atom for Ridberg Statea and Subsequent Ioni-
zation of it by an Electric Field Pulee:
~1 1~ 555.6 lm; I~ Z� 680 nm; ~ 3� 577-595 nm
Key: '
1. Singlete 2. TripleCs
The experimental dependence, of the ytterbium ion outpuC per pulse on the
furnace temperature is presented in Figure 4, a. The calculated dependence
of the number of Yb atoma in the recording zone (dashed curve) ie a18o
presenCed in this figure. The difference in the course of the experimental
from the calculated curve at ~.mperatures below 250oc; ia apparently related
to the fuct that the atomic pait in the furnace was uneaturated under the
experimental conditions. The sCrcn~ instability of the ion eignal was
cauaed by fluctuationa of the numbpr of atoms in Che excitation zone with
the loweat possible atomic beam inCet:sities. Actually~ the probability ~f
two atome impinging in the excitatton zone simultaneously will be much
lesa then the probability of one +~tom impinging in this zone under these
conditione. In this case the recor~'.i�~g system will reaponi in moat caees _
upon the appearance of ~nly a sin~le atom and the probability ~f recording �
k-ions P(k) will be determined by Poisaon distribution. This ie clearly
confirmed experimentally. The values of P(k) for ytterbium atome at two
different ~.alues of the average number of atoms N in the excitation zone
are preaented in Figure 4, b and c. The Poisson disCributions (solid lines), :
calculated for the same values of k, are preaented for comparison.
SO ~
;
FOR OFFICIAL USE dNLY
S
v
~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
, _ , . - -
,r
~
FOR AF~'iCIAL UgB ONLY
ATOYNYA IIyMpN .
, ~ ~ 3n~NrPo~?y Yb
i ~ :
. '
. ~ Nrnyn~o e~~- `
`j~ \ 4~ONOrO OAN '
\
C~\ ' ~
f . ,
\ ~
/ `
~ .
. ~ /
~ / ~ ~ ~ ~
~ . . . ~ '
~f
n.s.p�r� ,~rw, (4 ) ~
i �
~
. ~w~ ~5) ' ~
~
Figure 3. Detecting Cuvette of Experimental Inatallation for ~
Photoionization Detection of Single Atoma 7hrough
Ridberg States (the Configuration of the Atom Beem '
Through the Laser Beams and the Beam of the Formed
Ione in the Region of Interaction of Atoms with the
Laeer Beama and the Electric Field Pulee is given)
Key~
.
1. Atomic beam 4. Laser beama
2. Electrode 5. Furnace
3. Electric field pulse
51
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
.,.,,v. . . , , _
,
. , . . ~
~
. _ . s'~
. . . _ . . . . . ~ ; ~4
. . . - - . . . . . ~ ~ ~ F.
_ FOtt OFFICIAL USE ONLY .
to~ " . a . prw~ b ; ~
0~3 ~ ~
10~ 0,9 ~ ~ 1:~`
0,1 ~ ~ ;
~ . ~ ~ ~
o ~ z a ~ eK
io ,
,
.
. , ~
~ ~
~ . ~
/ P (wl ~ ~
/ � ;
~ 0~0 ' `
O~1 ~ , ~ ;
/ 0,4 , "
~ . ' O,~ '
!
~
OA1 0
160 ~00 ~160 300 T; 0~ 1 Z 3 I ; i
~ ;i:
. s
Figure 4. Experimental Data which Demonatrate Photoionization
Detection of Single Yb Atoma: a-- dependence of
number of atome in Che observation zone on furnace
temperaturea; b-- dietribution of nwaber of ion
counts during 5 seconds of obaervation with average '
number of atome in the obaervation zone of A~ 0.04;
c-- the same during 30 aeconds of obeervation at ~`r
N = 0.003 i_
~
Inveatigating Nucl~Accesaible in Small Quantititee by -
Laser Spectroacopy Methode
The high eensitivity and reaolution of laeer spectroacopy a?ethode permit
investigation of the characteristice of nuclei accessible only in very '
small quantities. Succeasful experiments on investigation of radioacCive ;y
nulcei. far from the stability boundary have b~en conducCed in a number
of foreign laboratories (Switzerland, Weat Germany and France). Aliaostall 4;
i'
the experiments were carried out by the aimplest method of fluorescence
excitation of atoma using the emission of a retuned dye laeer. Specifi-
cally, an installation ie operating at CERN for investigation of the auper- t
fine structure of the optical lines of atoma with short-lived nuclei on
a mass separator operating on a eingle line with a aynchrocyclotron rated
at 600 MeV ~the ISOLDE installation).
.
52 �
FOR OFFICIAL USE ONLY -
~
.
_ _ . . _ . - ~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060055-4

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-44850R000100064455-4
,
1~'Ok ~11~'t~' [C 1 AL 1131~ ONLY
A proton b~am irradiaees a target from which unatiabie nuciei formed in
the volume 8re evaporated after diffusion. The flux of eveporated atom~
enteYe an ioni~er gnd then the mase eep~rator at whose outpuG fluxee of .
monoieoeope aroms with intenaiey of 107-1011 ione per second are achieved,
The flux of maee-eeparated ione is efther coliected ~n ehe soutice of radio-
aceive atome if the nuclear decay time is eufffcienCly high ~hour~ or day~)
nr ie neutralized directiy during the charge-exchaege procese if Che nuclear
decay time ie ehort (eeconde or minutes). The aCome in the form of an atomic
beam or in a gae m8y then be completely investigated by laeer spectroecopy
methode, ~
Let us preeeat ae an example the resulte of ~nveetfgating ehorti-lived ~
nuclei produced on the lfne of rhe ISOLDE mase separator by leser epectro- +
acopy methods.s Even neurron-deficient unetabie i.eotopes of Hg182-190 Were ;
inveatiggted on thfe installation~ The ion flux inteneity of theee ieoCopes ;
with lffetfines from 1 to 60 minutea comprieea 107-109 fone per 0econd, The !
euperfine structure (ST8) of the 2,537 R line of inercury~ aesuming a rotel
of 108 etome of the mercury ieotope, can be meaeured within 1-2 minutes during
flunrescence excit~tion ueing pulae dye leeer emiesion whoee second emiesion ~
harmonic is retuned to the region of 11 ~ 2,537 The ieotope shift of the
_ optic~l transition of one ieotope A with respect Co anoCher B ie meaeured ;
from 3T5 data:
~~~e rcZe~~ I V~ p 8