JPRS ID: 10192 USSR REPORT ELECTRONICS AND ELECTRICAL ENGINEERING (CORRECTED COPY)

• Document Type: CREST
• Collection: JPRS
• Document Number (FOIA) /ESDN (CREST): CIA-RDP82-00850R000400080046-9
• Release Decision: RIF
• Original Classification: U
• Document Page Count: 150
• Document Creation Date: November 1, 2016
• Sequence Number: 46
• Content Type: REPORTS

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000440080046-9
FOR OFFICIAL USE ONLY
JPRS L/ 10192
16 December 1981
- ERRAT[1M: This cover should be substituted for
cover on JPRS L/10192 of 16 December 1981 USSR
Report ENGINEERING AND EQUIPMENT (FOUO 12/81).
USSR Re ort
p
ELECTRONICS AND ELECTRICAL ENGINEERING
~FOUO 12/81)
FBIS FOREIGN Bl~OADCAST INFORMATIO~i SERVICE
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/42/09: CIA-RDP82-00850R000400084446-9
NOTE
JPRS publications contain information primarily from foreign
newspapers, periodicals and books, but also from news agency
transmissions and broadcasts. Materials from foreign-language
sources are translated; those from English-language sources
are transcribed or reprinted, with the original phrasing and
other characteristics retained.
Headlines, editorial reports, and material enclosed in brackets
are supplied by JPRS. Processing indicators such as [Text]
- or [Excerpt] in the first line of each item, or following the
last line of a brief, indicate how the original information was
processed. Where no processing indicator is given, the infor-
mation was summarized or extracted.
Unfamiliar names rendered phonetically or transliterated are
enclosed in parentheses. Words or na.mes preceded by a ques-
tion mark and enclosed in parent:~eses were not clear in the
original but have been supplied as appropriate in context.
Other unattribut~d parenthetical notes with in the body of an
item originate with the source. Times within 3.tems are as
given by source.
The contents of this publication in no way represent the poli-
cies, views or atti*_udes of the U.S. Government.
COPYRIGHT LAWS AND REGULA.TIONS GOVERNING OWNERSHIP OF
MATERIALS REPRODUCED HEREIN REQUIRE THAT DISSEMINATION
OF THIS PUBLICATION BE RESTRICTED FOR OFFICIAL USE ONLY.
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000440080046-9
FOR OFFICIAL USE ONLY
JPRS L/10192
16 December 1981
USSR REPORT
~ ELECTRONICS AND ELECTRICAL ENGINEERING
(FOUO 12/81)
CONTENTS
CERTAIN ASPECTS OF PHOTOGRAPHY, MOTIUN PICTURES AND TELEVISION
Experimental Three-Matrix Color Television Camera Using
Charge-Coupled Devices With 580x532 Elements 1
COr4:UNICAT.IONS, COMMUNICATION EQUIPMENT, ?tECEIVERS AND
TRANSMITTERS, NETWO?tKS, RAI)IO PHYSICS, DATA TRANSMISSION
AND PROCESSIPIG, INFORMATIQN THEORY
Improving Noise Immunity of Pulse-Time Aircraft
Instrument Landing Systems 16
Parameter Substantiation Technique for Electromagnetic
Interference Simulators 22
Applying Posinomial Lstimate to Efficiency Determination
of Equipment With High Electroma.gnetic Compatability
Indicators 27
MICROELECTRONICS
Magnetically Tuned Semiconductor Microwave Devices 30
PUBLICATIONS, INCLUDING COLLECTIONS OF ABSTRACTS
Analog Integrated Circuits 38
Annotation and Abstracts From the Journal ~HIGH-VOLTAGE
TECHNOLOGY' 43
Annotation and Abstracts from Colle~tion 'Improving
Tractional Electric-Drive and Power Supply Systems' 48
- a- [III - USSR - 21E S&T FOUO]
F'OR OFFIC'~AT. USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9
FOR OFFICIAL USF. ONLY
~
Annotation and Abstracts From Journal ~M~THODS AND DEVICES
FOR PRODUCING AND PROCESSING RADIO SIGNALS~ 57
Annotation and Abstracts From Collection 'Methods and Means
for Optimization of Electromechanical Elements and Systems'.......... 63
Annotation and Abstracts From Callection 'Physics of
Semiconductor Materials and Devices' 70
Cryogenic Electronics in Marine Radio Equipment 76
Design and Production Technology for Microelectronic Digital
Measuring Instruments 80 ~
.
Digital Information Transmission Via Low-Speed
Communication Channels 84
Electrical Engineering Handbook 88
Impurities And Point Defects in Saniconductors 100
Neuristor And Other Functional Circuits With Volume Coupling........... 104
Non-Destructive Test Methods To Detect Faulty Radin ,
Equipment....~ 1.09
Nonlinear Hydroacoustics 113
Operation of Radio Systems 118
Precision Standard Time Services 122
Problems of Radio Signal Processing 127
Radiocommunication Channels for ASU TP 129
Reflector Scanning Antennas 132
Secondary Power Supplies for Radio Electronic Equipment 136
Semiconductor Multiplier Diodes 140
Square-Wave Generators on MOS Elements 143
~
" - b - ,
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400084046-9
FOR OFFICIAL USE ONLV
CERTAIN ASPECTS OF PHOTOGRtiPHY,
MOTION PICTURES AND TELEVISION
UDC 621.397.61:621.397.132
EXPERIMEI3TAL THREE~IATRIX COLOR TELEVISION CAMERA USIlJG CHARGE-COtJPLED DEVICES WITH
580x532 ELEMENTS
Moscow TEKHNIKA K IlVO I TELEVIDENIYA in Russian No 6, Jun 81 pp 30-38
[Article by Ye. V. Kostyukov, A. N. Markov, N. K. Milenin, B. Ya. Nepomngashchiy,
Ye. A. Polonskiy and A. D. Tishchenko, All-Union Scientific Research Institute of
Television and Radio Broadcasting]
CText] Much progress has been made here and abroad recently in developing Iarge-
format matrices of charge-coupled devices (CCD), making it possible to
use them for building models of all-semiconductor one-, two- and three-matrix color
TV cameras [1, 4-8].
In the USSR we have developed both p- and n-channel large-format CCD matrices with
frame transfer of charges and 580x532 elements [3]. These matrices are capable of
operation at the 625-line standard and contain a atorage aection, memory section
and output regiater (Figure 1) with three-phase electrode systems in the fortn of a
three-layer, part~y overlapped polysilicon structure in which the electrodes of a
single phase match up with ~each layer of polysilicon ,~ich makes it possible
to improve the technological effectivenesa of CCD fabrication [2, 3]. The area of
the storage section is 9.5x12.8 m~n, and that of the memory section 6.7x12.8 mm.
The overall size of the crystal is 17.8x14.7 mn.
Figure 1. Diagram of Larg~-Format CCD
Matrix With 580x532 Elements
mNl
ceKuua miz 2 Ke :
~ 1~ ?mKoneeMUa m~ y 1. Stora e section
/290x5321 8
2. Fstl, 2, 3
Ce~ruup 4ni
~ 3~ ~yAmu ~ 3. Memory section
1290~5J21 m~~ ~4~ ~'m~ ~ 2~ 3
( 5)
ea.~ooNVU aezucmo ~d~~ ~8~ 5. Output register
( 9) eaxov 6. FTegl, 2, 3
,7am~op ( 7) 7. ShutteY'
. Rc ~0~~6~~v3 8. Background charge input
9. Output
- 1 -
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400084046-9
FOR OFFICIAL USE ONLY
The image elements in the storage section measure 33x24 ~un, in the memory section
24x24 pm, and in the output register 24x80 ~m. Surface channels are used for
charge transfer in the matrices. The CCD matrices which have been developed are
assembled in a cermet case with 32 leads.
Saturation exposure with respect to light xesponse amour.ts to about 4.03 lux�sec
for the large-format matrices. The typical spectral response curve of the CCD
matrices is shown in Figure 2. The sharp drop in the blue region of the spectrum
_ is caused by the absorption of light in the polysilicon electrodes, the 0.5-~un
thickness of ~ahich is in accordance with the specified value of their resistance.
The rate of charge transfer from storage section to memory section is governed by
the time constant of the RC electrode system. A high rate of charge transfer can
lead to a darkening of the image in the center of the scanning pattern if the elec-
trode system has bilateral power supply. Hence, RC is predetermined, and when
C=(12-16)103pF the thickness of the electrodes, practically, can not be less than
0.5 }un.
~ Ea*"
/,0
Figure 2. Spectral Response Curve of
o.s a 580x532-ElemenL CCD Matrix
Key:
~~,SI ' . i.
relative
2. il, ~
a~
a.z
~ ~ ~ ~ ~
S00 700 900 t f00 A HH
The inefficiency of charge transfer in the output register at an operating frequen-
cy of 10 Mfiz amounts to E= 6�10'4, which leads to a difference of the frequency
contrast characteristics (FCC) of the matrices at the left and right edges of the
sca~~ning pattern. On the left edge of the scanning vattern, where inefficiency of
charge transfer can be ignored, a decline is observed :in the FCC due to finite
geometrical size of the elements, diffusion of charge carriers, an irregularity in
the matrix output apparatus and so on (Figure 3).
Tt~e output apparatus of the large format CCD matrices has two outputa--primary and
compensating. The output apparatus has a floating diffusion region.in the primary
ct~annel and also integrated MOS-transistors for charge clearing, and integrated
outflow repeaters in both the primary and comQensating channels. The compensating
channel is used for suppreasing the interference from the operating pulsea in the
primary channel.
The functional diagram of a developed and fabricated experimental model color TV
camera using three native large format matrices (with 580x532 elements) is shown
in Figure 4. Signals from the output transistors of the matrices 5 are read by
- 2 -
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/42/09: CIA-RDP82-40854R040400080046-9
FOR OFF[CIAL USE ONLY
preamplifiers 9 and fed to the inputs ef balancing amplifiers 10, in which occur
video signal coupling, b?ack Zevel regulation, supplemental amplification and com-
pensation of light diffusion. The amplified signal goes to regulated amplifiera
11, which effect operation of the white level automatic balance system; thereafter,
the signals pass through gamma correction units 15, limiters and blanking circuits
16. In the R and B channels the signals are handled in a 1.S~IIiz band whereas, in
the G channel, including the balancing amplifiers, the full bandwidth--5 MHz--is
nf~ (1)AeBoru Kpau u,wd~n~ce~a Figure 3. FCC of a CCD Large Format Ma-
trix at an Operating Freqeuncy
~ Ue~mp of 10 MHz
o,~s g) np~a~v Kvau Key:
o,so 1. Left edge of image
2. Center
o,zs 3. Right edge
4. f, MHz
0 I 2 J 4 S 6 f MJt~~ ~F~
maintained, and just ahead of the buffer amplifier 14 the signal goes through a
low pass filter 12 with a pass band of 1.5 MHz. The G signal is used to form the
high frequency part of the signal, a;:d the aperture correction eignal is also
formed from it in aperture corrector 13. The high frequency part of the G signal
' and the apertur, correction signal are added to the R, G and B low frequency sig-
nals in summing amplifiErs 17.
A preamplifier schematic is shown in Figure 5. The purpose of the preampl.ifiers
is amplification of the useful signal and suppression of switching interferen:e.
The CCD outputs are the outputs of FET~s, one of them carrying a signal and switch~
_ ing noise and the other switching noise only. The signals from the matrix outputs
proceed across decoupling repeaters VT1 and VT~ to the inputs of differential am-
plifier A1 which suppresses synphase switching noise. Amplifier A1 is an opera-
tional amplifier (200 V/~sec, amplification factor 3000). This amplifier is load-
ed on a fifth ordez Cauer low pass filter, fr~m the output of which the signal
proceeds across emitter repeater VT3 to the video processing board. To reduce
stray currents the preamplifiers are enclosed in ehields a~1d positioned next to
the CCD matrices. The output signal from the preamplifiers has an amplitude of
200-300 mV.
The balancing amplifiers (Figure 6) perform a number of functions including ampli-
fication and coupling of the signal to the black reference level. It is usually
not possible to isolate ~nformation on the black level durfng a horizontal quench-
ing pulse because of thF differential reading of the signal from the CCD matrices
- and the dependence of the blanks' level on the control system for the matrices.
Hence, coupling of tlne signal is performed at the black level derived from blacked
out elements. Several elements on each line are covered up for the purpose. This
method of coupling does have ane defect however. Between the cryatal face and the
- 3 -
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000440080046-9
FOR OFFICIAI. USF ONLY
R Rs
9 f0 11 13 f6 f7
5 .
- ~ 6 Gr
- J 9 f0 f2 !f f3 f6 f~ f8
6
ZQ fJ
8
9 f0 f1 ~ 13 16 f7
6 6 6 fg
7
B
Figure 4. Functional Dir~gram of the Three Matrix Color TV Camera
Key:
1. 350PF7-lA lens 11. Regulating amplifiers for automatic
2. Infrared filter ~ white balance system
3. Color-separating prism 12. Low pass filter
4. Neutral light filters 13. High frequency signal ahaper and
S. CCD large format matrices aperture corrector
6. Output pulse amplifiers-switches 14. Buffer emplifier
, (drivers) 15. Gairnna correction units
7. Apparatus for CCD matrix control 16. Limiters and blanking circuits
8. Mas~ar oscillator and aynchroni- 17. Swnming amplifiere
zation system 18. Col~r monitor
9. Freamplifiers 19. Automatics t~stems
10. Balancing am~lifiers 20. Diaphragm drive
protective gla~s of the matrix case is a gap through which light comes in part way
under the darkening strip, leading to a distortion of the "black~~ level. An illu-
mination compensating circuit is used to reduce these distortions. From here nn '
it is proposed tn apply a darkening coating directly to the CCD matrix crystal.
Besides coupling and amplifying, the balancing amplifiera also perform gain switch-
ing from field to field. The switching is necessary because the levels of the sig-
nals in adjacent fields are unequal. In the first field the signal buildup~takes
place at the electrodes of only the first phase, but in the second field i.t occurs
simultaneously at the electrodes of the second and third phases.
The foundation of the balancing amplifier (Figure 6) ia a broad band operational
anplifier A1 in G channel; a 574UD1 can be used in R and B channels. Gain switch-
ing from field to field is done by a divider R2, R8 controlled by a switch on FET
VT2, to which the field freqtiency pulses go. The coupling circuit is made from
- 4 -
FOR OFFICIAL USE 1DNLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9
F'OR OF'FICIAL USE ONLY
~ (1)
~~T, +F * ~ �
, KnJOJf ( ~
( ~ RS R7 R~f 2
j R/ RJ 1K fOR Cl C2
~ JK JK E
' . ~ 1_ - A' I[! I t
~�t
� ~ C +f + ~ '~~J =t6 =CS
~ VTZ ,
~ KnJO~f
. ~ ~ .B6 +f
R2 4 1R RB V~
JK JK !OK KTJ6d
*e" ~ R11
/K
Figure 5. The Preamplifier Circuit
Key:
1. Part number at VTl and VT2 is KP303Ye
2. Capacitive feedback, identified in text
R8
!K
~ 2~ un Rt
� (1) BxcB � KnJOJt{ , ( 6)
+
1K A~ Ba.coB
R4 RJ6
~K R9 l00
aoK Rf7 u~
RS R~ *E ' ?OK
~a~ !,f
~ � A2+ ~ Kn
o,7E ~ 5) '
~
~ ( 4) v� R~oE R~2 . Rt~
Kn.ro~e ~ k c2
( 3~ R~ CJ 0,1 ,
20K af Rff RfJ
E
- R6 +E ~J~ RfB
6 R7 fooK 13rf
R~s -E R?0
. . -E ~ >2~, JJK
~ =t,0
Ifar~WCCrtuA
~ 7) ~ac~emK~
Figure 6. The Balancing Amplifier Circuit
Key:
1. Input 4. KP3038 7. Illumination com-
2. Field frequency pulses 5. KP303Ye pensation
3. Clamping pulses 6. Output
- 5 -
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9
FOR OFF[CIAL USE ONLY
operational amplifier A2, a type 153UD6. A switch on FET t;'T3 is opened by a shift-
ed clamping pulse at the beginning of an active line. The signal level correspond-
ing to the black strip is stored at capacitor C4. This signal is compared with
zero potential and amplified by operational amplifier A2. The amplified error sig-
nal goes to the inverting inrut of amplifier A1. The operational amplifier A3, a
type 153UD6, performs the functi~ns o� illumination compensation and black level
regulation. Regulation of the black level R18 is accomplished by means of a dis-
placement applied to the inverting input, and the compensati~n signal is forn:ec', via
integration at capr~citor C5. The degree of compensation is regulated by resistor
R20. Amplifier A3 produces an error signal which is applied to the input of ampli-
fier A1, and FET VT1 cuts off the compensation signal for the time of the coupling
operation.
Signals having an amplitude of 2 V and coupled to the black level from the balanc-
ing amplifier outputs go to amplifiers with a regulated amplification factor (Fig-
~ ure 7). The regulating element in the amplifiers is a 525PS1 four-square multi-
plier. The signal from collector loads R10, R11 is picked up by broadband opera-
tional amplifier A2 and then passed to the automatics system. This system com-
pares R, G, and B signals and produces error signals. The error signals proceed
across a divider R9, R12, inputting into a feedback circuit, to the amplf.fier~s
controlling input. Resisto~ R8 establishes the nominal gain and the range of auto-
matic balance. At the ci.rcuit output the signal output is 5 V, which is ample for
normal operation of the very simple gamma correction unit employing resistive di-
viders and diodes. ,he signal, processed by the gamma correction unit, proceeds
across the black-and-white-levels-limiters circuit (Figure G). The hybrid integrat-
ed circuit ef the limiters provides white level limitation of 2 V, black level of
60 mV, and also performs the blanking operation. After these operations, the high
frequency part of the signal and the aperture correction signal are introduced in-
to the RY, GY and BY lov~ frequency signals. The swmning is performed in broadbanQ
operational amplifiers with powerful output stages.
The formation of the high frequency portion of the R, G and B signals and the aper-
ture correction signal takes place in the aperture correction unit (Figure 8).
- The G signal (5 MHz band) goes to the input of this cascade from the output of the
balancing amplifier. The broadband signal passes across a delay line and a low
pass filter with a pass band of 1.5 MHz. The high frequency portion of the signal
is formecl by operational amplifier A1 as the difference between the broadband and
narrow band signals. ror noise reduction purposes this signal is processed by a
minimum limiter made up of traasistors VT1 and VT2. The aperture correction signal
is generated by operdtional amplifier A2 and is likewise limited with respect to
minimum (VT3 and VT~i). The correction unit includes provision for regulating the
degree of cor.rection by means of resistive dividers. The high frequency and aper-
ture correction signals are added to the Rr,GY and Br loa frequency signals in sum-
ming amplifiers.
The drop in the FCC of the CCD matrices is caused by aperture distortions which
crop up due to the ultimate geometrical dimensions of the image elements, by the
integrating properties of the output apparatus, by inefficiency of charge transfer
[9] and so on. The circuits in Figure 8 are not sufficient for total FCC correc-
tion. The model therefore includes F~iditional correction circuits. Since an equi-
valent circuit of the output apparatus can be represented, accurately enough, as an
integrating RC circuit, an FCC drop attributable to output apparatus deficiency can
- 6 -
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-00850R004400080046-9
FOR OFFICIAL USE ONLY
u
R3 R4 RS R/0 f ~
~f~ ~ JK /R
RI 1.1 0 fo lI ~+E t Z~
~ ~p q~ 1 ~t BerxaO
1~&a~
~reaeo too 9 J ~ e` 4 E~ u~ 3~
ff/ RlJ JK
6 ld lp,r e.~oa
E ,6~ b
~ ~ 4~ RJ6lcrtna.
~t/t exad Nonp 1K aemnw.
aonaHSc
,~IK
Figure 7. Circuit of Amplifier With Regulated Amplification Factor
Key:
1. Video input 3. Output to automatics system
2. Output 4. Balance voltage input
&pd RI R! p p .
f3rua ~av ~ , s~ v~/
Kr,~a
~ 1 Rt a Rt~ ( 3~
~
exoa R Rw sy
f(!M/u K R!0 ` ~
J
6 ~12
( 2) nK ~K y .
a a
RI R
~ YTJ
R4 ?BK KTJ6a
+ Rzf ~ 4)
~t~ R1 R1d AK
s~ Rn i ~ v~~ ~
RTJ26
Rb 6x 20K ~
`
Figure 8. Circuit of the Aperture Correction Unit
Key:
1. Input, f 5 I~Iz 3. High frequency output
2. Input, f 1.5 N~iz 4. Aperture correction output
can be corrected by a capacitive feedback (C*) in operational amplifier A1 of the
preamplifier (Figure 5). The circuits for corrections of the FCC drop attributable
to inefficiency of. charge transfer are examined in detail in [9, 10]. In these
circuits the amplitude of the correction signal added to the primary signal is
automatically regulated according to the sawtooth (or a more complex) law, depend-
ing on the number of charge transfers from a given image element to the output
apparatus.
'To form pulse trains with a variable PRF, use ie made of circuita dividing one
com~non reference frequency fo=29.75 I~iz. A functional diagram of the logic cir-
cuits for the control of the CCD matrices and part of the synchronizing generator
- 7 -
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9
FOR OFF[CIAL USE ONLY
r------------ ~3~---------------
fo=?9,7SMru ~ Cucmena GltN!(pOMtlJQt C;
4. Primary gating, IIK, K L< M;
5. [see figure];
6. mlEn~ = misn~~
7. Secondary gating, s(t); +
8. Measurement of tdelay f[s (t)].
The spatial selection step takes up an insignificant portion of the landing time.
Thus, to service an aircr aft descending at a speed of up to 160 lm~/hr, a radio
beacon with a directional pattern scanning frequency of 13.5 Hz at a range of 40 km
expends more than 105 scan periods. In the primary gating step, according to (4),
about 100 periods will be used when M= 100.
- 19 -
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9
FOR OFFICIAL USE ONLY
In the secondary gating step (blocks 5, 6 and 7 in Figure 3), the "form" of the
_ signal s(t) is restored, which was damaged by L~~e interference NE(t) = N~,~(t) +
+ N~.a(t) + s~.o(t) and the channel with the radio beacon is gated in accordance
w1rl~ tl~~ ti~~r~mdr~ry ~~,:~tin}; };~~~~~~r;it~?r ~l};nr~l. I~~~c~nuHe ~f ~hn nonHtei~dy-~tnle n:~t~irc~
of N~(t), tt~e technique of digital adaptive interference compensation is promising,
in which the mean square error (SKO) is minimized in the disc-ete real time.
t= nTp, n= 0,1, N- 1(using "empty" channel intervals ascertained during
the primary gating stage) between the process x(n) and the signal y(n) = N=ef(n)�
�{t~l, tu~}, generated by an adaptive filter (ADF) with tunable coefficients
{wi} from the reference interference Nref(n). The minimum of the mean square
error in the channel with the radio beacon is:
min ~m ~`n~) = min {~n ((x� - Yn)~l) = min (m ((sn NEn-Yn)=1}-sm ~s,~,~ ; ~5)
in channels occupied only by interference, min{m[e2]} 0. The presence of intra- ~
path interference s~o(n), which distorts s(n) and makes it difficult to compensate
for N~.~(n) + N~.a(n), in NE(n) impedes the efficient utilization of the well kr_own
algorithm for seeking a minimum of the mean square error of [8]. The influence of
s~.o(n) can be eliminated and the compensation processes speeded up by segragating
the pulse response (IO) h(n) of the medium from the mixture x(n), generating the
interference by the technique of homamorphous filtering [9, 10]. In this case,
the mixture of the signal and interference
x (n) - s (n) ~ N: (n) = s (n) h (n) _ ~j s (ni) h (n - m) ~6)
c~~
is converted to the frequency region c~~= k~w, k= 0,1, N- 1,
N--~
,r (k) ~ x (n) W"k - S (k) H (k), lp - exp j2r.~Jy),
nc0
Then the inversc transform of the logarithmic spectrtna is found:
N-i N-t
X (n) - ~V ~j I In X (k)~ l~-"k = N ~j ( ln S (k) In N (k)~ = s (n) h (n).
Acl k=1
~
The localized capsters s(n) and h(n) are segregated, in which case hN(n), which
def ines the noise N~,~ and N~.a, is inserted in the adaptive filter for the oper-
ational correction of the coefficients {t~i}, i.e., to boost the speed of the
adaptive compensator. As an analysis of the mixture on a YeS-1020 computer has
shown:
-20-
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400084046-9
, FOR OFF[CIAL USE ONLY
a~ (~l = U~~~ Isin St~kl ~ 2cst P sin IQer ~t - t.)~/~SeK ~t - t~~~
where u= p= 0.5 to 1, t3 = 0 to n/i2~R and F~ = 13.5 Hz, the capster li(t) is a
a decay~ng pulse train with a period of t3, is separated from the "continuous"
capster s(t).
BIBLIOGRAPAY
,
1. "MLS Multipath Studies", V. 1., Lincoln Laboratory, I~IIT, Lexington, Massachusets,_
1976.
2. Levin B.R., "Teoreticheskiqe osnovy statisticheskoq radiotekhniki" ["Theoreticai
Principles of Statistical Radio Engineering"], Book 3, Moscow, Sovetskoqe Radio
Publishers, 1976.
3. Kuz'min S.Z., "Osnovy teorii tsifrovoq obrabotki informatsii" ["Principles of
Digital Information Processing Theory"], Moscow, Sovetskoqe Radio Publishers,
1974.
4. Tartakovskiy G.P., et al., "Voprosy statisticheskoq teorii radiololcatsii"
["Problems in Statistical Radar Theorq"], Moscow, Sovetskoye Radio Publishers,
1963.
5. Chechetkin V.D., "Tez. Dokl. ~IXXIII Vsesoyuz. Naucn. sessii, posvyashchennoy
Dnyu radio" ("Abstracts of Reports to the 33rd All-Union Scientific Conference
Devoted to 'Radio Day Moscaw, Sovetskoye Radio Publ~~hers, 1978.
6. Brawn, Gery S., IEEE TRANS., 1977, Vol. 25, No. 1.
7. Akimav P.S., RADIOTEKHNIKA [RADIO ENGINEERING], 1977, Vol. 32, No. 11.
8. Widrow B., Glover D., et al, TIIER [PROC. IEEE (Translated into Russian)],
1975, Vol. 63, No. 12.
9. Rabiner L., Gould B., "Teoriya i primeneniqe tsifrovoq obrabotki signalov"
["Theory and Applications of Digital Signal Processing"], Moacow, Mir Publishers,
1978.
10. Oppenheim A., INFORMATION AND CONTROL, 1967, Vol. 11, Nov-Dec.
COPYRIGHT: Radiotekhnika, 1981
8225
CSO: 1860/361
- 21 -
FOR OFFICUL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400084046-9
FOR OF:~'ICiAL USE ONLY
UDC 621.391.82
PARAMETER SUBSTANTIATION TECHNIQUE FOR ELECTROMAGNETIC INTERFERENCE SIMULATORS
Moscow RADIOTEKHNIKA in Russian Vol 36, No 6, Jun 81. (manuscript received 16 Apr 80)
pp 74-76
[Paper by V.V. Kuznetsov, A.A. Lyubomudrov and L.F. StefanovichJ
[Text] Radioelectronic equipment (REA) is frequently subjected to various kinds
of electromagnetic interference (EP) (lightning, industrial interference, etc:),
which leads to operational dropouts or irreversible failures [1]. When protecting
radioelectronic equipment against such interference, tests of noise immunity oc-
cupy a special place, which are being conducted with increasing frequency by means
of specially designed interference simulators [2]. The problem of selecting the
optimal interference simular pulse parameters for the tests has not been solved at
the present time. The difficulties involved in its solution are due to the random
nature of electromagnetic interference, the crnnplexity of generating pulses with
parameters close to the actual ones as well as the diversitq of the radioelectronic
equipment to be tested. We shall find the optimal parameters of the test pulses
based on a probabilistic statistical approach to the tests.
The most complete criterion of radioelectronic equipment efficiency operating in
the presence of noise is the successful operation probability, Psuc.~ Which should
be somewhat greater than a certain probability specified in the technical specifi-
cations, PSFeC., with a confidence level of Pcon.~ i.e. [3, 4]:
> p
P~Psuc. spec.~ - Pcon. 1
For this reason, it is expedient to simulate the impact of interference on radio
electronic equipment, taking into account ihe�final goal of the test, expressed
mathematically by means of (1).
Let ~.~(tI) be the distribution density of the random characteristic P oP an
interference field at the radioelectronic equipment input. Depending on the situ-
ation, such a characteristic can be the maximum electrical or magnetic field
intensity, or the current or voltage induced by the electromagnetic field in
cables or interassembly connections [5, 6]. The distribution density ~~(II)
depends on the interference source power and its distance from the radioelectronic
equipment, which are, as a rule, of a random nature. The distributions of the
-22-
r'OR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-00850R004400080046-9
FOR OFFICiAL USE ONLY
characteristic of II for the case of lightning interference have been found experi-
mentally and theoretically in a nwnber of papers [7, 8] .
,
We shall assume that the timewise characteristics of the interence are either
determinate or averaged. In the case where this condition is not met, to avoid
omissions of frequency resonance, the timewise characteristics of the test field
should run through the entire range of variation in the time characteristics of
the interference which are possible in the given situation during the testi~ig
process.
The parameters of electromagnetic interference simulators can be substantiated
by means of the f ollowing method . We f i.rst f ind the minimum permissible level of
s}stem i~unity to interference, correspo.^.~liz~g to the successful operation proba-
b ility P SPec, Then, taking into account the statistical scatter in the interfer-
ence generated by the simulator, its level is determined for the test. Thus, the
essence of the method consists in determining the minimum interference level which
confirms the minimum permissible successful operation probability.Pspec�
By way of example, we shall consider the case where the transformation of the
electromagnetic field of the interference in the shields and circuits of radio
electronic equipment and the failures which occur are linear processes. In the
case of linear conversion of the randam quantity II, there is a change in the
scale of the curve ~~(n), while the overall shape of the distribution curve does
not change. The distribution density of the random quantity II reduced by a factor
of x times is:
~i (Mt) - ~Po (XMt) X. ~ ( 2)
where Mi = II/x is the amplitude induced in the i-th radioelectronic component.
Let the equipment have a"weak" link, the ~mmunity of which to el~otrical overloads
is substantially less than the imnunity of the others. The immnunity level M3
[Me] of the "weak" link component to electrical overloads has a statistical scatter
which can be described by the probability density ~e(Me). Fram the equation [9]
a, df.
pspeC - pT. a- S T� (M,) dM. S~n (X~ Mt).x~ dM~, ~3)
taking (2) into account, we sh all find the minimum permissible attenuation factor
x~ of the amplitude of the interference II, at which the specif ied probability of
successful operation of the equipment PspeC is just achieved.
Let the laws governing the scatter ~N(IIH~ of the interf erence amplitude from the
simulator be known relative to any nominal value mH specified beforehand. We find
the desired interf erence level mH,~ for the ~ests fram the following equation:
-23-
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-00850R004400080046-9
FOR CFFICiAL USE ONLY
(4)
� Afs .
pT. a~ S~~ ~Ma) dA9~ S~Px ~X~'Mle) x~ dMlx~
where `t1~" �
For normal distributions of ~,~(II), ~e(Me) and ~H(IIN) with mean values and mean ,
square deviations qf m.~ and me and de and mH and QH, we obtain the following
from (3) and (4) respectively following transformations:
r ,n. - (m~lXm)1 �
pr.a ~ ~ S~/e;+(Q~x~)'J~ (5)
~ 1'
m. -r (~N. m/xm) ~
pr. s ~ ~
{ vo; + cQ.. ~iX~r ' ' ~6~ :
where ~{y} is the probability integral; aH.~ is the mean square deviation from
the nominal value of mH,~. An analysis of equations (5) and (6) shows that with
an increase in PSpec, x~ and mH,~ also increase.
~
For normal distribution ~f the load and immunity, the praposed technique can
also be used in the case of an unknown i~unity of the components to electrical
overloads. In this case, it is essential to known only the coefficient of vari-
atii_on ~e/me, the value of which depends on the perfection of the production tech-
nology for the radioelectronic equipment components. By dividing the numerator
and denominator of the expressions inside the curly braces in equation (5) and (6)
by m.~ [me] and writing x~me = x~, we obtain:
~ _ m~~X~ 1 _ ~N: ~~X~ ~
~ ~
I'T. a! m }/(Q,/n?�)'~-(Qn/X~)' . Pr, a ~ j/ ~Q~/~i~'-F ~QN. m/X~I'
In the case of unknown coefficients of variation ve/me and 6H/mH we find the
quantity mH,~ froin (7). As we see, an increase in the ratio Qe/me, mH.~ decreases.
Consequently, for an unknown immunity of the system components to electrical over-
loads, the customer designating the interference level for the tests should use
as the basis the smallest possible coefficient of variation of component immunity.
It is expedient to determine the latter in laboratory tests. Moreover, one can
make use of the data for similar components.
-24-
FOR OFFICIAL USE ONLY �
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-04850R000400080046-9
FOR OFFICIAL USE ONLY
In the case of a sophisticated production technology for the components, equations
(3; 3nd (4) are simplified:
Me ~ M~ .
pr. a m~~a ~x~ Mt) x~ dM~. ~�s.s~ pM (x~~~fix)x~ d~'1ix�
~ . . ~
With sufficiently good repeatability of the interference amplitude from the
simulator, equation (8) assumes the form: mH.~ = Mex~.
The number of samples for the tests is:determined from Pearson's and Klopper's
formula [4] .
The transformation of the electromagnetic field in radioelectronic equipment can,
~in the most general case, be the consequence of nonlinear processes. The immunity
of nonlinear camponents to electrical overloads is no longer described by the
amplitude of the induced voltage (as in the case of linear systems), but rather
by the iriduced energy or power. When intarf erence flows through nonlinear systems,
the distributions governing the interference characteristics change their form.
For this reason, even if the distribution of the i~unity of the individual ~
system components to electrical overloads is known, the law governing the distrib-
ution of the nonlinear system 3mmunity ~~T(IICT) is most often unknown. We approx-
imate the distribution of nonlinear system immunity to interference with a normal
distribution, by specifying a sufficiently low coefficient of variation cr~/m~.
We find a certain boundary distribution density from (3), ~~T.~ ~n~T) having a
mean yalue of m~T.~ and a mean square deviation of ~ ~c~Tm~T,~ , correspond-
ing to the successful operation probability: Q~*�~ ~ m~ ~
_ ' ~ (~cT "'er.~~' ~ '
ee ' ~ r�e*~
er. ~ la ner ~
PspeC - pT. s-`QeT ~er. m e ~ ct 1 d�~ aS ~n ~/1) d17.
a�
~~t . ~
We determine the desired interf erence level for the tests, mN.~, the scatter in
which is characterized by the distribution density ~H.~(IIH), from an equation
analogous to (4): , ~
�cr"'~eT. . , .
m Z r�er
�~er.
~ 1 . ner .
r. J aer mei. ro e ` QI7
e s ~ 4 N. m~R w~ d/T
o f/2~
~T
-25-
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000440080046-9
FOR OFFICIAL USE ONLY
Moreover, when testing nonlinear systems ta find the possible amplitude resonances,
~ it is necessary to act on one of the samples with a stepped changing load in a
range of variation of II from zero to m;~.~.
BIBLIOGiAPHY
. 1. Nawnov Yu.Ye., Avayev N.A., Bedrekovskiy M.A., "Pomekhoustoychivost' ustroystv
na integral'nykh logicheskikh skhemakh" i"Tkie Interf erence Immunity of Devices
Using Logic IC's"], Moscow, Sovets~coye Radio Publishers, 1975.
2. Galkin A.P., Lapin A.N., Samoylov A.G., "Modelirovaniye kanalov sistem svyazi"
"Modeling the Channels of Com~nunications Systems"], Moscow, Svyaz' Publishers,
1979.
3. Gurvich I.S., "Zashchita EVM :t vneshnikh pomekh" ["Protecting Computers against`,
External Interfer-~nce"], Moscow, Energiya Publishers, 1975.
4. Pupkov K.A., Ko~tyuk G.A., "Otsenka i planirovaniye eksperimenta" ["The Evalua-
tion and Planning of an Experiment"], Moscow, Mashinostroyeniye Publishers,
1977.
5. Alizady A.A., Khydyrov F.L., ELEKTRICHESTVO [~ZECTRICITY], 1978, No 9.
6. Bazutkin V.V., Zaporozhchenko S.I., ELEKTRICHESTVO, 1975, No 1.
7. Bronfman A.I., "Rezhimy raboty ventil'nykh razryadnikov pri grozovykh perenap-
ryazheniyakh" ["Operational Modes of Diode Dischargers in the Case of Lightning
Induced Overvoltages"], Moscow, Energiya Publishers, 19771
8. Alizade A.A., Muslimov M.M., Khydyrov F.L., ELERTRICHESTVO, 1976, No 11.
9. Kapur L., Lamberson L., "Nadezhnost' i proyektirovaniye sistem" ["Systems Design
' and Reliability"], Moscow Mir Publishers, 1980.
COPYRIGHT: Radiotekhnika, 1981
8225
CSO: 1860/361
-26-
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2407/02/09: CIA-RDP82-00854R000400080046-9
~ FOR OFFICIAL US~ ONLY ~
UDC 621.396.669
APPLYING POSINOMIAL ESTIMATE TO EFFICIENCY DETERMINATION OF
EQUIPMENT WITH HIGH ELECTROMAGNETIC COMPATABILITY INDICATORS
Moscow RADZOTEKHNIKA in Russian Vol 36, No 6, Jun 81 (manuscript received 2 Jan 80)
PP 76-77
[Article byrA.D. Kaluzhskiy]
[Text] The electromagnetic compatability (II~iS) indicators of equipment can be
improved in the general case by means of incorporating additional devices~(DU)
and by changing the characteristics of the equipment itself, for example, ~he
linearization of operational modes of amplifiers and additional shielding of
assemblies. This entails a change in a number of equipment indicators, and conse-
quently, in equipment efficiency.
The problem of obtaining a function for the change in equipment efficiency when
improving its electromagnetic compatability indicators is a particularly acute
one now, since an improvement in electromagnetic compatability indicators of equip-
ment is accompanied by an increase in equipment complexity, size, weight as we11
. as a degradation of a number of other characteristics, something which at a
certain point leads to a reduction of equipment efficiency as a whole [1J. When
deriving such a function, it is necessary to choose an optimum design variant for
the equipment for each value of the electromagnetic compatability indicator and
correspondingly determine the weighting factors for each of its indicators. The
specif ic features of this task are those situations where an improvement in'electro-
magnetic compatibility indicators leads to a slight change in some of the equip-
ment indicators, while others are constant. In this case, a nonlinear estimate
of the efficiency is needed which makes it possible to ascertain and not lose these
changes. It is expedient to use posinomials as such estimates: a nonlinear esti-
mate proposed by R. Daffine, et al. [2], and used to estimate the efficiency of
communications systems by Yu.M. Vozdvizhenskiy [3]. A special case expression for
such an estimate has the form:
. LA 1 1 ~11R ~ 1~
i -
where k is the number of draft designs of the equipment, each of which has its
own electromagnetic compatibilitq indicator; Lk is the efficiency of the k-th
-27-
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400080046-9

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000440080046-9
FUR OFb'ICIAI. USF: ONLY
design; nik is the coefficient of success of the i-th indicator of the k-th equip-
ment design; ~i is a coefficient which is defined as the weighting factor of the
i-th equipment indicator. In (1), the quantity rlik can be defined by the relation-
ships of [4] :.+jt~?~a,~~.,,;l~�rnr?=a~*~a~ ~a=. where aik is the value of the i-th indica-
tor of the k-th draft design of the equipment; ai min and a i~X are the best values
of the i-th indicator from among the k designs considered for the equipment.
, To determine the value of ~i, we plot the family of .
curves rt~ for various values of ~(see the figure) . .
o'ol Since small deviations of ai from ai max ~d ai min
qs ' .
Q8 ap are assumed, and consequentl~, also small deviations
4~ - of r1 from 1, and a significant ~:hange in n~, then ~
q ~8a'~ - should vary fram 1 to i.e., 1