TRIP REPORT - NOISE MODULATED (SANITIZED)

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP78-03330A004100060013-2
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 2
• Document Creation Date: December 22, 2016
• Document Release Date: July 31, 2012
• Sequence Number: 13
• Publication Date: October 31, 1958
• Content Type: MEMO

Declassified in Part -
* _ STANDARD FORM NO. 8~
~~CC M2Y~ZOY'GLYIdZLYIZ UNITED STATES GOVERNMENT
FROM
susJECr: Trip Report -Noise Modulated
~ ~1 DATE : ~, October 1958
OR G C@ p~n~ R~'V2 DATg~~ BY O (o_ Yj'yn
@R!G CLASS ~-'~+~ @~, ~ TYFE ~_.-"~-~x
+~- PAG~s Z--
JUST ~ U~XT BEV ~Q~~ EIaSS ,lC~.
At3T14; 10@ t~
Sanitized Copy Approved for Release 2012/07/31 :CIA-RDP78-03330A004100060013-2
discussions concerning this program were:
as r , u o n+o se mo ted Participating in
On October visit was maaie to
to monitor t e progress o L26~
2. revie~red the theory of noise modulation and
detection. He pointed out that the correlation detection employed in
such systems. utilises the total ener63r present in a signal regardless
of its waveform and that the waveform of a signal ma}r vary great]y
without aay boss in total energy transmitted. anti-3am noise systems
employ a rawpidly chang~.ng waveform which the er~ay finds extreaee],y'
difficult to duplicate, and without ari eataat replica of the signs]:,
demodulation is in~ossihle. If the ener? in the signal is uniform]y
distributed in both time and frequency, it sounds like ryarma7, back-
ground noise to conventional intercept equipment. The ability of the
correlation receiver to integrate widely dispersed pulses makes it
possible to recover sigiala well beneath the noise level, thus pro-
viding another avenue of protection against unfriendly intercept.
3. The extent to which the system will operate below the. ambient
noise level is determined by the WT product -the combination of
band length and bandwidth. The greater this WP products the greater
protection a signal has against 3amni:~g or intercept. ~ Since arty s e
pulse, regardless of its shape, has a WT product of 1, it is aeaesaary
to tramm~dt a string of pulses for~each.baud to gain a Wr advantage.
said that .two techniques are in current use for generating
such a string of pulses.
1~. The first method, used i.n th is a multiplier-
integrator system which transmtts the output of_one of two key genera-
tors representing mark and space. An exact duplicate of the key generator
signal used at the transmitter must_be avsilah7.e at the receiver for
comparison purposes and it is here that the principal shortcoming of
the multiplier-integrator technique is evident. Precise time information
CpNFIDENTIA~
Q
[~` (~
Declassified in Part -Sanitized Copy Approved for Release 2012/07/31 :CIA-RDP78-03330A004100060013-2
 Declassified in Part -Sanitized Copy Approved for Release 2012/07/31 :CIA-RDP78-03330A004100060013-2
~I`D~NT~~L ~~.
ie needed to keep the two reference signals in synchronism, since the slightest
drift causes the correlation detector output to fall off sharply, and with
very much plos8 of ayncn the intended receiver is no awre able to demodulate
the incoming noise signal than is the enea~- intercept station. Attempts have
been made to transmit the reference signal to the receiver on another frequency,
according to but the path losses and distortion introduced make sash
a solution pre worthless.
5. In the second method, an impulse is sent into a comgleat filter, and
the output of the filter, which appears as a badly distorted noise s3.gnal,
is transmitted to the reaeivero There it is fed into a compl.emsntary filter
and emer es as a reasonable facsimile of the original pulse. These two
filters, observed, are merely a split delay line. The structure
of the delay line is changed constantly according to a prearranged node to
frustrate enemy analysis. Precise time synchronism is not necessary with
matched filter systems, but the complexity of the filters makes it doubtful
that this is a suitable a roach for our noise modulated
sa? d t~~ has recently designed a aia c e er s staQn
for and that it was being constructed at
It is a 60 ~ system using a 10 kc bane .
6. said that it was nc~w ossible to redid certain thic~gs
abort the propos noise modulated that were not certain
when the program began. A crystal oven in the field unit near appears aLrost
inevitable, since the stability of the transmitted signal is e~ctremely
critical, If awry than about 10 cycles of frequency drift occurs, according
to preliminary aalculations~ the message could be missed completely. FS$
modulation will probably be used and about 23 db of signs], compression.is
anticipated. It may be advisable to begin transmitting the signal gradually
so that there is no abrupt increase in noise level on the operating frequency.
strop urged that a similar to that used to
be incorporated o ?r~~e equipment to combat
pa a said that 7 to 8 db of signal. power would be lost if multipath
reflections - which aontain a siaable amount of the transmitted energy -are
not recovered and correlated. The contractor inquired whether a short burst
of CW could precede the transmitted message for time and frequency 8ynchroni-
zation, He was told that shah a solution was highly unsatisfactory and asked
to itTVastigate the suitability of a 140 millisecond recognition signal using
a very simple cods which would Derry little or no information but which might
solve the reaogrsit3.?n problem. He was told that the base station wr~uld
know
~ONFID~~T~AL
~`~~~~~'
Declassified in Part -Sanitized Copy Approved for Release 2012/07/31 :CIA-RDP78-03330A004100060013-2