US2401402A - Secret communication system - Google Patents

Description

June 4, 1946. A. v. BEDFORD SECRET COMMUNICATION SYSTEM 6 Shets -Sheet 1 Filed Aug. 29. 1942 n m 6;: M f d 6mm H u..e L. L. 6S8 B Mum d L A a 6mm Q\Qbk k 5 3m A c BEN a 1 mm. QM H Q\ E w June 4, l9

Fig.5.

v. BEDFORD SECRET COMMUNICATION SYSTEM Y Filed Aug. 29, 1942 6 Sheets-Shet 3 234 a a 7 a 810 '11 12131415161 0 flair; 5.5

Fit $.12.

ffi G. 10.0

can; area #1; h

enter rd Q 3 HZda-VBedfo attorney June 4, 1946 A. V. BEDFORD I 2,401,462

SECRET COMMUNICATION SYSTEM v Filed Aug. '29, 1942 6 Sheets-Sheet 5 Zmnentor a The invention lother embodiment of the invention;

code- Patented June 4, i946 2,401,402 SECRET COMIWUNHCATION SYSTEM Aida V. Bedford, Collingswood, N. J...

Radio Corporation of America,

of Delaware assignor to a corporation Application August 29, 1942, Serial No. 456,578

(Cl. l79-=-l.5)

19 Claims. 1

My invention relates to communication systems and particularly to systems for transmitting information so that it is unintelligible to anyone not provided with the prearranged code information.

An object of the invention is to provide an improved means for and method of transmitting information secretly.

A further object of the invention is to provide an improved means for and method of utilizing a code signal for making a communication signal unintelligible to an enemy or other unauthorized person.

A still further object of the invention is to provide improved means for making code discs or the like for use in a system of the above described character.

In practicing my invention the communication signal S is multiplied by the code signal K before transmission. Thus, an unauthorized person can determine only with considerable difficulty. if at all, how much of the signal amplitude at a certain instant is produced by the communica tion signal and how much by the code signal. Since the transmitted signal SK is a product of the two signals, there will be no transmission of signals (and, therefore, no transmission of code signal) when there is no communication signal.

At the receiver the signal S is obtained by multiplying the received signal SK by the reciprocal of the code signal, i. e., SK 1/K=S. The code signal l/K is produced at' the receiver by suitable means such as a code disc which is held in synchronism with the code-signal-producing means at the transmitter.

will be better understood from the following description taken iii-connection with the accompanying drawings in which Figure 1 is a block and circuit diagram of one I embodiment of the invention,

Figure 2 is a circuit diagram of a modulator unit that may be utilized in the system of Fig. 1,

Figure 3 is a group of graphs that are referred to in explaining the operation of the modulator of i Figure 4 is a, block diagram illustrating another way of synchronizing the code signal producing means at transmitter and receiver,

Figure to '13 are graphs that are referred to in explalni'ng the invention, I

Flgur'el isa block diagram illustrating an- Figure'15-is a perspectlveyiewfof apparatus for discs-for. thetransmitter and re- Figures 16 to 18a show, by way of example, the shapes of certain dies for cutting contours on code discs to obtain the code signal K and its reciprocal l/K, and

Figure 19 is a block and circuit diagram showing another embodiment of theinvention.

In the several figures, like parts are indicated by similar reference characters.

Referring to Fig. 1, the invention is shown applied to radio apparatus that may be switched for operation either as a transmitter or as a receiver. The several switches are shown in the po-' sition for transmitter operation. This is the operation of the apparatus thatwill first be described. A microphone and a speech amplifier are-shown at I and 8, respectively. The signal S is applled.through a switch 9 to the multiplier unit [4 which is enclosed by broken lines. The code signal K may be produced by means of a code disc IS, a mask II, a light source la, a condensing lens I9, and a photoelectric cell 2 I. Signal K is supplied through an amplifier 22 and a switch 23 to the multiplier unit It.

In the specific example of Fig. 1, the signal S (Fig. 5) is split into the two portions that are positive and negative with respect to the A. C. axis of the signal by means of diodes or other suitable rectifiers 28 and 21. The signal is supplied to the diodes 26 and 21 through a coupling capacitor H and a grounded grid leak resistor l2 of suitable low resistance. The positive portion A (Fig. 6) is applied to a modulator 28. The negative portion B (Fig. 7) is reversed in polarity by an amplifier stage 29 having unity gain and applied to a modulator 3|. A carrier-frequency signal F is applied to modulators 28 and SI from a source 32 whereby carrier waves modulated by signals A and -B, respectively, appear in their output circuits. The output of modulator 28 (after being filtered by high-pass filter 39) is shown in Fig. 9.

Modulators 28 and 3| may be of any suitable type, but, preferably, modulate linearly so that no signal distortion is introduced.

through resistor 31. Thus the location of the carrier signal F with respect to the tube characteristic 36 is varied by the signal S as shown in Fig.

One suitable modulator is shown in Fig. 2. It comprises a vacrespectively. The outputs 3 and a modulated carrier. wave is produced. action is, of course, well known in the art.

The outputs of modulators 28 and 2| are passed through high pass filters 88 and 4|, respectively, to remove the original audio signal S and are then applied to multiplying units 42-48 and 44-48, respectively. As explained below, each multiplier unit has applied to it either the positive portion or the negative portion of the code signal K. Each of the units 42, 48, 44 and 48 may be a unit like the modulator shown in Fig. 2 where the tube characteristic is like that shown in Fig. 3, or stated differently, each of the units 42, 42, 44 and 48 may be a variable mu tube which has its gain varied by the code signal.

The code signal K (Fig. 8) is split into the two portions that are positive and negative with respect to the A. C. axis of the signal. This may be done by applying signal K through a coupling capacitor 41 and a low impedance grid leak resistor 48 to a pair of diodes 48 and ii. The positive portion M (Fig. 8) is supplied through a conductor 52'to the multiplier units 42 and 44. The negative portion N (Fig. 8) is reversed in polarity by an amplifier stage 58 and supplied through a conductor 54 to the multiplier units 43 and 48. Assuming, for example that the multiplier units 42, 43, 44 and 48 are like the modulator circuit of Fig. 2, it will be understood that at each unit the code signal (either M or N) is applied to the control grid through a resistor while the signal modulated carrier wave from either modulator 28 or modulator 8| is applied through a coupling capacitor to the control grid.

The resulting output of unit 42 is the carrier signal or wave F modulated by signal A times signal M, that is, the output is F modulated by AM. Similarly, the outputs of units 48, 44 and 48 are, respectively, F modulated by -AN, F modulated by -BM and F modulated by EN.

The facts that the outputs of the multiplier units are the carrier signal F modulated by the product of the communication signal and the code signal will be apparent if it will be kept in mind (referring-to Fig. 3) that the carrier-frequency outputof a modulator or multiplier unit is proportional to the applied low frequency signal. For example, the carrier-frequency output of modulator 28 is proportional to signal A (the positive'portion of S) and the carrier-frequency output of multiplier unit 42 is proportional to the code signal M. More specifically, if the tubes in units 28 and 42 each have acharacteristic such as to provide linear modulation, the carrierfrequency output of unit 28 is CFA where C is a constant, and the output of unit 42 is CCFAM where C is a constant.

The outputs of multiplier unts 42, 43, 44 and 48 are passed through high pass filters 58, 51, 58 and 59, respectively, to filter out any of the the "average frequency" original or low frequency code signal that may be present.- These outputs are next applied to.

detectors 8 l 82, 88 and 84, respectively, to produce the detector outputs AM, AN,

of. detectors 82'and 83 are reversed in polarity by amplifier stages 88 and 81, respectively, to produce outputs AN and BM, respectively. Figs. of example, the signal products AN and BN'for the signals S and K shown in Figs. 5 and 8.

The signals AM, AN, IBM and EN are added in a suitable mixercircuit 88- whereby the mixer output is the complete signal SK illustrated in Fig. 12. This is passed through a low pass filter 88 to remove any carrier frequency that may be 10 and 11 illustrate, by way current also modulates a present. It is then amplified by an amplifier II and supplied through a switch 12 to a radio transmitter 13 or to a wire line, if preferred.

From the foregoing it will be apparent that the transmitted signalv SK may be made 'to bear so slight a resemblance to either the communication signal S or the code signal K that neither one of these signals can be determined by unauthorized persons except by laborious "cut and try methods. The coded wave SK has neither the same shapes, frequencies or amplitudes as the original waves. v

The code signal K should provide enough reversals of the communication signal S (speech signal for example) to destroy any semblance'oi the original tones and harmonics. For example, if signal 8 contains no frequencies higher than 3000 cycles per second, the coding disc l8 may be provided with 50 positives and 50 negative lobes and rotated at 20.revolutions per second whereby about 1000 cycles per second. It the disc i6 is three inches in diameter each lobe would average about one-tenth inch in length along the disc periphery so that no difilculty should be encountered in obtaining adequate accuracy in making the coding and decoding discs with simple equipment.

The receiver for decoding the signal SK may be the same apparatus as in Fig. l with the switches 8, 22 and I2 thrown to their lower contact positions R. The multiplier i4 is unchanged. The signal SK is supplied from a radio receiver 18 through the switch 8 to the multiplier l4. A decoding signal l/K produced by a disc 11 and associated light source 18, condensing lens 18, mask 8i and photoelectric cell 82 is supplied through an amplifier 88 and the switch 28 to the multiplier l4.

The code disc l8 at the transmitter and the code disc 11 at the receiver are held in synchronism and in the proper phase relation by suitably synchronizing means. For example, at the transmitter a GOO-cycle per second current from a source 84 may be supplied through switch arms and 85 to a synchronous motor 88 which rotates the code disc i8 and 11. The 600-cycle radio transmitter 81 for the transmission of synchronizing signal to the receiver. At the receiver the switch arms 80 and 85 are in their lower positions R whereby the received 600 cycle current is supplied from a radio receiver 88 to the synchronous motor 88. In some cases it may be preferable to transmit the synchronizing signal over a wire line.

The received signal SK is decoded in the receiver multiplier i4 since the multiplier output is SK(l/K) :8. The signal S may be supplied through the switch 12 to headphones or to a loudspeaker 89. No detailed discussion of the decoding operation of the multiplier I4 is required since it has previously been shown how it operates to multiply the signal supplied to diodes 28 and 21 by the signal supplied to diodes 48 and 5|. If these two signals are SK and 1/K, respectively, it' follows that their product is the. desired signal S. v

It maybe noted that if the modulators 28 and 8! are biased to cut-off as shown in Fig. 3, the diodes 28 and 21 may be omitted since the modulator tubes themselves act as rectifiers and are not affected by the negative portion of the signal. It will also be evident that the code signal for transmission may be l/K and the decoding sigof signal K would be i nal K instead of vice versa as assumed in the foregoing description.

For the purpose f illustration the ordinate values for the waves of Figs. 5 to 13 are given in the table below:

QAOLMQ gether that they run at speeds that difier slightly, say by one percent. The combined coding wave KK will then repeat (assuming the particular disc design previously described) only after the g passage of some 5,000 to 10,000 lobes of the coding Time s A B p K M N AM AN BM BN BK UK .9 .9 o 1.8 1.8 0 1.8 0 o o 1.6 .126 c o o .4 .4 o o o o o o 2.5 o o o 1.3 1.3 o o o o o o .77 -.e o .c 1.7 1.7 c c o -1.o2 o -1.o2 .aa 1 .I5 0 -.5 -.s o .s o o o .40 .1 1.25 .o .5 o -2.0 0 -2.0 o -1.0 o 0 -1.0 -.s .e .6 o -1.5 o -1.5 0 .o 0 o -.9 -.e7 0 o o -.7 o -.7 o c o o 1 0 -1.43 -.s o .5 -1.1 o -1.1 o o 0 .55 .115 -.91 -.4 o -.-1 o -.4 o o o .10 .10 -u -.7 0 -.7 1.2 1.2 0 o 0 .s4 0 o 83 1a -1.0 o -1.0 1. 1.9 o c 0 -1.9 o o .53 14 -.4 0 1.9 1.9 o o 0 .7c 0 o .53 16 .a .a o 1.5 1.5 c .45 o o o 0 .117 16 .7 .7 c 0 o -1.1 o .77 o o o -.91 17 .c .c o o o -1.7 0 0 0 0 -.s11

Fig. 4 illustrates another way in which the wave KK', or about once every five seconds. This coding and decoding discs may be synchronized. would make an early solution of the code by an The transmitter carrier wave is modulated by a enemy unlikely, especially when it, is considered synchronizing signal having a frequency, such that he must not only obtain a substantially coras 4000 cycles per second, outside the frequency rect wave shape, but he must phase it correctly. band of the speech or other communication sig- Fig. 15 illustrates apparatus that may be emnal, the upper limit of this band being deterployed for making coding and decoding discs. mined by the low pass filter I0. The 4,000-cycle Two such discs are shown at III and H2, respecsource, which is indicated at 9|, also supplies so tively, in the process of being cut to the desired synchronizing signal over a conductor 92 and shape. A plurality of male dies H3 and II 3a through a switch 93, a high pass filter 90 and an are mounted around the periph y of a rotatable amplifier 98 to the synchronous motor 08. supporting disc IIt. One of the dies II3 is shown Similar apparatus at the receiving station is pulled above its normal "rest position to show put in condition for reception by moving switches the lower die I I6 more clearly. The correspond- 9, 23, I2 and 03 to the contact positions R. The lug female dies II8 are formedina disc II? which radio receiver 91 now passes the signal SK rotates with the disc IM. Levers H0 and H0 are through a low pass filter 98 to the multiplier It pivoted on a supporting frame IZI so that when and passes the 4000-cycle synchronizing signal they are depressed the dies cut contours on the through the high pass filter 04 to the motor 88. 1 discs III and 2 that will generate the signals Thus the coding and decoding discs at trans-mit- K and 1/K, respectively. ter and receiver are held in synchronism in a In the specific design of Fig. 15 alternate dies system requiring the transmission of only one H3 ut a t ur to p uce s al K while th carrier wave. other dies II3a cut a contour to produce signal Figure 14 shows an embodiment of the inven- /K. A notched disc I22 and a positioning arm tion designed to transmit messages that will be I28 prope y 13051111011 the K and 8 i h still more difiicult for a person not provided with respect to discs III and H2 each time the supthe code key or code disc to decode. An addiporting discs II and H6 are rotated to make tional multiplier IN is employed which may be a new cut. a duplicate of the multiplier Id. The output of The supporting shafts I28 and I21 are held in multiplier IOI is fed into the multiplier It the a definite rotatable relation by means of sprocket same as the code signal K is fed. into it in the wheels I28 and I20 connected by a chain I3I. embodiment of Fig. 1. A toothed wheel I32 and a positioning arm I83 Multiplier IOI supplies the product of two code insure rotation of the code discs III and H2 a signals K and K to the multiplier Id. The .code whole number of spaces or distance units 11 meassignal K may be produced by means of a. photoured along the periphery of the discs. Since the electric cell I02 and a code disc I 08 that is driven, length of each die is a whole number of said preferably through gears I00, by the synchronous units, as indicated in Figs. 16 to 18, this facilimotor 88. Signal K is supplied through an emtates positioning the discs I II and H2 for the plifier I00 and a. switch I08 to the multiplier IOI. next cut to be made. by the dies H3. It will be When the code signal K is also supplied to mulevident that various modifications may be made tiplier it! its output is m. This is assumed in the code uttin apparatus. F r xampl to be the condition for transmission. The rethe code discs II I and II 2 may be diametrically suiting signal obtained for transmission is SKK'. opposed with respect to the supporting disc H0 For reception, the several switches are moved and all K dies located along one half or side of to the R position and the signals l/K' and l/K disc II 8 and all l/K dies located along the half are supplied to the multiplier IOI. The decodor side of disc Ild. Any particular coding disc ing signal l/K' is produced by a disc I01 and ay be cut by usin ither a few or all of the associated photoelectric cell I08. Thus the sigdies H8 and in any random sequence. nal supplied to the loudspeaker 89 is Figs. 16, 17 and 18 show dies I I 8 of three separate shapes or wave forms K10, K11 and K12. SKK K 4 Figs. 16a, 17a and 18ashow the dies H00 having T make the double coding y tem of Fig. 14 the corresponding reciprocal wave forms l/Kio, most efiective; the shafts carrying the code discs I/Kn and l/Kia. The positive and negative Ii-W and nus-m should be so geared to- 13 value of K and I/K are measured from the broken lines in Figs. 16 to 18, these lines representing a certain radial distance from the center of the code discs Ill and H2. It may be noted that where the value of K is zero, at a point 2:, in Fig. 16, the corresponding correct value of UK i infinity but no serious distortion is caused by making l/K some convenient definite value at this point.

It will be understood that the code signals K and l/K may be generated in various ways. For example, phonograph recordings of K and UK might be used and the signals taken oil by the usual phonograph pickup. Another method is to use a disc of film or other photographic material located in front of a cathode ray tube, the code having been recorded on the film as variable density. The disc is then scanned by sweeping the luminous spot on the tube fluorescent screen in a circle and a photoelectric cell picks up the light transmitted through the film.

liLv invention is not limited to any specific apparatus for multiplying a signal by the coding Wave K or the decoding wave l/K. Fig. 19 shows another type of multiplier 14A that may be substituted for the multiplier i4 in Fig. l. The components A and -B of signal S are applied to light sources such as neon lamps 2" and 2i1 respectively, that may be modulated. Batteries 216 and 2H may be provided to maintain a suitable D.-C'. potential across the lamps 2l6 and 2H.

Light from lamp M6 is directed over two paths as by means of mirrors 22l and 222. Lens systems 223 and 224 concentrate the light on slits in 'diaphragms 226 and 221., These slits are imaged by lens systems 223 and 22! upon slits in diaphragms Hi and 232, respectively, after reflection from galvanonieter mirrors 233 and 234. Each of the mirrdrs 233 and 234 is rocked about a pivot point whereby the amount of light passing through the slits oi diaphragms 23i and 232 and falling upon photoelectric cells 235 and 236 varies in accordance with the signal rocking the mirror. The mirrors 233 and 234 are rocked or vibrated by the code signal components M and -N, respectively, these components passing through amplifiers 231 and- 233 to the mirror driving coils 239 and 2. The structure for the galvanometer mirrors and their driving mechanism may be as described in Dimmick Patent 1,936,833. The resulting outputs of photoelectric cells 235 and 236 are AM and AN, the output -AN being reversed by a reversing amplifier stage 240.

Similarly, light from lamp 2 is directed by mirrors 242 and 243 through lenses 244 and 246, diaphragms 241 and 246, lenses 249 and 2 to the galvanometer mirrors 252 and 253, and from galvanometer mirrors 252 and 253 through the slits of diaphragms 254 and 256 to photoelectric cells 251 and 260. The outputs of photoelectric cells 251 and 266 are -BM and EN. The signal BM is reversed by an amplifier 26l. The signals AM, AN, BM and EN are added to produce the complete coded signal SK.

I claim as my invention:

The method of communication which comprises producing a, communication signal and a distorting signal, multiplying the values of said signals as measured from their alternating curent axes by each other to obtain their product, transmitting said product to the point of reception and there multiplying said product by a sig nal that is substantially the reciprocal oi said distorting signal, said last multiplication also be- 8 ing a multiplication of the signal values as measured from their alternating-current axes.

2. The method of communication which comprises producing a communication signal and a distorting signal each oi which comprises frequency components lying within a common frequency band, multiplying said signals by each other to obtain their product, transmitting said product signal to the point of reception and there multiplying said product signal by a signal that is substantially the reciprocal of said distorting signal. v

3. A system for transmitting a communication signal from a -transmission-point to a reception point which comprises means at the transmission point for producing a distorting signal, means for multiplying said communication signal by said distorting signal to obtain their product, and means located at the point of reception for multiplying said product by a signai that is substantially the reciprocal of said distorting signal, both of said multiplications being a multiplication of the signal values as measured from the alternating-current axes of the signals.

4. In a system for secret transmission 01' a communication signal, a communication unit that includes means for producing a distorting signal and means for multiplying the instantaneous amplitude 01 said communication signal as measured from its alternating-current axis by the instantaneous amplitude of said distorting signal as measured from its altemating-current axis to produce a product signal.

' 5. In a system for secret transmission of a communication signal comprising frequency components that lie within a certain frequency band,

a communication unit that includes means for producing a distorting signalcomprising frequency components that lie within a frequency band at least a portion of which is common to a portion of said certain frequency band, and means for multiplying said communication signal'by said distorting signal to produce a product signal, said multiplication being a multiplication of the signal values as measured from the alterhating-current axes o! the signals.

6. A system for secret transmission of a communication signal from a transmitter to a receiver, said transmitter including means for producing a coding signal and means for multiplying said communication signal by said coding signal to produce a product signal, means for transmitting saidproduct signal to the receiver, said receiver including means for producing a decoding signal that is substantially the reciprocal of said coding signal, and means for multiplying said product signal by said decoding signal, both of said multiplications being a multiplication of the signal values as measured from the alternatingcurrent axes of the signals.

7. The invention according to claim 6 wherein means is provided for maintaining said code signal producing means and said decoding signal producing means in synchronism.

8. A transmitter-receiver unit for the transmission and reception of a communication signal, said unit comprising means for producing a distorting signal, multiplier means for multiplying signals by each other when applied thereto, switching means for applying said communication signal and said distorting signal to said multiplier means to obtain their product, means for producing adecoding signal that issubstantiallr the reciprocal of said distorting signal, and

switching means tor supplyin a received product signal and said decoding signal to said mu]- tiplier means to obtain the original communication signal. r

9. In the method of communication, the steps which comprise producing a communication signal having an A.-C. axis, producing a distorting signal having an A.-C. axis, multiplying that part of thecommunication signal located on one side of its A.-C. axis by that part of the distorting signal located on one side of its A.-C. axis, separately multiplying said part of the communication signal by the part of the distorting signal located on the other side of its A.-C. axis, separately multiplying the other part of the communication signal located on the other side of its A.-C. axis by said first part of the distorting signal, separately multiplying said other part of the communication signal by said second part of the distorting signal, and adding said products.

10. The invention according to claim 9 wherein said multiplication is accomplished by modulating a carrier wave successively by the two signal parts and then demodulating the resulting modulated carrier wave.

11. The methodpf communication which comprises producing a communication signal and a plurality of distorting signals, combining said distorting signals, multiplying said communication signal by said combined distorting signal to obtain their product, and transmitting said product signal to the point of reception and there multiplying said product signal by a signal that is substantially the reciprocal of said combined distorting signal.

12. The method of communication which comprises producing a communication signal and a plurality of distorting signals, multiplying all of said signals by each other to obtain their product, transmitting said product to the point of reception and there multiplying said product by the reciprocal of the product of said distorting signals, each of said multiplications being a multiplication of the signal values as measured from the alternating-current axes of the signals.

13. A system for secret transmission of a communication signal from a transmitter to a receiver, said transmitter including code signal producing means, said code signal producing means comprising a plurality of code discs which are geared together for rotation at slightly difiering speeds, and means for multiplying said communication signal by said code signal to produce a product signal, means for transmitting said product signal to the receiver, said receiver including means for producing a decoding signal that is substantially the reciprocal of said coding signal, and means for multiplying said product signal by said decoding signal.

' aeorsoa 14. In a system for secret transmission of a communication signal, a communication unit that includes means for producing a code signal, a-

second means for producing a code signal, means for causing the time relation of said code signals to change with respect to each other slowly in a predetermined manner, and means for combin mg said code signals, and means for multiplying said combined code signal and said communication signal by each other to obtain their product, said multiplication being a multiplication of the signal values as measured from the alternatingcurrent axes of the signals.

15. The invention according to claim 14 wherein each of said code signal producing means comprises a code disc and wherein said code discsv are geared together for rotation at slightly different speeds.

16. In'a system for secret transmission'of a communication signal, a communication unit that includes means for producing a code signal, a second means for producing a code signal, means for causing the time relation of said code signals to change with respect to each other slowly 'in a predetermined manner, and means for multiplying said code signals and said communication-signal by each other to obtain their product.

17. The method of multiplying a. communication signal bya code signal which comprises modulating a light beam by one of said signals and modulating the resulting modulated beam by the other of said signals.

18. Apparatus for obtaining the product of two signals which comprise a lamp which may be modulated, means for modulating said lamp by one of said signals to obtain a modulated light beam, means including a galvanometer mirror for modulating said modulated light beam in accordance with the other signal, and a photo-electric positioned in said path.

position of the path of the reflected beam. and

means for vibrating said mirror in accordance with the other signal whereby said amount of light is varied, and a photoelectric cell positioned to receive the light passing through the means ALDA v. emroan.

Images