US2425066A

US2425066A - Pulsed multiplex system employing different width and repetition frequencies for each channel

Info

Publication number: US2425066A
Application number: US579723A
Authority: US
Inventors: Labin Emile; Donald D Grieg
Original assignee: Standard Telephone and Cables PLC
Current assignee: STC PLC; Federal Telephone and Radio Corp
Priority date: 1945-02-26
Filing date: 1945-02-26
Publication date: 1947-08-05
Anticipated expiration: 1964-08-05
Also published as: GB623415A; FR952462A; ES182437A1

Description

E. LABIN ET AL 2,425,06 v SYSTEM EMPLOYING DIFFERENT WIDTH AND REPETITION FREQUENCIES FOR EACH CHANNEL 5 Sheets-Sheet l PULSED MULTIPLEX Filed Feb. 26, 1945 Falsi /arf/ Aug. 5, 1947.

Aug- 5, 1947. E. LABIN ETAL PULSED MULTIPLEX SYSTEM EMPLOYING DIFFERENT WIDTH AND REPETITION FREQUENCIES FOR EACH CHANNEL Filed Feb. 25, 1945 5 Sheets-Sheet 5 1V/f /5 @d IN VEN TRS fM/ ZE H5/N DONA/0 D. GIP/EG A T TONEY Patented Aug. 5, 1947 UNITED STATES PATENT GFFlC-E 2,425,066 PULSED wlmlrfri.FLEXV SYSTEM Aiilvni"'Lirif1`1iii DIFFERENT WIDTH AND REBETIITION FREQUENCIES FOR EACH CHANNEL Emile Labin, New York, and Donald D. Grieg,

Forest Hills, N. Y., assgnors to Federal Tele"- phone and Radio Corporation, New York, N. Y., a corporation of Delaware Application February 26, i945, serial ne. Statale 18 Claims. 1

This invention relates to 'transmission 'of' intelligence by radiant energy and to systems for broadcasting information and entertainment programs, such as sound, facsimile, picture, etc., in a multiplex method over the same or nearly the same carrier frequency.

It has been proposed heretofore to transmit simultaneously a number of channels of intelli'- gence over a common carrier frequency. In such systems, it is the usual practice to interleave in time thesignal pulses of the different channels. This is usually accomplished by synchronizing the signals of the different channels for diiferent time occurrences so that they interleave in serial order to form a single train of discrete pulses. I

In many situations it is difficult to provide proper synchronism for the signals of different channels thereby necessitating diierent carrier frequencies in order to avoid interference loetween the channels. One situation where proper synchronism is difficult Without special 'equipment is the insertion of channel pulses from m'obile craft such as airplanes and railroad trains into a train of channel pulses at relay points along the path of a series of relay stations.

It is an object of the present invention to p'rovide a method and a system for transmittingor broadcasting a plurality of channels of intelligence over a common transmission medium Without the need of synchronism between the signals of the different channels.

Another object is to provide a ymethod and means for inserting in a given carrier medium a channel of signal pulses unsynchronized with the channel pulses already present in said medium, in a manner so that the signal pulses of the different channels may be separated at a receiving point With a minimum of interference from the inserted signal pulses. Y

Another object of the inventionis to provide a system for transmitting or broadcasting a plurality of channels of intelligence ona common carrier Without synchronism between the different channels and means for unscrambling the pulses of the diierent channels at a receiving point with a minimum of interference from the pulses of channels overlapping the pulses of a desired channel. l

According to one of the features of our invention, the different channels of signal pulses may be produced at Widely diierent points Without any synchronizing connection and broadcast at a common carrier frequency at these points. The signal pulses of `the different -channels`,nof course, may be applied toa common transmission 2 medium sich 'as a wire line 'or coaxial cable at different peints therealong, either direct or in the forni of a modulated carrier of a common frequency. n,

The signal pulses of different channels are distinguished by differences in pulse Width and by greatly minimised by use of time modulation of thel pulses as distinguished from amplitude mod'- ul'ation. The time modulation may take any one of the plurality of different forms. For example, the pulses `may be displaced in time relative to a given time position according to substantiall'y the instantaneous values of a signal wave or the successive pulses may be displaced toward and yati/'ay from 'each other in push-pull manner according to the modulating signal. At the receiver the pulsesqof a zgiven channel are unscrambled from the pulses of other channels, preferably vaccording to their 'distinguishing Width characteristic although other methods of separation may be employed. For example, the pulses 'may not only be separated according to Widtlfirfof the pulses but also by their average repetition rate `in onjunction With the Width distinguishing characteristic.

The above and other objects and features of the invention will become better understood upon consideration of vthe following detailed description to be read in connection with the accompanyingj ydrawings' in which:A

Fig. lis a diagrammatic illustration of a plu- 'rality of broadcasting stations and one receiver for selective reception according to the principles of our invention:

`2 is a graphical illustration useful in explaining the selective' receiving function of the receiver of Figs. 1 and 5;

3 is a lschematic Wiring and block diagram of the transmitting circuit of one of the broadcasting stations;

Fig. 4 is a graphical illustration useful in eX- plainlng the operation 4of the pulse time modulator circuit shown Fig. 3;

,Fig` 5 is a. block diagram of a receiver capable of selecting `any one of a plurality of signal channels broadcast simultaneously over a common carrier vfrequency 'according to the broadcasting arrangement of Fig'. 1;

Fig. 6, is1 a block diag-rain of a broadcasting systeinnfor' a pluralityv of signal czharinels;` and Fig'. *l is adiagrammatic illustration of a relay fr communicating system of the character to which our invention is applicable.

Referring to Fig. l, a plurality of transmitter stations I, 2 and 3 are shown for broadcasting on a common carrie-r frequency F1l and an additional transmitter station 4 for broadcasting intelligence on a carrier frequency F2. A receiver 5 is also shown provided with two controls 6 and i for carrier frequency tuning and pulse width discrimination. The signal pulses transmitted by stations I, 2 and 3 are distinguished in pulse Width and the average repetition Vrate of the pulses. For example, the signal pulses of station l are of a given width Wi and the average repetition rate is selected as R1. The signal pulses of

stations

2 and 3 are of diierent widths W2 and W3 and diiferent average repetition rates R2 and R3 respectively. Since transmitter station il operates on a carrier frequency diierent from the carrier frequency of stations I, 2 and 3, the signal pulses thereof need not distinguish from the pulses of stations I, 2 and 3 but may have the saine width and repetition characteristics as one of them.

In Fig. 2, graphs a, b and c show three trains of channel pulses A, B and C having different widths and repetition rates such as may be assigned to the transmitter stations I, 2 and 3 respectively. Since these signals may be transmitted simultaneously on a common carrier frequency F1, they will form a composite wave in the ether such as represented by graph d. The pulses of these channels may be chosen of Various widths ranging from a fraction of one microsecond to two or more microseconds as may be desired. For satisfactory sound transmission a channel may have a pulse repetition rate anywhere from about 10,000 pulses per second to as high as 20,000 or more, but for other services the rate may be either higher or considerably lower than the 10,() rate mentioned above depending upon the type of intelligence transmitted.

The receiver of Fig. 1 which is shown in more detail in Fig. maybetuned to receive any desired carrier frequency such as Fi and may be further tuned for signal pulse selection according to pulse width. In this way, the programs of the transmitter I, 2, 3 and 4 may be selectively received at 5 and, as described in greater detail hereinafter, with substantially no interference from the pulses of the channels not selected.

Before describing the receiver function of the invention the details of a transmitter will be described. In Fig. 3, a transmitter is shown provided with one form of P. T. M. (pulse time modulator) 8, such as may be employed for production of signal pulses at a desired repetition rate and modulated in time according to the intelligence to be transmitted. An oscillator 9 provides the basic wave for the modulator, the frequency of oscillation controlling the repetition rate of the pulses produced as will. be clear from the following description. The base wave of oscillator 9 is represented at I0 o-n graph :I: of Fig. 4. The wave I6 is applied to the primary II of an input transformer I2 in parallel with the signal voltage from audio source I3 which is applied to primary coil I4. The modulator circuit includes two secondary coils I-5 and I6 coupled to the control grids of two vacuum tubes II and I8 in pushpull arrangement similar to a full wave rectier. The modulator amplifies and, in eiect, fully rectifies the wave vIIB to obtain a cusper Wave 20, graph y of Fig. 4. While the rectication of wave Iii may be symmetrical with respect-to the zero e axis I9, it is shown for purposes of illustration as unsymmetrical by means of different biasing potentials 2l and 22, Fig. 3. This gives the effect of an offset axis 23, graph about which modulation takes place. The signa-l intelligence operates, in effect, to vary the wave I0 relative to its offset axis 23 as respects the full wave rectication. The maximum limits of relative variations between the wave and the axis 23 are indicated by the upper and lower modulation limits 24 and i9, the lower limit being in this instance selected the same as the original zero axis of wave I0.

When the input signal varies the relative relation between the oiset axis 23 and the wave l to the upper limit 24, the cusper wave 20 is displaced as shown by the broken line 25, and when varied to the opposite limit I3, it is displaced as shown by broken line 26. It will be observed that the signal wave thus varies the time positions of the cusps 2l', 2B, 29 and 3! in push-pull manner toward and away from each other thereby decreasing and increasing the time intervals therebetween.

For transmission purposes, the cusps are clipped from the wave 2E) and shaped to form substantially rectangularly shaped pulses in pulse widthshaper 3i. The Shaper 3I may be of any known type capable of clipping the wave 2i] between limits 32 and 33.r Since circuits of this character are well known a detailed showing is unnecessary. The output pulses of the shaper 3| may be repeated as shown in graph z, Fig. 4.

The pulses thus produced are used as the modulating voltage for an RF translator 34 for transmission at the chosen carrier frequency such as F1, for example. Carrier frequency pulses are radiated by antenna 35.

For a further discussion of P. T. M. modulators of the :character shown at 8 and also of pulse shapes, reference may be had to our cov pending application Serial No. 455,897, nled August 24, 1942, and assigned to Federal Telephone and Radio Corporation. It will be understood, of course, that other forms of P. T. M. modulators may be employed.

In Fig. 5 the receiver 5 0f Fig. 1 is shown in greater detail. The receiver includes an RF amplifier and detector 36 of broad band characteristics for receiving the train of pulses broadcast on a selected carrier frequency such as Fi.

Broad band receiving circuits of this character are well known and the details thereof need not be described. The detector circuit however should be sufliciently linear to preserve the pulse shapes upon detection even where there is considerable overlapping of the pulses as indicated lby the composite wave of graph d, Fig. 2.

The composite wave output of the detector is applied to a diiferentiator 3l of the character adapted to produce derivative pulses in response to each abrupt change in the leading and trailing edge portion of the wave. That is to say, for each change in the leading edge from an approximately steady voltage condition to a sharp voltage increase, a needle-like positive pulse is produced, and for each abrupt change in voltage from a substantially steady condition to a sharp decrease, a negative pulse of needle-like shape is produced. Graph e, for example, represents the derivative pulses derived from differentiation of the composite wave form of graph d. Differentiation of pulse 3S of channel A (see graphs a and d), for example, produces

derivative pulses

38a and 38a the two derivative pulses corresponding in time to the leading and trailing edges of the pulse 38. Differentiation of pulse 3S of channel C likewise produces derivative pulses 39e and 39e. Differentiation of pulse which results from an overlapping of pulses from channels A and B produces positive derivative pulses im and tot corresponding to the abrupt changes in the leading edge of the pulse di), and negative derivative pulses lita and ilb' corresponding to the abrupt changes in the trailing edge of the pulse ll. The composite pulses di, l2 and i3 when dierentiated produce positive derivative pulses die, da, 5l-2d, 42h, 43o-c and 33h and negative derivative pulses Mc', da', 52a', 212', A311', 153D', and 153e as shovm in graph e. It is clear that all of the derivative pulses of graph e correspond to leading and trailing edges of the channel pulses A1, A2, A3, A4, A5, B1, B2, B3, C1, C2 and C3 as shown by their time relation in graph a, b and c.

Energy of the derivative pulses represented by graph e is applied over connection li, Fig. 5, to clipper 5 and over connection d.: to an inverter l which inverts the polarity of the derivative pulses and applies the inverted pulse energy to a delay device 4S, the output of which is applied to the clipper d5. Referring to graph f of Fig. 2, the inversion of graph e is shown in its retarded condition as it appears in the output of delay device 48, the device i8 being adjusted, for example, to impose a delay equal to the width W1 of the pulses of channel A. This adjustment is made with the View to selecting the pulses of channel A to the exclusion of the pulses of the other channels.

When the derivative pulse energy represented by graph e is applied over circuit de to clipper #35 and the energy represented by graph f is applied by the delay device 58 to the clipper 55, the pulse energy appears therein substantially as indicated by graph g. It is seen from graph g that certain of the derivative pulses of graphs e and f coincide and produce coincidence pulses such as indicated at i9 for the derivative pulses 38a' and l-Sa. It will be clear from a comparison of graphs e, f and g that coincidence pulses 5B, 5l, 52 and 53 are derived from pulses lilla and !-la, dia' and I-tlia, lZa and l-a, la' and i-dBa-c respectively. By properly biasing the clipper l5 to clip the coincidence pulses at clipping level 5d, a series of pulses ai, a2, as, etc., is obtained corresponding exactly with the relative time positions of the pulses A1, A2, A3, etc., oi channel A.

Since the pulses a1, a2, as are of the same time position as the corresponding pulses of channel A, they carry the time modulation imposed upon the pulses of channel A. The output of clipper d5 may therefore be applied to any suitable T. M. demodulator 55 capable of translating the time displacement of the pulses into amplitude displacements which when passed through a low pass filter 56 may be applied to speaker 5l or other utilization apparatus.

Assuming that the delay device i8 is adjusted to retard the inverted pulses an amount equal to W2, the width of the pulses of channel B, the inverted pulses will have the timing indicated in graph i of Fig. 2. When these pulses are applied to clipper d5 with the pulse energy received over connection 44 such as indicated by graph e, the resultant may be represented by graph 7' wherein certain of the pulses coincide in the manner hereinbefore described. It will be noted however that the negative coincidence pulses 5e, 5i! and 50 are derived from pulses Mib and Z-dilh lb and 2-42b, A322 and 2l3b respectively of graphs e and z'. It will also be noted that pulses 39C of graph e and pulse 2-38a of graph i coincide to produce a coincidence pulse 6i of graph j. This pulse however is of positive polarity as contrasted to the coincidence pulses resulting from the pulses of channel B and therefore is eliminated when the pulses of graph 7' are clipped at level 62, whereby pulses b1, b2, b3, etc., of graph k are obtained, which correspond exactly in the time relation of the channel pulses B1, Bz, B3, etc.

Assume that the device t3 of Fig. 5 is adjusted for retarding the inverted pulses an amount W3 corresponding to the Width of the pulses of channel C. Graph m represents the output of the delay device i8 and graph n represents the pulse energy resulting from the pulse output of device 48 and the pulses received over connection 44. It is believed clear from a comparison of graphs e, m and 1L that by the same operation hereinbefore described a train of pulses c1, c2, ca, etc., will be obtained having the exact timing relationship of the corresponding pulses of channel C, graph c.

From the foregoing, it is clear that the receiver of Fig. 5 is capable of segregating the pulses of a given channel regardless of distortion due to the overlapping thereof by pulses of other channels. It is recognized that there may be occasional instances where a false pulse indication of the desired width may be produced by an overlapping relationship between pulses whereby a composite pulse is created having leading and trail'- ing edge portions corresponding to edges of the two pulses, spaced apart a time interval equal to the width of the desired pulse. The occurrence of such false indications, however, is very infrequent and moreover will as often as not occur at the proper time position for the modulating signal, and it can be stated that the receiver of Fig. 5 will unscramble the pulses and select substantially only the pulses of a desired Width regardless of distortion due to overlapping by ypulses of other channels.

For a further discussion of pulse width selector circuits of the character shown in Fig. 5, reference may be had to the copending application of E. Labin. Serial No. 467,509, iiledvDecember 1, 1942, which is assigned to the Federal Telephone and Radio Corp. Other pulse width discriminating circuits may, of course, be used for unscrarnbling the channel pulses. Another such circuit is disclosed in a copending application of D. D. Grieg, Serial No. 511,404, filed November 23, 1943, and assigned to Federal Telephone and Radio Corp.

While Fig. l shows the diiferent channels transmitted from different stations, it will be clear that the transmission may occur from a single station such as illustrated in Fig. 6. In this embodiment the diierent channels may originate at studios located various distances from the broadcasting station. According to the principles of this invention such studios need not be synchronized, one with respect to the other, but may be permitted to have independent signal pulse repetition rates R1, R2 and R3 such as indicated for the pulse modulators E3, 64 and 55. rihese diierent rates, of course, will be controlled by the frequency of the oscillator associated with the modulator such as indicated at $5 for modulator E53. The pulse widths for each channel are determined by the pulse Shaper -associated with the output of the modulator as indicated at 6l, 68 and 69 for the three channels. I The pulses lof `7 each channel may thus be produced at the studio and transmitted by Wire or wireless to the main broadcasting station where they are mixed in a mixer circuit l and applied to an RF translator H for modulating a carrier of a given frequency.

Fig. 7 illustrates a further application of the principles of our invention. The system shown comprises a series of relay stations l2, 13, M and 15, the stations 12 and l5 being shown as terminal or sub-terminal stations. This system may comprise relays of the character adapted to transmit a multiple of channels on a common carrier freqency which is beamed from one repeater station to the next. The signal pulses of the diierent channels may be timed to interleave, thereby forming a single train of pulses or they may be unsynchronized according to the transmitter arrangement of Fig. 6.

As hereinbefore stated, it is difficult to insert a channel of pulses, either for communication, reporting position or other purposes, from a mobile device, such as a railroad train or such as airplane 16, into the train of pulses relayed over a series of relay stations Without interference with the channels being relayed. According to our invention, it is possible to transmit signal pulses from craft 16 for reception by antenna il of the relay stations whereby they are relayed along with the pulses of other channels. It will be understood of course that the system herein described operates at U. I-I. F. and at relatively loW power so that the transmission does not carry much further than about 30 miles or so, and that, in line-of-sight transmission. It will thus be clear that while the signal pulses from a plane 16 overlap occasionally with pulses of other channels, the rate of the signal pulses Will be difierent and also the Width thereof will be diiierent from the pulse rates and Widths of other channel pulses. The pulses of the different channels may be unscrambled and the desired channel selected at receiving points as described in connection with the receiver of Fig. 5.

While We have shown and described the principles of our invention in connection with specic apparatus, it should be understood that the specific apparatus examples are given for illustration purposes only and not as a limitation of the scope of the invention as set forth in the objects and the appended claims.

We claim:

1. In a method of multi-channel transmisn sion, the steps of producing for each channel a train of electrical pulses of a Width and at a rate of recurrence diierent from the pulse widths and recurrence rates of the pulses produced for the other channels, and transmitting the trains of pulses thus produced over a common transmission medium.

2. In a method of multi-channel transmission, the steps of producing for each channel a train of electrical pulses of a Width and at a rate of recurrence diierent from the pulse Widths and recurrence rates of the pulses produced for the other channels and amplitude modulating a carrier wave of a given frequency according to the trains of pulses thus produced.

3. In a method of multi-channel transmission, the steps of producing for each channel a train of electrical pulses of avWidth and at a rate of recurrence diierent from the pulse widths and recurrence rates of the pulses produced for the other channels, transmitting the trains of pulses thus produced over a common transmission medium, and selectively receiving from said common transmission medium the electrical pulses of a desired one of the channels according to the pulse width and average repetition rate characteristics of such channel.

4. In a method of multi-channel transmission, the steps of producing for each channel a train of electrical pulses of a Width and at a rate of recurrence different from the pulse Widths and recurrence rates of the pulses produced for the other channels, transmitting the trains of pulses thus produced over a common transmission medium, and at a receiving point, receiving the resulting pulse wave from said common transmission medium, differentiating the pulse wave to obtain a series of derivative pulses corresponding to abrupt changes in amplitude of said Wave, the derivative pulses corresponding to changes increasing in amplitude being of a rst polarity and those corresponding to changes decreasing in amplitude being of a second polarity, producing selecting pulses disposed at a time interval equal to the Width of the pulses of a desired channel from the derivative pulses of one of said polarities. and producing a pulse output each time a selecting pulse coincides with a derivative pulse ofthe other of said polarities.

5. A method of simultaneously transmitting the intelligence of a plurality of sources, comprising producing for each source of intelligence a train of electrical pulses of a Width and at a rate of recurrence different from the pulse Widths and recurrence rates of the pulses produced for the other of said sources, modulating in time a characteristic of the pulses oi each train according t-o substantially the instantaneous values of the corresponding source oI" intelligence, and transmitting the trains of pulses thus produced over a common transmission medium.

6. A method of simultaneously transmitting the intelligence of a plurality of sources at a given carrier frequency, comprising producing for each source of intelligence a train of pulses of a Width and at a rate of recurrence different from the pulse widths and recurrence rates of the pulses produced for the other of said sources, modulating in time a characteristic of the pulses of each train according to substantially the instantaneous values of the corresponding source of intelligence, and amplitude modulating a carrier Wave of said given frequency according to the trains of pulses thus produced.

'7. A method of simultaneously transmitting from a plurality of transmitting stations at a given carrier frequency for selective independent reception comprising producing at each station a. train of pulses of a Width and at a rate of recurrence different from the pulse Widths and recurrence rates of the pulses produced at the other of said stations, modulating in time a characteristic of the pulses of each train according to substantially the instantaneous value of an intelligence, and at each station, amplitude modulating a carrier wave of said given frequency according to the train of pulses produced for such station.

8. In a method oi simultaneous transmission of the intelligence of a plurality of sources, the steps of producing for each source of intelligence a train of electrical pulses of a Width and at a rate of recurrence different from the pulse Widths and recurrence rates of the pulses produced for the other of said sources, modulating in time a characteristic of the pulses of each train according to substantially the instantaneous values of the corresponding source of intelligence, selectively receiving from the common transmission medium the signal pulses of a desired one of said sources to the exclusion of the other signal pulses, and demodulating the time modulation of the received pulses.

9. In a system for transmitting a plurality of channels of signal pulses on a common carrier over a chain of relay stations; the method oi adding an additional channel of intelligence to said common carrier at any desired relay point along the chain of relay stations, comprising producing a train of signal pulses of a width and at a rate of recurrence different from the pulse Widths and recurrence rates of the 'signal pulses of the other channels, modulating a carrier of a frequency substantially the same as the frequency of said common carrier, and transmitting said carrier for reception by at least one of said relay stations, whereby said added channel of signal pulses is relayed by said stations on said common carrier.

lo. In a system of signal pulse, multi-channel communication on a common carrier over a chain of relay stations; the method of adding an additional channel of intelligence t said common carrier at any desired point along the chain o1" relay stations, comprising producing a train of signal pulses of a Width and at a rate of recurrence diierent from the pulse Width and recurrence rates of the signal pulses of the other channels, modulating a carrier of a frequency substantially the same as the frequency of said common carrier, transmitting said carrier for reception by at least one of said relay stations, whereby said added channel of signal pulses is relayed by said stations on said common carrier; and at a receiving point associated with said chain of relay stations, the steps of producing a Wave representing the envelope of said carrier,l

and selectively obtaining from said wave pulse energy corresponding to the signal pulses of a desired channel to the exclusion of the other signal pulses.

1l. In a system of multi-channel transmission, means for producing for each channel a train of signal pulses of Width and average repetition rate differing from the Widths and repetition rates of the signal pulses produced for the other channels, and means for transmitting the signal f' pulses of the channels over a common transmission medium.

l2. In a. system of multi-channel transmission, means for producing for each channel a train of signal pulses of Width and average repetition rate diering from the Widths and repetition rates of the signal pulses produced for the other channels, and means for amplitude modulating carrier Wave of a given frequency according to the trains of pulses of said channels.

13. In a system of multi-channel transmission, means for producing for each channel a train of signal pulses of Width and average repetition rate differing from the Widths and repetition rates of the signal pulses produced for the other channels, means for transmitting the signal pulses of the channels over a common transmission medium, and means for selectively receiving from said common transmission medium the signal pulses of a desired one of said channels telligence of each source including means for producing signal pulses of Width and average repetition rate different from the Widths and repetition rates of the signal pulses produced for the intelligence oi the other sources, and separate means for transmitting the signal pulses of said sources over a common transmission medium.

l5. In a communication system having a plurallty of sources of intelligence, pulse modulator means for the intelligence of each source including means for producing signal pulses of Width and average repetition rate diierent from the widths and repetition rates of the signal pulses produced for the other sources, means for transmitting the signal pulses of said sources over a common transmission medium, and means at a receiving point for receiving from said common transmission medium the signal pulses of a desired one of said sources according to the pulse width and average repetition rate characteristics of such channel.

16. ln a communication system having a plurality of sources of intelligence, pulse modulator means for the intelligence of each source including means for producing signal pulses of Width and average repetition rate different from the widths and repetition rates of the signal pulses produced for the other sources, and means for transmitting the signal pulses of said sources over a common transmission medium; and means at a receiving point for receiving the pulse Wave resulting from the signal pulses of said plurality of sources, means for differentiating the pulse wave to obtain a series of derivative pulses corresponding to abrupt changes in amplitude of said wave, the derivative pulses corresponding to changes increasing in amplitude being of a rst polarity and those corresponding to changes decreasing in amplitude being of a second polarity, means for producing selecting pulses disposed at a time interval equal to the Width of the pulses 0f a desired channel from the derivative pulses of one of said polarities, and means responsive to coincidence between a selecting pulse and a derivative pulse of the other of said polarities to produce an output pulse.

i7. In a systemior transmitting a plurality of channels of signal pulses on a common carrier over a chain of relay stations; means for adding an additional channel of intelligence to said comcarrier at any desired relay point along the chain of relay stations, comprising means for oducing a train of signal pulses of a Width and a rate of recurrence different from'the pulse Widths and recurrence rates of the signal pulses of the other channels, means for modulating a carrier oi a frequency substantially the same as the ircduency of said common carrier, and means lor transmitting said carrier for reception by at least one of said relay stations, whereby said added channel of signal pulses is relayed by said stations cn said common carrier.

In a system for transmitting a plurality of channels of signal pulses on a common carrier over a chain of relay stations; means for adding an additional channel ci intelligence to said common carrier at any desired relay point along the chain of relay stations comprising means for producing a train of signal pulses of a width and at a rate of recurrence difierentgfrom the pulse Widths and recurrence rates of the signal pulses ol the other channels, means lor modulating a carrier of a frequency substantially the same as t frequency of said common carrier, and means 12 sponding to the pulse envelope shapes of said carrier, and means for selectively obtaining from the pulse Wave pulse energy corresponding to the signa1 pulses of a desired channel to the exclu- 5 sion of the other signal pulses.

EMILE LABIN. DONALD D. GRIELG.