US2408063A - Multiplex receiver - Google Patents

Description

Sept. 24, 1946.

*.D. D. GRIEG MULTIPLEX RECEIVER Filed July 29, 1944 2 Sheets-Sheet lA 7D CHA/wa SI"4 fre.

' ArTaR/vfx sep't. 24,1946.l

D. D. GRIEG MULTI'PLEX RECEIVER Filed July 29',4 1944 2 sheets-Sheet 2 Patented Sept. 24, 1946 MU LTIPLEX RECEIVER -Donald D. Grieg, Forest Hills, N. Y., assignor to Federal Telephone and Radio Corporation, "N ew York, N. Y., a corporation of Delaware l Application July 29, 1944, SerialNo. 547,123

This invention relates to4 multi-channel communication systems and more particularly to the selective reception of a plurality of ltransmitted channels.

One of the'objects of this invention is to provide anv improved `method and means for the selectivereception in a multi-channel communication or broadcasting system of a plurality of intelligence conveying channels, the different channels being distinguishd by different identifying characteristics.

A further object Aof the invention is to provide a method and vmeans at the receiving pointsof acommunication' system for selectively receiving any one'or more-of a pluralityof signal channels transmitted in time-,spaced relation over a common transmission medium, such as a transmission line or a given carrier radio frequency Wave.

Another object of the invention is to provide methods and means for selective multi-channel reception, wherein signal intelligence is conveyed over each channel by means of a series of pulses which are modulated with respect to any one of a number of their characteristics such as time position, for example, in accordance with the instantaneousV value of the intelligence, and wherein the pulses of the individual channels have different identifying widths.

An additional object of the invention includes the provision of a method and means for `the selective reception of. the respective channels in multi-channel communication systems in accordance with the characteristic pulse Widths-thereof.

According to certainfeatures' of the invention, a number of channels,` conveying signal intelligence by means of a series of pulses, eachy channel. characterizedby anidentifying pulse width and the pulses of which are modulated with respectto one of their characteristics in accordance with. the signal intelligence to be conveyed are received in time-spaced relation as a single train ofpulses at the receiving terminal and applied to a pulse-width-to-amplitude-translator the incoming pulses, having different widthsifor the various channels, are translatedinto undulations ofzsimilar width, the/amplitudes of which are made to vary as the-widths of the correspondingpulses. Froml the translator, the undulations now characteri'Zed-foreach 'channel byV an identifying amplitude may then be fed as a single'train of undulations-'to a' peak riding clipperfandV to one'or more of aplurality of mixer and peak riding clipper circuits, representing, the various channels of Vcommunication respectively. In the, i'lrsty peak-ridingV clipper circuit,

11 Claims. (Cl. 179-15) the undulationshaving relative maximum amplitude are segregated by clipping a given portion of their amplitude which is ampliiied and demodulated for recovery of the original signal conveyed over this channel.

The clipped portions, thus obtained from the first peak riding clipper, may also be used to substantially eliminate from: the train of undulations, before the train of undulations is applied l to subsequent peak riding clipper circuits, which. act to successively segregate by clipping of their peak portions, the undulations havingA relative maximum amplitudes in sequential order according to their values. In order to achieve this effect, the clipped portions of a given channel or channels having relative -maximum amplitude undulations, may be iirst inverted and'thenmixed with the train of undulations from the original 'translator or apreceding channel, whereby the firsty orv preceding maximum amplitude undulations-are substantially cancelled out. This brings in line for segregation by the peak riding clippers in the subsequent channel circuits the undulations of the channel having the second or succeeding maximum amplitude undulations.

This principle may be employed to successively eliminate lone channel after another before applying the train of undulations to the peak riding clipper of the next channel. l

The. above and other objects and features of the invention will become more apparent upon consideration of the following, detailed description to be read in connection with the accompanying drawings, in which:

Fig. 1A is a block diagram of a receivingv terminal of a multi-channel communication system according to my invention;

Fig. 1B is a block diagram of avmodifled form l of thereceivingterminal of Fig. 1A.l

Here,

Fig. 2 is a schematic circuit diagram ofA a width-to-amplitude translator and of a peakriding clipper circuit of the receiving terminal circuits of Fig. l; and

Fig. 3 is a graphical illustration of the operaktions of thecircuits of Fig. 2.

3 and a peak-riding clipper 4, and the third channel includes a similar mixer and peak riding clipper 6. The output of the peak riding clipper 2 is applied to the mixer 3, mixer 5 receiving the output of mixer 3 and that of the peak riding clipper 4. The outputs of the peak riding clippers 2, 4 and 8 may be applied to democlulators, as desired. A further channel `Il, which is not shown, may be similarly supplied from mixer 5 and clipper Ii, as indicated.

In Fig. 1B, the receiving terminal is made u of the same major circuit elements as the form of Fig. 1A, except that in this case, the adjustable translator I feeds all of thevchannel circuits in parallel, and the output of ther peak riding clippers 2, i and 6 are fed into a common connection P, from which the mixers are individually supplied, the various channel circuits beingA separated by de-couplers D. Further channel circuits, as required, may be similarly connected to the translator I and the peak riding clipper connection P. y

Fig. 2 illustrates circuit I and 2, as shown in block form in Figs. 1A and 1B.

The circuit I is utilized for translating a train of pulses of various widths into undulations having corresponding amplitude Variations, as described, for instance, in the copending application of E. Labin and myself, Serial No. 487,072, iiled May l5, 1943, wherein the circuit isvdisclosed as a pulse selector according to pulse width.

The circuit includes a limit clipping stagev 'i as an input coupler, which limits all input pulses to substantially the same amplitude and, also, inverts the input pulses from a positive polarity, as indicated by the pulses of curve a, Fig. 3, to negative polarity indicated by the pulses of curve b. The output pulse energy from stage I is applied through a resistor R to a tuned shock-excitable L.-C. circuit 8. Connected across the tunable circuit 3 is a vacuum tube 9, the cathode II] of which is connected to the input side of the circuit 8, while its anode II is connected to the opposite side I2 of the tunable circuit. The side I2 is also connected to a source of anode potential at I3. The pulse energy, as in curve b, Fig. 3, from an anode connection I4 of the tube 1 is applied to the grid of the tube 9, so as to block conduction between the cathode Ill and the anode II, while pulse energy is applied to the circuit 8. The undulations, which are produced in circuit 8 in response to pulseenergy coming in over connection I4, are taken 01T through a connection I5 for application to the peak riding clipper stage 2. The undulations in connection I5' arefed over a condenser I6 to a grid I1 of tube I8." The grid circuit includes aresistor I9 and -abiasing resistor 20, the latter being shunted by the usual by-pass condenser 2|'. An anode 22, througha load resistor 23, is connected to the positive end 24 of an anode potential source B+ at I3. Potentials developed across resistor '23 are delivered as output to the next stage at connection 21. Branching oi from connection I5 is a connection 2B which serves to feed undulati'ons from the width-to-amplitude translator I to the circuit of channelVv 2 as in Fig. 1A, or to the circuit of channels 2, 3 and others as in Fig. 1B.

The operation of the circuits of Fig.v2 will now be described in connection with the curves of Fig. 3.

Assume for the purpose of illustration, the width of the pulses for the'three channels in curves a and b to be .of the character indicated by the reference characters W1, W2 and W3. As-

4 sume, also, that circuit 8 is tuned for a maximum response, that is for selection of pulse width W1. In that case, curve c represents the output of circuit 8 when the circuitis tuned for a maximum response to pulses having the width W1, illustrating a series of output undulations as obtained fromthe different pulse Widths of curve a.v In order to clarify this, it may be mentioned at this point, that when the leading edge of the pulse W1 is applied at negative polarity, as in curve b, to circuit 8, an initial undulation 29 is j produced, Vwhich is normally followed by undulations 38 and 3|, and so on in the form of a damped wave.r 4 When the. circuit 8 is tuned to a frequency, the period of which is exactly twice the width W1, the trailing edge of the W1 pulse occurs where the initiated, oscillatory energy crosses the zero axis fromv undulation 29 to the undulation 30. Since the trailing edge of this pulse shock-excites the circuit 8 in the same direction at this point, the undulation 32 produced thereby in circuit 8 adds algebraically to the undulation 30 to produce a resultantundulation 33. The next succeeding pairs of undulations produced by the leading and trailing edges of pulse W1 would normally tend to produce a negative undulation 34, which would tend to continue as a damped wave. Damping tube 9, however, eliminates any such trailing oscillations so that they do not interfere with the undulations produced by subsequent pulses applied to circuit 8.

A pulse Width less than pulse Vwidth W1, such, for example, as the width W3 will not produce undulations as great ras undulation 33, if the circuit tuning is adjusted to correspond to pulse Width W1. This is illustrated by undulation 35, produced in response to the pulse width W3.l The reason for this is readily apparent, because shock excitations, produced by the leading and trailing edges of the pulses of lesser width than W1, will in part be opposed to each other. Undulation 36, in response to the greater pulse width W2, is likewise smaller than undulation 33, since here again, the oscillations, produced in response to the leading and trailing edges of theA greater pulse width, are in part opposed-to each other, so that the algebraic summation thereof is less than in the case of undulations produced in response to pulse width'Wi. y

Curve c of Fig. 3 therefore represents a series of output undulations corresponding to the train of pulses, as in curve a, which are present in the output connection I5, and which may be fed to the peak riding clipper stage 2 and to mixer 3 as inv Fig,` 1A,V or in parallel to all channels as in Fig. 1B.

rAs the `operating characteristics'of a peak follower or peak-riding clipper are known in the art, only a brief description ofsuch operation'will now be given. If, for example, a positive potential, as represented'by the undulation 33 is applied to the side'31 of the condenser I6 (Fig.'2'), grid current will flow through resistanceV I9 charging' the condenser I6 and resultingV in a negative potential on side 38, thereof. This negative potential is effective on the grid I1, the value of such potentialbeing proportionate to the voltage amplitude of the undulation applied to condenser IS. `The paramentersof the circuit, represented lby condenser I6 and its associated'resister I9, are such that the resultant time constant of the circuit will permit the condenser to lose its charge only very slowly and Aonly to a certain point. 'I 'he effect of the charge on condenser I6 is to maintainY a cut-'off bias on the tube I8 at willbe kept replenished byeach succeeding unf dulation. The loss'of the charge and'- thereby the' cut-off level is, of course,- also a'function ofthe repetition rateof the-undulationsresponsible lfor the establishment'of the level; v'It becomesk clear,- therefore, that only portion` of the-undulation 33 appearing above theaforesaid'level39 will be passed by the-tube I8, and will appear `in its output-atei. The function of such a peak follower or peak riding 4clipper as at 2, is to tend to-'clip oilv aiixed amountfrom the crest of thoseapplied undulations having maximum amplitude,- regardless of the actual amplitude of such undulation. Iffundulations of lesser amplitude than thatfrepresented by the level 39 are applied, suchundulations would not Ibe represented inthe outputl becausefofthe peak clipping'level line 39 which is established by undulations -having a maximum relative amplitudek extending above the level V39, This circuit tends, therefore, to select maximum amplitude undulations in each case. 'The level represented byline 39, as will be now understood, varies withthe maximum amplitude of the undulations receiv,ed,'so that, if undulation 33 were eliminated :from the series ofv undulations shown in curve'c,the undulations having the next maximum relativeamplitude, would be thosereferred l to'by'reference 36 curve f, and the clipping level would automatically be established at 4i, accordingly. In each case, therefore, the peakV clipping level will beyestablished automaticallyA to 'select thecrest of the undulations having a maximum amplitude for a' given series ofV undulations.

This characteristic I make use of by using the clipped peaks of themaximum amplitude vundulations to successively eliminate such maximum undulations whereby the peak riding clipper'circuit in the subsequent channels are enabled to select undulations having the-next highest amplitude, in succession.

The clipped peaks, for instance, of undul-ation 33, ind-icated by curve d, may be amplified and inverted in the clipper circuit` 2, to assume the form as shown in curve e which are then mixed (at mixer3 with the series ofundulations represented in curve c. The resulting series'of undulations, as obtained from mixer 3, is shown in curve f where itis seen that the undulations 33 have now practically disappeared, and undula- 'This train of undulations (curve f) is applied to .the peak riding clipper circuit ll, Fig. 1A, where, automatically, a clipping level 4l is established curve f, and the peaks 42 of the undulations 36 are clipped off as shown in curve g. This output is analogous to that obtained from peak riding clipper 2 for channel I, and may be similarly used to obtain a reproduction of the original signal for channel 2 by demodulation. Portions 42 may also be used to eliminate undulations 36 by inversion and mixing in mixer 5 with the output of mixer 3. This enables peak riding clipper 6 in the circuit of channel 3 to establish 6 automatically a peak .clipping levelcorre'sponding to undulations 35,.which havenow comeltobethe maximum amplitude undulationsjinfthe, series applied to-clipper'. y v i 4Inthe form offthe circuit Iarra'ngementshown in- Fig.'A flB-,the successive elimination` of the maximum amplitude undulations Yoff preceding chan- 33 and 36 are reduced to negligible size :and the.r

clipper 6 receives the remaining undulation-35Lto acton. The decouplers D between channels-serve to Vprevent the feedback of the clipperoutputto- Ward-thepreceding channels. l A A Further similar clipping and mixing operations may be employed to obtainfthe-separationof the undulations of additionalchannels. l f It will be apparent that by varying the tuning of the resonant circuit-8 for eiective `maximum response to other pulse widths, the sequen-ceeofv the selection of thev channels may be changedat wi1l,if desired. Y f 1 ItA is thus seen that by a` combination of? the circuits'hereinabove described, aneiective pulse width selector has been provided, vin accordano with the objects set forth heretobefore. v While I have described above the principles of this invention Vin connection with specicapparatus, vit is to be clearly understood that this description is made Vonly by Way of example, and not as a limitation on thek invention vand the iscope of accompanying claims.

Iclaim: f

1. A methodl of selectively receiving a plurality of channels of 'communication from a multi- `channelvtrain of pulses wherein the pulses of each channel differ from pulses of otherchannels by a diierent` identifying width; comprising translating said pulses into a series of corresponding undulations the amplitudes of ywhich vary substantially as the respective wid-th ofthev corresponding pulses; andv successively selecting the undulations of each channel in accordance with .the relative values of'the'amplitudes thereof.

2. A method according to claim 1, wherein the Selected. f

3. 'A method according to claim l, wherein the operation of translating includes the steps of producing periodic oscillations by means of said pulses under conditions of resonance at a frequency having a period proportionate to the width of the pulses of a given first channel.

4. A method according to claim 1, wherein the `operation of selecting includes the steps of successively clipping the peaks of the undulations in the order of the values of the amplitudes thereof.

5. A method of selectively receiving a plurality of channels of communication from a multichannel train of pulses wherein the pulses of each channel diier from pulses of other channels by a different identifying width; comprising translating said pulses into a series of corresponding undulations the amplitudes of which vary substantially as, the respective width of the corresponding pulses; and selecting the undulations oa given channel by successive maximum peak clipping the undulations of the channels having the larger amplitudes, employing the resultant of said maximum peak clipping operationto eliminate said undulations having ,the larger amplitudes fromithe .said series', and maximumv peak clippingV the maximum amplitude vundulations thus obtained. Y

l 6. A method of selectivelyireceiving a 'channel of communication from a multi-channeltrain .of pulses Wherein'the pulses of each channel differ from pulses of other channels bya dierent identifying width; comprising translatingl said pulses into aseries of corresponding'undulations the amplitudes of Which varysubstantially as .the respective Width ofv the corresponding pulses; .and selecting the undulations of a given vchannel by maximum peak clipping the undulations of the channelshaving the larger amplitudes; invertingfamplifying and mixing the resultant of said maximum peak clipping operations with the` said series-ofundulations representing all the channels, and maximumpeak `clipping the maximum amplitude undulations thus obtained.

A method-of -selectively receiving ia plurality; ofchannels of communication from` a multichannel trainof pulses wherein thepulsesof each-rchannel'diifer from pulses of otherchannels by a different identifying width; comprising translating said pulses into a series of correspending undulations of the amplitudes of which vary Isubstantially vas the respective Width of the corresponding pulses; andr selecting the undula-l tions any-'of said 'channels by successiveY maximumpeak clipping the undulations of the channels preceding those of any given channel with respect toy their amplitude values; .mixing thatrain of undulations obtained for the channel immediately preceding any given channel, before said train has been subjected'to maximum'peak-clipping, with the inverted-and amplified output of the maximum peak clipping-operation as. applied to 'said preceding channel, .and .maximum peak clipping the series of undulations -thus .i obtained.

8. A multiplex system. forfselectively and simultaneously receiving any. of a number .of chan'- nels of communicationfroml -a` multi-channel train of-v pulses wherein the pulses of each charinel diiler from. pulses ofother channels ibyia diierent identifying Width; comprisin'gfm'eansior translating saidv pulses,v into a. series.` of corre.- sponding undulations the amplitude` of` which Vary substantially as the respective width ofthe corresponding pulses common to all channels; and for each channel, means for selecting from said series the undulations of a given channel' in accordance with thev relative amplitude thereof.

*9.y A system accordingto claim 8, wherein said means for translating includes a vresonant circuit tuned to a frequency having a vperiod proportionate to the'width ofv the pulses of agiven first channeli i s l0. A system for selectively. receiving a plurality of channels of communication from-a multi-channel train of-` pulses wherein 'the pulses of each channel diifer--fromvpulses of other channels by a different'identifyingwidth; comprising means for translating said pulses into a series of corresponding undulations .the amplitudes of Which vary substantially as the respective Width of vthe corresponding pulses, including aresonant Vcircuit tuned to a frequency having a period proportionate to lthe Width of the pulses of a given rst channel, a plurality of receiving channels each having circuit means for selecting the vundulations of a channel when the undulations thereof arev of maximum amplitude, including peak-rider clipping means inl each-receiving channel, and-mixer means-in all but the first receiving channel; means for applying said series of undulations to thel circuit means of the first and second receiving channels; and means for applying the output of the circuit means of each receiving channel to the. mixing means of the succeeding receiving channel. v

ll. A systemor-selectivelyreceiving a plurality of channelsof communication from a multi-channel train ofA pulses whereinthepulses of each channel differ-from pulses of other channels by a different identifying width; comprising means for translating said pulses into a series of corresponding undulations the amplitudes of which vary substantially as the respective width of Ythe corresponding-pulses, including a resonant circuit tuned tov a frequency having a period proportionate to the Width of the pulses of a givenchannel; a plurality of receiving channels each having circuit means for selecting the undulations-of a channel Iwhere the undulations thereof are of maximum amplitude, including peak rider clipping means in each receiving channel, and mixer means in all but the first channel; means for applying said series of undulations to each of said channel circuit Amea'r'is;` means for receiving the output of the circuit means of each of the channels separated into channel sections; and means for applying the cumulated output of the preceding circuit means' to the mixer means of a given channel.

- DONALD D. GRIEG.HV

Images