|Publication number||US2732431 A|
|Publication date||Jan 24, 1956|
|Filing date||Apr 17, 1951|
|Publication number||US 2732431 A, US 2732431A, US-A-2732431, US2732431 A, US2732431A|
|Inventors||J. E. Boughtwood|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (3), Referenced by (14), Classifications (6)|
|External Links: USPTO, USPTO Assignment, Espacenet|
Jan.- 24, 1956 J. E. BoUGHTwOoD 2,732,431
CROSS-TALK SUPPRESSION Filed April 17, 1951 5 Sheets-Sheet l yJim 24, 1956 .1. E. BoUGHTwooD 2,732,431
CROSS-TALK SUPPRESSION Filed April 1'7, 1951 5 Sheets-Sheet 2 FIG. 2 J
o r A B om? D E ol 5 1 y E A B c D E E b 3 (1. 2
LLI D D l...- C :I 0. 2
L|.| D D t Ol a' 2 2 TIME- LIJ D D l: J 0.. b 2
l/TlME- LLI D D t C a 2 l INVENTOR.
T1 1E- 1. E. BouGHTwooD ATTORNEY Jan 24, 1956 J. E. BoUGHTwooD 2,732,431
CROSS-TALK SUPPRESSION Filed April 17, 1951 5 SheebS-Shee'I 3 FIC-3.3
AMPLITUDE TIME #E lo o FREQUENCY LLI D D t C .J D. 2
J. E. BOUGHTWOOD ATTORNEY Jan. 24, 1956 J. E. BoUGHTwOoD CROSS-TALK SUPPRESSION 5 Sheets-Sheet 4 Filed April 17. 1951 TIME- P .INVEN TOR.
J. E. BOUGHTWOOD fm( Wg. wmf
l ATTORNEY Jan. 24, 1956 Filed April 17, 1951 R EC.
5 Sheets-Sheet 5 l I Yl i C I D DETECTOR AMPLIFIER I I sTAGEs C a. D -II- g I I l I ll- I 1 l I l E I l F I DETECTOR AMPLIFIER I M I STAGES E aF l I I I- l I I I I I l G I I l DETECTOR AMPLIFIER I I STAGES G H I l I -II- l INVENTOR. 65/ J. E. BouCI-ITwooD ATTORNEY United States Patent O CROSS-TALK SUPPRESSION Jann E. Boughtwood, Halesite, N. Y., assignor to The Western Union Telegraph Company, New York, N. Y., a corporation of New York Application April 17, 1951, Serial No. 221,470
' 5 Claims. (Cl. 179-15) The present invention relates to pulse modulation systems and more particularly to suppressing cross-talk between the channels thereof.' K
In pulse modulation systems wherein the transmitting time is divided into discrete intervals divided among the differentv intelligence channels, it is important that the signals representing intelligence samples do not overlap, or if they do overlap that such overlapping is suppressed before the intelligence samples are applied to the respective receiving channels. If such overlapping is not suppressed, intelligence intended for one channel will appear' as interference in one or more other channels. This interference, which is commonly termed cross-talk, may be sutiiciently strong to render the systemunusable orf at least seriously to impair the operation thereof.
The chief cause of cross-talk in pulse modulation systems is distortion of the pulse wave shapes resulting from the non-linear phase-frequency and amplitude-frequency characteristics of the transmitting medium and the filters and other circuits of the transmitter and receiver.
Accordingly, it is an object of the invention to provide means for suppressing cross-talk between channels of a pulse modulation system.
More particularly, it is an object of the invention to provide means for suppressing cross-talk in a pulse modulation system resulting from distortion of the received pulse wave shapes.
Another object of the invention is to provide simple and easily adjustable apparatus for suppressing cross-talk between channels of a pulse modulation system.
Further objects of the invention will appear from the following description.
In accordance with the invention, these objects are achieved by providing means to shift the phase of a portion of the high frequency components of the received pulses to correct the shape of the leading edges of the received pulses, differentiating the received pulses to correct the shape of the trailing edges thereof, and by adding to the received pulses a modulation component of equal amplitude and in phase opposition to the long term crosstalk present in the respective receiving channels.
The invention will now be described in greater detail with reference to the appended drawing in which:
Fig. l illustrates the receiver of a pulse amplitude modulation system;
Figs. 2, 3 and 4 are diagrams illustrating the operation of the circuit of Fig. 1;
Fig. 5 illustrates the productionof long term crosstalk; and i Fig. 6 shows a circuit for suppressing long term crosstallt.`
Referring now to the'drawing and more particularly to Eig; 1, the receiver illustrated'therein is substantially the same asthe receiver of the pulse amplitude modulation system described in my copending U. S. patent application Serial No. 221,469, tiled concurrently herewith and now Patent No. 2,672,517.
2,732,431 Patented Jan. 24, 1956 i 2 The receiver of Fig. l comprises a radio receiver 10, the output of which is applied to an amplifier and low pass tilter 11, the output of which is, in turn, derived from a copending patent application.
potentiometer 12 and applied to the control grid 13 of an amplifier tube 14 through a capacitor 15. The amplified output of tube 14 is applied to control grid 16 of an electron discharge tube 17 through a coupling capacitor 18. Anode 19 of tube 17 is coupled to a source of Vpositive potential through a resistor 20, while cathode 21 is coupled to ground through a resistor 22. Resistors 20 and 22 are preferably equal in magnitude so that the signal Avoltages developed thereacross will be substantially equal and in phase opposition.
One end of a variable capacitor 23 is connected to anode 19. One end of a variable resistor 24 is connected to cathode 21. The free ends of capacitor 23 and resistor 24 are connected together and to the control grid 25 of an electron discharge tube 26 through a coupling capacitor 27.
Assuming a signal voltage having a given frequency npplied to control grid 16, the phase of the voltage applied to control grid 25 can be adjusted through substantially by varying the respective values of capacitor 23 and resistor 24 because the voltage applied to grid 25 is de termined by the relative proportions and phase relation ships of the voltages at anode 19 and cathode 21 applied to grid 25. If the signal voltage applied to control grid 16 comprises a plurality of frequency components, the
'phase relationships between the components can be varied by varying the relative values of capacitor 23 and resistor 24 Anode 28 of tube 26 is coupled to a source of positive potential through a resistor 29. Cathode 30 of tube 26 is coupled to ground through two series connected potentiometers 31 and 32. The tappings of potentiometers 31 and 32 are coupled to ground through variable resistor 33 and variable capacitor 34 and variable resistor 35 and variable capacitor 36, respectively. Circuit elements 31 through 36 constitute a diierentiating network, the function of which will be described more fully hereinafter.
Anode 28 is coupled to ground through a variable capacitor 37 and to a synchronous gate 38 through coupling capacitor 39. Cathode 30 is coupled to frequency divider system FD, which is in turn coupled to synchronous gate 38 and to an electronic receiving distributor 40. The outputs of receiving distributor 40 are applied, respectively, to detector-amplifiers 41A through 41H. The outputs of detector-amplifiers 41A through 41H are applied, respectively, to output terminals OA through OH through low pass iilters 42A through 42H, respectively. The output of detector-amplifier 41A is also applied to a band pass filter 43 which is coupled to distributor stepper 44. The output of stepper 44 is applied to frequency dividers FD.
The circuits of the detector-amplifiers 41A through 41H will be described more fully hereinafter in connection with Fig. 6 The circuits and functions of synchronous gate 38, frequency divider system FD, receiving distributor 40, low-pass filters 42A through 42H, band pass filter 43 and stepper 44 are fully described in my copending patent application referred to hereinbefore.
In Fig. 2A there are illustrated idealized band pulses for application to amplifier tube 14 of Fig. 1. These pulses are substantially identical to the sending Vdistributor output after slicing in a-gating circuit as described in my said In Fig. 2 .the pulses for channels A through' E are indicated, channel C having modulationpresent thereinindicated `by crosshatching. The pulses of Fig. 2a contain a large number of harmonic` components, so that transmission through circuits and media having restricted `band width will produce changes in wave shapes.
`might actually be present at'the output of amplifierand filter 11. It will be observed that the steep leading and trailing -edges'of the pulses ofl Fig. 2a have .beenlost. Furthermore, the `waves 'corresponding to-thedifferent channels overlap,`which will tend to produce cross-talk in the outputs of receiving distributorv 4t). Fig. 2c illus- `trates the wave shapes of Fig. 2b without modulation in channel C. yThe time intervals outlined by dashed lines A through E indicate the portions of the waves sliced bysynchronous gate '33. ln other words, only those Aportions of the pulseslying within the dashed lines will be applied to receiving distributor A40. Use of a. gating circuit materiallyI reduces cross-talk, but, 'as is evidentfrom Fig. 2c, overlapping during the gating intervals still occurs.
v Fig. 3aillustr-ates the wave-shape oflthe channel C pulse at thefoutput ofiilter 11 and the gating intervals B', C and D of channels B, C and D, respectively. The leading and trailing edges of pulse Chave slopingv Ashapes instead of the Vdesiredsharp rise and fall of Fig.
2a. The principal causes of this loss of desired shape of the leading edge are the non-linear amplitude-frequency and phase-frequency characteristics ofthe circuits and media through which the pulse has been transmitted.
In particular, restricted bandwidth and limited transient responses produce variations in amplitude and phase of the high frequency pulse components with respect to the low frequency pulse components.
Adequate suppression of crosstalk does not require the received pulses to be returned to their ideal shapes, but only that the pulse amplitude of augiven channel be sharply reduced during the gating intervals of other channels. This is accomplished with respect to the leading edge ofthe waves byvshifting the phase Vof a portion of the high frequency components. Correction of the leading edge of the waves is accomplished in the circuit of Fig. 1 by tube M17 and its associated phase shifting circuit.
The phase-f`requency characteristic of this phase shifting circuit is illustrated in Fig.A 3b, which' is a plot 0f phase vs. frequency. If the phases '.at anodev 19 and cathode 21 be assumed -as +90 and 90", respectively, at low frequencies capacitor 23 will appear as an open circuit, so the voltageapplied to'grid 25 will' have a phase of -V9G". As thefrequency is increased,"the veffect of capacitor 23 becomes more pronounced, shifting the phase of the voltage `applied to grid 25. At high frequencies, capacitor' 23 acts substantially 'as a short circuit, so that the phase of the voltage.applied`to grid 25 will -be substantially +90". WhenA pulse C is applied to the phase shifting network, the different frequency components thereof will 'be shiftedlin phase in accordance with the characteristic shown in Fig. 3b. These phase shifted components 'vary 'the' wave shapeiof pulse C, producing a 4corrected waveshape at the leadingedge of the Wave.
The corrected pulse RC is shown in Fig. 3c. f From lFig. 3c it can be seen that lthe leading edge of wave RC does net appearv duringjtheVV gating interval: B. lThis phase shiftingcircuit vvill have a'negligible effect'onthe trailing edge of thewave.
Fig. 3d is a vector diagram of the voltages across the velements of the phase 'shifting circuit. The lphase of resultant voltage VR, which .is the voltage-developed between the' junction of resistor 24 and capacitor'i23 and groundjcan be varied through"'180 along locus curve L by relative 'adjustment of resistor 24and capacitor v23,
Tilt/bichadjustment "va'ries voltages V24 and' V231tinmagnitudeuand` phase. vCapacitoi'v 23'lrenders theV circuit-freiquencyy sensitive, so that the phaseshift producedV can be materially differentk at l high and.l lovv frequencies', jas indicatcdin Fig'.- 3b.
Corrected lwave RCstill overlaps. gatingffintervalf` D of channel D, as shown in Fig. 4a. This overlap .can be suppressed by differentiating pulse RC to produce a pulse `TC," and combining pulse YTO with 'pulse' RC l to produce pulse RC. The pulses RC, TC and RC are illustrated in Figs. 4a, 4b and 4c, respectively. The differentiating 5 network for producing the trailing edge correction comprises circuit elements 31 through 36 in the cathode circuit of tube 26. This differentiating circuit also tends to teepen the leading. edge` of pulse R'C, asshown in Fig.
"l" he trailing edge correcting circuit `will-not be aseffective if the trailing c dge of pulse Cl exhibits-oscillatory excursions, as illustrated in Fig. 4d. However,.;these oscillatory excursions,as. illustrated in Fig. 4d, are compensated for in the circuit of Fig. 1 by partially decoupling anodej'ZS with respect y.to'high frequencies by means of capacitor 37. Capacitor 37 effectively converts the wave of Fig. 4d into the Wave of Fig. 3a.
Cross-talk may also be produced through thesuppression of low frequency components 'of the pulses. In Fig. 5a there is shown an ideal vpulse without modulation. Fig. 5b illustrates the pulse of'Fig. 5a after having passed through a network which strongly attenuatesthe lower frequency components. It will be noted'that the pulse of Fig. 5b exhibits` a negative excursionV of relatively-short duration and relatively large amplitude. 'If-suppression of low frequency components is reduced'throughfcareful design of the system, a pulse as shown' in Fig.5c'inwhich the negative excursion iseXaggerated, will result. This negative excursion is 'of relatively'small and substantially constant amplituden but long duration -relative 'to Athe .sampling4 intervals. Distortion Vof this type'yields long term cross-talk, that is, cross-talk appearing in'each of the channels With substantially 'equal amplitude.
Ain accordancev with the invention, long term cross-talk is suprressed'by adding a'm'odulation component of equal amplitude` and in p'haseopposition to lthe long' term crosstalk' in each of Athe receiving channels. A suitable circuit is.shown" in Fig. 6, which illustrates'in detail thcdetectorarnplier stages 41A'through 41HY of Fig.I l.
' etectoramplifier stage 41A comprises a diodeerectiier 50, the cathode 51 of which is coupled to ground through a resistor 52. The channel A signal .pulses are .applied vtov cathode 51 through coupling capacitor 53. Anode 54 ofv diode 'S0' is coupled to control grid 55Uof amplifier tube 56 through a coupling capacitor 57. Anode 54 is also coupled'to ground throughtheparallel'combination of capacitorSSmand'resistor 59. Cathode 60 of tube 56 is coupied to ground-through-thelseries combination of'biasing resistorsland 62. Control grid SS is coupled to the junction of resistors 61and-62 'through a voltage'divider comprising resistors Y63and 64. The junction of resistors63and 64-is coupled to-ground through a `variable capacitor 65. "-Anode'66 of tube 56r is coupled to a source 4of positivel potentialwthrough a'resistor 67and' to terminal-A.
f The -otherA detector-amplifier stages Aare'identical with the v'stage described'and hence will-inet lne-.described in detail.l lt is' important tonote,` however,fthatrxthepoint in each stagey corresponding to -thejunctionY ofresistors 63 andfttjis'coupledfto the same terminalofcapacitor 65. Therefore,-a vportionof the output signal-ofJeach amplifier tube will appear across capacitor 65 and-Will be .applied to the control grids of each of the amplifier'tubes.
The magnitude of the voltage'thus applied may'bead- 65 justedby varying the Valuenof capacitor 65. jThe'phase of this voltage is such as to suppress the long termfcrosstalk in each ofthe stages. A.It is evident that lthe vmore nearly equal the cross-talk in each'stage'produc'ed by distortion vinone channel, the more complete vvillfbelthe `suppression lof long term cross-talk.
l' Whilethe invention has been describedin a'pa'rticular embodiment thereof and in a particular 'use,-' itvv is-not desired, that it be limited thereto, forobvious'Amodifications thereof will occur to1 those'skilled inthe-arthwithout departing from the spirit and scope of the invention as set forth in the appended claims. n
What is claimed is:
l. A multi-channel pulse modulation system receiver, comprising phase shifting means having an input circuit and an output circuit and being arranged toshift the phase of frequencies lying within a'frst given range with respect to the phase of frequencies lying within a second given range lower than said rst Vgiven range, means to apply received pulses having sloping leading and trailing edges and including frequency components lying with in said rst and second given ranges to the input circuit of said phase shifting means thereby substantially to steepen the leading edges of said received pulses, a differentiating network coupled to the output circuit of said phase shifting means, said network including means to diiferentiate said pulses thereby to produce correcting pulses, means to combine said correcting pulses and said received pulses thereby substantially to steepen the trailing edges of said received pulses, a plurality of receiving channels, means including an electronic distributor arranged to commutate said receiving channels, means coupled to said combining means to apply said received pulses to said electronic distributor, means coupled to said receiving channels to derive therefrom modulation components proportional to long-term cross-talk present in said receiving channels, and means to apply said modulation components to each of said receiving channels in a sense to suppress long-term cross-talk therein.
2. A multi-channel pulse modulation system receiver, comprising phase shifting means having an input circuit and an output circuit and being arranged t shift the phase of frequencies lying within a rst given range with respect to the phase of frequencies lying within a second given range lower than said rst given range, means to apply received pulses having slopingleading and trailing edges and including frequency components lying within said iirst and second given ranges to the input circuit of said phase shifting means thereby substantially to steepen the leading edges of said received pulses, capacitive means coupled to the output circuit of said phase shifting means selectively to attenuate the frequency components of said received pulses lying Within said first given range, a dilerentiating network coupled to the output circuit of said phase shifting means, said network including means to differentiate said pulses thereby to produce correcting pulses, means to combine said correcting pulses and said received pulses thereby substantially to steepen the trailing edges of said received pulses, a plurality of receiving channels, means including an electronic distributor arranged to commutate said receiving channels, means coupled to said combining means to apply said received pulses to said electronic distributor, means coupled to said receiving channels to derive therefrom modulation components proportional to long-term crosstalk present in said receiving channels, and means to apply said modulation components to each of said receiving channels in a sense to suppress long-term cross-talk there- 1n.
3. In a- .multi-channel pulse modulation system receiver, comprising circuit means to receive signal pulses having a sloping trailing edge, an electron discharge tube having a n control grid coupled to said circuit means, means including a differentiating network connected in the power circuit of said electron discharge tube, said network including means to differentiate said signal pulses thereby to produce correcting pulses, means to combine said correcting pulses and said Vreceived signal pulses thereby to steepen the trailing edges of said received signal pulses, an electronic distributor, second circuit means coupled to the output of said electron discharge tube and to said electronic distributor, a plurality of receiving channels, means coupling said receiving channels to said electronic distributor.
4. The combination set forth in claim 3 including means coupled to said receiving channels to derive therefrom modulation components proportional to long-term cross-talk present in said receiving channels, and means to apply said modulation components to each of said re- Y ceiving channels to suppress long-term cross-talk.
5. In a multi-channel pulse modulation system receiver, the combination comprising circuit means to receive signal pulses, a phase shifting network having an input circuit coupled to said circuit means and an output coupled to an electron discharge tube, a differentiating network connected in the cathode circuit of said electron discharge tube, a plurality of receiving channels, an electronic distributor arranged to commutate said receiving channels, means coupled to the output circuit of said electron discharge tube to apply said received pulses to said electronic distributor, means coupled to said receiving channels to derive therefrom modulation components proportional to long-term cross-talk present in said receiving channels, and means to apply said modulation components to each of said receiving channels to suppress long-term cross-talk.
References Cited in the file of this patent UNITED STATES PATENTS 2,524,251 Bradley Oct. 3, 1950 2,553,572 Frum May 22, 1951 FOREIGN PATENTS 605,128 Great Britain July 16, 1948 g OTHER REFERENCES Theory and Application of Electron Tubes, 1944, Reich, p. 357. y
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US2524251 *||Oct 26, 1948||Oct 3, 1950||Philco Corp||Pulse-modulation system|
|US2553572 *||Nov 10, 1947||May 22, 1951||Int Standard Electric Corp||Cross talk reduction in pulse multiplex receiver systems|
|GB605128A *||Title not available|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7072415 *||Sep 27, 2001||Jul 4, 2006||Rambus Inc.||Method and apparatus for generating multi-level reference voltage in systems using equalization or crosstalk cancellation|
|US7093145||Jul 30, 2004||Aug 15, 2006||Rambus Inc.||Method and apparatus for calibrating a multi-level current mode driver having a plurality of source calibration signals|
|US7456778||Mar 29, 2006||Nov 25, 2008||Rambus Inc.||Method and apparatus for calibrating a multi-level current mode driver having a plurality of source calibration signals|
|US7859436||Oct 24, 2008||Dec 28, 2010||Rambus Inc.||Memory device receiver|
|US8199859||Oct 4, 2010||Jun 12, 2012||Rambus Inc.||Integrating receiver with precharge circuitry|
|US8320494||Jun 15, 2006||Nov 27, 2012||Rambus Inc.||Method and apparatus for generating reference voltage to adjust for attenuation|
|US8634452||Jun 7, 2012||Jan 21, 2014||Rambus Inc.||Multiphase receiver with equalization circuitry|
|US8861667||Jul 12, 2002||Oct 14, 2014||Rambus Inc.||Clock data recovery circuit with equalizer clock calibration|
|US9544169||Jan 17, 2014||Jan 10, 2017||Rambus Inc.||Multiphase receiver with equalization circuitry|
|US20020075968 *||Sep 27, 2001||Jun 20, 2002||Jared Zerbe||Method and apparatus for generating multi-level reference voltage in systems using equalization or crosstalk cancellation|
|US20060186915 *||Mar 29, 2006||Aug 24, 2006||Carl Werner||Method and apparatus for calibrating a multi-level current mode driver having a plurality of source calibration signals|
|US20060233278 *||Jun 15, 2006||Oct 19, 2006||Rambus Inc.||Method and apparatus for generating multi-level reference voltage in systems using equalization or crosstalk cancellation|
|US20090097338 *||Oct 24, 2008||Apr 16, 2009||Carl Werner||Memory Device Receiver|
|US20110140741 *||Oct 4, 2010||Jun 16, 2011||Zerbe Jared L||Integrating receiver with precharge circuitry|
|U.S. Classification||370/201, 370/310|
|International Classification||H04J3/02, H04J3/10|