Abstract: The present invention consists of a multirate data receiver which will recover data at any of N data rates with no predetermined restrictions on the value of N. The data rate at which the receiver operates is selected by a single 1-of-N switch selection. An equalizer section is constructed for each of the N data rates but are all controlled by the same (common) automatic adjustment circuitry, which controls gain and frequency compensation. The output from one of the equalizer sections is selected by an analog multiplexer constructed of field effect transistor (FET) switch circuits. The recovered waveforms from the equalizer sections are processed by common circuitry to recover the transmitted data. Clock recovery may be accomplished, for example, by a digital phase locked loop constructed such that the characteristics of the loop filters and divide-by-M feedback counters are easily modified by logical signals controlled by the 1-of-N switches.

Abstract: Bandpass filters are used to sample the output signal from the equalizer. These samples are amplified in variable gain amplifiers and summed with the input signal to achieve equalization. The amount of gain applied to each sample is based on the difference between each sample's magnitude and a reference signal. In one embodiment, the reference signal is derived from the adjacent bandpass sample. In another embodiment, one bandpass sample is the reference for comparison with all other samples. In the third embodiment, bandpass filters are also connected to the input signal in order to achieve a feed-forward rather than feedback configuration for the equalizer. In all of the embodiments, the magnitude signals from the bandpass samples are weighted before being compared to the reference signal. The weighting factors allow the filter to be designed for a desired transfer function and adapted to an input signal spectrum.

Abstract: A fractionally spaced adaptive equalizer is disclosed wherein the coefficients (C.sub.-2 to C.sub.+2) are updated synchronously by an integral -T error circuit (40) and alternately by a fractional =T error circuit (48). This arrangement features a single unique minimum for the adaptive structure thereby preventing the coefficient drift to very large values so coefficient tap leakage is not required to minimize coefficient values. The adaptive characteristic is well suited for linearly dispersive channels, such as exhibited during terrestrial radio transmission during tropospheric multipath propagation. The inventive principles are broadly applicable to any of the well known algorithms for adapting digital filters. A configuration of fractionally spaced equalizers (70.sub.1 -70.sub.4) provides the benefits of Nyquist-rate coefficient updating for a dual rail quadrature amplitude modulated (QAM) signal application.

Abstract: SMSK data link degradation in future and immediate past bits produced by carrier phase error and small errors in the sampling instant are reduced by a recursive adaptive equalizer having a feedforward stage with adjustable weight for each future bit to be corrected and a feedback stage with adjustable weight for each immediate past bit to be corrected. The weights are controlled by a controller which samples the output and adjusts each weight until the output is correct.

Abstract: In an adaptive equalizer of the decision feedback type for reducing intersymbol interference in subscriber loops of a telephone system, the corrected digital signal is applied to means for detecting the occurrence in the signal of a zero state following a non-zero state. When that occurs, a correlator samples the signal during the "zero" bit period and applies the sample to an integrator. The integrator output varies the coefficients of the feedback signal applied to the input means for addition to the incoming digital signal to provide the corrected signal. The equalizer may have several stages, each for detecting a "zero" state in a different bit period following the original pulse. Also the circuitry may be duplicated for each polarity if the input signal is bipolar.

Abstract: A method and apparatus for equalizing a scanned noisy digital signal with an automatic and adaptive decision feedback equalizer comprising a nonrecursive part and a recursive part. A switch is disposed in the feedback loop of the equalizer between a decision device and the recursive part of the equalizer for allowing the determination of a step size factor. For this purpose the switch is in such a position that a reference signal can be fed into the feedback loop and at the same time the gradient coefficients are fed to a device for setting the filter coefficients. The gradient coefficients are determined at that point in time where the switch was set to the other switch setting. The number of the required interation intervals to approaching the minimum of the equalization error can be decreased by employing the method of the present invention. In addition, the filter configuration can be employed in a multiplex operation.

Abstract: A terminal for a duplex transmission system for digital signals, which includes an adaptive quantized equalizer for reception and a pulse-shaping equalizer for transmission. The correction factors for the transmission equalizer are derived from the adaptively adjustable coefficients of the reception equalizer, the derivation being effected in accordance with recursive expressions which signify minimum intersymbol interference between successive pulses in a received digital signal.

Abstract: A transmitter for transmitting a carrier whose frequency is modulated by pseudo five-level signals. In order to obtain a simple receiver and an improved signal-to-noise ratio, a substantially three-level signal is generated in the transmitter at the detection instants t=(2m-1)T/2 by means of a pre-modulation filter.

Abstract: In a data receiver, a complex data signal after demodulation and sampling is applied to an equalizer which provides an improved signal which is applied to a phase and amplitude correction circuit for correction utilizing a complex reference vector. The corrected signal is applied to a decision circuit the input and output of which are applied to a difference determinator circuit, the output of which is a complex residual error signal utilized in conjunction with a plurality of gain factors generated in a gain factors generator to determine the reference vector. The gain factors vary in time from transmission initialization such that optimal compensation for transmission disturbances is achieved both at initialization and after stabilization.

Abstract: An automatic correction circuit is responsive to suitable bit patterns in a bit stream being received from a transmission medium, to generate an error amplitude map of signal distortion being introduced by the transmission medium. The error amplitude map is digitally stored or recorded. After each non zero bit occurrence in the signal stream a synchronized output signal corresponding to the recorded error amplitude map is generated and added to the bit stream being received to reduce signal distortion, whereby information recovery from the received bit stream is improved.

Abstract: A circuit for selectively combining an input signal with a selected fraction of the input signal, twice differentiated, is comprised of two amplifier circuits, each having two input stages and a dual output stage with one output stage of each amplifier circuit being gain controllable in response to a control signal. One of the gain controllable output stages of one amplifier and the non-gain controlled output stage of the other amplifier are connected to a single combined output terminal. The other output stages are connected to a circuit ground. An input signal is supplied to one input stage of each amplifier circuit and the same input signal, after being double differentiated by means of an inductor and capacitor, is supplied in noninverted form to the other input stage of one of the amplifier circuits and is supplied in inverted form to the other input stage of the other amplifier circuit.

Abstract: A carrier and timing recovery technique is disclosed for combating the phase effects of multipath fading occurring during radio transmission. The technique is based on a known mathematical model of multipath fading. A test variable generator (26) develops a number of test variables whose relationship characterizes the multipath distortion. A processor (27) derives a plurality of coefficients of weighting measures from the test variables. The coefficients serve to adjust a matched filter (28) to provide greater use of those portions of the received radio signal yielding higher signal-to-noise ratios than those portions with lesser yields, as well as to compensate for multipath phase perturbations. The technique is also compatible with maximum-likelihood sequence detection techniques to provide further compensation for multipath fading effects.

Abstract: A line equalizer including a .sqroot.f equalizer for compensating a .sqroot.f-characteristic of a transmission line, a BT equalizer connected in series with the .sqroot.f equalizer for removing an echo component caused by a bridged tap (namely, BT) on the transmission line, and a circuit for controlling the .sqroot.f equalizer is disclosed. A signal applied to the .sqroot.f equalizer is subjected to over-equalization to make the time domain length of impulse response at the output of the .sqroot.f equalizer small. The equalization state of the output of the .sqroot.f equalizer is judged by signals formed in the BT equalizer, and the gain of the .sqroot.f equalizer is controlled on the basis of the result of judgement. In this manner, the .sqroot.f equalizer is controlled without suffering any interference between a control loop of the .sqroot.f equalizer and a control loop of the BT equalizer.

Abstract: In a system in which a central modem individually polls a number of remotely located modems for data, an adaptive equalizer in the central modem is initially trained to minimize communication link interference from a remote modem and develops a set of equalizer coefficients based on the initial training which is transmitted to the remote modem and stored there for future use while the central modem goes on to do the same with other remote modems. When a remote modem is polled, the stored equalizer coefficients are used to set the adaptive equalizer of the remote modem so the adaptive equalizer of the central modem does not have to be retrained for each individual remote modem.

Abstract: An equalizer arrangement comprises a fixed-characteristic equalizer which is supplemented by a variable-amplitude notch equalizer to enable proper equalization of signals transmitted via cables of arbitrarily mixed cable gauges. Automatic control of the notch amplitude is provided by detecting a predetermined signal level in the equalized signal, comparing this with its correct value, and controlling an up/down counter accordingly. The count of the counter determines a current flow through a diode which constitutes a variable resistance of the equalizer.

Abstract: An automatic equalizer includes two wave memories. These wave memories jointly constitute a wave integration memory for storing amplitude information of an input signal. The wave form of input signal to be equalized is divided into fragments along time base, and the amplitude components (analog) of respective fragments are sequentially converted to digital forms in a manner that these fragmentary digital components are alternatively distributed to respective wave memories. For instance, the digital components of odd number fragments are stored in one wave memory, and the digital components of even number fragments are stored in the other wave memory. In this case, the read/write speed of each wave memory may be half of the read/write speed of a prior art wave integration memory which should store all fragments of the digital components.The fragmentary digital components corresponding to the input signal wave form are divided into at least two groups, e.g. an odd group and an even group.

Abstract: A transversal equalizer has a transversal filter which performs a convolution of an unequalized signal and a tap weight. In the automatic equalization mode, the tap weight is updated according to the result of a correlation of the unequalized signal and an error signal, and the updated tap weight is applied to the transversal filter. In the fixed equalization mode, a fixed tap weight stored in a memory is applied to the transversal filter.

Abstract: An automatic equalizer has a transversal filter circuit for equalizing an input signal to be equalized and providing an equalized output according to the value of tap coefficients thereof. The transversal filter circuit includes a low-pass filter (LPF) at its output stage. The equalized output is subjected to a phase shift by the LPF. The equalized output is applied to a sign circuit as well as a wave integration circuit. The sign circuit generates a sign signal indicating a rate of change of the equalized output to time. The wave integration circuit generates a differentiated wave signal indicating the change of wave form or the change of amplitude information of the equalized output. The sign and wave signals are applied to a correlator which performs the correlation operation of the sign and wave signals and providing a tap coefficient control signal. This control signal is used for determining the tap coefficients of the transversal filter circuit.

Abstract: An equalizer apparatus suitable for use in a MODEM for receiving a signal having passed through a number of carrier-band lines is disclosed in which the number of carrier-band lines is detected from a training signal, and electrical connection of a fixed equalizer to a variable equalizer is controlled on the basis of the number of carrier-band lines, in order to form a precise, simple automatic equalizer apparatus.

Abstract: In a digital system, especially where PCM coded speech is conveyed over twisted pair cables optimized for analogue speech, a pulse is received distorted, the distortion being especially bad on the trailing edge of the pulse. This decays slowly enough to cause intersymbol interference with the next pulse.This is overcome by applying the signal from an amplifier (11), FIG. 4, to one input of a subtractor (12) via a delay and attenuation circuit (13-15) and to the other input of the subtractor directly. The delay (13) is such as to optimize the peak of the received pulse, and the attentuation level is calculated to cancel out the tail of the received pulse so that it is zero at the center of the sampling point of the next bit.

Abstract: A wide-band ghost reduction circuit apparatus for a television receiver comprises a ghost detector, two transversal filters having delay characteristics different from each other, a switch for selecting one of the filters and an automatic gain control circuit responsive to the output of the ghost detector for controlling the filters and the switch. Each of the transversal filters includes tap amplifiers whose gains are controlled by the automatic gain control circuit and having outputs connected, respectively, to tap inputs of delay devices formed of such as a CCD each provided with a tap and included in the associated transversal filter. Output end of one of the transversal filters having a smaller delay is connected to a selected one of the input tap of the other transversal filter through a switch so that input video signal components join together at every input terminal or every individual tap position of the delay devices.

Abstract: An adaptive equalizer performs equalization at IF, demodulates the input signal and then generates the equalization coefficients at baseband from the demodulated data. Coefficient updating can be done on a non-real time basis and coefficients for individual bursts stored in a coefficient memory for retrieval when the same burst occurs in the next frame, thus obtaining a pseudo-continuous equalization for each burst. The error signal is derived from demodulated equalized data and correlated with demodulated unequalized data to generate the correlation coefficients, and individual error signals from each channel can be combined into a single composite error signal for correlation purposes.

Abstract: An improved Kalman equalizer for providing a tap weight of a transversal filter for equalizing a transmission line has been found. A complex number fast Kalman equalizer (FIG. 27), which is the improvement of the prior fast real number Kalman equalizer (see; "Fast calculation of gain matrices for recursive estimation schemes" by Lennart Ljung, et al, INT. J. Control, 1978, vol. 27, No. 1, 1-19), has a complex conjugate generator (C.sub.2) for processing a complex input signal so that an equalization can be performed on the bandwidth division concept, and a transmission line having a complex impulse response can be equalized. Then, a bandwidth division equalization in which a whole bandwidth is separated into a plurality of sub-bandwidth each being equalized by a related sub-equalizer becomes possible. Also, a QAM signal (Quadrature Amplitude Modulation) which has a complex impulse response can be equalized. A bandwidth division fast Kalman equalizer (FIG.

Abstract: A non-linear equalizer corrects for intersymbol interference in a digital data transmission system by introducing baud rate samples into an N-stage tapped delay line and provision is made for storing a sequence of M prior decisions in a prior decision register. Both the samples from the tapped delay line and the contents of the prior decision register are coupled to a weighting matrix processor which is updated to permit adaptation to different channels conditions. When initially placed into operation, a training sequence is coupled to a reference register and differences between arrived-at decisions and the true symbols are employed to provide an error vector which is then used to update the weighting matrix in the processor. Decisions on the data estimates are hard-limited to form a set of M prior decisions which are placed into the prior decision register.

Abstract: In adaptive filters, under certain incoming signal conditions the tap coefficients tend to drift toward relatively large values. This condition has been countered by introducing so-called leakage into the tap coefficients which tends to drive them toward zero. Introduction of leakage is desirable under certain incoming signal conditions and not others. Indeed, during intervals that partial band energy, e.g., single frequency tones, multifrequency tones and the like, is incoming to the filter there is a need to introduce a relatively large leakage value into the tap coefficients. However, during intervals that whole band energy, e.g., speech or noise, is incoming to the filter no leak is desired or needed. This is realized by detecting intervals when partial band energy is incoming to the filter and controllably increasing the leakage value during those intervals.

Abstract: A small-sized LSI variable equalizer is provided for accurately equalizing waveforms of signals transmitted via transmission lines of different distances. Variable equalizer units capable of stepwise changing the equalizing characteristics thereof are connected in series with each other with the variable equalizer units having variable step widths different from each other. An output signal of the variable equalizer units is compared with a reference signal to convert the comparison output signal to a digital signal which includes an upper order bits and lower order bits, whereby one of the equalizer units which has a wide variable step width is controlled by the upper order bits and the other of the equalizer units which has a narrow variable step width is controlled by the lower order bits.

Abstract: An equalizer of a transversal filter type is given a substantially constant gain at a preselected frequency (f.sub.p) in a predetermined frequency band of an input and an output signal. For this purpose, the output signal is given by multiplying the routine transversal filter output by a reciprocal of an absolute value of a sum of complex tap gains (C.sub.1 to C.sub.N+1). Alternatively, the input signal and successively delayed signals (IN and D.sub.1 to D.sub.N) may be multiplied by the reciprocal before summation. It is possible to approximate the reciprocal by omitting those of the tap gains which are near both ends (as C.sub.1 and C.sub.N+1).

Abstract: A characteristic control system for a digital equalizer comprises a digital equalizer constructed from a digital filter, for directly giving a desired equalizing characteristic with respect to a digital pulse-modulated signal, and a control part for changing the equalizing characteristic of the digital equalizer from a first arbitrary characteristic to a second arbitrary characteristic so that one or a plurality of intermediate equalizing characteristic is obtained before reaching the desired second arbitrary characteristic.

Abstract: A signal processing system detects the transmission characteristics of a channel thereby to compensate the output waveform of the channel into the most-optimum state. The impulse response of the channel is detected by transmitting a code having a keen autocorrelation from the transmission end and by determining the correlation between the received signal of the code transmitted and the same code of the aforementioned code at a reception end.

Abstract: A method and apparatus for the adaptive equalizing of phase and amplitude distortion in a received signal by an N-stage digital delay line system and comprising the steps of digitally sampling the received signal and advancing each digital sample through successive stages of the N-stage digital delay line, producing a continuously up-dated scaled error signal after each advance of the digital samples in the delay line, multiplying the digital sample in each delay line stage by the scaled error signal after each sampling advance to produce a first scaled signal in each stage, separately accumulating the first scaled signals of each stage after each sampling advance, scaling and accumulating the accumulations of the first scaled signals of each stage after each M accumulations thereof to produce a weighting tap signal for each stage of the delay line, multiplying the received samples by the weighting tap signal of each stage to produce a weighted sampling for each stage, summing the weighted samplings to produc

Abstract: A low cost automatic equalizer is provided for reducing the amplitude and phase distortion produced by close-in echos. The color subcarrier burst, or some other portion of the video signal having predetermined amplitude and phase characteristic prior to transmission, is gated into detection circuitry. A regenerated color subcarrier, phase shifted to be in phase with the subcarrier burst, is used to detect the in-phase and quadrature components of the subcarrier burst, which are indicative of amplitude and phase distortion. The signals indicative of amplitude and phase distortion are applied to a three tap delay line equalizer, where the amplitude and polarity of the first and third tapped signals are controlled in unison to reduce amplitude distortion, and where the amplitude of the first and third tapped signals is controlled in unison, but the polarity of the first and third tapped signals is controlled oppositely to reduce phase distortion.

Abstract: A radio signal will often suffer from amplitude and group delay distortion caused by multipath fading. An adaptive equalizer, which effectively cancels the amplitude distortion effects, is based upon the use of control voltages derived from selected portions of the IF band; the control voltages being selectively applied to bump and slope equalizer sections to cancel the effects of amplitude distortion. Cancellation of group delay distortion is predicated upon its being proportional to amplitude of the slope correction required. Thus, group delay correction is introduced in relation to the slope correction introduced to compensate for distortion introduced during normal multipath propagation conditions.

Abstract: A fast start-up system is disclosed for a transversal equalizer comprising a transmitter capable of generating a periodic symbol sequence with a period L, the auto correlation function of said sequence being 0, except for its center. The system generates and transmits a test signal based upon the generated symbol sequence. A receiver generates the same periodic symbol sequence as that generated in the transmitter in a proper phase relation thereto according to a received signal. The correlation between L sampled values and a sequence obtained by recirculating the periodic symbol sequence for each of consecutive M shifted phases, and for each of M' consecutive shifted phases is calculated. The calculated M+M' correlation values are multiplied by a constant. A discrete Fourier transform is obtained for the M+M' multiplied correlation values at N points equal to or greater than M+M' in number.

Abstract: An adaptive equalizer for a twisted pair telephone line comprises a differential input/output amplifier, with the line connected to its positive input via a resistor and DC blocking capacitors. The junction of these capacitors is grounded via an adjustable impedance.The amplifier has a feedback loop with another adjustable impedance also connected to the negative input of the amplifier.Line impedance variations vary the impedance of both of the adjustable impedances, which, include diodes to separately compensate for the line's attenuation and frequency characteristics.

Abstract: Disclosed is a constrained adaptive equalizer for use in a digital communication system receiver combining transversal and decision feedback equalization techniques. The constrained adaptive equalizer uses a constraint on the taps of the tranversal section of the equalizer which has both leading and lagging taps. These constraints force the transversal equalizer to correct only the leading echoes of a received digital signal in a communication system and treat them as if they resulted only from phase distortion. The constrained adaptive equalizer minimizes noise amplification while reducing intersymbol interference. The transversal segment of the constrained adaptive equalizer approximate the phase equalizer by constraining its multiplying coefficients i.e. taps, in accordance with the symmetry of the differential delay characteristic of the transmission medium around a signal carrier frequency.

Abstract: The dispersive effects of frequency selective fading in a digital, FM, or AM radio system are reduced by means of an adaptive equalizer (11) comprising a cascade of feed-forward stages (1,2, . . . N), each of which includes: a first parallel wavepath (1-1, 1-2, . . . 1-N) including a first adjustable attenuator (20-1, 20-2, . . . 20-N); a second parallel wavepath (2-1, 2-2, . . . 2-N) including a second adjustable attenuator (21-1, 21-2, . . . 21-N) and delay means (22-1, 22-2, . . . 22-N); and means (23-1, 23-2, . . . 23-N) for combining the signals in said wavepaths and for coupling said combined signal to the next stage. By a suitable selection of parameters, according to two unique relationships, a transfer function can be realized which can compensate for amplitude and delay distortions caused by minimum and nonminimum phase fades.

Abstract: In a data receiver (100), sampling circuitry (120, 125) forms samples of a received data signal representing a succession of data symbols. The samples, which are formed at twice the symbol rate, are multiplied by respective ones of a queue of coefficients in a fractionally spaced equalizer (150). Further circuitry (155, 160, 165, 170) forms decisions in response to the resulting products as to the values of the tansmitted symbols and generates error signals. The values of the coefficients are updated in the equalizer in response to the error signals. Timing recovery circuitry (230) within the equalizer periodically identifies the largest of the coefficients and either advances or retards the phase of the sampling circuitry depending on whether that coefficient is or is not within a predetermined portion of the queue. The magnitude of the amount by which the phase is advanced or retarded is determined by the position of the largest coefficient relative to the center of the coefficient queue.

Abstract: An adaptive system in a digital data receiver providing compensation for amplitude and phase distortions introduced by the data transmission channel comprises, at the output of the transmission channel (1) which produces a signal vector X.sub.k, a transversal filter (2) with N weighting coefficients, followed by a decision circuit (3). For the direct determination of the N coefficients, the system comprises a circuit (4) for estimating the autocorrelation matrix A of the vectors X.sub.k, a circuit (5) for approximating the estimated matrix by a circulant matrix R.sup.e a circuit (6) for calculating a vector whose components are the eigenvalues of R.sup.e, a circuit (7) for estimating a vector V which expresses the correlation between the output siganls of the channel (1) and the decision circuit (3), calculating circuits (8) and (9), and a circuit (10) for estimating a vector C.sup.R which represents an approximation of the vector of the N optimum weighting coefficients of the filter.

Abstract: The input video signal including a ghost signal is supplied to a first transversal filter weighted by a first weighting signal. The output of the first transversal filter is supplied to a difference circuit to detect the ghost signal component which is supplied to a second transversal filter weighted by the second weighting signal derived from the input video signal. The output of the second transversal filter is supplied to a circuitry which includes a sign detecting circuit and a tap gain memory to produce the first weighting signal.

Abstract: Equalizer apparatus is provided which includes a number of successively cled lattice stages, each of the stages comprising a selected configuration of adjustable electrical components, the input to the first stage of the successively coupled lattice stages comprising the output of a data transmission channel. The equalizer apparatus further includes a subtractor device coupled to the lattice stages for providing a succession of error terms, each of the error terms comprising an accumulation of the squares of a number of error quantities, each of the error quantities indicating the difference between an undistorted training signal, in a sequence of training signals which are received by the subtractor device, and the same training signal after it has travelled through the data transmission channel and is received by the lattice stages.

Abstract: An adaptive system in a digital data receiver providing compensation for amplitude and phase distortions introduced by the data transmission channel having, at the output of the transmission channel which supplies a signal vector X.sub.k, a transversal filter having N weighting coefficients, followed by a decision circuit and further a summing circuit and a multiplying circuit. To determine the N coefficients iteratively the system comprises a circuit for determining an estimated matrix A.sub.k of the autocorrelation matrix A of the signal vectors X.sub.k, a circuit for approximating A.sub.k by a circulant matrix R.sub.k, a circuit for calculating the diagonal matrix G.sub.k whose diagonal elements are the eigenvalues of R.sub.k, a number of calculating circuits and a circuit for up-dating the vector C.sub.k which represents the N weighting coefficients of the filter at the instant t.sub.o +kT.

Abstract: An automatic equalizer comprises a transversal filter connected to receive an input signal such as a television signal in which a predetermined periodic reference signal is present and having variable tap gains, and a correlator for forming the correlation between the input and output signals of the transversal filter and correcting the tap gains of the transversal filter to cancel out distortion components such as ghost component contained in the input signal. For the purpose of preventing the slow convergence and fluctuation of tap gains of the transversal filter due to non-periodic noise components contained in the input signal, a circuit is further provided to integrate a predetermined portion of the waveform of the reference signal contained in the input signal. An output signal of the waveform integrator circuit is correlated with the output signal of the transversal filter by the correlator.

Abstract: In a device for equalizing multiphase and/or multiamplitude modulated data signals, which device includes a decider connected to receive a demodulated input signal and to produce a demodulated output signal, an equalizer, and a difference forming member connected to produce an output signal representative of the difference between the demodulated decider output and input signals, the difference forming member output signal being supplied to the equalizer as a control value for adaptively setting the equalizer coefficients, a multiplier is connected between the decider and the equalizer for multiplying the signal supplied thereto by a positive scalar value.

Abstract: A recursive automatic equalizer is described for implementing the telephone equalization function at a line circuit which may be multiplexed between a plurality of telephone subscriber sets. A recursive digital filter structure having programmable coefficients minimizes the error between the equalizer input and a reference signal. The recursive filter transfer function is variable via feedback coefficient update, with respect to its input and the reference signal. The recursive filter coefficients are adaptively changed to rapidly converge to a final value based upon a mean square error algorithm. The desired filter transfer function can be achieved with a low number of coefficients, for example, five, rather than the heretofore high number of coefficients required in non-recursive filter structures.

Abstract: Tap coefficient drift in a fractionally spaced equalizer (10) is minimized by adding signal energy (n(t)) to the received data signal (s.sub.r (t)) at frequencies where the sampled channel transfer function has substantially zero gain.

Abstract: Two microwave signals of identical frequency, possibly differing in amplitude and/or phase from each other, are combined by feeding them to respective inputs of a directional coupler with outputs each carrying an unshifted component of one and a quadrature-shifted component of the other input signal. The composite output signals of the coupler are fed to a hybrid junction which combines them cophasally into a sum signal and with relative phase inversion into a difference signal. The sum signal, besides energizing a load, is applied directly to a reference input of a first coherent detector and through a 90.degree. phase shifter to a reference input of a second coherent detector each receiving the difference signal on a monitoring input. The first detector controls a phase shifter upstream of one of the coupler inputs to correct for phase deviations while the second detector controls a phase shifter downstream of one of the coupler outputs to correct for changes in relative amplitude.

Abstract: An automatically adjusted equalizer circuit includes an operational amplifier having a parallel resonant circuit, a voltage controlled resistor, and an amplifier output signal monitoring circuit in a feedback path for the adjustable equalizer. The presence of the parallel resonant circuit in the feedback path causes the amplifier gain to vary with frequency. An input signal to the operational amplifier will be distorted depending upon the particular loss characteristic of a telephone wire to which the input is connected. Automatic adjustment of the voltage controlled resistor, and the presence of the resonant circuit, causes the slope of the gain of the amplifier to inversely compensate for the loss characteristic of the particular line in question. Within the variety of lines, each line will have different loss characteristics as a function of length and gauge of the different telephone lines.

Abstract: An improved equalizer receives the output of a data transmission channel at he first stage of a number of successively coupled lattice stages which are each made up of a selected configuration of adjustable electrical components. A subtractor device is coupled to the lattice stages for providing a succession of error terms. The error terms are an accumulation of the squares of a number of error quantities, which are the differences between a sequence of undistorted training signals or estimates of the transmitted signal which are received by the subtractor device, and the same training signals after they have travelled through the data transmission channel. A delayed stage and two dimensional vector matrices are coupled in each of the lattice stages for iteratively adjusting their components in accordance with a least squares procedure, that is, the components are adjusted each time a signal is received and the iterative adjustment continues until a prespecified error limit is reached.

Abstract: An equalizer corrects cross-polarization distortion which occurs between two series of digital signals which are transmitted on individually associated ones of two mutually orthogonal cross-polarized waves. The distortion is corrected when a circuit discriminates between the two series of digital signals and gives an output signal which represents a value judgment as to the amount of distortion which has occurred. The output signal serves as an address for selecting one of many memory element circuits, each element circuit having a predetermined value stored therein. The value stored in the selected memory element circuit is subtracted from one series of the digital signals. The idea is that, during transmission, something (e.g., raindrops) transfers energy from one to the other of the cross-polarized waves. The equalizer circuit passes judgment as to how much energy has been transferred, and then subtracts that amount of energy in order to return to the original signal.

Abstract: A receiver for a quadrature amplitude modulated data signal impaired by linear and nonlinear distortion, phase jitter and additive noise includes circuitry which compensates for these impairments. In particular, the receiver includes a processor (FIG. 1 44; FIG. 2, 44') which subtracts a feedback nonlinear signal (FIG. 1, D(n); FIG. 2, D'(n)) from each sample of the received signal, either prior or subsequent to demodulation, providing compensation for nonlinear intersymbol interference. The feedback nonlinear signal subtracted from each sample is comprised of a weighted sum of products of individual data decisions and/or the complex conjugates of data decisions, each such product, in turn, being multiplied by a predetermined harmonic of the carrier frequency. In an illustrative embodiment, compensation for second- and third-order intersymbol interference is provided by including two- and three-multiplicand weighted products in the feedback nonlinear signal.