Abstract: In the MPEG2 Advanced Audio Coder (AAC) standard, Temporal Noise Shaping (TNS) is currently implemented by defining one filter for a given frequency band, and then switching to another filter for the adjacent frequency band when the signal structure in the adjacent band is different than the one in the previous band. The AAC standard limits the number of filters used to either one filter for a “short” block or three filters for a “long” block. In cases where the need for additional filters is present but the limit of permissible filters has been reached, the remaining frequency spectra are simply not covered by TNS. This current practice is not an effective way of deploying TNS filters for most audio signals. We propose two solutions to deploy TNS filters in order to get the entire spectrum of the signal into TNS. The first method involves a filter bridging technique and complies with the current AAC standard. The second method involves a filter clustering technique.

Abstract: An eye opening measurement technique, that does not interrupt a receiver's normal operation, is used as a metric for optimizing any selected parameters of the receiver's operation. If eye opening size decreases, as a result of a change to a receiver parameter, the polarity for stepwise changes is reversed such that the next change is in the opposite direction. Other types of search procedures can be used. Eye opening size is the difference between the eye's upper and lower edges. Measurement of eye opening size is accomplished using a data and auxiliary slicer that find each “edge” of an eye opening based upon the slicers' level of agreement. Depending upon the level of agreement, and whether symbols of the upper or lower region of the eye are counted, the threshold of the auxiliary slicer can be adjusted in the direction necessary to converge on the eye edge sought.

Abstract: A method and apparatus for performing channel equalization in a communication system may potentially reduce power consumption in channel equalizers of communication systems. A filtering circuit filters a received data sequence as a plurality of data values to be stored in a plurality of filter cells. Each filter cell may store at least one data value and may contain a coefficient related to the stored data value. A coefficient updating circuit may update the coefficients based on at least one parameter, and may compare the updated coefficients to a threshold. Based on the comparison, filter cells of selected coefficients may be selected for restoring the received data sequence to its original state.

Abstract: The present invention is related to a method and apparatus for enhancing processing speed for performing a least mean square operation by parallel processing. The input data are subdivided into a plurality of groups and processed by a plurality of adaptive filters in parallel. A Jaber product device is utilized for rearranging the processing results. A subtractor subtracts the output from the Jaber product device from a desired result to generate an error signal. A feedback network adjusts the adaptive filters in accordance with the error signal.

Abstract: An efficient configurable signal filter. The filter includes a first mechanism for receiving a first signal of a first type and a second signal of a second type. A second mechanism selectively filters the first signal during a first mode of operation, and filters the second signal during a second mode of operation. A third mechanism generates control signals. A fourth mechanism automatically configures the second mechanism to operate in the first mode of operation or the second mode of operation based on the control signals. In a specific embodiment, the first type of signal is characterized by a first rate, and the second type of signal is characterized by a second rate. The first signal and the second signal are digital ADC outputs. The second mechanism includes plural filter blocks, each having one or more Multiply-Accumulate (MAC) pipes. Each of the one or more MAC pipes include one or more MAC blocks that are each associated with a coefficient memory data structure of a coefficient memory.

Abstract: In active noise reduction, at least one input signal (25) is fed to a computing unit (18), which passes on the at least one input signal (25) to at least one additional computing unit (19), wherein the at least one input signal (25) is processed for the generation of at least one output signal (26) in the at least one additional computing unit (19). Therein, a kind of processing for processing in the additional computing unit (19) is set by the computing unit (18). Finally, the generated at least one output signal (26) is fed to the computing unit (18). Furthermore, apparatuses for carrying out the method are disclosed.

Abstract: An interpolation filter without a FIFO memory is configured as a cascade arrangement of simpler interpolation sub-filters that are operated in reverse order. The interpolation sub-filter that produces the highest sampling frequency is operated first, followed by interpolation sub-filters that operate at successively lower sampling frequencies. Computational independence of the cascaded sub-filters is guaranteed by adding delays to sampled and filtered signals. Delays are implemented by operating each of the cascaded sub-filters using prior filtering results that are computed during a previous sampling interval. A small increment to random-access memory is required for storing the successively delayed signals. The digital signal processor performing the filtering process is stalled for one clock cycle at the time a filtered signal sample is outputted so that the outputted signal sample can be produced without a timing conflict.

Abstract: Presented herein is a system and apparatus for accelerating arithmetic decoding of encoded data. In one embodiment, there is presented a symbol interpreter for decoding CABAC coded data. The symbol interpreter comprises a first memory, a CABAC decoding loop, and a syntax assembler. The first memory receives a bitstream comprising the CABAC coded data at a channel rate. The CABAC decoding loop decodes the CABAC symbols at the channel rate, and comprises an arithmetic decoder for generating binary symbols from the CABAC coded data at the channel rate. The syntax assembler decodes the binary symbols at a consumption rate.

Abstract: The present invention provides an apparatus for channel equalization and method thereof, by which a digital signal received via a plurality of antennas is equalized. The present invention includes receiving digital transmission signals using a plurality of antennas, respectively, initializing equalizer coefficients and equalizing the received signals respectively, adding the equalized signals together, predicting a noise amplified in the equalizing step, and generating a final output signal by removing the predicted noise from a value resulting from adding the equalized signals together. Therefore, the present invention performs equalization using a plurality of antennas, thereby enhancing the signal to noise ratio of the final output and facilitating the equalization of the severely distorted channel.

Abstract: Systems, methods and apparatus are provided through which in some embodiments of recursive hierarchical segmentation of data with any number of spatial dimensions. Some embodiments of the recursive hierarchical segmentation include computationally efficient parallel implementations and other embodiments of the recursive hierarchical segmentation include computationally efficient serial implementations.

Abstract: Computing optimal Linear Equalizer (LE) coefficients gopt from a channel estimate h. A channel impulse response h is first estimated based upon either a known training sequence or an unknown sequence. The channel estimate is formulated as a convolution matrix H. The convolution matrix H is then related to the LE coefficients in a matrix format equation, the matrix format equation based upon the structure of the LE, the convolution matrix, and an expected output of the LE. A Fast Transversal Filter (FTF) algorithm is then used to formulate a recursive least squares solution to the matrix format equation. Computing the recursive least squares solution yields the LE coefficients using structured equations.

Abstract: A signal processing device for processing a passband signal to generate an equalized signal includes a passband adaptive equalizer for generating the equalized signal according to the passband signal, including at least one feed-forward equalizer (FFE) and one feedback equalizer (FBE), and a multilevel quantizer coupled with the passband adaptive equalizer for selectively utilizing a single predetermined threshold or a plurality of multiple predetermined thresholds to quantize the equalized signal in order to generate a sliced signal.

Abstract: A FIR filter for use in an adaptive multi-channel filtering system, includes a first memory for storing data, and a second memory for storing filter coefficients. The second memory stores only non-zero valued coefficients or coefficients that are above a predetermined magnitude threshold such that the overall number of coefficients processed is significantly reduced.

Abstract: A correlation device is provided that includes an adder for adding an input signal sequence and an auxiliary signal sequence to obtain an addition signal sequence, and a delay element for delaying the addition signal sequence to obtain the auxiliary signal sequence, whereby the delay element has a plurality of coefficient outputs for providing addition signal sequence coefficients. The correlation device comprises further a linking element for the coefficient-wise linking of an addition signal sequence coefficient with a linking coefficient to obtain a correlation result.

Abstract: An adaptive noise reduction method and apparatus capable of reducing efficiently variable period noise from a main input is provided. The pitch of a noise waveform to be reduced can be variable with a change in a period of motor noise occurring when the revolution period is changed by disc motor control of DVD-RAM, by revolution speed control of other motors, and revolution period change on starting a motor and the like. Therefore, the renewal of adaptive filter coefficient becomes almost unnecessary, thus allowing noise reduction to be performed without degrading a noise canceling effect.

Abstract: An adaptive filter device, including a finite impulse response (FIR) filter which is based on filter coefficients, which are determined based on a predetermined iterative adaptation algorithm for determining filter coefficients of an adaptive filter, wherein, in at least one iteration step of said predetermined iterative adaptation algorithm a sum value is determined, wherein each summand of said sum value depends on one of said filter coefficients, and, if said sum value is above a predetermined threshold, the filter coefficients are modified.

Abstract: A reduced-complexity, fast converging adaptive filter may be used for network echo cancellation applications, including applications having sparse echo paths. The new filter, referred to as a selective-partial-update proportionate NLMS filter, may be based on a proportionate NLMS (PNLMS) technique and selective partial updating of the adaptive filter coefficients. The new PNLMS filter may exploit sparseness of a communications channel to speed up the initial convergence of the NLMS technique included in the filter by weighting regressor data proportionately with an estimated magnitude of the channel impulse response. Selective partial updating is essentially a data selection method to reduce the computational complexity. The performance of the selective-partial-update PNLMS filter compares favorably to an adaptive filter using standard PNLMS for echo paths specified in ITU-T Recommendation G.168.

Abstract: The accumulated change in values representative of actions taken by a processor, such as the number of email messages processed by an email server, in a given time period is determined. Actions are represented as data points on a plot. Look-ahead intervals are defined for each point. Candidate pairs of points are determined for each look-ahead interval by comparing the first value in the look-ahead interval with other values in the look-ahead interval. A candidate pair comprises the first point and another point having a lesser value. If a candidate pair has a value therebetween, the candidate pair is discarded. If, however, a candidate pair has no value therebetween, the first value of the candidate pair is a peak value for the look-ahead interval. The accumulated change is determined by calculating the sum of the peak values, plus the final value, minus the initial value, for the given time period.

Abstract: Signal processing systems and methods are described that include a multi-analog receiver front end with adaptive filtering. The multi-analog receiver front end uses two or more analog-to-digital converters (ADCs) to remove additive electrical noise present in the analog front end. The multiple ADCs are followed in the signal processing path by digital statistical signal processing. The multi-analog receiver front end adaptively determines the passband of a digital filter in a system with input signals having a wide frequency range of interest, and controls filtering of the input signals to the narrow frequency range that includes an input signal. The multi-analog receiver front end, through removal of additive noise, provides higher signal-to-noise ratios for a given power dissipation and chip area when compared to receiver front ends which do not use the multiple ADCs.

Abstract: An adaptive signal equalizer with a feedforward filter in which the feedback error signal and corresponding incoming data signal are dynamically aligned in time using signal interpolation, and further, to control the precursor/postcursor filter taps configuration, thereby producing more adaptive filter tap coefficient signals for significantly improved and robust signal equalization.

Abstract: A continuous time electronic dispersion compensation architecture using feed forward equalization and a non-linear decision feedback equalization forms an output signal by a linear combination of successively delayed versions of the input signal and the sliced output signal weighted by appropriate coefficients. A selected number of taps in the mixer used to generate a corresponding number of coefficients for use in the feed forward equalizer are held to a selected voltage to ensure that the coefficients associated with these two taps do not drift. This causes the other coefficients to converge to a unique minimum square error value. In one embodiment the selected voltage is the maximum system voltage.

Abstract: Adaptively analyzing an observed signal to estimate that part of the signal that best corresponds to a steering vector. Modifying the steering vector by convolution of the steering vector with a vector estimating the effect of multipath on the observed signal.

Abstract: A method, system, and computer program are provided for filtering a signal. The method, system, and computer program receive a sample of a signal being filtered and identify a bias associated with the sample. The bias includes a cushion and an increment. The method, system, and computer program also output an expected value for the sample of the signal being filtered combined with a portion of the bias. The portion of the bias is based at least partially on a size of the cushion.

Abstract: An equalizer is provided which is capable of making a filter factor to be set in the equalizer having an equalizing filter converge rapidly and a method is provided for setting an initial value for the rapid convergence of the filter factor. In the equalizer having a filter factor computing device to compute a filter factor for an equalizing filter, and a differential detecting circuit to generate a differential signal between a signal output from the equalizing filter and a common pilot diffusing code, an initial value for the filter factor computing device is generated and set by a multipath timing detecting circuit, a reverse diffusing section, and a channel estimating device being operated based on a received signal.

Abstract: An eye opening measurement technique, that does not interrupt a receiver's normal operation, is used as a metric for optimizing any selected parameters of the receiver's operation. If eye opening size decreases, as a result of a change to a receiver parameter, the polarity for stepwise changes is reversed such that the next change is in the opposite direction. Other types of search procedures can be used. Eye opening size is the difference between the eye's upper and lower edges. Measurement of eye opening size is accomplished using a data and auxiliary slicer that find each “edge” of an eye opening based upon the slicers' level of agreement. Depending upon the level of agreement, and whether symbols of the upper or lower region of the eye are counted, the threshold of the auxiliary slicer can be adjusted in the direction necessary to converge on the eye edge sought.

Abstract: In order to attain an optimally compressed, narrow pulse peak at the filter output of a correlation filter for the purpose of reception, the interfering secondary maxima of the autocorrelation function of binary codes must be as small as possible. The invention uses specially designed signal codes which are used to generate the associated complementary signal code from the received sequence by means of evaluation in the reception filter. The subsequent parallel formation of the autocorrelation functions of the received signal code and the complementary signal code exhibits secondary maxima having an opposite mathematical sign, thus resulting in the desired prefect pulse peak having secondary maxima which are equal to zero during summation at the filter output.

Abstract: A Decision Feedback Equalizer (DFE) system includes a DFE and a DFE coefficients processor. The DFE receives an uncompensated signal and operates upon the uncompensated input using DFE coefficients to produce an equalized output. The DFE coefficients processor formulates a channel estimate as a convolution matrix H. The DFE coefficients processor determines a Feed Back Equalizer (FBE) energy constraint based upon the channel estimate. The DFE coefficients processor relates the convolution matrix H to the DFE coefficients in a matrix format equation, the matrix format equation based upon the structure of the DFE, the convolution matrix, an expected output of the DFE, and the FBE energy constraint. The DFE coefficients processor formulates a recursive least squares solution to the matrix format equation and computes the recursive least squares solution to the matrix format equation to yield the DFE coefficients.

Abstract: Directly computing Feed Forward Equalizer (FFE) coefficients and Feed Back Equalizer (FBE) coefficients of a Decision Feedback Equalizer (DFE) from a channel estimate. The FBE coefficients have an energy constraint. A recursive least squares problem is formulated based upon the DFE configuration, the channel estimate, and the FBE energy constraint. The recursive least squares problem is solved to yield the FFE coefficients. The FFE coefficients are convolved with a convolution matrix that is based upon the channel estimate to yield the FBE coefficients. A solution to the recursive least squares problem is interpreted as a Kalman gain vector. A Kalman gain vector solution to the recursive least squares problem may be determined using a Fast Transversal Filter (FTF) algorithm.

Abstract: A system and a method are disclosed for converting analog signal to digital. An analog to digital converter system comprises an analog to digital converter (ADC) and an adaptive filter coupled to the ADC, wherein the adaptive filter is coupled to the ADC in a feedforward configuration. A method of generating a digital signal based on an analog signal comprises applying the analog signal to an analog to digital converter (ADC) to generate an ADC output and sending the ADC output to an adaptive filter, wherein the adaptive filter is coupled to the ADC in a feedforward configuration.

Abstract: A method for effecting aliasing cancellation in an audio effects algorithm using a delay modulated signal, derived from interpolation of a delay modulator at an instantaneous sampling frequency, including: determining the instantaneous sampling frequency 1/Tisf and band limiting an input signal, to which the audio effects algorithm is to be applied to ½ Tisf prior to interpolation.

Abstract: A method of signal processing comprises receiving an unknown input signal that includes a distorted component and an undistorted component, the unknown input signal having a sampling rate of R; and performing self-linearization based at least in part on the unknown signal to obtain an output signal that is substantially undistorted, including by generating a replica distortion signal that is substantially similar to the distorted component, the generation being based at least in part on a target component having a sampling rate of R/L, L being an integer greater than 1.

Abstract: Operation of an adaptive filter includes determining a first filter coefficient and determining a second filter coefficient based on the first filter coefficient, in one embodiment, and in another embodiment includes comparing pseudo-error counts generated for each of several time intervals and setting filter coefficients based on the comparison. Adaptive filter coefficient generating apparatus includes a pseudo-error count generator adapted to receive the output of the adaptive filter and to generate counts of pseudo-errors occurring during successive time intervals, and a filter coefficient generator generating filter coefficients for the adaptive filter in response to the pseudo-error counts.

Abstract: A digital filtering apparatus and method for digitally filtering out undesirable or invalid data from data signal lines. The digital filtering apparatus includes a digital delay element having one or more outputs, a comparator connected to the outputs of the digital delay element, and a final stage connected to the output of the comparator and the outputs of the digital delay element. The digital filtering apparatus recognizes and filters out invalid data from data received by the digital delay element, and allows valid data to pass through the filter. Data is considered invalid data if its logical data state transition has a duration less than the clock setting of the digital filtering apparatus. The clock setting can be established by the number of active delay components in the digital delay element. The inventive digital filtering apparatus represents an improvement over conventional analog filters, e.g., in manufacturing efficiency and filtering performance.

Abstract: An active noise control system is provided which cancels a noise using a secondary noise from a speaker that is operated in accordance with an output from an adaptive controller. The system is configured such that microphone monitor interrupts a switch to thereby stop the secondary noise from being produced, when an error signal delivered by a microphone used for an adaptive computation in an LMS processing portion has the same sign for a predetermined duration. This allows the system to prevent the user from hearing an abnormal acoustic noise resulting from an abnormal operation or divergence of an adaptive filter even when the output signal from the microphone used for the adaptive computation is indicative of an abnormal level.

Abstract: A digital filter is disclosed, the filter comprising a device for determining the initial condition of a partial filter input, using a partitioned filter input signal, and partitioned filter input coefficients and a device for determining the initial condition of a partial filter output, using a partitioned filter output signal, and partitioned filter output coefficients, and a device for realising said digital filter as a sum of outputs of Finite Impulse Response (“FIR”) filter elements, wherein inputs of the FIR filter elements are dependent upon said partial filter input initial condition and said partial filter output initial condition, and said current block of filter input signal values.

Abstract: An adaptive equalizer finite impulse response (FIR) filter for high-speed communication channels with modest complexity, where the filter is iteratively updated during a training sequence by a circuit performing the update: h(t+1)= h(t)+?[sgn{d(t)}?sgn{z(t)?Kd(t)}]sgn{ x(t)}, where h(t) is the filter vector representing the filter taps of the FIR filter, x(t) is the data vector representing present and past samples of the received data x(t), d(t) is the desired data used for training, z(t) is the output of the FIR filter, ? determines the memory or window size of the adaptation, and K is a scale factor taking into account practical limitations of the communication channel, receiver, and equalizer. Furthermore, a procedure and circuit structure is provided for calibrating the scale factor K.

Abstract: The invention is a method and apparatus for frequency-domain adaptive filtering that has broad applications such as to equalizers, but is particularly suitable for use in acoustic echo cancellation circuits for stereophonic and other multiple channel teleconferencing systems. The method and apparatus utilizes a frequency-domain recursive least squares criterion that minimizes the error signal in the frequency-domain. In order to reduce the complexity of the algorithm, a constraint is removed resulting in an unconstrained frequency-domain recursive least mean squares method and apparatus. A method and apparatus for selecting an optimal adaptation step for the UFLMS is disclosed. The method and apparatus is generalized to the multiple channel case and exploits the cross-power spectra among all of the channels.

Abstract: An equalizing apparatus includes an equalizer which has a plurality of adjustable tap weights that equalizes a received signal based on values of the adjustable tap weights, a tap weight update calculation unit coupled to the equalizer and which determines tap weight updates for use in adjusting the tap weights during operation of the equalizer, an offset memory that stores one or more tap weight update offset values and a summer coupled to the tap weight update calculation unit and to the offset memory. The summer combines each of the tap weight updates with one of the tap weight update offset values to produce modified tap weight updates which, in turn, are provided to the equalizer to adjust the tap weights.

Abstract: In an exemplary embodiment, a multi-rate data conversion circuit receives digital data at varying data rates, receives a data rate input corresponding to the digital data and converts the digital data to a converted output based upon the data rate input. A direct digital synthesis circuit receives the converted output and synthesizes a modulated output signal based upon the converted output. A multi-rate converter receives the digital data, the data rate input and a clock signal and converts the digital data to converted digital data. A multi-rate digital data filter receives the converted digital data and produces a filtered digital output. An output scaler receives the filtered digital output and produces a scaled and filtered digital output. Finally, an adder combines the scaled and filtered digital output with a center frequency input and produces the converted output.

Abstract: An adaptive filter is implemented by a computer (10) processing an input signal using a recursive least squares lattice (RLSL) algorithm (12) to obtain forward and backward least squares prediction residuals. A prediction residual is the difference between a data element in a sequence of elements and a prediction of that element from other sequence elements. Forward and backward residuals are converted at (14) to interpolation residuals which are unnormalized Kalman gain vector coefficients. Interpolation residuals are normalized to produce the Kalman gain vector at (16). The Kalman gain vector is combined at (18) with input and reference signals x(t) and y(t), which provides updates for the filter coefficients or weights to reflect these signals as required to provide adaptive filtering.

Abstract: A flexible adaptation engine includes a coefficient adaptation circuit that implements multiple adaptation algorithms, and/or multiple coefficient selection algorithms, to adapt the filter coefficients of one or more digital filters, such as the transversal filters of a receiver. In one embodiment, a controller selects the filter coefficients to be adapted, and the adaptation algorithm(s) to be used to adapt the selected coefficients, based on various criteria such as convergence status data, clock recovery status signals, the current load on a processor that adapts the coefficients, and/or manual control signals. In one embodiment, the architecture supports the ability to vary the number of coefficients that are updated at a time, and to concurrently apply different adaptation algorithms to different subsets of filter coefficients. The flexible adaptation engine may be implemented in application-specific hardware and/or as a processor that executes software.

Abstract: A method for processing a signal includes receiving a signal from a channel at a channel speed and providing error adjustment to the signal. The method includes sampling the signal at a speed less than the channel speed to yield sampled scalar data. The signal is sampled at a first phase. The method includes determining a sampled scalar error associated with the signal at a second phase. The difference between the first phase and the second phase comprises a delay. The method also includes forming a cross-correlation vector from the sampled scalar error and the sampled scalar data at a vector speed. The vector speed is less than the speed at which the signal is sampled. The method also includes determining compensation information for the error adjustment. The compensation information is based on the cross-correlation vector, and the compensation information is determined at the vector speed.

Abstract: In some embodiments, an adaptive filter employs two adaptation modes, where during one adaptation mode the adaptive filter is updated only when the received training sample is a first binary value and during the other adaptation mode the adaptive filter is updated only when the received sample is a second binary value. Each adaptation mode provides a set of filter weights, and these two sets of filter weights are averaged to provide an adapted set of filter weights. The use of two adaptation mode allows for a clock boundary in which the digital portion of the filter operates at a lower clock rate than the analog portion. In other embodiments, a filter architecture is described for providing the algebraic signs of the received data samples, important for sign-sign least means square filtering algorithms. In other embodiments, a filter architecture is described in which efficient use is made of voltage-to-current converters so as to achieve a high throughput rate during filtering.

Abstract: An inverse filtering method, comprising: generating a first filtered signal based on an input signal; and combining the first filtered signal with the input signal for obtaining a residual signal. The generating comprises: generating at least two second filtered signals, each of said second filtered signals not significantly delayed in time relative to each other, the generating being stable and causal; and amplifying at least one of the second filtered signals with a prediction coefficient.

Abstract: Closed loop wireless communication of signals using an adaptive transmit antenna array, in which multiple copies of signals to be transmitted by the transmit antenna array are produced with delays and weights that are functions of the multi-path transmission channel characteristics from the transmit antenna array to a receive antenna array of a receiver and are combined before transmission by the transmit antenna array. The delays and weights of the transmit copies for each transmit antenna element are functions of the respective multi-path transmission channel characteristics from that transmit antenna element to the receive antenna array such that the multi-path signal components propagated to each receiver element are received with distinguishable delays according to the propagation path. The receiver combines the received signal components from each receive antenna element with delays and weights that are respective functions of the multi-path transmission channels.

Abstract: Subtraction of a signal 111 from a pulse response 110, where signal 111 provides a good approximation of the tail of pulse response 110, can provide a method for canceling the tail of pulse response 110. For continuous data streams, signals X(t), 223 and Y(t) can correspond to, respectively, signals 110, 111 and 112. X(t) differs from 110 in being a continuous data stream. 223 differs from 111 in being the low pass filtering of X(t), such low pass filtering accomplished by a unit LPF 211. Y(t) differs from 112 in being a continuous stream of equalized data, produced by subtracting the signal at 223 from X(t). A measurement unit 213, analysis unit 214 and decision unit 215 can act to continuously adapt LPF 211 such that tail cancellation equalization is continuously achieved. Measurement unit 213 can construct, from Y(t), a set of correlation measurements that can be used to adapt LPF 211.

Abstract: This invention describes an apparatus and method to improve the performance of a decision feedback equalizer (DFE) for time-varying multi-path channels. For minimum-phase channels, the equalization is performed in a time-forward manner. For maximum-phase channels, the equalization is performed in a time-reversed manner. More specifically, for maximum-phase channels, the filter coefficients are computed based on the channel estimates reversed in time, and the filtering and equalization operations are performed with the received block of symbols in a time-reversed order. In the context of this invention, the term “minimum-phase channel” implies that the energy of the leading part of the channel profile is greater than the energy of the trailing part. The term “maximum-phase channel” implies that the energy of the leading part of the channel profile is less than the energy of the trailing part.

Abstract: An adaptive filter, employing an adaptively controlled forgetting factor, includes a first circuit for determining a gradient of an error signal and a second circuit for determining a value of the forgetting factor. The value of the forgetting factor is updated based on comparing the gradient of the error signal to the forgetting factor. Equations to update the forgetting factor may be solved using a recursive least squares algorithm. Comparing the gradient of the error signal to the forgetting factor may include, for example, dividing the error signal gradient by a compliment of the forgetting factor. A method for updating the forgetting factor includes determining the gradient of the error signal, determining a value of the forgetting factor, and updating the value of the forgetting factor based on comparing the gradient of the error signal to the forgetting factor. The forgetting factor may be stored in the adaptive filter.

Abstract: A system and method facilitating signal enhancement utilizing an adaptive filter is provided. The invention includes an adaptive filter that filters an input based upon a plurality of adaptive coefficients and modifies the adaptive coefficients based on a feedback output. A feedback component provides the feedback output based, at least in part, upon a non-linear function of the acoustic reverberation reduced output. Optionally, the system can further include a linear prediction (LP) analyzer and/or a LP synthesis filter. The system can enhance signal(s), for example, to improve the quality of speech that is acquired by a microphone by reducing reverberation. The system utilizes, at least in part, the principle that certain characteristics of reverberated speech are measurably different from corresponding characteristics of clean speech. The system can employ a filter technology (e.g., reverberation reducing) based on a non-linear function, for example, the kurtosis metric.

Abstract: A channel equalizing and carrier recovery system for home phoneline networking alliance (HomePNA) receiver and method thereof are provided. The channel equalizing system includes a frequency diverse quadrature amplitude modulation (FD-QAM) equalizer and a quadrature amplitude modulation (QAM) equalizer. The FD-QAM equalizer receives an FD-QAM signal, determines FD-QAM tap coefficients, and equalizes the FD-QAM signal using the FD-QAM tap coefficients. The QAM equalizer receives a QAM signal, determines QAM tap coefficients, and equalizes the QAM signal using the QAM tap coefficients. The QAM equalizer receives the FD-QAM signal and determines the QAM tap coefficients during a predetermined header period. The carrier recovery circuit includes a phase detector, a loop filter, and a numerically controlled oscillator (NCO).