Abstract: An echo cancellation device (ECD) comprises an echo canceller (EC) including a transfer function estimator (EST, H) and a subtractor (ADD) and a residual echo suppression device (G, ADD2). The residual echo suppression device (G) comprises a residual echo filter (G) having an adjustable filter function (g). This filter function (g) can be adapted to either remove from the subtractor output (TNE′) the spectral characteristics relating to the reception signal (RFE) and/or to emphasize in the subtractor output signal (TNE′) a background signal spectral content relating to the transmission signal (TNE). A noise generation means (NGM′) can be provided at the output of the adaptable filter (G) for injecting a noise process in to the filter output signal (TNE′) prior to a speech coding in a speech coder (COD). The noise process masks in the filter output signal a spectral content relating to the reception signal (RFE).

Abstract: An echo canceller having an adaptive algorithm for canceling an echo signal comprises an adaptive filter, a renewal controller and an initial operation controller. The adaptive filter generates a pseudo echo signal by renewing tap coefficients based on a far-end input signal, a near-end input signal and a residual signal, the renewal controller determines a time average of a renewal amount for each tap coefficients and estimates an initial delay of the echo signal by detecting the time average indicating a maximum renewal amount. The initial operation controller operates the adaptive filter to perform an initial operation to eliminate the tap coefficients corresponding to the initial delay of the echo signal and to renew the remaining tap coefficients corresponding to the echo signal.

Abstract: A two-step training method for the estimation filter in the echo cancellation (EC) path of the analog front-end (AFE) circuit for a modem, such as an asymmetric digital subscriber line (ADSL) modem. During the first step, the coefficients of a filter that is preferably equivalent to the combination of the EC path and the receive (RX) path in the AFE circuit are determined with the EC estimation filter held fixed and the transmit (TX) path in the AFE circuit disabled. During the second step, the TX path is enabled and the coefficients of the EC estimation filter are determined using the path equivalent filter generated during the first step. The two-step training method takes into account variations in the operating characteristics of components in the various processing paths in the AFE circuit to converge on satisfactory coefficients for the EC estimation filter within an acceptable number of training symbols.

Abstract: An echo canceller (110) for a full-duplex communication system (100) such as a speakerphone determines a gain factor to be used in sub-band adapter coefficient updates by using both time domain and sub-band signals to determine the state of the system (100). In one embodiment a gain factor controller (120) uses the sampled time domain signal to determine if the system (100) is in the silence state, and if not uses the sub-band signals to determine whether the system (100) is in the talk, double-talk, or listen state. The gain factor controller (120) selectively changes the gain factor and selectively performs coefficient updating based on the state. In another embodiment the gain factor controller (120) determines the state and selectively updates coefficients using state information developed on a band-by-band basis.

Abstract: The magnitude processing circuit 607 non-linearly processes the received signal 2 and then outputs a non-linear processed signal. The delay processing circuit 300 further delays the non-linear processed signal and then outputs a delayed processed signal. The delayed processed signal is time-multiplexed with the non-linear processed signal. The multiplexed signal is used instead of the received signal 2. The adaptive filters 122 and 124 operate alternately with the non-linear processed signal or with the delayed processed signal obtained by non-linearly processing and then delaying the received signal 2 as the input signal. Adaptive filter coefficients can be obtained by using twice the number of conditions compared with the number is twice that for the case where only the received signal 2 is used as the input signal.

Abstract: A communication system includes an echo canceller and double-talk detector. The echo canceller includes an adaptive finite impulse response filter that generates an adaptive filter weight vector. The double talk detector computes a squared norm of the filter weight vector and detects an increase of the squared norm in order to detect double-talk status. Upon detection of double-talk status, operation of the echo canceller is suspended.

Abstract: The Least Mean Square (LMS) and Normalized LMS (NLMS) algorithms commonly employed in adaptive filters of echo cancelers are further optimized. Finite impulse response (FIR) filters are used to estimate a transfer function of an echo channel in a communications link. The LMS and NLMS algorithms are used to adapt the filter coefficients of the estimated transfer functions. By including the echo channel energy gain in the LMS or NLMS update equation, adaptation speed is increased by making adaptation responsive to the channel energy gain. An algorithm for estimating the echo channel energy gain adapts the estimate based on measured system parameters, such as a measured instantaneous channel gain and a near-end voice level. By considering in the NLMS algorithm, the average energy of either the microphone signal or the error signal, as well as the standard reference signal, a higher nominal update gain can be used. With a higher nominal update gain, the NLMS algorithm will converge more quickly.

Abstract: In methods and apparatus for echo cancellation, a noise estimate is selectively added to a downlink communications signals to improve the convergence speed and stability of an echo-canceling adaptive filter. The added signal content enables the echo-canceling adaptive filter to more quickly track echo path changes during user communications and prevents divergence of the adaptive filter during periods in which the downlink signal does not contain information (e.g., speech) sufficient to develop a quality echo path estimate. To prevent performance degradation from a user perspective, the added noise is made to resemble existing system noise (e.g., either near-end or far-end background noise). In exemplary embodiments, filter convergence speed and stability are further improved by whitening the spectrum and increasing the power level of the added noise whenever a near-end user is active and masking the added noise.

Abstract: The invention is a method and apparatus for performing adaptive filtering, and particularly echo cancellation, utilizing an efficient and effective adaptive algorithm. The invention is particularly useful in connection with network echo cancellation but is more broadly applicable to any situation where an adaptive estimate of a signal must be generated in real-time. The invention includes an improved proportionate normalized least mean squares algorithm for generating an impulse response estimate that is useful for generating an echo cancellation signal to be subtracted from the echo containing signal.

Abstract: Methods and apparatus for compensating for the background noise modulation caused by the operation of a non-linear processor on an audio signal include a controller that receives operating status information from the non-linear processor and generates a noise reduction control signal. The methods and apparatus further include a noise reduction circuit that receives the noise reduction control signal from the controller. The noise reduction circuit reduces the background noise in the audio signal when a noise reduction control signal is generated by the controller and injects synthesized background noise into the audio signal when a noise reduction control signal is not generated by the controller.

Abstract: A bidirectional communication device for transmitting and receiving communications signals in a system which presents a transmission path for conducting a communications signal from the device, a reception path for conducting a communications signal to the device, and an echo path which conducts echo signals from the transmission path to the reception path, the device having an echo canceler connected between the transmission path and the reception path for minimizing echo signals in the transmission path, the echo canceler comprising an adaptive filter for filtering the communications signals from the device according to a filter characteristic having a plurality of filter coefficients, wherein a first group of the coefficients, constituting less than all of the coefficients, has finite filter coefficient values and the remaining ones of the plurality of coefficients have values of zero.

Abstract: An enhanced near-end voice signal may be generated in a hands-free environment by receiving an audio signal, generating an estimated acoustic echo signal, and generating a processed signal by removing the estimated acoustic echo signal from the audio signal. A near-end enhanced spectrum is then determined, that has one or more ranges of contiguous frequencies over which the detector spectrum takes on its largest values, wherein the range of contiguous frequencies are those associated with a relatively high echo return loss in the processed signal. The processed signal is filtered in accordance with the near-end enhanced spectrum, thereby generating an enhanced near-end voice signal. The enhanced near-end voice signal may then be applied to any of a number of components that are intended to process near-end speech. For example, when applied to a voice activity detector, the amount of energy contained in the enhanced near-end voice signal is then measured.

Abstract: A hybrid echo suppresser and associated methods of use provide echo suppression in a bi-directional communications system, while minimizing desired-voice signal disruption. The hybrid echo suppresser includes a linear echo canceller (LEC), a non-linear processor (NLP), and a control circuit. The LEC operates on an echo-causing signal to produce an estimated-echo signal, which is subtracted from an input echo-containing signal, thereby producing an echo-cancelled signal. The NLP operates on this echo-cancelled signal to eliminate residual echo arising from system non-linearities. The control circuit repeatedly compares the echo-containing signal and the estimated-echo signal to determine if the echo-containing signal includes a desired-voice component. If not, the control circuit operates the NLP in an echo-voice mode having substantial signal attenuation.

Abstract: A compander circuit (20) for echo cancellation in telecommunications apparatus with an estimator module (21) for estimating an echo coupling and a delay of a line echo and a characteristic module (23) with an associated multiplier (24) for controlling the short-time output level of zout(k) as a function of the short-time input level of zin(k) is characterized in that at all inputs and outputs of the compander circuit (20), a plurality of N communications channels are applied and outputted, respectively, at an N-fold clock rate as compared with a single communications channel; that all memory units in the compander circuit (20) are designed as multiplexer-buffer-demultiplexer (MBD) units (30); and that a modulo-N counter (26) is provided which controls all MBD units (30) so that in each of the MBD units (30), always the same memory cell (351, 352, 35N) assigned to a particular communications channel is addressed.

Abstract: Full acoustic coupling and echo cancellation is provided in implementations where there is acoustic coupling between acoustic zones.

Abstract: In handsfree communications devices including a linear echo or noise canceler, both analog and digital gain factors are adjusted in response to user volume control inputs, the analog gain factor being frozen during active communications (e.g., during telephone calls). By providing dual user controls, one of which is disabled when a call is in progress, embodiments of the invention provide the wide dynamic range required in most handsfree applications without degrading performance of the linear echo canceler.

Abstract: An equipment (1) for the suppression of whole echoes composed of several partial echoes in telecommunications (TC) equipment, such as end devices, transmission systems or exchange equipment, with the help of adaptive FIR (=Finite Impulse Response) filters, which reproduce the whole echo and extract it from the echo-laden useful signal and which has been transferred to the respective TC equipment, in which the homogeneous FIR filters or partial filters (10), each of which are able to reproduce a whole echo or partial echo, are implemented on a single ASIC (=Application Specific Integrated Circuit) (9) for the m parallel and independent TC channels n to be served; and in which the whole echoes or partial echoes to be reproduced can be calculated in real time in the ASIC (9) and in which a digital signal processor (4) is designed to control the ASIC (9) and set the filter settings, especially after the coefficient required to reproduce the echo is calculated on the ASIC (9).

Abstract: To be able to successfully compensate the echos occurring in bidirectional telecommunications systems, comprising for example 2-wire/4-wire converters and/or AD/DA converters and/or satellite transmission links, the echo coupling factor and echo delay time for a system must be determined with a very high degree of accuracy and independently of the double talk situation. For this purpose, in accordance with the invention a process is provided wherein, by analyzing a dialogue, states and time segments are determined in which the probability of an incorrect evaluation of the coupling factor between the transmission paths and the echo delay time is very low. To improve the measurement accuracy, auxiliary quantities dependent upon the speech dynamics and the characteristic curve of the speech signal are introduced.

Abstract: An echo canceling apparatus suitable for multiplexed transmission lines is provided that achieves a short convergence time and a small number of arithmetic operations. Adaptive filters are located in respective transmission lines. Each adaptive filter adaptively controls tap positions such that tap coefficients are located in dispersive regions which are obtained by removing fixed delays from the impulse response in an echo path. A controller receives tap position information from the adaptive filters to mutually compare the convergence degrees among the adaptive filters. Thus, the controller respectively supplies coefficient update control signals to the adaptive filters so as to suppress coefficient adaptation of an adaptive filter whose convergence is leading compared with the others.

Abstract: Residual echo components resulting from an adaptive echo canceler are minimized by employing a subband echo suppressor. The residual echo components are minimized by individually, controllably adjusting the echo suppressor attenuation level in each of the subbands to gradually attenuate echo peaks having a magnitude lower than a clipping threshold with the magnitude of the attenuation increasing for residual echoes of decreasing magnitude. Additionally, the threshold of suppression engagement is dynamically matched to the signal gain in the echo path for each subband. If the signal gain in a subband is high (strong echo), the suppressor must attenuate residual echo peaks in the subband of higher energy to maintain a constant quality of service.

Abstract: An object of the present invention is to provide an echo canceller capable of generating, when the double-talk detection precision is degraded and the learning of a filter coefficient swings by noises, an echo replica signal using a filter coefficient having a low degree of error at the time of detecting a double-talk state. A backup filter coefficient is calculated by adding a predetermined weighting factor times a filter coefficient obtained by the learning and a predetermined weighting factor times the preceding backup filter coefficient stored in second storage means.

Abstract: A system and method for detecting and controlling severe echo. The system includes an echo canceller having an adaptive filter responsive to a set of coefficients, a first memory storing a software program operative to evaluate these coefficients to determine the severity of echo existing in the system, and a second memory accessible by the echo canceller and the first memory which stores the coefficients. Upon evaluating the coefficients, the system and method enables the proper mode of echo cancellation in the echo canceller.

Abstract: In accordance with the invention, from the characteristic curve of the speech signal there are derived pulse sequences which are dependent upon the frequency but not the amplitude of the speech signal, so that in the case of a real echo, independently of the attenuation, identical pulse sequences occur in the case of the transmitted signal and, with a time shift, in the case of a received echo. The correspondence between the pulse sequences of transmitted signal and received signal is an indication that the received signal is an echo of the transmitted signal.

Abstract: A telephone system including a base station and a mobile unit in communication with each other is provided with an acoustic echo canceler. A sound signal is input. A reference signal that matches the sound signal is formed. A plurality of encoders and decoders receive the reference signal and form a modified reference signal. The plurality of encoders and decoders match a set of encoders and decoders that operate on the sound signal in the base station and the mobile unit. An adaptive filter receives the modified reference signal and forms a replica signal. A subtractor subtracts the replica signal from an echo signal received by the base station.

Abstract: An adaptive filter (and a corresponding adaptive filtering method) for use in an echo cancellation environment, comprising a finite impulse response filter having a vector of adaptive coefficients, and an adaptive coefficient vector update module which modifies the vector of adaptive coefficients in response to an error signal and based on a predetermined probability density function for said error signal. In accordance with several illustrative embodiments of the present invention, the probability density function is advantageously non-Gaussian, and has certain advantageous characteristics which provide for robustness in the adaptive algorithm based thereupon.

Abstract: A system and method for controlling echo using zero echo path, ringing, and off-hook detection. When conditions for zero echo path are detected, the present embodiment bypasses echo cancellation by replacing the output signal with the input to the echo canceller. In addition, the present embodiment also provides ringing and off-hook detection which additionally evaluate when the output signal should be replaced by the input to the echo canceller. The system and method provides additional detection and control benefits to echo cancellers and can be implemented in echo cancellation systems without substantial change to the existing echo canceller.