Geert Rombouts

Adaptive feedback cancellation in hearing aids

In general feedback cancellation setups, standard adaptive filtering techniques fail to provide a reliable feedback path estimate if the desired signal is spectrally colored because of the presence of a closed signal loop. In this paper, several approaches for improving the estimation accuracy of the adaptive feedback canceller in hearing aids will be reviewed, including constrained adaptation and bandlimited adaptation of the feedback canceller as well as adaptation with the prediction-error method (PEM) using a fixed or adaptive model of the desired signal. Partitioned-block frequency-domain implementations of these algorithms will be compared for acoustic feedback paths measured in two commercial behind-the-ear hearing aids. In addition, it is shown that the tracking performance of the PEM-based feedback canceller with adaptive signal model can be improved by the so-called shadow filter approach known from echo cancellation.

Double-Talk-Robust Prediction Error Identification Algorithms for Acoustic Echo Cancellation

The performance of an acoustic echo canceller may be severely degraded by the presence of a near-end signal. In such a double-talk situation, the variance of the echo path estimate typically increases, resulting in slow convergence or even divergence of the adaptive filter. This problem is usually tackled by equipping the echo canceller with a double-talk detector that freezes adaptation during near-end activity. Nevertheless, there is a need for more robust adaptive algorithms since the adaptive filters convergence may be affected considerably in the time interval needed to detect double-talk. Moreover, in some applications, near-end noise may be continuously present and then the use of a double-talk detector becomes futile. Robustness to double-talk may be established by taking into account the near-end signal characteristics, which are, however, unknown and time varying. In this paper, we show how concurrent estimation of the echo path and an autoregressive near-end signal model can be performed using prediction error (PE) identification techniques. We develop a general recursive prediction error (RPE) identification algorithm and compare it to three existing algorithms from adaptive feedback cancellation. The potential benefit of the algorithms in a double-talk situation is illustrated by means of computer simulations. It appears that especially in the stochastic gradient case a huge improvement in convergence behavior can be obtained

Optimally regularized adaptive filtering algorithms for room acoustic signal enhancement

In many room acoustic signal processing applications, a room impulse response identification is needed to eliminate undesired effects such as echo, feedback, or reverberation. This is typically done using an adaptive filter driven by a speech or audio input signal. However, such signals exhibit poor excitation properties, which cause standard adaptive filtering algorithms to be very sensitive to disturbing signals, especially in the underdetermined case. A popular remedy is regularization, which is usually implemented with a scaled identity regularization matrix. This type of regularization is governed by a single regularization parameter, the value of which is often chosen in an arbitrary way. We propose to regularize the adaptive filter using a non-identity regularization matrix, in which prior knowledge on the unknown room impulse response may be incorporated. When knowledge of the disturbing signal is also used to add prefiltering and weighting in the adaptation, a new family of regularized adaptive filtering algorithms is obtained, which is shown to be optimal in a mean square error sense. Existing regularized algorithms can then be obtained as special cases, assuming limited or no prior knowledge is available. When combined with a recently proposed method of extracting prior knowledge from the acoustic setup, our algorithms exhibit superior convergence behaviour compared to existing algorithms in different simulation scenarios, while the additional computational cost is small.

Generalized sidelobe canceller based combined acoustic feedback- and noise cancellation

We propose a combination of the well-known generalized sidelobe canceller (GSC) or Griffiths-Jim beamformer, and the so-called PEM-AFROW algorithm for joint estimation under closed loop conditions of a room impulse response and a desired speech signal model, resulting in a system for multimicrophone combined acoustic feedback and noise cancellation. For specific applications (e.g. public address systems), the computational complexity may be reduced dramatically compared to state-of-the-art proactive acoustic feedback cancellers, while feedback cancellation performance is only marginally degraded.

Combined Feedback and Noise Suppression in Hearing Aids

In this paper, solutions for combined feedback and noise suppression in hearing aids are developed. The techniques presented are based on the generalized sidelobe canceller (GSC) and adaptive feedback canceller (AFC), with a prediction error method (PEM) adaptation to avoid speech distortion. Two possible cascades of GSC-based noise reduction and AFC, namely an ldquoAFC firstrdquo and a ldquoGSC first,rdquo as well as a truly integrated solution that jointly suppresses feedback and noise are discussed. The integrated solution (called PEM-GFIC) achieves optimum synergies between noise and feedback suppression at the lowest computational cost. In addition, it cancels more feedback than the generalized echo and interference canceller, a joint solution for echo and noise suppression. In the cascaded solutions, the feedback and noise suppression filters are not always optimally exploited. For high input SNRs, ldquoAFC firstrdquo scheme generally may achieve better feedback cancellation because of its larger number of degrees of freedom. However, noise reduction by the GSC-stage seriously affects the feedback cancellation performance. At low SNRs, ldquoGSC firstrdquo generally achieves more feedback cancellation than PEM-GFIC at the expense of worse noise reduction. At high hearing aid gains and/or large SNRs, the noise reduction stage however negatively affects the performance of the feedback cancellation filter, resulting in a worse feedback and noise suppression compared to PEM-GFIC.

Fast QRD-lattice-based unconstrained optimal filtering for acoustic noise reduction

We derive a fast QRD-least-squares lattice (QRD-LSL) based unconstrained optimal filtering algorithm for multichannel acoustic noise reduction. As known from the literature, the unconstrained optimal filtering approach is an alternative to the popular GSC beamforming, which does not rely on a priori information and hence possesses improved robustness. The optimal filtering problem involved is special in that the desired response signal is not known explicitly. The derivation of the QRD-LSL algorithm is based on a significantly reorganized version of a QRD-RLS-based unconstrained optimal filtering scheme. Overall an order of magnitude complexity reduction is obtained without any performance penalty, which makes this new approach affordable for real time implementation.

An integrated approach to acoustic noise and echo cancellation

We describe an approach to speech signal enhancement where acoustic echo cancellation and noise reduction, which are traditionally handled separately, are combined in one integrated scheme. The optimization problem defined by this scheme is solved adaptively using a QRD-based least squares lattice (QRD-LSL) algorithm. We show that the performance of the integrated scheme is superior to the performance of traditional (cascading) schemes, while complexity is kept at an affordable level.

A fast exact frequency domain implementation of the exponentially windowed affine projection algorithm

Fast versions of the so-called affine projection algorithm are based upon a specific approximation which theoretically is valid only if no regularisation is applied. We derive a fast frequency domain version of affine projection that does not need this approximation. As shown, this can be used to implement regularisation by means of an exponentially windowed updating procedure.

Implementation of an RTLS blind equalization algorithm on DSP

Blind equalization has been a very active area of research during the last years. Research is mostly focused on performance without too much attention on the complexity of the presented techniques. However, the high complexity of these blind algorithms coupled with the high data rates of mobile telecommunications may hamper a practical implementation. Recently we presented a recursive total least squares (RTLS) algorithm which has a reduced computational complexity (P. Vandaele and M. Moonen, 1997). We integrate this algorithm into a transmitter/receiver structure and present some results on the implementation of the algorithm in DSP.

Detection-Guided Fast Affine Projection Channel Estimator for Speech Applications

In various adaptive estimation applications, such as acoustic echo cancellation within teleconferencing systems, the input signal is a highly correlated speech. This, in general, leads to extremely slow convergence of the NLMS adaptive FIR estimator. As a result, for such applications, the affine projection algorithm (APA) or the low-complexity version, the fast affine projection (FAP) algorithm, is commonly employed instead of the NLMS algorithm. In such applications, the signal propagation channel may have a relatively low-dimensional impulse response structure, that is, the number m of active or significant taps within the (discrete-time modelled) channel impulse response is much less than the overall tap length n of the channel impulse response. For such cases, we investigate the inclusion of an active-parameter detection-guided concept within the fast affine projection FIR channel estimator. Simulation results indicate that the proposed detection-guided fast affine projection channel estimator has improved convergence speed and has lead to better steady-state performance than the standard fast affine projection channel estimator, especially in the important case of highly correlated speech input signals.