Adaptive Transversal Filter Research Articles

Isolation of resonance in acoustic backscatter from elastic targets using adaptive estimation schemes

The problem of underwater target detection and classification from acoustic backscatter is the central focus of this paper. It has been shown that at certain frequencies the acoustic backscatter from elastic targets exhibits certain resonance behavior which closely relates to the physical properties of the target such as dimension, thickness, and composition. Several techniques in both the time domain and frequency domain have been developed to characterize the resonance phenomena in acoustic backscatter from spherical or cylindrical thin shells. The purpose of this paper is to develop an automated approach for identifying the presence of resonance in the acoustic backscatter from an unknown target by isolating the resonance part from the specular contribution. An adaptive transversal filter structure is used to estimate the specular part of the backscatter and consequently the error signal would provide an estimate of the resonance part. An important aspect of this scheme Lies in the fact that it does not require an underlying model for the elastic return. The adaptation rule is based upon fast Recursive Least Squares (RLS) learning. The approach taken in this paper is general in the sense that it can be applied to targets of unknown geometry and thickness and, further, does not require any a priori information about the target and/or the environment. Test results on acoustic data are presented which indicate the effectiveness of the proposed approach.

A parallel binary structured LMS algorithm for transversal adaptive filters

The adaptation process in digital filters requires extensive calculation. This computation makes adaptation a slow and time consuming process. Simple, but versatile, parallel algorithms for adaptive filters, suitable for VLSI implementation, are in demand. In this paper a regular and modular parallel algorithm for an adaptive filter is presented. This parallel structure is based on the Gradient Vector Estimation Algorithm, which minimizes the Mean Square Error. In the parallel method, the tap weights of the adaptive filter are updated everys steps, whereas in the recursive algorithms, the tap weights are updated at each step. Fors step update, bit strings of lengths are used to derive the terms with which the tap weights of the adaptive filter are calculated. The algorithm presented computes the tap weights at timen+s as a function of the tap weights at timen, the inputs from timen+1 ton+s−1, and the desired output from timen+1 ton+s−1. The algorithm also can be mapped to a VLSI architecture that is both regular and modular and allows either expansion of the order of the filter or the degree of parallelism obtainable. A comparison between the performance of the sequential LMS algorithm, Fast Exact LMS algorithm, and the parallel binary structured LMS algorithm is presented.

Influence of quantization error on the performance of an adaptive transversal filter

We report on the results of a theoretical analysis of the influence of quantization error on the performance of an adaptive transversal filter minimizing the root-mean-square error by means of a gradient least-mean-square algorithm. An expression for the covariance matrix of weight coefficients is defined. The dependence of the output excess noise power on the adaptive system parameters is analyzed.

Rejection of an FH signal in a DS spread-spectrum system using complex adaptive filters

The performance of a direct sequence QPSK spread-spectrum receiver using adaptive filters in the presence of frequency hopped interference is analyzed. The analysis includes both the adaptive prediction error filters and the adaptive transversal filters with two-sided taps. If the product of the instantaneous frequency offset /spl Omega//sub l/, between the jamming signal and the carrier of the spread-spectrum signal, and the sampling period /spl Delta/ is 360/spl deg/ (/spl Omega//sub l//spl middot//spl Delta/=360/spl deg/), the filter gain is reduced to zero. The filter gain G highly depends on the filter adaptation rate /spl mu/. Depending on /spl mu/, G can vary from zero to more than 20 dB for a jammer/signal power ratio (J/S) of 20 dB. If /spl Omega//sub l//spl middot//spl Delta/ is small enough (/spl les/10/spl deg/), the performance of the transversal filter is better than that of the prediction error filter, in the case when /spl mu/ is small. For larger values of /spl mu/ or /spl Omega//sub l//spl middot//spl Delta/, these performances are approximately the same. Numerical results for the hopping sequence of the jamming signal are also presented. Besides the filter gain the analysis of the adaptation rate (time constant) filter misadjustment and the system bit error probability is also included. >

Analysis of finite length acoustic echo cancellation system

The acoustic echo canceler using an adaptive transversal filter and the least mean-square (LMS) algorithm is the most effective technique to reduce acoustic echoes in a hands-free telephone system. However, the requirement of a very high order filter for each microphone results in difficulties in convergence and hardware implementation. In this paper, the performance of the finite length adaptive filter is studied. A formula which relates the echo cancellation to the filter size, N, is established. Detailed analysis shows that this finite filter length will have better performance using speech than white noise.

Structural system identification using Least Mean Square (LMS) adaptive technique

This paper presents an approach to identify structural parameters using the Least Mean Square (LMS) adaptive transversal filter. This method features easy computer simulation and fast data processing. This approach is effective even in the presence of input noise when it is applied in communications; however, this advantage is not obvious when it is used in structural systems, since the signal bandwidht it deals with in communications is much wider than the bandwidth in structure systems. With a low-pass filter added to a data acquisition system, this approach functions effectively for both linear and nonlinear structural systems. Numerical examples and experimental tests have been included to demonstrate the technique for both linear and nonlinear systems.

Adaptive enhancement of microphone array signals

The signal-to-noise ratio of a speech signal picked up by a microphone array can be improved by adaptive post processing. Enhancement techniques known from single microphone or dual microphone signal processing, like noise canceling and spectral subtraction can be extended to a multimicrophone array system. The noise canceling technique and derived structures try to model the room impulse response by an adaptive transversal filter. Thus the performance of these algorithms is limited by the ratio of filter length to reverberation time and by the capability to track the nonstationary impulse response. Reduction of the noise of approximately 8 dB can be achieved with acceptable filter length in a stationary environment, but precautions must be taken to avoid canceling of the desired speech signal. The spectral subtraction method yields higher improvements in signal-to-noise ratio up to 12 dB, but due to false estimation of magnitude short time spectra the processed speech signal contains distortion known as ‘‘musical tones.’’ Different structures for adaptive post processing of microphone array signals based on both enhancement techniques are presented and discussed. Their performance in stationary and nonstationary environments is characterized by means of improvement of the signal-to-noise ratio and subjective speech quality.

Development and analysis of distributed acoustic echo cancellation microphone system

Abstract Acoustic echo cancellation using an adaptive transversal filter and the least mean square algorithm is the most effective technique of reducing acoustic echoes in a hands-free telephone system. However, a very high-order adaptive filter and multiple echo paths for each microphone result in difficulties in algorithm convergence and hardware implementation. This paper presents a distributed system which performs acoustic echo cancellation in the microphone unit. Each acoustic echo cancellation microphone unit has two closely spaced directional microphones pointing in opposite directions and a much lower order adaptive filter than in the earlier systems. Computer simulations were conducted based on real speech data collected from a typical conference room. Finally, a robust double-talk detector, which identifies three modes of operation, was developed and tested in real-time

Vectorization of the DLMS transversal adaptive filter

The subject of high sampling rate realizations for transversal adaptive filters is addressed. In particular, a vectorized version of the delayed least mean squares (DLMS) algorithm is derived using look-ahead computation techniques. The resulting parallel algorithm is then mapped onto a linear array of highly pipelined processing modules, which can accept an input vector of arbitrary length, and compute the corresponding output vector in a single clock cycle. The proposed system is shown to be capable of implementing transversal adaptive filters at sampling rates which are theoretically without bound. The performance of the proposed system is analyzed and simulation results are presented to verify the convergence properties of the algorithm under varying degrees of vectorization.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Adaptive allpass filtering for nonminimum-phase system identification

Efficient algorithms for the recursive identification of nonminimum-phase systems using an adaptive allpass filter together with an adaptive transversal filter are introduced. Both the output and input sequences are assumed to be available, and the proposed technique basically uses a linear predictor at the output of the system to equalise its magnitude response and an allpass filter to match its phase response. The mirror image property of the numerator and denominator polynomials in an allpass transfer function reduces the number of adaptive parameters needed compared to an unconstrained recursive identification scheme. Computer simulations are used to support the claims made in the paper.

Recursive prediction of chaotic time series

Considerable progress has been made in recent years in the analysis of time series arising from chaotic systems. In particular, a variety of schemes for the short-term prediction of such time series has been developed. However, hitherto all such algorithms have used batch processing and have not been able to continuously update their estimate of the dynamics using new observations as they are made. This severely limits their usefulness in real time signal processing applications. In this paper we present a continuous update prediction scheme for chaotic time series that overcomes this difficulty. It is based on radial basis function approximation combined with a recursive least squares estimation algorithm. We test this scheme using simulated data and comment on its relationship to adaptive transversal filters, which are widely used in conventional signal processing.

Adaptive acoustic echo cancellation microphone

In teleconferencing, speaker-phone systems have an annoying acoustic echo problem. The acoustic echo canceller using an adaptive transversal filter is an effective solution. However, very-high-order adaptive filters have computational and convergence problems. Furthermore, acoustic echoes picked up by multiple microphones make the real-time application very difficult to implement. In this paper, a new approach to these problems is proposed. This new acoustic echo cancellation microphone system consists of two closely positioned directional microphones pointed in opposite directions and an adaptive filter. The adaptive filter converges to the transfer function difference between the two echo paths, which are similar since the two microphones are close to each other and far away from the loudspeaker. Therefore, a very-low-order adaptive filter is sufficient to reduce the echo. The steady-state analysis of this new system shows significant improvement over the traditional system. The system is tested using real speech data collected from a typical conference room. The experimental and simulation results show that more than 26 dB of cancellation can be achieved during the receive mode with the filter order equal to 64.

The gamma-filter-a new class of adaptive IIR filters with restricted feedback

Generalized feedforward filters, a class of adaptive filters that combines attractive properties of finite impulse response (FIR) filters with some of the power of infinite impulse response (IIR) filters, are described. A particular case, the gamma filter, generalizes Widrow's adaptive transversal filter (adaline) to an infinite impulse response filter. Yet, the stability condition for the gamma filter is trivial, and LMS adaptation is of the same computational complexity as the conventional transversal filter structure. Preliminary results indicate that the gamma filter is more efficient than the adaptive transversal filter. The authors extend the Wiener-Kopf equation to the gamma filter and develop some analysis tools. >

Joint time-delay estimation and adaptive recursive least squares filtering

A general estimation model is defined in which two observations are available: a noisy and a noise-filtered and delayed version of the transmitted signal. The delay and the filter must be estimated. The joint estimator is composed of an adaptive delay element operating in conjunction with an adaptive transversal filter. The delay is updated using a form of derivative, with respect to the delay, of the sum of squared errors. The adaptive delay is limited to integer values and is defined as the lag. The lag value is computed and updated so that the optimum least-squares solution is attained. The joint algorithm is obtained by combining the lag update relations with a version of the fast transversal filter RLS algorithm. Simulations show that both stationary and time-varying delays are effectively tracked and that the adaptive filter properly estimates the reference filter impulse response. >

Multiple-Microphone Acoustic Echo Cancellation System with the Partial Adaptive Process

The acoustic echo canceler using an adaptive transversal filter and the least-mean square (LMS) algorithm is the most effective technique to reduce acoustic echoes in a hands-free telephone system. However, the requirement of a very high order filter for each microphone echo path results in difficulties in convergence speed and hardware implementation. In this paper, a partial adaptive process (PAP) which utilizes the exponential decay characteristics of the acoustic echo path impulse response is developed. Simulations are conducted to show that the performance of the PAP method could be acceptable in practical use. A combination FIR and IIR filter structure and a new time-weighted LMS (TW-LMS) algorithm to complement the PAP are also proposed. This new adaptation algorithm and filter structure further improve the performance of the PAP method in terms of echo reduction and computational savings.

Joint gradient-based time-delay estimation and adaptive minimum mean-squared-error filtering

A general estimation model is defined in which two observations are available; one is a noisy version of the transmitted signal, while the other is a noisy filtered and delayed version of the same transmitted signal. The time-varying delay and the filter are unknown quantities that must be estimated. A joint estimator is proposed. It is composed of an adaptive delay element in conjunction with a transversal adaptive filter. The same error signal is used to adjust the delay element and the filter such that the minimum mean squared error is attained. Two joint gradient-based adaptation algorithms are studied. The joint steepest-descent (SD) algorithm is first investigated. The possibility of a multitude of stable solutions is established and a condition of convergence is presented. A stochastic implementation of the joint SD algorithm, under the form of a joint least-mean-square (LMS) algorithm, is then presented. It is analysed in terms of convergence in the mean and in the mean square of both the delay estimate and the adaptive filter weight vector estimate. The conditions of convergence of the joint LMS algorithm are established as a function of the power spectral densities of the observed signals and the minimum mean squared error.

Exponentially weighted stepsize NLMS adaptive filter based on the statistics of a room impulse response

A normalized least-mean-squares (NLMS) adaptive algorithm with double the convergence speed, at the same computational load, of the conventional NLMS for an acoustic echo canceller is proposed. This algorithm, called the ES (exponentially weighted stepsize) algorithm, uses a different stepsize (feedback constant) for each weight of an adaptive transversal filter. These stepsizes are time-invariant and weighted proportionally to the expected variation of a room impulse response. The algorithm adjusts coefficients with large errors in large steps, and coefficients with small errors in small steps. A transition formula is derived for the mean-squared coefficient error of the algorithm. The mean stepsize determines the convergence condition, the convergence speed, and the final excess mean-squared error. Modified for a practical multiple DSP structure, the algorithm requires only the same amount of computation as the conventional NLMS. The algorithm is implemented in a commercial acoustic echo canceller, and its fast convergence is demonstrated. >

The effect of I,Q mismatch errors on adaptive cancellation

The effects of in-phase (I) and quadrature-phase (Q) amplitude errors and low-pass-filter (LPF) errors on adaptive cancellers are investigated. I,Q errors occur because of errors in the synthesis process of the mixers and LPFs designed to be identical for each input channel. These I,Q errors among the channels result in cancellation degradation. Tapped delay line transversal filters have been proposed as a way to compensate for these errors and thus improve cancellation performance. However, it is shown that if there is any LPF mismatch, then transversal filtering has a small effect on improving canceler performance. The use of individual I,Q adaptive transversal filter weighting is suggested as a means of completely eliminating the phase amplitude errors, and making the canceler performance responsive to transversal filter compensation. >

Joint recursive optimality—a non-probabilistic approach to adaptive transversal filter design

It is common practice to design and interpret adaptive filter algorithms in terms of some optimality principle. The predominant approach purports to miminize a quadratic measure of the filter's error signal. This approach is, however, tied to the assumption of a time-invariant applications environment so that operational algorithm structures (such as LMS) are only obtained from ad hoc modifications aiming at continuous tracking. Joint recursive optimality is a unified framework for the “optimal” design of adaptive transversal filters. It covers a wide class of adaptation algorithms including standards like least mean squares and recursive least squares algorithms. Contrary to existing optimality principles, a deterministic criterion is proposed which operational algorithms meet exactly for each time step. The criterion features a trade-off between time variance of the filter coefficients and error signal power. The paper discusses implications for the interpretation of a variety of algorithms and concludes with guidelines for the incorporation of prior statistical or coefficient modeling information into the adaptation algorithm.

Adaptive LMS transversal filters with controlled length

Dynamic control of the filter length during initial adaptation for the adaptive LMS transversal filter is proposed. Theoretical analysis of the influence of the filter length on the convergence speed and the steady-state error is presented for the channel estimator and the channel equaliser. Simulation results performed for the channel estimation and equalisation of a telephone channel show that a significant gain in the convergence speed can be achieved over that for the constant length adaptive filter.