SQUARE-ROOT FORM BASED DERIVATION OF A NOVEL NLMS-LIKE ADAPTIVE FILTERING ALGORITHM (original) (raw)
Related papers
Journal of Algorithms, 2000
In this paper, a new multichannel recursive least squares (MRLS) adaptive algorithm is presented which has a number of very interesting properties. The proposed computational scheme performs adaptive filtering via the use of a finite window, where the burdening past information is dropped directly by means of a generalized inversion lemma; consequently, the proposed algorithm has excellent tracking abilities and very low misjudgment. Moreover, the scheme presented here, due to its particular structure and to the proper choice of mathematical definitions behind it, is very robust; i.e., it is less sensitive in the finite precision numerical error generation and propagation. Also, the new algorithm can be parallelized via a simple technique and its parallel form and, when executed with four processors, is faster than all the already existing schemes that perform both infinite and finite window multichannel adaptive filtering. Finally, due to the particular structure of this scheme and to the intrinsic flexibility in the choice of the window length, the proposed algorithm can act as a full substitute of the infinite window MRLS ones.
A time-dependent LMS algorithm for adaptive filtering
2003
A novel approach for the least-mean-square (LMS) estimation algorithm is proposed. The approach utilizes the conventional LMS algorithm with a time-varying convergence parameter μ n rather than a fixed convergence parameter μ. It is shown that the proposed time-varying LMS algorithm (TV-LMS) provides reduced mean-squared error and also leads to a faster convergence as compared to the conventional fixed parameter LMS algorithm. This paper presents a performance study for the proposed TV-LMS algorithm and other two main adaptive approaches: the least-mean square (LMS) algorithm and the recursive least-squares (RLS) algorithm. These algorithms have been tested for noise reduction and estimation in single-tone sinusoids and nonlinear narrow-band FM signals corrupted by additive white Gaussian noise. The study shows that the TV-LMS algorithm has a computation time close to conventional LMS algorithm with the advantages of faster convergence time and reduced mean-squared error.
The square-root Schur RLS adaptive filter
[Proceedings] ICASSP 91: 1991 International Conference on Acoustics, Speech, and Signal Processing, 1991
A square-root normalized Schur RLS adaptive filter is presented which belongs t o the newly developed class of Schur-type algorithms for adaptive filtering and parameter estimation in the serialized data case of recursive least squares (RLS) processing. Schur-type algorithms can outperform many of the now well-known "fast" adaptive filtering algorithms due t o their inherent ability t o work with arbitrary recursive windowing of the data. Key features of the square-root Schur RLS adaptive filter are a fully pipelineable structure and excellent numerical properties. A systolic array of CORDIC processors for implementation of the square-root Schur RLS adaptive filter is presented and its performance is illustrated with a typical example.
A general class of nonlinear normalized adaptive filtering algorithms
IEEE Transactions on Signal Processing, 1999
The normalized least mean square (NLMS) algorithm is an important variant of the classical LMS algorithm for adaptive linear filtering. It possesses many advantages over the LMS algorithm, including having a faster convergence and providing for an automatic time-varying choice of the LMS stepsize parameter that affects the stability, steady-state mean square error (MSE), and convergence speed of the algorithm. An auxiliary fixed step-size that is often introduced in the NLMS algorithm has the advantage that its stability region (step-size range for algorithm stability) is independent of the signal statistics. In this paper, we generalize the NLMS algorithm by deriving a class of nonlinear normalized LMS-type (NLMS-type) algorithms that are applicable to a wide variety of nonlinear filter structures. We obtain a general nonlinear NLMS-type algorithm by choosing an optimal time-varying step-size that minimizes the next-step MSE at each iteration of the general nonlinear LMS-type algorithm. As in the linear case, we introduce a dimensionless auxiliary step-size whose stability range is independent of the signal statistics. The stability region could therefore be determined empirically for any given nonlinear filter type. We present computer simulations of these algorithms for two specific nonlinear filter structures: Volterra filters and the recently proposed class of Myriad filters. These simulations indicate that the NLMS-type algorithms, in general, converge faster than their LMS-type counterparts.
Generalized square-root adaptive algorithms
International Conference on Acoustics, Speech, and Signal Processing, 2000
In this paper we extend the QR-based recursive least squares algorithms to a more general square-root formulation, which involves array transform using orthogonal and possibly non orthogonal transforms e.g. Gauss transform. The adaptive algorithms, thus obtained, have better numerical properties than the traditional pseudo-inverse-based algorithms. This iiew update scheme covers an extended set of algorithms it connects the recursive least squares to gradient based algorithms. Moreover, we show that the recently derived Block diagonal QR based algorithms [1][2] belongs to this general square-root class.
A FAST Algorithm for Adaptive Filtering
2009 16th International Conference on Systems, Signals and Image Processing, 2009
In this paper, we evaluate the possibility to develop algorithms of adaptation for the applications system of acoustic echo cancellation, while maintaining equilibrium between its reduced calculation complexity and its adaptive performances. We present new algorithms versions of fast recursive least squares numerically stable (NS-FRLS). They are obtained by means of redundant formulas, available in the fast recursive least squares (FRLS) algorithms, to estimate numerical errors and to retroact them in an unspecified point of the algorithm in order to modify its numerical properties. These algorithms represent a very important load of calculation that needs to be reduced. we propose a new (M-SMFTF) algorithm for adaptive filtering with fast convergence and low complexity. It is the result of a simplified FTF type algorithm, where the adaptation gain is obtained only from the forward prediction variables and using a new recursive method to compute the likelihood variable. This algorithm presents a certain interest, for the adaptation of very long filters, like those used in the problems of echo acoustic cancellation, due to its reduced complexity, its numerical stability and its convergence in the presence of the speech signal. Its calculation complexity is of 6L (L is the finite impulse response filter length) and this is considerably reduced to (2L+4P) when we use a reduced P-size (P<<L) forward predictor.
Advanced algorithms for adaptive filtering
2009
In this paper, we evaluate the possibility to develop algorithms of adaptation for the applications system of acoustic echo cancellation, while maintaining equilibrium between its reduced calculation complexity and its adaptive performances. We present new algorithms versions of fast recursive least squares numerically stable (NS-FRLS). They are obtained by means of redundant formulas, available in the fast recursive least squares (FRLS) algorithms, to estimate numerical errors and to retroact them in an unspecified point of the algorithm in order to modify its numerical properties. These algorithms represent a very important load of calculation that needs to be reduced. we propose a new (M-SMFTF) algorithm for adaptive filtering with fast convergence and low complexity. It is the result of a simplified FTF type algorithm, where the adaptation gain is obtained only from the forward prediction variables and using a new recursive method to compute the likelihood variable. This algorit...
Fast Algorithms with low Complexity for Adaptive Filtering
2009
The numerically stable version of fast recursive least squares (NS-FRLS) algorithms represent a very important load of calculation that needs to be reduced. Its computational complexity is of 8L operations per sample, where L is the finite impulse response filter length. We propose an algorithm for adaptive filtering, while maintaining equilibrium between its reduced computational complexity and its adaptive performances. We present a new (M-SMFTF) algorithm for adaptive filtering with fast convergence and low complexity. It is the result of a simplified FTF type algorithm, where the adaptation gain is obtained only from the forward prediction variables and using a new recursive method to compute the likelihood variable. This algorithm presents a certain interest, for the adaptation of very long filters, like those used in the problems of echo acoustic cancellation, due to its reduced complexity, its numerical stability and its convergence in the presence of the speech signal. Its c...