The use of the modified escalator algorithm to improve the performance of transform-domain LMS adaptive filters

Parikh, V.N.; Baraniecki, Anna Z.

doi:10.1109/78.661330

Cited by 14 publications

(4 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The frequency magnitude responses of the coloring filters are shown in Figs. 6(a) through 6(d) [4]. These The MSE learning curves for the four inputs are shown in Figs.…”

Section: System Identificationmentioning

confidence: 98%

“…Instead of estimating the cross-correlation functions by using the simple low-pass filter, we can use the steepest descent method with time-varying convergence parameter ) (k i µ [4] as the following:…”

Section: Escalator Coefficient Adaptation By Mse Criterionmentioning

confidence: 99%

“…Though the ESC structure orthogonalizes the input signals, its coefficient adaptation algorithms have not been sufficiently studied for faster convergence speed. The LMS algorithm for the ESC in which mean squared local estimation errors are minimized is currently used in the ESC filtering problems [3], [4]. In this paper, we present a new fast adaptation algorithm for the ESC structure by introducing the least squares (LS) criterion to the local errors of the structure.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Least Squares Approach to Escalator Algorithms for Adaptive Filtering

Kim¹

2006

ETRI J

View full text Add to dashboard Cite

In this paper, we introduce an escalator (ESC) algorithm based on the least squares (LS) criterion. The proposed algorithm is relatively insensitive to the eigenvalue spread ratio (ESR) of an input signal and has a faster convergence speed than the conventional ESC algorithms. This algorithm exploits the fast adaptation ability of least squares methods and the orthogonalization property of the ESC structure. From the simulation results, the proposed algorithm shows superior convergence performance.

show abstract

“…The frequency magnitude responses of the coloring filters are shown in Figs. 6(a) through 6(d) [4]. These The MSE learning curves for the four inputs are shown in Figs.…”

Section: System Identificationmentioning

confidence: 98%

Section: Escalator Coefficient Adaptation By Mse Criterionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Least Squares Approach to Escalator Algorithms for Adaptive Filtering

Kim¹

2006

ETRI J

View full text Add to dashboard Cite

show abstract

“…Three kinds of input signals, obtained by filtering white noises with three 32-tap linear-phase FIR filters (i.e., lowpass, band-pass and high-pass filters as in [6]), are used in the simulation to test the converging performance of the proposed method even in case of input processes with various statistics. It is known that DCT-LMS has close-tooptimal converging performance for low-pass processes [7]. Also, the signal-to-noise ratio (SNR) at the output of the model system is chosen to be 40 dB.…”

Section: Dj(k) = J F ( T ) M J ' 2 P ( M J Rmentioning

confidence: 99%

Adaptive volterra filtering using M-band wavelet transform

Kim

Lee²,

Park³

et al.

ISSPA '99. Proceedings of the Fifth International Symposium on Signal Processing and Its Applications (IEEE Cat. No.99EX359)

View full text Add to dashboard Cite

A new LMS adaptive Volterra filtering in the M-band wavelet transform domain is presented, where the input is preprocessed with MDWT (M-band discrete wavelet transform) being followed by power normalization. In particular, the preprocessing procedure leads to effective reduction of the eigenvalue spread of a Volterra input auto-correlation matrix, and thus improves the convergence rate of the adaptation process even in case of wide classes of input processes. To demonstrate the performance of the proposed approach, some simulation results are provided.

show abstract

Convergence analysis and a synchronized learning algorithm for a joint lattice predictor and FIR adaptive filter

Tokui

Nakayama

Hirano

2003

Electron Comm Jpn Pt III

View full text Add to dashboard Cite

SUMMARYSince the transfer function of a two-stage adaptive filter combining a lattice predictor and a FIR filter includes the reflection coefficients and the filter coefficients, a conventional learning algorithm does not integrate the updates of the reflection coefficients and the filter coefficients. Therefore, error reduction could not be guaranteed. To solve this problem, we proposed a learning algorithm that corrects the coefficients of the adaptive filter synchronized to updating the reflection coefficients. The filter coefficients w(n + 1) are corrected so that the transfer function based on the filter coefficients w(n + 1) updated by using the reflection coefficients k(n) updated at sample time n is identical for the reflection coefficients k(n + 1) updated at sample time n + 1, and the filter coefficients w(n + 1) are used in the output calculation for sample n + 1. The computational complexity required for updating is 2 (prediction order × filter order). We conducted a simulation to estimate an unknown system by using white noise, colored noise, and speech as the input signal and compared the performance to a conventional method. We verified the following. The conventional method that does not correct the filter coefficients in response to fluctuations in the reflection coefficients cannot adequately reduce the estimation errors. The proposed method converges 4 to 5 times faster than NLMS for speech input, and its errors are equivalent to those of RLS. When the prediction order << the filter order, the computational complexity can be substantially less than RLS.

show abstract

The use of the modified escalator algorithm to improve the performance of transform-domain LMS adaptive filters

Cited by 14 publications

References 12 publications

A Least Squares Approach to Escalator Algorithms for Adaptive Filtering

A Least Squares Approach to Escalator Algorithms for Adaptive Filtering

Adaptive volterra filtering using M-band wavelet transform

Convergence analysis and a synchronized learning algorithm for a joint lattice predictor and FIR adaptive filter

Contact Info

Product

Resources

About