Robust least mean logarithmic square adaptive filtering algorithms

Xiong, Kui; Wang, Shiyuan

doi:10.1016/j.jfranklin.2018.10.019

Cited by 53 publications

(12 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Essentially, unlike the traditional VSS methods, the proposed VSS scheme in this paper uses the statistics of the error, which can significantly improve the convergence rate and steady-state filtering performance of RNLMAT in impulsive noises. In addition, (16) can be seen as the improvement of the estimate method proposed in [40] since smoothing factor γ in (16) emphasizes the influence of the recent values ofσ e min (i) and forgets the past ones, thus improving the traceability of data statistical variations. Finally, we summarize the proposed variable step-size RNLMAT (VSSRNLMAT) as follows:…”

Section: B Vssrnlmat Algorithmmentioning

confidence: 99%

Robust Normalized Least Mean Absolute Third Algorithms

2019

Self Cite

View full text Add to dashboard Cite

This paper addresses the stability issues of the least mean absolute third (LMAT) algorithm using the normalization based on the third order in the estimation error. A novel robust normalized least mean absolute third (RNLMAT) algorithm is therefore proposed to be stable for all statistics of the input, noise, and initial weights. For further improving the filtering performance of RNLMAT in different noises and initial conditions, the variable step-size RNLMAT (VSSRNLMAT) and the switching RNLMAT (SWRNLMAT) algorithms are proposed using the statistics of the estimation error and a switching method, respectively. The filtering performance of RNLMAT is improved by VSSRNLMAT and SWRNLMAT at the expense of affordable computational cost. RNLMAT with less computational complexity than other normalized adaptive filtering algorithms, can provide better filtering accuracy and robustness against impulsive noises. The steady-state performance of RNLMAT and SWRNLMAT in terms of the excess mean-square error is performed for theoretical analysis. Simulations conducted in system identification under different noise environments confirm the theoretical results and the superiorities of the proposed algorithms from the aspects of filtering accuracy and robustness against large outliers.

show abstract

Section: B Vssrnlmat Algorithmmentioning

confidence: 99%

Robust Normalized Least Mean Absolute Third Algorithms

2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…e constrained least mean logarithmic square (CLMLS) based on a relative logarithmic cost function and its variants were proposed in [19], and they were used in the application of sparse sensor array synthesis achieving the desired beam pattern with much less senor elements. In [20], a robust least mean logarithmic square (RLMLS) algorithm and its variable step-size variant were presented to combat impulsive noises, and its theoretical mean square performance was also analyzed with the stationary white Gaussian inputs.…”

Section: Introductionmentioning

confidence: 99%

Kernel Least Logarithmic Absolute Difference Algorithm

Wei

Shi

et al. 2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Kernel adaptive filtering (KAF) algorithms derived from the second moment of error criterion perform very well in nonlinear system identification under assumption of the Gaussian observation noise; however, they inevitably suffer from severe performance degradation in the presence of non-Gaussian impulsive noise and interference. To resolve this dilemma, we propose a novel robust kernel least logarithmic absolute difference (KLLAD) algorithm based on logarithmic error cost function in reproducing kernel Hilbert spaces, taking into account of the non-Gaussian impulsive noise. The KLLAD algorithm shows considerable improvement over the existing KAF algorithms without restraining impulsive interference in terms of robustness and convergence speed. Moreover, the convergence condition of KLLAD algorithm with Gaussian kernel and fixed dictionary is presented in the mean sense. The superior performance of KLLAD algorithm is confirmed by the simulation results.

show abstract

“…Although conventional LMS algorithms have good performance against Gaussian noise, their convergence performance is largely affected by other noises, especially impulse noise. Various adaptive algorithms for impulse noise [3], such as the sign algorithm [4], bias-compensate algorithms [5], and the family of logarithmic cost algorithms [6,7], have been investigated to improve system robustness. Some adaptive algorithms use a step-size scaler presented by modifying the tan h cost function to exclude the effects of impulsive samples [8,9].…”

Section: Introductionmentioning

confidence: 99%

Affine projection Weibull M‐transform least mean square algorithm against impulsive interference

2021

View full text Add to dashboard Cite

The well-known affine projection sign algorithm is one of the classic adaptive filtering methods for denoising and channel equalisation, which can achieve robust performance against coloured input and impulse noise. This work proposes the affine projection Weibull M-transform least mean square algorithm to improve the steady-state performance of affine projection algorithms. The proposed algorithm combines the Weibull M-transform cost function with the L2-norm constraint of weight vector to improve the performance of existing affine projection algorithms and achieve a better convergence rate and better misalignment. This novel algorithm also preserves good performance against impulsive interference and coloured input. Simulation results demonstrated that the proposed algorithm has a remarkable improvement in convergence performance compared with other algorithms.

show abstract

Robust least mean logarithmic square adaptive filtering algorithms

Cited by 53 publications

References 36 publications

Robust Normalized Least Mean Absolute Third Algorithms

Robust Normalized Least Mean Absolute Third Algorithms

Kernel Least Logarithmic Absolute Difference Algorithm

Affine projection Weibull M‐transform least mean square algorithm against impulsive interference

Contact Info

Product

Resources

About