KLMAT: A Kernel Least Mean Absolute Third Algorithm

The kernel methods provide an interesting way to improve the performance of nonlinear adaptive filters. In this paper, a kernel least mean absolute third (KLMAT) algorithm is proposed for accelerating the convergence speed of the kernel adaptive filter. Combining the benefits of the kernel method and the least mean absolute third (LMAT) algorithm, the proposed KLMAT performs robustly against noise with different probability densities. To further improve the convergence property of the KLMAT algorithm, a variable step-size (VSS) version is designed by a Lorentzian function, which is called the VSS-KLMAT algorithm. Moreover, the stability and convergence property of the proposed algorithms are analyzed. Simulation results in the context of time series prediction show that the KLMAT and VSS-KLMAT algorithms provide faster convergence rate than LMAT and kernel least mean square (KLMS) algorithms with the same steady-state error in the various noise probability densities.

[1]  Weifeng Liu,et al.  Kernel Adaptive Filtering: A Comprehensive Introduction , 2010 .

[2]  Haiquan Zhao,et al.  A Novel Normalized Sign Algorithm for System Identification Under Impulsive Noise Interference , 2015, Circuits Syst. Signal Process..

[3]  Weifeng Liu,et al.  Kernel Affine Projection Algorithms , 2008, EURASIP J. Adv. Signal Process..

[4]  C. K. Michael Tse,et al.  Kernel Affine Projection Sign Algorithms for Combating Impulse Interference , 2013, IEEE Transactions on Circuits and Systems II: Express Briefs.

[5]  Qianqian Liu,et al.  A novel adaptive harmonic detecting algorithm applied to active power filters , 2010, 2010 3rd International Congress on Image and Signal Processing.

[6]  Tyseer Aboulnasr,et al.  A robust variable step-size LMS-type algorithm: analysis and simulations , 1997, IEEE Trans. Signal Process..

[7]  Michael J. Black,et al.  The outlier process: unifying line processes and robust statistics , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[8]  Sung Ho Cho,et al.  Performance of least mean absolute third (LMAT) adaptive algorithm in various noise environments , 1998 .

[9]  H. Tong,et al.  Threshold Autoregression, Limit Cycles and Cyclical Data , 1980 .

[10]  William A. Sethares,et al.  Adaptive algorithms with nonlinear data and error functions , 1992, IEEE Trans. Signal Process..

[11]  Weifeng Liu,et al.  The Kernel Least-Mean-Square Algorithm , 2008, IEEE Transactions on Signal Processing.

[12]  Yi Yu,et al.  A new normalized LMAT algorithm and its performance analysis , 2014, Signal Process..

[13]  Q. Y. Miao,et al.  Kernel least-mean mixed-norm algorithm , 2012 .

[14]  Lehel Csató,et al.  Sparse On-Line Gaussian Processes , 2002, Neural Computation.

[15]  Xiangping Zeng,et al.  Identification of Nonlinear Dynamic System Using a Novel Recurrent Wavelet Neural Network Based on the Pipelined Architecture , 2014, IEEE Transactions on Industrial Electronics.

[16]  Brian D. O. Anderson,et al.  Stability of adaptive systems: passivity and averaging analysis , 1986 .

[17]  John C. Platt.  A Resource-Allocating Network for Function Interpolation , 1991, Neural Computation.

[18]  Yi Yu,et al.  An improved variable step-size NLMS algorithm based on a Versiera function , 2013, 2013 IEEE International Conference on Signal Processing, Communication and Computing (ICSPCC 2013).

[19]  Weifeng Liu,et al.  An Information Theoretic Approach of Designing Sparse Kernel Adaptive Filters , 2009, IEEE Transactions on Neural Networks.

[20]  S. Haykin,et al.  Kernel Least‐Mean‐Square Algorithm , 2010 .

[21]  V. J. Mathews,et al.  Improved convergence analysis of stochastic gradient adaptive filters using the sign algorithm , 1987, IEEE Trans. Acoust. Speech Signal Process..

[22]  José Carlos M. Bermudez,et al.  Stochastic analysis of the Least Mean Kurtosis algorithm for Gaussian inputs , 2016, Digit. Signal Process..

[23]  Yi Zhou,et al.  New partial update robust kernel least mean square adaptive filtering algorithm , 2014, 2014 19th International Conference on Digital Signal Processing.

[24]  Shie Mannor,et al.  The kernel recursive least-squares algorithm , 2004, IEEE Transactions on Signal Processing.

[25]  C. Richard Johnson,et al.  Adaptive algorithms with filtered regressor and filtered error , 1989, Math. Control. Signals Syst..

[26]  Eweda Eweda,et al.  Stochastic Analysis of a Stable Normalized Least Mean Fourth Algorithm for Adaptive Noise Canceling With a White Gaussian Reference , 2012, IEEE Transactions on Signal Processing.

[27]  Aaas News,et al.  Book Reviews , 1893, Buffalo Medical and Surgical Journal.

[28]  Badong Chen,et al.  Quantized Kernel Least Mean Square Algorithm , 2012, IEEE Transactions on Neural Networks and Learning Systems.