A Novel Normalized Sign Algorithm for System Identification Under Impulsive Noise Interference

To overcome the performance degradation of adaptive filtering algorithms in the presence of impulsive noise, a novel normalized sign algorithm (NSA) based on a convex combination strategy, called NSA-NSA, is proposed in this paper. The proposed algorithm is capable of solving the conflicting requirement of fast convergence rate and low steady-state error for an individual NSA filter. To further improve the robustness to impulsive noises, a mixing parameter updating formula based on a sign cost function is derived. Moreover, a tracking weight transfer scheme of coefficients from a fast NSA filter to a slow NSA filter is proposed to speed up the convergence rate. The convergence behavior and performance of the new algorithm are verified by theoretical analysis and simulation studies.

[1]  V. John Mathews,et al.  A stochastic gradient adaptive filter with gradient adaptive step size , 1993, IEEE Trans. Signal Process..

[2]  Shin'ichi Koike,et al.  Convergence Analysis of Adaptive Filters Using Normalized Sign-Sign Algorithm , 2005, IEICE Trans. Fundam. Electron. Commun. Comput. Sci..

[3]  M. Razaz,et al.  A normalized gradient descent algorithm for nonlinear adaptive filters using a gradient adaptive step size , 2001, IEEE Signal Processing Letters.

[4]  S. Douglas.  A family of normalized LMS algorithms , 1994, IEEE Signal Processing Letters.

[5]  PooGyeon Park,et al.  Variable Step-Size Affine Projection Sign Algorithm , 2012, IEEE Transactions on Circuits and Systems II: Express Briefs.

[6]  C. P. Kwong,et al.  Dual Sign Algorithm for Adaptive Filtering , 1986, IEEE Trans. Commun..

[7]  Rodrigo C. de Lamare,et al.  A low-complexity strategy for speeding up the convergence of convex combinations of adaptive filters , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[8]  E. Eweda,et al.  Analysis and design of a signed regressor LMS algorithm for stationary and nonstationary adaptive filtering with correlated Gaussian data , 1990 .

[9]  Tania Stathaki,et al.  A normalized robust mixed-norm adaptive algorithm for system identification , 2004, IEEE Signal Processing Letters.

[10]  Jia-Sien Soo,et al.  A multistep size (MSS) frequency domain adaptive filter , 1991, IEEE Trans. Signal Process..

[11]  A. Constantinides,et al.  Least mean mixed-norm adaptive filtering , 1994 .

[12]  S. B. Jebara,et al.  Variable step size filtered sign algorithm for acoustic echo cancellation , 2003 .

[13]  Shinichi Koike.  Variable Step Size Normalized sign Algorithm for fast convergent adaptive filters with robustness against impulse noise , 2000 .

[14]  Danilo P. Mandic,et al.  A normalized mixed-norm adaptive filtering algorithm robust under impulsive noise interference , 2003, 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03)..

[15]  D. P. Mandic,et al.  NNGD algorithm for neural adaptive filters , 2000 .

[16]  Robert Price,et al.  A useful theorem for nonlinear devices having Gaussian inputs , 1958, IRE Trans. Inf. Theory.

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

[18]  Aníbal R. Figueiras-Vidal,et al.  Plant identification via adaptive combination of transversal filters , 2006, Signal Process..

[19]  Zhang Jingjing,et al.  Variable Step Size LMS Algorithm , 2012 .

[20]  S. Koike,et al.  Analysis of adaptive filters using normalized signed regressor LMS algorithm , 1999, IEEE Trans. Signal Process..

[21]  Jonathon A. Chambers,et al.  An adaptive step-size code-constrained minimum output energy receiver for nonstationary CDMA channels , 2003, 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03)..

[22]  Lu Lu,et al.  A novel convex combination of LMS adaptive filter for system identification , 2014, 2014 12th International Conference on Signal Processing (ICSP).

[23]  Paulo Sergio Ramirez,et al.  Fundamentals of Adaptive Filtering , 2002 .

[24]  Vincent K. N. Lau,et al.  The Mobile Radio Propagation Channel , 2007 .

[25]  A. Figueiras-Vidal,et al.  Adaptive combination of normalised filters for robust system identification , 2005 .

[26]  Ali H. Sayed,et al.  Mean-square performance of a convex combination of two adaptive filters , 2006, IEEE Transactions on Signal Processing.

[27]  S.C. Douglas,et al.  Analysis and implementation of the max-NLMS adaptive filter , 1995, Conference Record of The Twenty-Ninth Asilomar Conference on Signals, Systems and Computers.

[28]  Anthony G. Constantinides,et al.  LMS+F algorithm , 1995 .

[29]  Dong-Jo Park,et al.  Convergence analysis of sign-sign LMS algorithm for adaptive filters with correlated Gaussian data , 1995, 1995 International Conference on Acoustics, Speech, and Signal Processing.

[30]  Jacob Benesty,et al.  A variable step-size normalized sign algorithm for acoustic echo cancelation , 2010, 2010 IEEE International Conference on Acoustics, Speech and Signal Processing.

[31]  Danilo P. Mandic,et al.  Towards the Optimal Learning Rate for Backpropagation , 2000, Neural Processing Letters.

[32]  Jacob Benesty,et al.  An Affine Projection Sign Algorithm Robust Against Impulsive Interferences , 2010, IEEE Signal Processing Letters.

[33]  S. C. Douglas.  Analysis and implementation of the max-NLMS adaptive filterAdaptive discrete cosine transform for feedback active noise control , 1995 .

[34]  J. Chambers,et al.  A robust mixed-norm adaptive filter algorithm , 1997, IEEE Signal Processing Letters.

[35]  J. Chambers,et al.  Convergence and steady-state properties of the least-mean mixed-norm (LMMN) adaptive algorithm , 1996 .