论文信息 - Exponential normalized sign algorithm for system identification

Exponential normalized sign algorithm for system identification

The sign algorithm is not influenced by impulsive noise, but it exhibits slow convergence rate. Therefore, in this paper, the exponential normalized sign algorithm is proposed for system identification. The proposed algorithm is obtained from the exponential function of the L1 norm of the output error signal. The results of computer simulations proved that the convergence rate of the proposed algorithm is better than that of the sign algorithm.

Jae Jin Jeong | Sang Woo Kim

[1] Anthony G. Constantinides,et al. A class of stochastic gradient algorithms with exponentiated error cost functions , 2009, Digit. Signal Process..

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

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

[4] C. L. Nikias,et al. Signal processing with fractional lower order moments: stable processes and their applications , 1993, Proc. IEEE.

[5] Woon-Seng Gan,et al. Subband Adaptive Filtering: Theory and Implementation , 2009 .

[6] E. Erzin,et al. Adaptive filtering for non-Gaussian stable processes , 1994, IEEE Signal Processing Letters.

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