Adaptive Filter by using Sparse Partial Update Extended Correlation LMS Algorithm in the Double-Talk Condition

Echo path estimation in echo canceling for teleconference system is a problem in double-talk condition. The correlation function based algorithms were defined by the authors to solve this problem. In this paper, in order to improve the convergence speed of correlation function based algorithm, we propose a new modified proportionate step-size adaptation method, and then implement it into frequency domain extended correlation LMS algorithm (FECLMS). The new algorithm is called proportionate frequency domain extended correlation LMS algorithm (PFECLMS). The computer simulation results support the theoretical findings and verify the robustness of the proposed algorithm in the double-talk situation.

[1]  Ioannis Pitas,et al.  Fast algorithms for running ordering and max/min calculation , 1989 .

[2]  Tyseer Aboulnasr,et al.  Complexity reduction of the NLMS algorithm via selective coefficient update , 1999, IEEE Trans. Signal Process..

[3]  Patrick A. Naylor,et al.  A short-sort M-Max NLMS partial-update adaptive filter with applications to echo cancellation , 2003, 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03)..

[4]  Milos Doroslovacki,et al.  New sparse adaptive algorithms using partial update , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[5]  Katsumi Yamashita,et al.  Design of echo canceler with frequency domain ECLMS algorithm , 2000, Globecom '00 - IEEE. Global Telecommunications Conference. Conference Record (Cat. No.00CH37137).

[6]  Katsumi Yamashita,et al.  Expanded CLMS algorithm for double-talk echo cancelling , 1999, IEEE SMC'99 Conference Proceedings. 1999 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.99CH37028).