An improvement of the "extended LMS" algorithm

作者：Gao, Y^*; Xie, SL

来源：Dynamics of Continuous Discrete and Impulsive Systems: Series B; Applications and Algorithms , 2003, 23-29.

• ELMS algorithm
• stereophonic acoustic echo cancellation
• cross-correlation

ELMS algorithm presented by J.Benesty(1995) is the first two-channel adaptive filtering algorithm that takes into account the cross-correlation between the two input signals. The algorithm does not preprocess input signals, so it does not degrade the quality of the speech. However, A lot of computer simulation results show that ELMS algorithm has a bad performance. Are there the effective algorithms based on J.Benesty's idea or not? In this paper, We answer the question. We first analyze the ELMS algorithm, then present a new algorithm by modifying the block matrix used in ELMS algorithm to approximate input signals self-correlation matrix. The computer simulation results indicate that the improved algorithm has a better behavior than the ELMS algorithm.

• 出版日期2003
• 单位广州大学; 华南理工大学