Adaptive Filtering Algorithm by Using Modified Decorrelation

Adaptive signal processing, an important branch in the Signal and Information Processing field, is widely used in the signal processing of digital communication. This paper is devoted to the study of an adaptive filtering algorithm by using modified decorrelation. The structure and theoretic analysis of this algorithm is presented. And its optimal weight vector is the solution of generalized Wiener-Hopf equation. The convergence speed and the stable misadjustment of this algorithm are compared with the algorithm in reference [7] for various nonnegative constants for processing uncorrelated signals. In computer simulations, adaptive echo canceller is used to evaluate the performances of this algorithm, the algorithm in reference [7] and DCR-LMS algorithm. Experimental results indicate that the improved algorithm performs best compared with the other two algorithms for processing strong correlated signals.