Experimental results of a modified architecture for oversampled, subband acoustic echo cancellers

The motivation for adaptive filtering in subbands stems from two well-known problems in LMS-based fullband adaptive filtering. First, the convergence and tracking can be very slow if the input correlation matrix is ill conditioned such as that found in speech. Second, very high order adaptive filters are computationally expensive. Adaptive filtering in subbands has been proposed to overcome these problems which are commonly encountered in acoustic echo cancellation applications. One problem with this technique is the slow, asymptotic convergence associated with oversampled architectures. A modification of the analysis bank is proposed to reduce the slow asymptotic convergence. The paper presents experimental results illustrating the benefits of this modified architecture