A novel interpretation and performance analysis of subband adaptive filter algorithms

We have recently introduced a framework within which a large number of classical adaptive filter algorithms can be viewed as special cases. Within this framework we have demonstrated that performance results for classical adaptive filters such as the least mean square (LMS), normalized least mean square (NLMS), affine projection (AP), and recursive least square (RLS) algorithms can be obtained as special cases of more general results obtained within the aforementioned framework. In this paper we extend our previous work by showing how modern subband adaptive filters can be formulated in such a way that the results of the generalized theory are directly applicable also to this important class of adaptive filters. We also point out that the results obtained are useful in the selection/design of filter banks for use in subband adaptive filters.