RLS algorithm for a new subband adaptive structure with critical sampling

Abstract

In a recent publication new family of adaptive structures with critical sampling of the subband signals, which yields exact modeling of FIR systems, was obtained. In this paper an algorithm based on the recursive least-squares (RLS) algorithm is derived for the updating of the adaptive coefficients of this new subband structure. An efficient implementation of the resulting structure is discussed, leading to computational complexity savings. Computer simulations illustrate the convergence behavior of the proposed algorithm.