A Novel Normalization and Regularization Scheme for Broadband Convolutive Blind Source Separation

Abstract

In this paper we propose a novel blind source separation (BSS) algorithm for convolutive mixtures combining advantages of broadband algorithms with the computational efficiency of narrowband techniques. It is based on a recently presented generic broadband algorithm. By selective application of the Szegö theorem which relates properties of Toeplitz and circulant matrices, a new normalization is derived which approximates well the exact normalization of the generic broadband algorithm presented in [2]. The new scheme thus results in a computationally efficient and fast converging algorithm while still avoiding typical narrowband problems such as the internal permutation problem or circularity effects. Moreover, a novel regularization method for the generic broadband algorithm is proposed and subsequently also derived for the proposed algorithm. Experimental results in realistic acoustic environments show improved performance of the novel algorithm compared to previous approximations.