An efficient robust stereophonic acoustic echo canceler based on parallel subgradient projection techniques

A robust adaptive filtering algorithm was established recently (see Yamada, I. et al., Proc. IEEE ICASSP, 2001), based on the interactive use of statistical noise information and ideas developed originally for efficient algorithmic solutions to convex feasibility problems. The algorithm is computationally efficient and robust to noise because it requires only an iterative parallel projection onto a series of closed half spaces highly expected to contain the unknown system to be identified and is free from the computational load of solving a system of linear equations. We show the potential applicability of the algorithm to the stereophonic acoustic echo cancellation (SAEC) problem that has commonly been recognized as one of the most challenging adaptive filtering problems due to the nonuniqueness problem. The numerical examples demonstrate that the proposed algorithm realizes stereophonic acoustic echo cancelers (SAECRs) showing dramatically fast and stable convergence for highly coloured excited speech-like input signals in severely noisy environments.