Rights Creative Commons: Attribution 3.0 Hong Kong License ROBUST RECURSIVE BI-ITERATION SINGULAR VALUE DECOMPOSITION (SVD) FOR SUBSPACE TRACKING AM) ADAPTrVE FILTERING

The recursive bi-iteration singular value decomposition (Bi-SVD), proposed by Strobach [l], is en efficient and well-structured algorithm for performing subspace tracking. Unfortunatelyy, its performance under impulse noise environment degrades substantially. In this paper, a new robust-statistics-based bi-iteration SVD algorithm (robust Bi-SVD) is proposed. Simulation results show that the proposed algorithm offers significantly improved robustness against impulse noise than the conventional Bi-SVD algorithm with slight increase in arithmetic complexity. For nominal Gaussian noise, the two algorithms have similar performance. 1