DECIMATION RATIO IN SUBBAND ADAPTIVE IDENTIFICATION OF A TIME-VARYINGAR MODEL

This paper presents some conditions to determine decimation ratio in s ubband adaptive identification of a time-varying system. It has been r eported in some experiments that a large decimation ratio is not suita ble for subband identification of a time-varying system. However, theo retical conditions for a maximally decimated ratio in a time-varying s ystem have not to date been reported. A short-time invariant model is the starting point for this research. The relation between a forgettin g factor used in the recursive weighted least-squares (RWLS) algorithm and a time variation of the time-varying system are derived. Then, ne w forgetting factors are derived for the multirate processing case. Us ing a convergence property provided by these forgetting factors, theor etical conditions are obtained having parameters estimated from decima ted signals that are as accurate as those from all observed signals. T he propriety of these conditions is demonstrated using simulations.