An adaptive filter design based on error estimation

In this paper, we present a novel weight adjustment method of adaptive filter that is based on error estimation. The new adaptive filter's adjustment of weight coefficient doesn't directly computed form the last time's error value, but is estimated based on history data. The proposed adaptive filter's structure is simple and can be widely applied. At the latter part of this paper, the stabilization of this new adaptive filter is proved. Simulation results show it has fast convergence rate and the better precision.