Wiener滤波新方法

A New Approach to Wiener Filtering

基于ＡＲＭＡ新息模型提出Ｗｉｅｎｅｒ滤波的一种新的时域方法。对带状态空间模型的线性离散随机系统，提出一种Ｗｉｅｎｅｒ状态估值器，可统一处理最优滤波、平滑和预报问题，且具有渐近稳定性，避免了求解Ｄｉｏｐｈａｎｔｉｎｅ方程和Ｒｉｃｃａｔｉ方程。

Based on the ARMA innovation model, a new time-domain approach to Wiener filtering is presented. The Wiener state estimators are presented for linear discrete stochastic systems with the state-space models. They can handle the optimal filtering, smoothing and prediction problems in a unified framework. They have the asymptotic stability, and avoid the solution of the Diophantine equations and Riccati equations. A simulation example shows the usefulness of the proposed results. 