时域Wiener状态滤波新方法  

New approach to Wiener state filtering in time-domain

作　　者：邓自立[1] 

机构地区：[1]黑龙江大学应用数学研究所黑龙江大学自动化系,黑龙江哈尔滨150080

出　　处：《控制理论与应用》2004年第3期367-372,共6页Control Theory & Applications

基　　金：国家自然科学基金项目(60374026);黑龙江省自然科学基金项目(F01-15).

摘　　要：基于稳态Kalman滤波器和射影理论,提出了统一和通用的时域Wiener状态滤波新方法,用它得到带非零均值相关噪声线性随机系统的渐近稳定的Wiener状态估值器和解耦Wiener状态估值器.它可统一处理状态滤波、预报和平滑问题.发现了Kalman滤波器和Wiener滤波器之间的变换关系,Wiener状态估值器可由Kalman估值器通过自回归滑动平均(ARMA)新息模型得到.一个仿真例子说明了其有效性.Based on the steady-state Kalman filter and projection theory,a new unified and general approach to the time-domain Wiener state filtering is presented,by which the asymptotically stable Wiener state estimator and decoupled Wiener state estimators are presented for linear stochastic systems with correlated noises having non-zero means.It can handle the state filtering,prediction and smoothing problems in a unified framework.The transformation relationship between the Kalman filters and Wiener filters is discovered,the Wiener state estimators can be obtained from the Kalman estimators by means of the autoregressive moving average (ARMA) innovation model.A simulation example shows its effectiveness.

关 键 词：随机系统 状态估计 WIENER滤波 KALMAN滤波 时域方法 

分 类 号：O29[理学—应用数学]