广义系统Wiener状态滤波新算法  

New algorithm to Wiener state filtering for descriptor systems

作　　者：许燕[1] 邓自立[2] 

机构地区：[1]北京印刷学院基础部,北京102600 [2]黑龙江大学自动化系,黑龙江哈尔滨150080

出　　处：《控制与决策》2003年第3期328-331,共4页Control and Decision

基　　金：国家自然科学基金资助项目 ( 697740 19);黑龙江省自然科学基金资助项目 ( F0 1- 15 )

摘　　要：应用时域上的现代时间序列分析方法 ,基于 ARMA新息模型和白噪声估计理论 ,由一种新的非递推最优状态估值器的递推变形 ,提出了广义系统 Wiener状态滤波的一种新算法 ,它可统一处理滤波、平滑和预报问题 ,且具有渐近稳定性。同某些算法相比 ,它避免了求解 Riccati方程和 Diophantine方程 ,且避免了计算伪逆 ,因而减小了计算负担。A new algorithm to Wiener state filtering for descriptor systems is presented by using the modern time-series analysis method in the time domain. The algorithm is based on the autoregressive moving average (ARMA) innovation model and white noise estimation theory, and exploits a recursive version of a new non-recursive optimal state estimators. It can handle the filtering, smoothing and prediction problems in a unified framework, and has the asymptotic stability. Compared with some existing algorithms, it avoids solving the Riccati equations and Diophantine equations, and avoids the calculation of the pseudo-inverse, which reduces the computational burden. A simulation example shows its effectiveness.

关 键 词：广义系统 Wiener状态估值器 滤波 平滑 预报 现代时间序列分析方法 

分 类 号：O211.64[理学—概率论与数理统计]