自校正信息融合Wiener滤波器及其收敛性  

Self-tuning Information Fusion Wiener Filter and Its Convergence

作　　者：王伟玲[1] 王强[1] 邓自立[1] 

机构地区：[1]黑龙江大学自动化系,哈尔滨15008

出　　处：《科学技术与工程》2009年第3期539-544,557,共7页Science Technology and Engineering

基　　金：国家自然科学基金(60874063)资助

摘　　要：对带未知噪声统计的多传感器系统,用求解相关函数矩阵方程组的方法得到噪声统计在线估值器,并提出了自校正Lyapunov方程。用现代时间序列分析方法,基于滑动平均(MA)新息模型的辨识,在按分量标量加权线性最小方差最优信息融合准则下,提出了自校正分量解耦融合Wiener滤波器,并用动态误差系统分析(DESA)的方法证明了自校正Lyapunov方程的收敛性,进而证明了自校正融合Wiener滤波器收敛于最优融合Wiener滤波器,因而具有渐近最优性。它的精度比每个局部自校正Wiener滤波器精度都高,且算法简单,便于实时应用。一个目标跟踪系统的仿真例子说明了其有效性。For the multisensor systems with unknown noise statistics,based on the solution of the matrix equations for correlation function, the on-ling noise statistics estimators can be obtained, and a serf-tuning Lyapunov equation is presented. Futher using the modem time series analysis method, based on the identification of moving average （MA） innovation models, under the linear minimum variance optimal information fusion criterion weighted by scalars for components, a self-tuming decoupled fusion Wiener filter is presented. The convergence of self-tuning Lyapunov equation is proved by the dynamic error system analysis（DESA） method, futher it is proved that the self- tuning fusion Wiener filter converges to the optimal fusion Wiener filter, so that it has the asymptotic optimality. Its accuracy is higher than that of each local self-tuning Wiener filter, the algorithm is simple, and is suitable for real time applications. A simulation example for a target tracking system shows its effectiveness.

关 键 词：多传感器信息融合 噪声统计估计 自校正Lyapunov方程 自校正Wiener滤波器 收敛性 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]