基于现代时间序列分析方法的多传感器信息融合Wiener滤波器

【作者】 李云；

【导师】 邓自立；

【作者基本信息】 黑龙江大学 ， 控制理论与控制工程， 2005， 硕士

【摘要】 应用现代时间序列分析方法,基于自回归滑动平均(ARMA)新息模型、白噪声估值器和观测预报器,在按标量加权线性最小方差最优信息融合准则下,提出了多传感器信息融合单通道白噪声Wiener反卷积滤波器,在按矩阵,按对角阵和按标量加权最优融合准则下,提出了带白色观测噪声和带MA有色观测噪声的ARMA信号的多通道多传感器信息融合Wiener滤波器,提出了多通道多传感器ARMA信号信息融合Wiener反卷积滤波器和多传感器信息融合Wiener状态滤波器。它们可统一处理融合滤波、预报和平滑问题。为了计算最优加权,提出了局部估计误差互协方差的计算公式。同单传感器情形相比,可提高滤波精度。本文提出的方法避免了求解Riccati方程和Diophantine方程,可减小在线计算负担。大量的跟踪系统仿真例子和数值仿真例子说明了其有效性和正确性,且说明了三种加权融合滤波器的精度无显著区别。因而采用按标量加权融合滤波器可显著减小计算负担,便于实时应用。更多还原

【Abstract】 Using the modern time series analysis method , based on the autoregressive moving average (ARMA) innovation model, white noise estimator and measurement predictor, under the linear minimum variance optimal information fusion criterion weighted by scalars, the multisensor information fusion single channel white noise deconvolution Wiener filter is presented. Under the optimal fusion rules weighted by matrices, diagonal matrices and scalars, the multisensor information fusion Wiener filters for multichannel ARMA signals with white observation noise and with moving average (MA) color observation noise are presented , the multichannel multisensor information fusion Wiener deconvolution filter is presented, and multisensor fusion Wiener state filter is also presented. They can handle the fused filtering , smoothing and prediction problems in a unified framework. The formulas of computing the variance and cross-covariance matrices among local estimation errors are presented, which are applied to compute the optimal weights. Compared with the single sensor case, the filtering accuracy is improved. The proposed methods, avoid the Riccati equation and Diophantine equation and can reduce the on-line computational burden. Many simulation examples for the target tracking system and numerical simulation examples show their effectiveness, and show that the accuracy distinction for three kinds of fused filters is not obvious, so that the fused filter weighed by scalars can obviously reduce the on-line computational burden, and is suitable for real time applications.更多还原