自校正信息融合Wiener反卷积滤波器

对于含有未知模型参数和噪声统计的多传感器信号反卷积系统，应用现代时间序列分析方法，基于自回归滑动平均（ARMA）新息模型参数的在线辨识，可在线估计噪声方差，进而提出了自校正信息融合Wiener反卷积滤波器。证明了它的渐近最优性，即若ARMA新息模型参数估计是一致的，则它收敛于当噪声方差已知时的最优融合Wiener反卷积滤波器。同单传感器情形相比，它可提高滤波精度。一个带三传感器的反卷积系统的仿真例子说明了其有效性。

Self-tuning Information Fusion Wiener Deconvolution Filter

For the multisensor signal deconvolution systems with unknown model parameter and noise statistics, by the modern time series analysis method, based on the on-line identification of the autoregressive moving average (ARMA) innovation model parameters, the noise variances can on-line be estimated, and a self-tuning information fusion Wiener deconvolution filter is presented. Its asymptotic optimality is proved, i.e. if the parameter estimation of the ARMA innovation model is consistent, then it will converges to the optimal fusion Wiener deconvolution filter with known noise statistics. Compared with the single-sensor case, its accuracy is improved. A simulation example for a deconvolution system with 3-sensor shows its effectiveness.