Covariance Intersection Fusion and Three Weighted Fusion Deconvolution Kalman Smoothers for ARMA Signal

Abstract

For multisensor autoregressive moving average (ARMA) signal deconvolution system, based on Kalman filtering and the augmented state space model, a covariance intersection (CI) fusion deconvolution Kalman smoother without cross-covariance of the local estimation errors is presented, which can significantly reduce the computational burden. Under the unbiased linear minimum variance (ULMV) criterion, three deconvolution Kalman fusers weighted by matrices, diagonal matrices or scalars are also presented respectively. It is proved that the accuracy of the CI fuser is higher than that of each local Kalman smoother and is lower than that of the optimal Kalman fuser with matrix weights. The geometric interpretation of accuracy relations based on covariance ellipses is given. A Monte-Carlo simulation example verifies the correctness of the theoretical accuracy relations, and shows that the actual accuracy of the CI fuser is close to that of the fuser with matrix weights, so that it has higher accuracy and good performance.