Covariance Intersection fusion Wiener signal estimators based on embedded-time-delay ARMA model

Different from those previous methods for solving the problems of time-delay systems, the time delays are embedded into the ARMA innovation models by modern time series analysis method. Using the measurement predictors and the white noise estimators, the local and the optimal information fusion Wiener signal estimators are presented for the two-sensor multichannel signal systems. Applying the CI (Covariance Intersection) method, the CI fused Wiener signal estimators are derived, which avoid the computation of the cross-covariance, therefore they can significantly reduce the computation burden. It is rigorously proved that their estimation accuracies are higher than those of the local Wiener estimators, and are lower than that of the optimal fusion Wiener signal estimators. The Monte-Carlo simulation results show that the actual accuracies of the presented CI fusion Wiener estimators approximate to those of the corresponding optimal information fusion Wiener estimators, and based on the covariance ellipse, the geometric interpretation of the accuracy relation is shown.