Self-tuning information fusion Wiener predictor and its convergence
Qiang Wang · Control theory & applications · 2009
For the multisensor systems with correlated measurement noises and unknown noise statistics,the on-line noise statistics estimators are obtained by the correlation method.Under the linear minimum variance optimal information fusion criterion weighted by scalars for components,by the modern time series analysis method,a self-tuning decoupled fusion Wiener predictor is presented based on the identification of the moving average(MA) innovation models.By using the dynamic error system analysis(DESA) method,it is proved that the self-tuning fusion Wiener predictor converges to the optimal fusion Wiener predictor,so that it has the asymptotic optimality.Its accuracy is higher than that of each local self-tuning Wiener predictor.Its algorithm is simple,and is suitable for real time applications.A simulation example for a target tracking system shows its effectiveness.