A NEW APPROACH OF OPTIMAL SELF-TUNING FILTERING FOR SINGLE OUTPUT SYSTEMS AND ITS APPLICATION TO TRACKING SYSTEMS

Using the modern time series analysis method, this paper presents a new self-tuning filtering approach for single output systems with unknown model parameters and unknown noise statistics. A self-tuning filter with asymptotically optimal behaviour is given by computing the self-tuning output predictors and observation noise filter based on ARMA innovation model. An application example to tracking system is given. Simulation results show usefulness of the new approach.