Frequency-weighted prefiltering for Kalman filter based target tracking

The effect of frequency-weighted prefiltering of the innovation sequence and the noisy measurement sequence on the performance of a Kalman filter-based target tracking algorithm has been studied. The simulation results of a conventional Kalman filter using a noisy measurement sequence have been compared to those of the Kalman filter using a prefiltered innovation sequence and a prefiltered measurement sequence. The results of the prefiltering of the innovation sequence for a particular frequency-band show a performance enhancement for the new variant Kalman filter in that frequency region, as compared to that of the standard Kalman filter. However, it is achieved at the cost of a phase lag. The paper also presents an analysis of the frequency-weighted prefiltering of the measurement sequence.