Tracking of a Target with Correlated Measurement Noise
Kyung Youn Kim · Journal of Electrical Engineering and Information Science · 1998
For many tracking applications such as modern radar systems, the measurement errors on successive observations are correlated significantly when the measurement frequency is high. Using nth-order AR model for the sequentially correlated measurement noise, we present a generalized decorrelation process and a simple parameter estimation technique to enhance the tracking performance. From Monte Carlo simulations for the target with/without maneuver, it can be seen clearly that the tracking performance can be improved considerably under the correlated measurement noise circumstances.