Effects of the crosscorrelation on track-to-track association

In order to improve the performance of track-to-track association (T2TA) in a distributed multisensor information fusion system, this paper studies the effects of the crosscorrelation on T2TA, including the crosscorrelation between tracks for the same target and the crosscorrelation across time of the state estimates, while the latter is often neglected in practice. Considering the crosscorrelation, the exact hypothesis tests for T2TA using a single frame or multiple frames of data were presented. Then powers of different hypothesis tests for T2TA were compared through qualitative analysis and Monte Carlo simulations. The simulation results show that the crosscorrelation in T2TA should not be ignored, and the exact multiframe test would have less power than the exact single time test if the crosscorrelation is strong. To quantify the strength of the crosscorrelation, a method of calculating the correlation coefficient was presented.