The error statistics of surveillance radar position measurements

This paper examines the error statistics of 2D radar position measurements when targets are extended relative to the resolution of the system. Monte Carlo simulations are used to show how the common assumption that range and angular errors are uncorrelated fail in some cases. It fails in surveillance systems that apply centroid image processing on extended targets. It is shown that there may be correlation which depends on the aspect angle of the target. Two new models that include this correlation are proposed. The proposed models can be used to improve the performance of target tracking and data fusion algorithms.