Abstract: This article makes a study of target track correlation with system errors existing in a radar network, analyzes theoretically how system errors affect the detected tracks of targets, and expresses the effect caused by radar system errors as rotation and mass motion between target tracks from radars. Starting from the analysis, a track registration-correlation algorithm prior to registration based on Fourier transform is presented. The algorithm describes the target track data from each radar as entile information, estimates and compensates the rotation and mass motion between the target tracks from the netted radars and fusion center by using the Fourier transform, and aligns the track data of the targets provided by the radars in the network with the fusion center, so the algorithm can associate the track accurately without prior estimation and registration of the system errors in radar network, and can provide reliable associated target tracks for later registration.

Key words: Fourier transforms, system error, sensor registration, track correlation, radar network