Abstract:

Passive sensor data association is a key issue in passive tracking systems, which can be solved by multidimensional assignment algorithm. The assignment result is depended on a selection of the cost function to some extent. However, because of the low accuracy of traditional least square estimation and the ignorance of fusion errors, the performance of the cost function based on the generalized likelihood ratio is poor. To solve this problem, a method based on distance weighted least square (DWLS) is proposed, which takes the information of distance into account, and constructs a modified cost function by integrating the fusion variance of target position. So the relationship between measurements and targets can be reflected more accurately. The simulation results show that the proposed algorithm improves the association performance with lower computational cost and has good application prospects.