A Measurement Pairing Method Based on MAP Criterion for Dense Targets in Dual-Sensor System

In dual-sensor multi-target data processing, it is crucial to pair the measurements originating from the same target. Up to now, a number of algorithms have been developed to deal with the issue. However, the local methods among them perform poorly in dense target scenarios, while common global algorithms are difficult to implement because of their huge computational complexity. In this paper, an efficient global method is proposed to deal with the measurement pairing of dense targets in a dual-sensor system. Referring to maximum a posterior probability(MAP) criterion, we show that the optimal pairing scheme is the one that minimizes the quadratic sum of the Euclidean distances between the paired measurements. Therefore, we convert measurement pairing into an assignment problem, which is a typical NP-hard issue in combinatorial optimization. The Hungarian algorithm is adopted to solve the converted problem and the simulation results verify the effectiveness of the proposed method.