A Belief Propagation Algorithm Based on Game Theory

According to the operational requirements and task assignment criteria of modern warfare, it is necessary to deploy a large number of sensor nodes on the battlefield. This introduces the needs for radar resource management (RRM). Considering that the traditional RRM algorithms only treat multiple targets as relatively independent individuals and track the targets independently, and hence these RRM algorithms don’t involve clutter processing and data association problems. In this paper, the game theory (GT) is introduced in the multi-sensor multi-target tracking (MMT) algorithm based on the Belief Propagation (BP) to realize RRM. A GT-based BP (GT-BP) algorithm is proposed to realize target allocation and clutter filtering in clutter environment. The simulation results show that the GT-BP algorithm realizes faster algorithm response and reduces the resource consumption and bandwidth requirements of the system while ensuring the tracking accuracy is close to that of conventional BP algorithm, when there exists a maximum limitation of radar resource in the system.