Efficient multi-target tracking with a two-tier hierarchical Wireless Sensor Network

Recently, a two-tier hierarchical wireless sensor network (WSN) was proposed to address the problem of multi-target tracking (MTT). The WSN estimated the multi-target states well, but kept no record of target identities. In this paper, the performance of the WSN is improved by activating certain sensors instead of all the sensors in the first level of the hierarchy at each time step. Furthermore, a track-to-estimate association scheme is also proposed to incorporate individual target identities and produce estimates of individual target tracks. The proposed tracking algorithm works well even when the target dynamics and/or measurement process is severely nonlinear. The efficiency of the algorithm is demonstrated in the application of bearing and signal strength tracking.