Collaborative target tracking of IoT heterogeneous nodes

Abstract A sensing system consisting of a single type of sensor has numerous defects in target tracking, for example, RFID sensor network has a limited positioning accuracy and camera sensor network has a large data communication. Thus, the target tracking of heterogeneous nodes has become the present research focus. Three types of sensor nodes, namely, RFID, camera, and passive sensors, were combined in this study for fusion of heterogeneous node data. Moreover, a target tracking algorithm of heterogeneous nodes was proposed. The proposed algorithm constructed a node sensing efficacy function through particle filtering, built a sensing energy efficiency function by balancing surplus node energies, and eliminated members in the perception group that participate in the target tracking process. Experimental results indicate that in comparison with existing algorithms, the proposed algorithm has high target tracking accuracy and has reached a favorable balance with regard to tracking accuracy and network energy consumption.