Genetic Fuzzy Tree Based Node Moving Strategy of Target Tracking in Multimodal Wireless Sensor Network

Abstract

Multimodal wireless sensor networks (WSNs) consist of variable types of sensor nodes can do many important applications, such as environment monitoring, health care, and target tracking. In this paper, we utilize the multimodal WSN to keep track of targets in 3-D space. All the nodes in the network are different. They can be pyroelectric infrared nodes, radio frequency nodes, nodes with cameras, and so on. Based on this, we assume each of them varies in battery, mobility, and target recognition performance. To achieve a better performance of target tracking in the multimodal WSN, we propose a node moving strategy. Genetic fuzzy tree is employed in this paper. It is a two-layer fuzzy tree system (FTS) optimized by the genetic algorithm. The first layer gives a score to each node and selects the moving nodes. The second layer then controls the moving distances. The Pittsburgh genetic algorithm is used to optimize the whole rule base and data base of the FTS. Simulation results prove that the tracking error can be reduced applying our approach.