Light-Weight Target Tracking in Dense Wireless Sensor Networks

Wireless sensor networks (WSNs) are widely used in detecting, locating and tracking the moving objects. However, some of the cheap, low-powered and energy-limited sensors that are deployed in large areas may use up their energy, which leads to the whole network failure finally. In order to reduce the energy consumption and prolong the network lifetime, (a) a new light-weight and energy-efficient locating scheme is proposed to estimate the current target location; (b) an energy-efficient parallel target tracking algorithm based on gene expression programming (P-GEP) is put forward for collaboratively mining the trajectory of the moving target, then, the future locations where the target will appear can be predicted within a given prediction accuracy, and sensor nodes that are far away from the predicted locations can be scheduled to be on/off finally; (c) the sliding window technique is adopted to discard some of the historical locations to balance the trade-off between the prediction accuracy and the energy consumption during the trajectory mining process. Extensive simulations show that the proposed methods can greatly improve the tracking efficiency and extend the network lifetime by around 39.4% and 94.2% compared with other tracking algorithms, i.e., EKF and ECPA.