An Empirical Study on Real-Time Target Tracking with Enhanced CPA Algorithm in Wireless Sensor Networks
Qing Yang, Alvin Lim, Kenan Casey and Raghu-Kisore Neelisetti
The original CPA (closest point of approach) algorithm can localize and track moving targets within a wireless sensor network that has a specific node configuration with respect to the target trajectory. As a target moves through a large network of randomly deployed sensors, the configuration of the nodes triggered along the target trajectory may not meet this requirement and will not localize and track the target correctly. To address this problem, we propose the enhanced CPA (ECPA) algorithm that can correctly compute the bearing of the target trajectory, the relative position between the sensors and the trajectory, and the velocity of the target. To validate ECPA, we designed and implemented the algorithm over a data-centric sensor network. This ECPA software also communicates over a collaborative mixed wireless sensor network with control software for controlling video sensor nodes that capture real-time images or video of the target at its predicted location. Our experimental results show that we can achieve our goals of detecting the target and calculating its location, velocity and direction of travel with reasonable accuracy. In addition, results from the target detection algorithm can be used to predict the future target location so that a camera can capture video of the moving target for identification purposes.
Keywords: Real-time target tracking, closest point of approach algorithm, collaborative mixed wireless sensor networks, distributed sensors, high-speed target tracking, distributed algorithm.