An Energy-Efficient Object Tracking Algorithm in Sensor Networks

Qianqian Ren, Hong Gao, Shouxu Jiang, Jianzhong Li

Abstract: Object tracking needs to meet certain real-time and precision constraints, while limited power and storage of sensors issue challenges for it. This paper proposes an energy efficient tracking algorithm (EETA) that reduces energy consumption in sensor network by introducing an event-driven sleep scheduling mechanism. EETA gives tradeoffs between real time and energy efficiency by making a maximum number of sensor nodes outside tracing area stay asleep. EETA reduces the computation complexity on sensors to O(N)by formulating the location predication of an object as a state estimation problem of sensor node, instead of building a complex model of its trajectory.EETA locates the object using modified weighted centroid algorithm with the complexity of O(N). We evaluate our method with a network of 64 sensor nodes, as well as an analytical probabilistic model. The analytical and experimental results demonstrate the effectiveness of proposed methods.