Energy-efficient real-time object tracking in multi-level sensor networks by mining and predicting movement patterns

A number of studies have been written on sensor networks in the past few years due to their wide range of potential applications. Object tracking is an important topic in sensor networks; and the limited power of sensor nodes presents numerous challenges to researchers. Previous studies of energy conservation in sensor networks have considered object movement behavior to be random. However, in some applications, the movement behavior of an object is often based on certain underlying events instead of randomness completely. Moreover, few studies have considered the real-time issue in addition to the energy saving problem for object tracking in sensor networks. In this paper, we propose a novel strategy named multi-level object tracking strategy (MLOT) for energy-efficient and real-time tracking of the moving objects in sensor networks by mining the movement log. In MLOT, we first conduct hierarchical clustering to form a hierarchical model of the sensor nodes. Second, the movement logs of the moving objects are analyzed by a data mining algorithm to obtain the movement patterns, which are then used to predict the next position of a moving object. We use the multi-level structure to represent the hierarchical relations among sensor nodes so as to achieve the goal of keeping track of moving objects in a real-time manner. Through experimental evaluation of various simulated conditions, the proposed method is shown to deliver excellent performance in terms of both energy efficiency and timeliness.