Object Tracking Sensor Networks using Sequential Patterns in an Energy Efficient Prediction Technique

Wireless Sensor Networks applications are attracting more and more research, especially in energy saving techniques/architectures which is the focal point of most researchers in this area. One of the most interesting applications of Wireless Sensor Networks is the Object Tracking Sensor Networks which are used mainly to track certain objects in a monitored area and to report its location to the application's users. This application is a major energy consumer among other Wireless Sensor Networks applications. There have been many techniques that assist in delivering the required data while maintaining a lower energy consumption than the early approaches. Our approach revolves around the ability to predict the objects' future movements in order to track it with the minimum number of sensor nodes, while keeping the other sensor nodes in the network in a sleep mode. Thus, achieving our goals while reducing significantly the network's energy consumption. The prediction technique used in our proposed solution is based on the inherited patterns of the objects' movements in the network and how to utilize data mining techniques, such as Sequential Patterns, in order to predict which sensor node the moving object will be heading next. We propose the Prediction-based Tracking technique using Sequential Patterns (PTSP), which is designed to achieve significant reductions in the energy dissipated by the OTSN network, while maintaining an acceptable missing rate levels. PTSP is tested against basic tracking techniques in order to determine the appropriateness of PTSP in various circumstances. We also test PTSP against some OTSN impacting factors, such as number of tracked objects, object speed, sampling duration and sampling frequency. We also test 3 different missing object recovery mechanisms implemented in PTSP to determine which is the most energy conservative. The experimental results had shown that PTSP outperformed all the other basic tracking techniques and contributed remarkable amounts of savings in terms of energy consumption of the entire network even through different circumstances.