Network lifetime improvement using landmark-assisted mobile sink scheduling for cyber-physical system applications

Abstract

Improving the lifetime of a wireless sensor network is a challenging issue in the development of energy-efficient cyber-physical systems, especially in the design of sensor data aggregation protocols. Mobile sinks are typically used in this context to increase the lifetime of the sensor network in a cyber-physical system. In this work, a landmark-assisted mobile sink scheduling scheme is proposed for designing an energy-efficient data aggregation protocol. A landmark-node identification method along with network clustering is first developed using random walks over network graphs. Subsequently, using the locations of landmark-nodes, a mobile sink scheduling method is developed for sensor data aggregation in an extended Kalman filtering framework. This method is robust since location information of the landmark-nodes is also continuously updated in the Kalman filtering framework. Experiments on identifying landmark-nodes and mobile sink scheduling are conducted on the Intel Berkeley Research Lab dataset and over a real WSN test bed with a Kobuki–Yujin robot. Experimental results obtained for network lifetime analysis and energy-efficiency for various network configurations indicate an improvement over standard methods and are motivating enough for use in practical cyber-physical system applications.