Energy Efficient Target Tracking in Heterogeneous WSNs Using a Combination of Activation Mechanism and Prediction Method

Tracking a mobile target is one of the primary applications in Heterogeneous Wireless Sensor Networks (heterogeneous WSNs). On the one hand, tracking without complex processing as well as achieving a high degree of accuracy and energy efficient consumption are critical requirements for these applications in a network area. On the other hand, artificial intelligence method provides adaptive mechanisms that present intelligent behavior in complex and dynamic environments like WSNs. In this paper, artificial neural networks are deployed to estimate target location. For this purpose, suggested beacon signals are able to facilitate distances estimation by which the network area is learned. Moreover, by analyzing the most probable region predicted for the next target location, a tracking window is created and sensors are dynamically clustered. Afterwards, sensors turn to the required mode to both preventing energy consumption and performing appropriate actions. The simulation results demonstrate the effectiveness of the proposed method.