Target tracking in noisy wireless sensor network using artificial neural network

The use of wireless sensor networks (WSN) in tracking applications is growing at a fast pace. In these applications, the sensor nodes discover, monitor and track an event or target object. Wireless sensor networks are by nature harsh, uncertain and dynamic, therefore there are many noise sources which malignantly impact on the performance and the efficiency of a wireless sensor network. On the other hand artificial intelligence method provides adaptive mechanisms that exhibit intelligent behavior in complex and dynamic environments like WSNs. In this paper we investigate application of artificial neural networks to tackle the noise interference in target tracking. Beacon signals help to estimate distances and learn Network Area. Computer simulations showed improvement in tracking accuracy in compare of traditional method.

I. Introduction

WSNs consist of a large number of small wireless nodes equipped with a variety of sensors including infrared, ultrasonic, pressure, cameras, sonar, and radar to support different types of sensing functionalities. Sensor nodes are fitted with at least one microcontroller, which provides the processing capability. They are also equipped with RF transceiver with usually an omnidirectional antenna to allow the communication with each other or a central unit. As for the power source, sensor nodes usually rely on small batteries with a limited lifetime. [1]. Sensors are generally deployed in an ad hoc manner in physical environment for observation of an event of interest. The sensors in the vicinity of an event must be able to monitor it and send data back to one or several sink. A sink sensor node has capability to communicate with outside world such as laptop and base station. WSNs are used in numerous applications such as environmental monitoring, habitat monitoring, prediction and detection of natural calamities, medical monitoring, and structural health monitoring [2].