Wireless sensors and neural networks for intruders detection and classification

The problem of automating the detection and classification of intruders in vast and hard to reach terrains has been an active research topic over the past decade. In this paper we present a new paradigm of intrusion detection and classification by combining the sensing potential of self-configuring and instantly-deployable wireless sensor networks (WSN's) with the reasoning capabilities of artificial neural networks (ANN's). The implemented system is referred to by WINIDC (Wireless Sensors and Neural Networks for Intruders Detection and Classification). WINIDC has been designed to detect intruders of many sizes and types. It can be considered as a smart and economical solution for the addressed surveillance problem. It is smart as it utilizes artificial neural networks to classify the different types of intruders. It is economical as it utilizes small, cheap wireless sensor nodes that are self-powered and self-configuring. Traditional solutions to such surveillance problem are expensive, complex and hard to maintain. The results show that integrating ANNs along with WSNs allow for real-time and accurate detection and classification of intruders. The proposed system can be utilized in monitoring intruders to protect assets and national borders.