Distributed Traversal Based Fault Diagnosis for Wireless Sensor Network

Abstract

Wireless Sensor Networks (WSNs) have become a new information collection and monitoring solution for the various application. Faults occurring to sensor nodes are prevalent due to the sensor device itself and the harsh environment, where the sensor nodes are deployed. To ensure the quality of service and to avoid further degradation of service, it is necessary for the WSN to be able to tolerant of the faulty nodes present in the network. The fault diagnosis techniques are classified based on the methods they employ to determine the faults. In this paper, we have proposed a traversal-based diagnosis algorithm that seeks to diagnose both permanent as well as intermittent fault in WSN. The proposed algorithm employs a special node called an anchor node to traverse the field. The traversal of the field is decided by a proposed traversal algorithm taking into consideration the length and breadth of the sensor field, and the transmission range of the nodes. The anchor node stops at defined positions in the deployment field where it executes the fault diagnosis algorithm taking into consideration the normal sensor nodes which are in its range. The diagnosis algorithm uses a timeout mechanism to identify hard faults and adjusted boxplot method to identify permanent and intermittent faults in the network. The adjusted boxplot method takes into consideration the skewness of the data generated by the nodes in the sensor field. The faulty sensor nodes are classified by using a Feed Forward Neural Net (FFNN) model with Gravitational Search (GS) learning algorithm. The proposed algorithm is implemented in the Omnet++ environment which shows very promising results. The performance parameters, such as detection accuracy, false alarm rate, false positive rate, and energy consumption of the proposed algorithm show significant improvement over the existing algorithms.