Fault Aware Wireless Sensor Networks

Wireless Sensor Networks (WSNs) collect information about the physical environment aiding to a wide variety of applications ranging from target detection to monitoring of harmful chemical gases. The drive to scale down the system size and cost has resulted in constraints in the quality of components. Reliable and accurate performance of sensors is necessary in critical applications. In this paper, we present statistical data analysis and signal processing techniques at the sensor node level to detect sensor faults and to eliminate noise. We also present the simulation of the proposed algorithm using real sensor data and demonstrate that the algorithm can distinguish between sensor faults and environmental events. Furthermore, we describe the real-time implementation of the developed algorithm. The information regarding the faulty sensor is broadcast to all nodes and the central processing base station node thereby achieving autonomous node level operation and a complete fault aware system.