Distributed fault detection of wireless sensor networks

Abstract

Wireless Sensor Networks (WSNs) have become a new information collection and monitoring solution for a variety of applications. Faults occurring to sensor nodes are common due to the sensor device itself and the harsh environment where the sensor nodes are deployed. In order to ensure the network quality of service it is necessary for the WSN to be able to detect the faults and take actions to avoid further degradation of the service. The goal of this paper is to locate the faulty sensors in the wireless sensor networks. We propose and evaluate a localized fault detection algorithm to identify the faulty sensors. The implementation complexity of the algorithm is low and the probability of correct diagnosis is very high even in the existence of large fault sets. Simulation results show the algorithm can clearly identify the faulty sensors with high accuracy.

Robust management of outliers in sensor network aggregate queries

“…When node S3 receives the values of its children nodes, the readings x6 of node S6 appear to be suspicious, since no other node in that subtree is aware of the fire. If one is to reject that reading (for instance using a voting protocol [3]), the monitoring application will lose a crucial observation. Techniques based on smoothing [14,15] will also obscure the outcome, especially if a lot more nodes are rooted at node S3.…”

“…Our methods differ from these techniques in that we do not try to "mask" abnormal readings, but instead promote them into first class citizens, on par with the requested aggregates, and make them available to the monitoring application. The work in [3] addresses the problem of identifying faulty sensors using a localized voting protocol. In section 2 we saw that local voting schemes are prone to errors when nodes that observe interesting events are not in direct communication.…”

“…In section 2 we saw that local voting schemes are prone to errors when nodes that observe interesting events are not in direct communication. Furthermore, while [3] requires a separate costly process that needs to run periodically in order to evaluate the network, our framework allows us to capture faulty as well as outlier nodes in real time during query evaluation and piggyback this information on messages used in query processing. Similarly, the work in [23] computes outliers in sensor networks in a manner that is decoupled from query processing and can thus lead to significant energy drain on the nodes.…”

Experiences and issues for environmental engineering sensor network deployments

“…Firstly, the cost of WSN devices means that devices can be destroyed or damaged without affecting the accuracy of data collection. Secondly, a large amount of effort [16], [17] in distributed systems research has focused on highly fault tolerant network stacks. WSN routing protocols are designed with fault tolerance and self-forming characteristics as primary goals which ensures that the failure of nodes will not alter the performance of the networks.…”

Remote Industrial Sensor Network Monitoring Using M2M Based Ethical Sniffers

“…Because of their tight coupling to often harsh physical environments, WSNs and other networks used in extreme conditions (e.g., in space [11]) are more likely to experience anomalies related to connectivity or hardware failures than conventional networks. Recent work also focuses on devising detection strategies that target network level [8,12], data level [13,14], or node and data level [15,16] anomalies.…”