An efficient approach to detect faulty readings in long-thin wireless sensor network using fuzzy logic

Wireless sensor networks (WSNs) consist of hundreds or thousands tiny sensor nodes that work together to do some special tasks. Detecting node with faulty readings is the one challenging issue in WSNs. Most of existing algorithms use correlation of two nodes to detect faulty readings, which resulted in high computational complexity. Also, algorithms that use inverse of distance as scale of similarity have weaknesses in accuracy of calculations. This paper proposed a new fuzzy-based algorithm to detecting faulty readings in WSNs. Using an effective fuzzy inference system can improve the decision-making algorithm, which use for detecting faulty readings in WSNs. We use of average sensed values instead of single value or multi sensed value. The computation complexity of our algorithm is low and the accuracy is in a high level.