An Efficient Approach to Detect Faulty Readings Using Fuzzy Logic and Read Vector Along The Substrate Access Wireless Long-Thin Sensor Networks

Abstract

Wireless sensor networks (WSN)s consist of hundreds or thousands tiny nodes that work together and connected to each other to do some special tasks. Detecting nodes with faulty readings is one the important issues in WSNs. The existing algorithms to detect faulty readings using weighted voting and are divided in tow category; existing algorithms that using correlation of two nodes read vectors as weight and algorithms that using inverse of distance as weight. The first category algorisms are costly and second category algorithms have weaknesses in accuracy of calculations. This paper proposes a new fuzzy-based algorithm to detecting faulty readings in WSNs. We propose the new method based on LTN however, it is applied in the most of WSN structure. Using an effective fuzzy inference system can improve the decision-making algorithm, which using for detecting faulty readings in WSNs. We use of entire read vector without any additional cost to the network. The experimental results show that the proposed algorithm imposes very low cost to the network; in addition, the accuracy of the results is improved when compared to the other algorithms