Diagnosis of fault node in wireless sensor networks using adaptive neuro-fuzzy inference system

Wireless networks (WSN) are sometimes inaccessible to humans and can be found in deep forests, diverse risky fields, high-mountainous areas and even underwater environments at least. Owing to continuous or instant changes in environmental parameters, failures in sensor networks are unavoidable. A fault can lead to a misread, which can harm the environment economically and physically. In several primary applications, wireless sensor networks (WSNs), including battlefield control, environmental monitoring and forestry fire monitoring are widely used. Research is being conducted to reduce energy usage, improve the longevity and life of the WSN network. This paper provides an ANFIS method for the summation, based on the neuro-fuzzy optimization model estimator of defect-tolerant WSNs. For WSNs, the scheme proposes the use of an intra-cluster and inter-cluster failure detection estimator adaptive neuro-fuzzy inference system (ANFIS). Traditional detection strategies for malfunction include a centralized mechanism for avoiding failure detection when the principal defect detector fails. In typical methods, the fault nodes are discarded. To solve these problems, we propose the identification and classification of distributed fault nodes by means of adaptive neuro-fuzzy inference method.