Novel Fault Management Framework Using Markov Chain in Wireless Sensor Networks: FMMC

Due to using wireless sensor nodes (WSNs) in inaccessible areas and applying limitations in making nodes to reduce costs, these networks are prone to faults. The performance and efficiency of the networks should not be affected by faults so that fault tolerance is a required feature. To improve fault tolerance and ensure optimal performance of network, fault detection and recovery or fault management is essential. This paper represents a fault management framework based on clustering algorithms to detect and recover faults in WSNs. In the proposed method, on self-detecting and diagnosing faults, all faults are modeled through Markov chain. In recovery phase, the status of nodes is defined based on the type of fault so that the faults are recovered. The results of simulation reveal that the proposed fault management framework results in improved energy consumption, increased number of alive nodes, improved detection accuracy, and reduced false alarm rate compared with other frameworks.