A novel hierarchical fault management framework for wireless sensor networks: HFMF

Wireless sensor nodes (WSNs) are employed to collect data for control and supervisory purposes in inaccessible areas. Applying sensors in inaccessible areas and their hardware limitations result in occurring faults and non-renewability of energy. Thus networks need a fault tolerant method to continue their optimal activity in the presence of faults. Here, through improving energy consumption and fault management, we propose a new hierarchical fault management framework to overcome the limitations. The proposed method complies with clustering algorithms. Hence, due to the importance of cluster head nodes a backup is employed to replace faulty ones. Also, data correlation among cluster members is used to cellularize the cluster nodes virtually. In this process, a cell remains in active mode as the representative of cell, and others are in sleep mode as spare ones. The purpose of this mechanism is to reduce the number of active cluster nodes and detect intermittent faults. To detect the permanent faults of nodes, self-detection method has been used. In addition, the proposed framework diagnoses and recovers faults in communication links between nodes. The results of simulation reveal that the proposed framework leads to improved energy consumption, alive nodes, and fault detection accuracy compared with other frameworks.