Research On Fault Diagnosis Algorithms Based On Neighbor Coordination In WSN

The wireless sensor network is a multi-hop,self-organizing network which consists of a large number of sensor nodes with computing,storage and wireless communication capabilities.With the advantages of low cost and low power consumption of sensor nodes,wireless sensor network has been widely used in many fields such as national defense and military,agricultural cultivation,medical and health,traffic management and so on.The sensor nodes in wireless sensor networks usually deployed in unattended and possibly hostile environments.The ill-disposed environment affects the monitoring infrastructure that includes the sensor nodes and the network.Sensor node's failures can seriously affect the stability and the lifetime of wireless sensor work.Therefore,the research of fault diagnosis is a very import and necessary subject for wireless sensor network.The fault diagnosis of the wireless sensor network can detect and replace the fault node in the system timely,which can effectively improve the stability and lifetime of the network.Many researchers have made a lot of research of fault diagnosis.However,few researchers are concerned about the adverse effects of the complexity deployment environment on the diagnostic results.At the same time,few scholars focus on the energy consumption of nodes during the diagnosis process in current researches.For the above weaknesses,we do the following work.(1)For the problem that it is difficult to accurately detect the faulty sensor node in complex environment,we presents a fault diagnosis algorithm based on clustering for wireless sensor network.Firstly,the nodes with similar characteristics in the wireless sensor network are classified into a class.Then,base on spatiotemporal correlation of the collected data,we diagnosis suspicious node by the neighbor nodes through two rounds.We have analyzed the performance of the diagnosis algorithm in this paper.The simulation results show that the algorithm designed in this paper has better diagnostic performance in complex environment than the traditional diagnosis algorithm.(2)For the problem that few people focus on the energy consumption of nodes during the diagnosis process in current researches,we propose an energy efficiently diagnosis algorithm for wireless sensor network in this paper.Compared with the traditional node fault diagnosis algorithm,the algorithm in this paper has taken into account the factors such as communication quality and residual energy,which can effectively reduce the number that the node join diagnosis with poor communication quality and low residual energy.In this paper,we analyze the energy consumption of the participating nodes when the system runs the diagnostic algorithm.Finally,we using Java finished the simulation experiment.The experimental results show that the fault diagnosis algorithm can effectively detect the fault node,and the algorithm has lower energy consumption compared with the traditional fault diagnosis algorithm.In summary,this paper studies the fault diagnosis of wireless sensor networks with the aim of improving the reliability and lifetime of the network.With the widely application of wireless sensor networks,the research on fault diagnosis is of great theoretical significance and practical value.