Research On Fault Tolerance Mechanism Based On Clustering In Wireless Sensor Networks

Wireless Sensor Networks (WSNs) have involved a wide range of disciplines whichhumans closely interact with the physical world. The distributed sensing capabilities of thesensor nodes and they can be deployed easily make WSNs be an important part of ourdaily lives. Nowadays, WSNs have been applied to a lot of fields, such as military,environment, health, home and industry. WSNs are hotspots of the international scientificresearch.The batteries which are difficult to replace provide the nodes with the energies. Andthe nodes’ storage, computing and communication capabilities are very limited. The linkswill breakdown or even the network lose masses of data because the sensor nodes arevulnerable on account of the environmental impacts and the restrictions of their owncharacteristics. Besides, the sensor nodes are often deployed in the unattended hostileenvironments, the failure rate of the nodes is high. The security has become the focuswhich we must concern. So Intrusion tolerance has been an essential indicator to measurethe performance of the network. Therefore, to strengthen the fault tolerance of WSNs andprolong the network lifetime have important theoretical and practical significances. Andhow to design less energy consumption, high fault tolerance and strong intrusion toleranceof wireless sensor networks become the subjects of our study.According to the problems mentioned above, the contributions of this thesis are asfollows：(1) In order to save the energy of the sensor nodes and thus maximize the networklifetime, we put forward a multi-path transmission mechanism based on clustering withdaemon nodes on the basis of the existing clustering algorithms, we use daemon nodes torelieve the loads of cluster heads. Then we adopt the expanding multi-path to assure thehigher reliability of the data transmission. Finally, the experimental results show that the new algorithm can reduce the energy consumption and enhance the fault tolerance of thewhole network to some extent.(2) In order to solve the security of the data transmission and strengthen the intrusiontolerance, we introduce the key management scheme after analyzing the security threats ofLEACH protocol. At the same time, we adopt the random key pair model and theevaluation mechanism to improve the confidentiality of the data transmission. Also, we usethe mixed multi-path mechanism among the clusters even if some cluster head is underattack during the process, the route can be repaired in time. Meanwhile, cluster heads cancooperate with each other to accomplish the basic tasks. Then the data will be decryptedaccording to the keys they carry when they arrive at the base station. Consequently, by thismeans can improve the integrity and authenticity of the data.