Research On Survivability Evaluation Of Cluster-Based Wireless Sensor Networks

Wireless Sensor Networks (WSN) are composed of a large number of sensor nodeswhich have functions of data sensing, gathering, handling and communicating. Due toits high mobility, flexibility, and wireless features, WSN have a promising future inmilitary affairs and civilian purpose.Network survivability is the ability to fulfill its mission in a timely manner in thepresence of attacks, accidents and failures. Sensor nodes have the features of limitedenergy, limited processing and storage capabilities and wireless transmission. Therefore,WSN of survivability has very important significance.Firstly, the thesis introduced the characteristics and applications of WSN. Second,we introduced the concept, the research background and the status of the survivability ofWSN, along with the evaluation standards, models and methods of quantitative analysis.Thirdly, a cluster-based survival analysis model of WSN is proposed. In this model, weassume that the cluster’s status is equivalent to the head node’s status, then we modeland analyze each head node of a hierarchical cluster-based WSN as a stochastic processbased on semi-Markov Process and Discrete Time Markov Chain. Based on theclassification of the status of head node, we analyze the isolation problem betweencluster head nodes and obtained the concept of k connectivity. Fourthly, wequantitatively analyze the survivability of WSN under Blackhole attack based on kconnectivity. We also quantitatively analyze the survivability under DoS attack based onthe ability of providing the basic services. Finally, we summarize the work of thesubject, and outlook to the next steps in the development and implementation of theresearch plan.