Research Of Intrusion Detection In Wireless Sensor Networks Based On The Structure Of Clustering

With the increasing development of the Internet of things, the wireless sensor network (WSN) has become one of the hot researches, and it is more and more widely applied to practical life. However, due to the unique nature of wireless sensor networks, such as the physical nodes exposure and lack of physical protection, making them easier to damage, and sensor nodes have limited energy, which result in the wireless sensor network security issues are more severe than the traditional wireless network. Therefore, the security problem of wireless sensor network has become an important research direction.Due to the unique nature of wireless sensor network itself, nowadays, some mature traditional network intrusion detection mechanisms are no longer applicable. Now, some research results of the wireless sensor network intrusion detection have existed, but those intrusion detection models analyze the attack characteristics on only a layer, not fully consider the impact of the data source of each layer, and without paying attention to energy consumption of the nodes in the analysis detection, which leads to the detection accuracy is not very high. In this paper according to the wireless sensor network node energy is limited the problem of low detection rate, on the basis of the existing clustering algorithms, an improved algorithm is presented, according to the location of the node, nodes are divided into clusters and in the clustering process considering the residual energy of sensor nodes and the distance between cluster head and member nodes to select the cluster head. It can reduce the network energy consumption and prolong the life cycle of the node. Based on the clustering structure, we propose an intrusion detection model, the node behavior information can be divided into cluster node information within and outside the cluster node information, when the nodes that are lurking in the internal network have abnormal behavior, firstly, through the information of nodes in the cluster to detect, when it is uncertain whether abnormal behavior of the node is malicious attacks, cluster head within the cluster and adjacent cluster head will exchange information, through cooperative detection among the cluster heads, to determine whether the abnormal behavior of the node is malicious attacks, so system can respond to deal with the condition in time and ensure the normal operation of wireless sensor network. The simulation results show that the improved clustering algorithm can reduce node energy consumption and prolong the network lifetime, cooperative detection model between cluster and cluster can provide high accuracy.The main research of this paper includes the following contents:Firstly, this paper introduces the basic concepts and characteristics of the wireless sensor network (WSN), etc., and introduce the related basic knowledge of the clustering algorithm and the related theory of intrusion detection.Secondly, the paper introduces several typical clustering algorithm, an improved algorithm is presented, according to the optimal cluster numbers formula, the whole network is divided into clusters and in the clustering process considering the residual energy of sensor nodes and the distance between cluster head and member nodes to select the cluster head.Thirdly, based on the clustering structure, this paper proposes an improved model of intrusion detection, and introduces the structure of the detection model, algorithm and the detection mechanism in detail.Finally, through simulation experiment to test the model, we get the simulation data and analyze them. At last, we obtain the final experiment results.