Research On Intrusion Detection Based On Hierarchical Structure For Wireless Sensor Network

With the advancement of microelectronics technology and wireless communication technology,wireless sensor network(Wireless sensor network,WSN)as a new type of network is increasingly needed and widely used by society.WSN is often used to replace human surveillance in remote or dangerous areas such as military battlefields,complex forests,and disaster areas.However,due to defects such as limited energy of sensor nodes and lack of computing resources,it is extremely vulnerable to attack and damage without supervision.Therefore,information security has always been the primary issue of wireless sensor networks.As an active defense security technology,intrusion detection can continue to provide reliable and effective protection for WSN.In recent years,intrusion detection algorithms for WSN have been proposed one after another,but there are still problems that the network energy consumption is not actually considered,and the detection accuracy and efficiency of the detection algorithm are low.In response to these problems,this paper proposes an improved layered routing algorithm and intrusion detection algorithm,the specific work is as follows:(1)Aiming at the defects such as the energy limitation of nodes in WSN that cannot be replenished,designing a more complex detection model will lead to accelerated death of nodes and shorten the life cycle of the network,so this paper uses a hierarchical structure intrusion detection model.The ordinary member nodes in the model are only responsible for collecting and forwarding information,the cluster head node is responsible for monitoring the behavior data of the member nodes in the cluster,and the uppermost base station is responsible for complex analysis and detection of the behavior data collected by the cluster head node.To minimize the resource consumption of intrusion detection in the network.(2)Considering that in the intrusion detection model,the cluster head node's monitoring and forwarding of member node's behavior consumes a part of energy,which may cause the node to die prematurely due to uneven network load.In response to this problem,this paper proposes a LEACH-R Hierarchical routing algorithm.In the algorithm,the cluster head election strategy is improved,so that the nodes with high residual energy and close to the base station have higher priority to become the cluster head,and the "cluster head reselection" mechanism is added to enable high energy nodes to Frequently elected cluster head.Through simulation comparison with LEACH and DEEC algorithms,it is found that the algorithm proposed in this paper can effectively extend the life cycle of nodes and networks,and improve energy utilization,providing a reliable guarantee for the operation of intrusion detection.(3)The key of the intrusion detection system is the intrusion detection algorithm.Aiming at the problems of high computational complexity and low detection accuracy of existing intrusion detection schemes,a discrete binary particle swarm optimization isolation forest algorithm(BPSO-i Forest)intrusion detection mechanism is proposed.Based on the idea of selective integration,the discrete binary particle swarm optimization algorithm is used to improve the initial forest generated by the isolation forest algorithm,select the isolation tree with high accuracy and large difference in the initial forest,construct the optimal isolation forest,and improve the detection accuracy and algorithm of intrusion detection Execution efficiency.On the WSN data set,compared with the traditional isolation forest algorithm,random forest algorithm and its improved algorithm,the results show that the detection accuracy and execution efficiency of this algorithm have been significantly improved,effectively ensuring the accuracy and real-time of intrusion detection in the model.