Research On Topology Construction And Intrusion Detection For Heterogeneous Wireless Sensor Networks

As an important part of Internet of Things(IOT),the heterogeneous wireless sensor networks(HWSNs)is becoming more and more widely used in various applications.The sensor nodes are vulnerable to external interference and malicious attacks because of their limited resources and a complex open-working environment.It is very meaningful to study the network and data security for HWSNs.To improve the network security of HWSNs,the security measures about the construction of a network topology is studied in this paper.A multi-path security topology control algorithm with intrusion tolerance is established.And a developed linear discriminant analysis(LDA)-based extreme learning machine(ELM)classification algorithm for intrusion detection is proposed based on the topology model.Through the further analysis of the detection results,a decision-response mechanism that can guarantee network security is carried out.The main work of this paper is as follows:1.A multi-path security topology control algorithm with intrusion tolerance for HWSNs(MPIT)is proposed.Considering the network energy consumption and network security,the algorithm constructs a non-uniform clustering network with the routing form which contains single-path and multi-path.Then the data is transmitted in a fragmented form using multiple paths with the(k,n)threshold secret sharing scheme.Finally,this algorithm balances the energy consumption and extends the network lifetime while ensuring the secure of transmission.2.A developed LDA-Based ELM classification algorithm for intrusion detection in HWSNs(DLECA)is proposed.The algorithm first uses the improved LDA to reduce the feature dimensions,and then uses the ELM algorithm to classify the data.The simulation result shows that the proposed algorithm ensures the accuracy of intrusion detection and improves the efficiency.3.A non-cooperative game based decision response mechanism for intrusion detection is developed for the false detections in the intrusion detection.This mechanism established a non-cooperative game intrusion detection model between attackers and defenders.Then the optimal strategies of the attacker and the defender respectively to reach the goal of improving network performance by analyzing the Nash equilibrium with their benefits from different strategies.This mechanism achieves the purpose of improving network performance and rational use of resources.