Research On Intrusion Detection Technology Of Internet Of Things Based On Voting

Under normal circumstances,the Internet of Things has limited node energy,network topology is easy to change,communication,computing and storage capabilities are also limited.At the same time,the number of sensor nodes is large,and the working environment and state of each node are different,which is easy to lose utility;In addition,there are still problems such as ambiguous network defense boundaries.In this case,how to establish an effective security system,detect and defend against various intrusions and malicious attacks,and ensure the reliability of the Internet of Things becomes a hot issue.Due to the diversity attack strategies,traditional passive detection technology cannot provide security for the current network before the attack occurs.Therefore,how to quickly and effectively discover the intrusion behavior becomes a prerequisite for the security of the Internet of Things.Intrusion detection technology can actively evaluate the overall network,detect the attack behavior and give a warning before the network attack causes damage,and intend to ensure the security of the network in an environment with insecure factors,so that the network has resistance to organized attacks.The ability to take risks to compensate for the deficiencies of passive cybersecurity mechanisms.Therefore,how to building an intrusion detection system with low energy consumption and high detection rate on the IoT nodes with limited energy to make the Internet of Things work safely and reliably for a long time becomes the research purpose of the thesis.By studying the technology of the existing IoT intrusion detection system,analyzes the intrusion detection technology and the overall framework of the existing IoT intrusion detection system,then proposes a material based on improved practical Byzantine Fault Tolerance(PBFT)voting networked intrusion detection algorithm.Firstly,the PBFT algorithm is introduced to select some nodes for active intrusion detection,and the feasibility of it in the IoT environment is proved.Then the PBFT algorithm is improved to make it suitable for the voting process of nodes in the Internet of Things.After that,the paper studies the merger process of the voting results in the voting process,and participates in the merger process of the voting results by taking the intrusion detection result and the credibility of each node as parameters.The comprehensive experiment shows that compared with IIDS(Integrated Intrusion Detection System(IIDS)),the energy consumption of the proposed votebased IoT intrusion method is reduced by 22.19% in the case of attack nodes accounting for 10%,and 15% in attack nodes.The reduction is 18.79%,and the average reduction is 20.49%.Compared with TDTC(Two-layer Dimension Reduction and Two-tier Classification Model,TDTC),the energy consumption is reduced by 10.11% when the attack node accounts for 10%,and the attack node accounts for 10.11%.In the case of 15%,the reduction was 8.87%,and the average reduction was 9.49%.It is proved that the proposed voting algorithm based on improved PBFT has obvious advantages in reducing the energy consumption of the IoT intrusion detection system.