Intrusion Detection Methods Of CBTC Systems Based On Cloud Computing | Posted on:2023-02-20 | Degree:Master | Type:Thesis | Country:China | Candidate:L R Hu | Full Text:PDF | GTID:2532306845498604 | Subject:Traffic Information Engineering & Control | Abstract/Summary: | PDF Full Text Request | The wide application of commercial software and hardware products in the Communication-Based Train Control(CBTC)system makes the information security threats to the system increasingly.Although cloud computing technology improves the performance of CBTC systems,it also brings new information security risks.The existing CBTC intrusion detection methods do not consider the threat characteristics in the virtualization environment and have the problem of single point of failure,which is difficult to meet the needs of intrusion detection systems of intrusion detection systems of oriented cloud computing CBTC.Therefore,timely research on information security and intrusion detection methods for oriented cloud computing CBTC systems has important theoretical and practical guiding significance for CBTC system cloud migration.By analyzing the characteristics of CBTC system and cloud computing,this paper proposes the deployment principle and basic architecture of CBTC system in cloud environment,and combines the characteristics of information attack to study a distributed intrusion detection system suitable for oriented cloud computing CBTC system.A detection method based on the combination of multi-dimensional relative entropy and random forest is designed to monitor traffic and packet anomalies.Establish a multi-point cooperative intrusion detection system based on blockchain and trust assessment,realize distributed cooperation between IDS nodes,avoid single point failure,and achieve attack source traceability while effectively improving intrusion detection performance.The main work of the paper is as follows:(1)The characteristics of cloud computing and the architecture and principle of CBTC system are analyzed,the deployment principles and architecture of oriented cloud computing CBTC system are put forward,the possible information attacks of CBTC system based on cloud computing are analyzed,and the scheme of intrusion detection based on intrusion detection of oriented cloud computing CBTC system requirements is designed;(2)An intrusion detection method based on multidimensional relative entropy and random forest is proposed.Based on the traffic characteristic distribution of cloud computing-oriented CBTC system,an improved multi-dimensional relative entropy traffic detection model is established to detect abnormal traffic.Based on the extracted packet features,a random forest-based intrusion detection method is used to identify abnormal packets;(3)A multi-point cooperative distributed intrusion detection system based on blockchain and trust assessment is established.An improved delegated proof of stake consensus algorithm and smart contract are proposed,a blockchain network for cloud computing CBTC intrusion detection system is established.an information chain that stores the detection results and trust degree of IDS nodes is designed,and the trust evaluation and distributed collaboration of intrusion detection nodes are realized to improve the performance of intrusion detection systems and achieve effectively identification of attack sources;(4)The overall performance of the cloud computing-oriented CBTC intrusion detection system is verified.An experimental environment for cloud computing CBTC system is built,cloud computing CBTC intrusion detection data set and trust evaluation data set are generated,and the overall performance and attack source tracing ability of cloud computing CBTC intrusion detection system are verified.The performance of the intrusion detection system proposed in this paper is evaluated under the normal operation and abnormal scenarios of the intrusion detection nodes.The experimental results show that the intrusion detection system can effectively identify different attack behaviors and attack sources.The attack detection rate reaches99.26%-99.33%,the detection rate of the attack source being a single device is 98.65%-98.74%.The intrusion detection based on multi-point cooperation improves the information security protection capability of cloud computing CBTC system. | Keywords/Search Tags: | CBTC, Cloud computing, Intrusion detection, Relativ entropy, Random forest, Blockchain, Source of attacks | PDF Full Text Request | Related items |
| |
|