A Network Attack Detection Model of Smart Grid Based on XGBoost Algorithm

2021 ◽  
pp. 481-488
Author(s):  
Yangyang Lian ◽  
Lifang Gao ◽  
Pengbo Fang ◽  
Pengpeng Lu ◽  
Liandong Chen ◽  
...  
Keyword(s):  
Smart Grid ◽  
Attack Detection ◽  
Network Attack ◽  
Detection Model ◽  
Grid Based

scholarly journals CONSTRUCTION OF SYSTEMS OF DETECTION OF INVASIONS INTO THE INFORMATI TON AND TELECOMMUNICATIONS NETWORK ON THE BASIS OF METHODS OF INTELLECTUAL DISTRIBUTION OF DATA

2020 ◽  
pp. 80-89
Author(s):  
S. Toliupa ◽  
O. Pliushch ◽  
I. Parhomenko
Keyword(s):  
Data Mining ◽  
Experimental Research ◽  
Intelligent Systems ◽  
Detection System ◽  
Attack Detection ◽  
Information Network ◽  
Distributed Information ◽  
Network Attack ◽  
Detection Model ◽  
Detection Systems

The article proposes a combinatorial construction of a network attack detection system based on selected methods of data mining and conducts experimental research that confirms the effectiveness of the created detection model to protect the distributed information network. Experiments with a software prototype showed the high quality of detection of network attacks and proved the correctness of the choice of methods of data mining and the applicability of the developed techniques. The state of security of information and telecommunication systems against cyberattacks is analyzed, which allowed to draw conclusions that to ensure the security of cyberspace it is necessary to implement a set of systems and protection mechanisms, namely systems: delimitation of user access; firewall; cryptographic protection of information; virtual private networks; anti-virus protection of ITS elements; detection and prevention of intrusions; authentication, authorization and audit; data loss prevention; security and event management; security management. An analysis of publications of domestic and foreign experts, which summarizes: experience in building attack detection systems, their disadvantages and advantages; of attack and intrusion detection systems based on the use of intelligent systems. Based on the results of the review, proposals were formed on: construction of network attack detection systems on the basis of selected methods of data mining and experimental research, which confirms the effectiveness of the created detection model for the protection of the distributed information network.

An Improved DDoS Attack Detection Model Based on Unsupervised Learning in Smart Grid

2021 ◽  
pp. 550-562
Author(s):  
Zhili Ma ◽  
Hongzhong Ma ◽  
Xiang Gao ◽  
Jiyang Gai ◽  
Xuejun Zhang ◽  
...  
Keyword(s):  
Smart Grid ◽  
Unsupervised Learning ◽  
Attack Detection ◽  
Detection Model ◽  
Ddos Attack ◽  
Model Based ◽  
Ddos Attack Detection

scholarly journals A Multi-Step Attack Detection Model Based on Alerts of Smart Grid Monitoring System

IEEE Access ◽  
2020 ◽  
Vol 8 ◽  
pp. 1031-1047 ◽  
Author(s):  
Hua Zhang ◽  
Xueqi Jin ◽  
Ying Li ◽  
Zhengwei Jiang ◽  
Ye Liang ◽  
...  
Keyword(s):  
Smart Grid ◽  
Monitoring System ◽  
Attack Detection ◽  
Detection Model ◽  
Model Based ◽  
Grid Monitoring

scholarly journals A Distributed Security SDN Cluster Architecture for Smart Grid Based on Blockchain Technology

2021 ◽  
Vol 2021 ◽  
pp. 1-9
Author(s):  
Ao Xiong ◽  
Hongkang Tian ◽  
Wenchen He ◽  
Jie Zhang ◽  
Huiping Meng ◽  
...  
Keyword(s):  
Smart Grid ◽  
Single Point ◽  
Cluster Structure ◽  
Cluster Head ◽  
Detection Algorithm ◽  
Attack Detection ◽  
Security And Privacy ◽  
Network Attack ◽  
Blockchain Technology ◽  
Distributed Security

This paper proposes a smart grid distributed security architecture based on blockchain technology and SDN cluster structure, referred to as ClusterBlock model, which combines the advantages of two emerging technologies, blockchain and SDN. The blockchain technology allows for distributed peer-to-peer networks, where the network can ensure the trusted interaction of untrusted nodes in the network. At the same time, this article adopts the design of an SDN controller distributed cluster to avoid single point of failure and balance the load between equipment and the controller. A cluster head was selected in each SDN cluster, and it was used as a blockchain node to construct an SDN cluster head blockchain. By combining blockchain technology, the security and privacy of the SDN communication network can be enhanced. At the same time, this paper designs a distributed control strategy and network attack detection algorithm based on blockchain consensus and introduces the Jaccard similarity coefficient to detect the network attacks. Finally, this paper evaluates the ClusterBlock model and the existing model based on the OpenFlow protocol through simulation experiments and compares the security performance. The evaluation results show that the ClusterBlock model has more stable bandwidth and stronger security performance in the face of DDoS attacks of the same scale.

scholarly journals Construction of Schemes, Models and Algorithm for Detection Network Attacks in Computer Networks

2019 ◽  
Vol 8 (12) ◽  
pp. 2234-2241
Keyword(s):  
Security Policy ◽  
Threshold Value ◽  
Attack Detection ◽  
Information Flows ◽  
System Of Differential Equations ◽  
Network Attacks ◽  
Network Attack ◽  
Detection Model ◽  
Detection Schemes ◽  
Analysis Of Time Series

This article is devoted to develop network attack detection schemes to search for vulnerable servers and the likelihood of determining the type of attacks by the contents of network packets, a network attack recognition scheme that allows to filter external network traffic by processing incoming requests is proposedas well a network detection model attacks as signs of detecting the position of a security policy is offered. Based on the analysis of time series, a network attack detection model that allows identifying network attacks by a threshold value is developed and a mathematical model for real-time recognition of network attacks is proposed.Model of the behavior of the information flows are shown that the linear model does not provide an adequate assessment of the current process to the critical states. Within the framework of developing models for detecting network attacks, an algorithm for detecting and identifying network attacks is proposed, which allows one to perform not only an exhaustive search for the classification features of network attacks, but to limit itself to a shortened search.The behavior of the queue of half-open compounds are described with an absorbing state, and a system of differential equations for state probabilities are obtained. Also new requests to belong to a particular cluster are analyzed

Sign in / Sign up

Export Citation Format

Share Document