Security Enhancement by Identifying Attacks Using Machine Learning for 5G Network

• Published in: International journal of communication networks and information security 2022-09, Vol.14 (2), p.124-141
• Main Authors: Keserwani, Hitesh, Rastogi, Himanshu, Kurniullah, Ardhariksa Zukhruf, Janardan, Sushil Kumar, Raman, Ramakrishnan, Rathod, Vinod Motiram, Gupta, Ankur
• Format: Article
• Language: English
• Subjects: 5G mobile communication; Accuracy; Artificial intelligence; Automation; Cellular telephones; Computers; Data encryption; Data transmission; Decision making; Machine learning; Model accuracy; Personal computers; Product development; Security systems; Social networks; Social research; Wireless networks; World Wide Web
• ISSN: 2076-0930; 2073-607X; 2073-607X; 2076-0930
• DOI: 10.17762/ijcnis.v14i2.5494

Need of security enhancement for 5G network has been increased in last decade. Data transmitted over network need to be secure from external attacks. Thus there is need to enhance the security during data transmission over 5G network. There remains different security system that focus on identification of attacks. In order to identify attack different machine learning mechanism are considered. But the issue with existing research work is limited security and performance issue. There remains need to enhance security of 5G network. To achieve this objective hybrid mechanism are introduced. Different treats such as Denial-of-Service, Denial-of-Detection, Unfair use or resources are classified using enhanced machine learning approach. Proposed work has make use of LSTM model to improve accuracy during decision making and classification of attack of 5G network. Research work is considering accuracy parameters such as Recall, precision and F-Score to assure the reliability of proposed model. Simulation results conclude that proposed model is providing better accuracy as compared to conventional model.