Attacks Detection Approach Based on a Reinforcement Learning Process to Secure 5G Wireless Network

• Published in: IEEE/CIC International Conference on Communications in China - Workshops (Online) pp. 1 - 6
• Main Author: Sedjelmaci, Hichem
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2020
• Subjects: 5G mobile communication; 5G wireless network; Computer architecture; Computer crime; Jamming; Machine learning algorithms; Monitoring; Network attacks detection; Reinforcement learning; Wireless networks
• ISSN: 2474-9133
• DOI: 10.1109/ICCWorkshops49005.2020.9145438

Fifth Generation (5G) wireless network will be a subject to a variety of cyber-threats from advanced and complex attacks. In this article, we aim to secure the 5G wireless systems from the most dangerous and advanced network attacks, e.g., jamming and Distributed Denial of Service (DDoS) attacks. We propose and develop a new cooperative attack detection based on a hierarchical Reinforcement Learning (RL) process to identify the network attacks. The cooperative detection is performed with a distributed detection systems executed at the different critical 5G network's organs such as access point, base station and servers. According to our experiments results, the proposed RL detection system enhances a detection of new misbehaviors attacks.