Enhanced security attack detection and prevention in 5G networks using CD-GELU-CNN and FMLRQC with HDFS-ECH-KMEANS

Security Attacks (SA) refer to a kind of malicious activity in which SA causes information destruction. Existing works have failed to concentrate on various attacks that limit the model’s performance. Therefore, this paper presents Cumulative Distribution-GELU-Convolutional Neural Network (CD-GELU-CNN) and Feature Map-based-Lattice Rainbow Quantum Cryp-tography (FMLRQC) techniques for SA detection and prevention. The proposed system originates from a source system. Firstly, log information is extracted from the 5G AD 2022 dataset. Then, the extracted information is vectorized and classified as attacked or not attacked. After that, the non-attacked data is transferred to the destination system. Then, data security and load balancing are performed. The load-balanced data is then checked in the Intrusion Detection System (IDS) to detect whether the packet is attacked or not. In the IDS, two datasets, namely 5G NDD and 5G SliciNdd datasets, are taken and pre-processed. Common features are extracted from the pre-processed data. Next, feature composition and feature selection are performed to select the best optimal features. Finally, the CD-GELU-CNN detects whether the packet is attacked or not. If the packet is not attacked, it can be transferred to the destination system. Otherwise, a notification is sent to the source system with a feature map to prevent the data. Therefore, the results prove that the proposed model achieved a high accuracy of 98.50%, outperforming prevailing techniques.