HIDE-6G: Advanced Intrusion Detection System for Secure 6G Network using Deep Learning
Unknown authors
Abstract
wireless networks are anticipated to undergo trials and installations as early as 2030, offering unprecedented capacity, dependability, and efficiency.However, attention is shifting towards the development of 6G networks to meet the demands of emerging applications.The transition to 6G brings new challenges, particularly in the realm of intrusion detection, where the sophistication of attacks necessitates advanced security solutions.To eliminate this challenge, a novel Hybrid Intrusion DEtection system for the 6G network (HIDE-6G) has been proposed to detect intrusion in the 6G network.The proposed method leverages advanced techniques such as Principal Component Analysis (PCA) for dimensionality reduction, a Spotted Hyena Optimization Algorithm for feature selection, and a Capsule Network-based Deep Autoencoder (CapsDA) for effective anomaly detection.The performance of the HIDE-6G is estimated using the NSL-KDD and CICIDS 2019 datasets, demonstrating superior results compared to existing techniques such as AD6GN, IDSoft, and LA-HLRW.According to the comparison analysis, the proposed HIDE-6G technique's detection rate is 6.10%, 22.27%, and 20.7% greater than the existing HADES-IoT, H3SC-DLIDS, and F-BIDS techniques respectively.