Optimal Mixed Kernel Extreme Learning Machine-Based Intrusion Detection System for Secure Intelligent Edge Computing
R. Pandi Selvam, T. Jayasankar, R. Kiruba Buri, P. Maheswaravenkatesh · Apple Academic Press eBooks · 2024
Edge computing (EC) technologies act as an important role in resolving the issues of remote EC, such as high delay, mobility, and location awareness. Every EC service can take place in a mutual way and be accessed by users through the Internet. A few of the attacks are used to root, remote login, denial of service (DoS), snooping, port scanning, etc., can occur in EC platforms because of Internet-enabled remote services. An intrusion detection system (IDS) is an effective way of protecting the network through the detection of attacks. In this view, this study introduces an optimal mixed kernel extreme learning machine-based intrusion detection system (OMKELM-IDS) for intelligent EC. The goal of the OMKELM-IDS 232 technique focuses on the detection and classification of intrusions in the EC environment. The OMKELM-IDS technique encompasses pre-processing, intrusion detection, and parameter tuning. Besides, the MKELM model can be designed to identify the occurrence of intrusions and classify it. Moreover, the quasi-oppositional cuckoo search algorithm (QOCSA) is applied for the optimal parameter tuning process of the MKELM model and thereby boosts the detection efficiency. The experimental result analysis of the OMKELM-IDS technique takes place using benchmark IDS dataset and the outcomes are studied under different aspects. The experimental outcome highlighted the enhanced performance of the OMKELM-IDS approach on the recent state of art approaches. The comparative outcomes portrayed the betterment of the OMKELM-IDS system with respect to distinct measures.