An In-depth Analysis of Intrusion Detection Systems with an Emphasis on Multi-Access Edge Computing and Machine Learning
Shruti Saxena, Nikunj Tahilramani · 2025
The exponential growth of Internet of Things (IoT) applications is leading to increased complexities in network data and computation. IoT devices with limited resources are susceptible to cyber threats, posing risks to networks. Multi-access edge computing (MEC) offers a solution by transferring tasks to the edge, enhancing efficiency and security. This study offers a detailed analysis of network intrusion detection systems (NIDS) for IoT, focusing on MEC and machine learning applications. Various datasets, metrics, and deployment strategies in NIDS design are compared, and a new MEC-based NIDS framework is proposed. The review covers the development of IoT architecture, specific intrusion detection techniques, and the ongoing need for strong security measures. By exploring anomaly-based IDS, edge challenges, and machine learning integration, this study showcases the progress and remaining obstacles in safeguarding smart environments. Furthermore, this study delves into the security landscape of Industrial IoT (IIoT), critically evaluating current security measures, emerging tools, simulations, and challenges unique to IoT/IIoT. Through this thorough analysis, this study aims to steer future research and practical applications in the realm of IoT security.