An Intrusion Detection Framework for MEC-Enabled IoT Networks

Abstract

Mobile Edge Computing (MEC) has emerged as a promising paradigm for supporting latency-sensitive Internet of Things (IoT) applications by bringing computational resources closer to data sources. However, the distributed nature of MEC environments increases exposure to network-based security threats, particularly network intrusions that can impact both system reliability and task offloading efficiency. This paper proposes a security-aware framework that integrates a machine learning–based Intrusion Detection System (IDS) with the DTOME (Dynamic Task Offloading with Hybrid Energy) scheme to enhance security in MEC-enabled IoT networks. A preprocessing security layer is deployed at the edge server to detect and filter malicious traffic before offloading decisions are executed. The proposed framework is evaluated using benchmark intrusion detection datasets and a comprehensive set of performance metrics. The results demonstrate robust detection performance and stable operation under edge computing constraints. The main contributions of this work include integrating machine learning–based intrusion detection with dynamic task offloading in MEC environments, conducting a multi-dataset experimental evaluation to improve result reliability, and highlighting the practical feasibility of intrusion-aware offloading for real-world edge systems.