INTELLIGENT THREAT DETECTION IN MOBILE EDGE NETWORKS USING MACHINE LEARNING

Abstract

Mobile Edge Computing (MEC) is an evolving paradigm that brings computation and data storage closer to end users, aiming to enhance the performance of mobile applications. However, the rise of MEC technologies has introduced significant security concerns, as malicious activities increasingly target mobile edge systems. These threats include data breaches, denial-of-service (DoS) attacks, and intrusion attempts that exploit infrastructure vulnerabilities, undermining the integrity, availability, and confidentiality of the system. Traditionally, mobile edge security relied on signature-based and rule-based detection methods. These systems depended on predefined patterns of known attacks, limiting their ability to detect novel or evolving threats. As cyber-attacks have grown more sophisticated, the need for advanced detection techniques has become critical. The integration of machine learning (ML) algorithms has transformed the landscape of mobile edge security by enabling real-time detection and prevention of both known and unknown attacks. Given the growing reliance on MEC for critical applications, it is essential to develop robust, adaptive security solutions that ensure data protection and system reliability. Earlier approaches faced challenges such as high falsepositive rates, limited detection capabilities, and the need for manual updates to threat databases. To overcome these limitations, ML-based systems are proposed to automate the detection of malicious activities in mobile edge environments. Using classifiers such as Decision Trees, Random Forests, and Deep Neural Networks, these systems can process large volumes of data, identify complex patterns, and accurately classify malicious behaviors. This approach enhances the accuracy, scalability, and efficiency of security mechanisms, aiming to build self-improving systems that continuously learn from new data and adapt to emerging threats.