A Deep Learning Approach For Detecting Malicious Activities For Mobile Edge Security

Abstract

Mobile edge computing (MEC) is an evolving paradigm that brings computation and data storage closer to the user, aiming to improve the performance of mobile applications. However, with the rise of these technologies, malicious activities targeting mobile edge systems have become a significant concern. Malicious activities in mobile edge security include attacks such as data breaches, denial of service (DoS), and intrusion attempts, which exploit vulnerabilities in the infrastructure and impact the integrity, availability, and confidentiality of the system. Historically, mobile edge security was addressed using signature-based detection methods and rule-based systems. These systems relied on predefined patterns of known attacks, which could only recognize threats for which signatures had been previously created. These approaches were limited in their ability to detect new or evolving attack strategies. Over time, the increasing sophistication of cyber-attacks necessitated the exploration of more advanced techniques. The introduction of machine learning (ML) algorithms to security systems has significantly changed the landscape of mobile edge security by enabling systems to detect and prevent both known and unknown attacks in real-time. Machine learning models, particularly those leveraging deep learning techniques, are capable of learning complex patterns in data and adapting to new threats as they arise. The motivation to develop an advanced solution stems from the need to overcome the shortcomings of earlier detection systems, particularly their inability to handle novel or evolving attacks. The growing reliance on mobile edge computing for critical applications has made it essential to develop robust security solutions capable of ensuring data protection and system reliability.