Deep Learning for Secure Mobile Edge Computing in Cyber-Physical Transportation Systems

Abstract

Key capabilities like sensing, communication, and traffic management are enabled by a cyber-physical transportation system (CPTS), which employs a wide variety of sensors and mobile wireless devices. These features are made possible by mobile edge computing (MEC), which enables a collaborative processing of real-time, compute-intensive activities directly at the edge of the network by a variety of linked devices, including moving automobiles and traffic sensors. As a result of MEC's ability to integrate computing, networking, and physical processes, cyber-physical systems have much more room to grow, especially in the transportation sector. However, there is an immediate need for strong security procedures since transport systems are becoming more interdependent and vulnerable to cyber threats due to the increasing use of edge computing. In response, this study introduces a deep learning model that actively seeks for and fixes security flaws using unsupervised learning methods. This project seeks to enhance the security of MEC in CPTS by using deep learning algorithms to predict network assaults and mitigate their negative impacts. We show how deep learning may be used to the real-world problem of cyber threat identification and mitigation by integrating state-of-the-art neural network designs into a safe computing framework for the edge.