REVIEW OF INTRUSION DETECTION SCHEMES IN FOG COMPUTING USING MACHINE AND DEEP LEARNING TECHNIQUES: CONCEPTS, CHALLENGES AND OPEN ISSUES

Abstract

Internet of Things (IoT) is emerging as new computing paradigm that is exhibiting enormous growth with the development of wireless communication technologies. The unprecedented growth of IoT devices has led to rapid increase in the generated data and computational load. Though cloud computing has unlimited data storage and processing capability. However, sending all the data directly to the cloud server for processing is not a suitable architecture for applications involving real time processing. In this regard, technologies including Edge Computing (EC) and Fog Computing (FC) have been developed to overcome these challenges. This review work aims to recapitulate the existing state of the security aspects in fog and edge computing and cover the machine learning and deep learning techniques for overcoming the associated threats. In this work, an attempt is made to review the studies using machine and deep learning techniques to address security problems in fog and edge computing. The paper discusses different types of attacks associated with fog and edge computing and corresponding mitigating technologies. Further, brief review of intrusion detection system is also presented. Finally, research challenges and open issues are discussed and possible solutions for the same are proposed.