Deep Learning Approach for Intrusion Detection and Mitigation in IoT Environment: A Comprehensive Study

The Internet of Things (IoT) has gained importance over the last decade due to the exponential growth in digital communication systems and networks. The future society relies on connecting and communicating everything through the internet. However, the likelihood of attacks such as Distributed Denial-of-Service (DDoS) attacks, man-in-the-middle attacks, and other routing attacks is one of the key issues in the IoT environment. To arrive at a solution to this issue, it is important for academicians and researchers to integrate the existing information and statistics about this serious threat. To fill this gap, this paper presents a comprehensive survey of deep learning approaches for intrusion detection mitigation in IoT networks. First, it provides deeper insights into the type of malware in various IoT layers. Also, it discusses the various attacks in cloud and SDN layers. Next, it compares the intrusion detection and mitigation performance of various deep learning techniques in the cloud and SDN platforms. Then, it details the performance analysis of deep learning techniques for benchmark datasets. Finally, it elaborates on the research trends and future direction of intrusion detection in the IoT environment.