Machine learning for security boosting in Internet of Things environments
Souhayla Dargaoui, Mourade Azrour, Ahmad El Allaoui, Azidine Guezzaz, Said Benkirane
Abstract
The continuous enhancement in communication technologies and the wide adoption of the Internet of Things (IoT) in different fields have significantly improved our lives. However, the restricted nature of IoT devices makes IoT networks a simple target for cyberattacks. During the last decade, considerable attempts have been made, and enormous security approaches have been proposed, to deal with security and privacy issues in IoT environments. Nevertheless, the traditional cryptographic mechanisms used to build the security solutions are not sufficiently efficient to deal with potential attacks. Consequently, various advanced mechanisms have been adopted recently. Machine learning (ML) is one of the most alluring techniques that may embed intelligence in IoT security and make it robust against all possible threats. This chapter, first, provides attack vectors in IoT architecture layers and presents various machine learning classes. Then, it surveys machine learning applications in IoT security. Finally, it discusses ML-based IoT security challenges and issues.