AI-Based Security Protocols for IoT Applications: A Critical Review

Abstract: The Internet Of Things (IoT) is a network of heterogeneous nodes that exchange data and critical information amongst themselves with minimum human intervention. The utility of this technology is large, thus it is used today in most modern applications. With billions of nodes connected across multiple applications, the area for potential security attacks is ever increasing. In addition to this, the nature of the data being transmitted also becomes more sensitive to the latest applications and this calls for effective security provisions. Due to its unique nature, traditional security provisions are not as successfully applicable in IoT networks. This leaves these networks vulnerable to malicious intruders. In such a scenario, Artificial Intelligence (AI) comes out as a powerful solution. This article serves to provide an overview of previously proposed AI-based solutions that can be applied to IoT networks to secure them. An industrywide overview of security provisions is provided by categorizing IoT applications into three broad sectors, namely, healthcare, smart grid, and smart city. The survey strives to give a clear industry-oriented vision of the available AI solutions and address the requirement of an application-ready security survey in the field.