Cyber Attack Detection in IoT using Deep Learning Techniques

Abstract

The Internet of things (IoT) consists of millions of digital devices which interact with each other through minimum user interaction. IoT is one of the most rapidly expanding computing sectors; however, it is vulnerable to many attacks. An emerging concern in the Internet of Things (IoT) space is attack and strange placement on the IoT framework. Attacks and dangers on these systems are also growing proportionally because of the expanding IoT foundation usage across all industries. In this paper, a review of previous work is conducted, and several deep learning techniques are proposed for accurately predicting attacks on IoT systems. Injection attacks, Man-in-the-middle attacks, Information gathering, Malware attacks, and DDoS/Dos attacks are such attacks and irregularities that might occur in an IoT framework. Identifying such attacks and malicious traffic is important for the Internet of things (IoT) network to block unwanted traffic and unauthorized access. The Edge-IIoTset Cyber Security Dataset and the VGG16 and VGG19 algorithms are utilized to evaluate the effectiveness of the proposed solution; F1 score, precision, recall, and accuracy are the assessment metrics used.