GAN Model for Network Intrusion and Malicious Detection in Network Security

Abstract

As of late, the utilization of Internet of Things (IoT) is persistently grow- ing, while IoT frameworks are experienced in applications from various spaces, similar to brilliant urban communities, industry, agribusiness, and so on. The Internet of Things is made up of billions of connected devices that can send and receive data over the internet. The Internet of Things (IoT) is at risk of cyberattacks because of this interconnection, which compromises its security. In this context A Malicious Attack  Detector (MAD) is developed for the Internet of Things (IoT) with the primary objective of preventing attacks on IoT systems. In order  to  first  learn about poisoning datasets and then identify malicious activities, MAD employs a GAN-based model. In this paper we present the accuracy per- formance of the discriminator on central generator using total number of epochs. We also highlight the recent risks and solutions for the malicious activities.