Internet of Things Intrusion Detection System Using Enhanced Deep Learning-based Feature Selection

Abstract

The Internet of Things (IoT) has become an integral part of our daily lives, with the increasing usage of interconnected devices. However, with this increased connectivity comes the risk of security breaches and intrusions. To address this issue, many researchers have proposed intrusion detection systems (IDS) that utilize deep learning techniques. Therefore, this study proposes an IDS for IoT networks, using enhanced deep learning-based feature selection. The proposed model analyzes network traffic patterns to distinguish between normal and abnormal behavior and adapts to changes in network environments to detect new attack patterns. The enhanced feature selection method identifies the most important features for detecting IoT network intrusions. This resulted in a significant improvement when compared to other existing IDS for IoT. The experimental results show that the model achieved a detection rate of 99.8% and a false positive rate of 2%, outperforming other state-of-the-art IDS for IoT. Furthermore, the propose model was able to detect both known and unknown attacks, making it highly effective in ensuring the security of IoT networks. The proposed model provides an efficient and accurate solution for detecting IoT network intrusions. This technology can greatly benefit the security of IoT devices and users, and hope it will be widely adopted in protecting IoT networks.

Keywords: Internet of Things, Deep learning, Intrusion detection systems, Feature selection, Network intrusion, Security, Attacks.