Strengthening Network Security: An Efficient DL Enabled Data Protection and Privacy Framework for Threat Mitigation and Vulnerabilities Detection in IoT Network

Abstract

The adoption of the Internet of Things (IoT) in smart manufacturing has recently seen a boost in economic and technological Advancement. As many network attacks have revealed how much detection matters for secure cyberspace. A data preprocessing step and a deep learning model are included in our novel system for identifying network attacks. We have developed a deep learning model, whose structures are based on CNN mechanisms. An evaluation of the model was done to see how they performed in detecting threats on the NSL-KDD dataset. Finding out about cyber security weaknesses within IoT devices before cybercriminals take advantage of them is increasingly difficult, but it is the main technology to secure these devices from attacks. The purpose of the research is to review the tools used for recognizing IoT vulnerabilities, using machine learning techniques with the datasets IoT. During the study, possible flaws in IoT architectures are highlighted on every layer, along with a description of how machine learning helps detect such flaws. An approach for finding and handling vulnerabilities in IoT using machine learning was first proposed and then a recap of recent studies is presented. The approach performs better than other DL- systems that use the NSL-KDD dataset. The accuracy was 81.2%, Recall was 96.30% and the system earned a Precision of 88%. It successfully counters all types of Active, passive, DoS, and DDoS attacks.

Keywords: Deep Neural Network, Internet Of Things Networks, Intrusion detection; CNN; BiLSTM; BiGRU