Unauthorized Access Prevention Model for Secure Data Transmission in Internet of Things-Based Network

Abstract

The Internet of Things (IoT) has revolutionized connectivity by enabling devices to interact seamlessly in various domains such as healthcare, smart cities, and industrial automation. However, this connectivity also introduces vulnerabilities, with unauthorized access posing significant threats to data security. This paper presents an unauthorised access prevention model for securing data transmitted in the internet of things-based network, with a focus on providing an Intrusion Detection System (IDS) that detects botnet attacks. The approach employs a Multi-Layer Perceptron (MLP) for fault detection, specifically designed to address the challenges inherent in IoT environments. The MLP architecture consists of three dense layers with Rectified Linear Unit (ReLU) activation functions, enabling the capture of intricate data relationships. Through sequential modelling, the MLP accurately identifies complex patterns indicative of botnet attacks. The training process of the MLP yields promising results, with significant accuracy and minimal loss values. The model achieves an accuracy of 99.99% for both training and validation (testing) data. Furthermore, the classification report highlights the model's exceptional precision and recall for classifying normal and IDS classes, demonstrating its reliability for detecting botnet attacks. To bolster security measures, the system incorporates encryption and access control mechanisms. Data encryption, utilizing the Fernet symmetric encryption algorithm, ensures the confidentiality of transmitted data. Access control mechanisms differentiate between authorized and unauthorized devices, enhancing overall security.