One of the primary security risks in IoT networks is address-based attacks, where hackers can exploit vulnerabilities by targeting specific IP addresses. To address this issue, various security techniques have been proposed, including address shuffling. Address shuffling is a security technique that involves randomly changing the IP address of an IoT device periodically, making it difficult for hackers to target the device. However, manually changing IP addresses can be cumbersome, especially when dealing with several IoT devices. We propose an AI-based address shuffling technique for IoT security. Our approach involves using machine learning algorithms to predict the best time to change the IP address of an IoT device based on various parameters such as accuracy, precision, throughput &time taken or speed to shuffle the IP address. We show that our approach can effectively reduce the risk of address-based attacks in IoT networks, while minimizing the impact on device performance and usability. Our experiments demonstrate that our approach outperforms existing techniques and can provide a more secure and efficient IoT environment.

IoT Security, IP Address Shuffling, Artificial Intelligence, IDS