AI-enhanced context-aware optimization of RPL routing protocol for IoT environments

The Internet of Things networks interconnect devices with sensing, communication, and data-driven capabilities, transforming various domains. However, IoT applications also encounter issues of network security, reliability and energy consumption. This paper addresses the vital challenge of designing efficient routing protocols for IoT networks. We focus on RPL, a widely used routing protocol for IoT, which has drawbacks such as ignoring context, causing congestion, and being vulnerable to attacks. To overcome these limitations, this paper proposes a novel AI-enhanced context-aware optimization method for RPL in IoT settings. The method involves a context-aware objective function that considers factors like energy, link quality, traffic, and mobility. An AI-enhanced optimization technique that employs reinforcement learning and deep neural networks adapts metric weights according to network changes and application needs. We compare our method with existing approaches (OF0 and MRHOF) through simulations, showing its better performance in improving network reliability and reducing energy consumption.