Design and Implementation of Integrated RPL Protocol and Deep Learning for Energy-Aware Wireless Sensor Networks

Abstract

A wireless sensor network (WSN) represents a significant and contemporary technology that has recently emerged, playing a crucial role in enabling the Internet of Things (IoT) to bridge the gap between the physical environment and the digital domain. The major challenges facing WSNs include limited power, data security and reliability, and node failure, along with limitations in data storage, processing, analysis, power management, and security. WSNs with limited physical capabilities employ the low-power lossless routing (RPL) protocol, which employs various processes to simplify communications and network design. Despite its importance and effectiveness, the RPL protocol faces several challenges, including handling high traffic and load balancing, which can lead to service interruption. This paper proposes a machine learning model to address the challenges of energy consumption and efficiency in the RPL protocol. The proposed model is based on the use of Random Forest (RF) and Support Vector Machine (SVM) to identify the optimal path from source to interface, enhancing the network lifetime and delivering data packets in an energy-efficient manner. The model is implemented in two scenarios, one with uniformly distributed nodes and the other with randomly distributed nodes. The results demonstrate that the proposed system outperforms the standard RPL protocol and other protocols in terms of extending the network lifetime and enhancing energy efficiency.