AI-Driven Energy-Aware Routing Protocol for Scalable Wireless Sensor Networks in 5G-Enabled IoT Environments

Abstract

Large-scale wireless sensor networks (WSN) deployment using Internet of Things (IoT) infrastructures enabled by 5G has been accelerated by its rapid growth. Nonetheless, the incorporation of dense WSNs into the 5G backhaul creates highly problematic issues in energy capacity, scalability, and latency controls. The traditional routing algorithms like LEACH and AODV are based on established systems and do not have adaptive intelligence, and they cannot fit dynamic 5G-IoT systems. The current energy-aware routing schemes do not collectively optimise the energy usage, delay, and path reliability. In this paper, we suggest an AI-assisted multi-objective routing architecture that employs reinforcement learning to provide scalable WSN operation in 5G-downstream IoT systems. The formulation of routing decisions is sequentially optimised whereby every node chooses the best next-hop based on the residual energy, path distance, and latency. With a weighted cost function and a Q-learning, adaptive state-aware routing is allowed to exist in the different network conditions. The presented simulation outcomes show that the suggested protocol is more likely to prolong the lifetime of a network by 30 percent, decrease the average energy usage by 22 percent, and minimise end-to-end latency than traditional protocols. This paper authorises that learning-based routing is a significant energy efficiency and scalability contributor in next-generation 5G-integrated IoT sensor networks.