Optimizing Cluster Head Selection and Routing in 5G WSNs: A Reinforcement Learning and Deep Learning Approach

The Internet of Things (IoT) and 5G wireless sensor networks (WSNs) have significantly transformed data transmis-sion and facilitated efficient inter-device communication. However, routing within these networks presents chal-lenges owing to constraints related to energy, latency, and packet loss. This study introduces an effective data trans-fer mechanism for IoT-based 5G WSN, utilizing a deep belief network (DBN) topology. The proposed framework incorporates a reinforcement learning (RL) approach for clustering and employs the Mantaray Foraging Optimiza-tion (MRFO) algorithm for cluster head (CH) selection, considering multiple objectives such as energy, delay, traffic density, and distance. Subsequently, a DBN-based routing protocol is applied for efficient data transmission. Exten-sive simulations indicate that the proposed architecture surpasses existing techniques in terms of network lifetime, throughput, number of active nodes, and energy consumption. Complexity analysis and statistical evaluation fur-ther substantiate the effectiveness of the proposed method. This study contributes to the advancement of energy-efficient and reliable routing protocols for IoT-based 5G WSNs, addressing the challenges associated with data transmission in these networks. The proposed framework has the potential to enhance the performance of various IoT applications, including smart cities, healthcare, and industrial automation, by ensuring efficient data transfer and extending the network lifetime.