Multi-objective intelligent clustering routing schema for internet of things enabled wireless sensor networks using deep reinforcement learning

The Internet of Things (IoTloT) is built on a foundation of wireless sensor devices that connect humans and physical objects to the Internet and enable them to interact with one another to improve the living conditions of citizens. Wireless Sensor Networks (WSNs) are widely utilized in systems based on loT to collect the data required by intelligent environments. However, loT-enabled WSNs encounter a variety of difficulties such as poor network lifespan, limited throughput, and long communication delays, due to the massive non-homogenous data streaming from numerous sensor devices. Therefore, a multi-objective intelligent clustering routing schema for loT-enabled WSNs utilizing deep reinforcement learning is proposed in this paper to overcome these shortcomings. The proposed schema partitions the entire network into various unequal clusters based on the present data load existing in sensor nodes, effectively preventing the network from dying prematurely. In addition, an unequal clustering mechanism is utilized to balance inter-cluster and intra-cluster energy consumption among cluster heads. The simulation findings demonstrate the effectiveness of the proposed schema in terms of energy efficiency, delivered packets, end-to-end delay, alive nodes, energy balancing, and network lifespan compared with the other two state-of-the-art existing schemes.