Improving the functionality of wireless sensor networks through the use of reinforcement learning and metaheuristic based energy efficient system.

To improve the routing and data transmission of wireless sensor networks, this study recommends utilising a new clustering mechanism that is energy efficient and intelligent. First, the positions of the sensor nodes in the network and energy values are acquired. Harris hawks optimization, a meta-heuristic that considers the distance, dispersion, and energy balance of sensors, is then used to do energy-aware clustering. To select the cluster head, a new reinforcement learning technique has been created. A fuzzy logic system that has been trained using the concepts of reinforcement learning is used to illustrate this approach. It seeks to select the cluster head with the greatest remaining energy, the fewest neighbouring nodes, and the shortest path to the information station's centre. The fuzzy rules for cluster head selection have been established using the wild horse optimization method. These rules take into account two criteria: maximum energy and lowest distance between neighbouring nodes. The results showed that the suggested strategy resulted in a 29% increase in network lifetime and a 46% increase in the volume of data delivered to the station when the simulation was repeated on many networks and compared to the Low-Energy Adaptive Clustering Hierarchy approach.