Energy Optimized Route Selection in WSNs for Smart IoT Applications

Interest in the IoT and smart cities has been growing as people learn about its potential applications in fields as diverse as healthcare, remote monitoring, and transportation. In these Internet of Things (IoT)-based systems, wireless networked sensors (WSNs) gather data critical to the operation of smart surroundings. IoT-enabled WSNs face challenges such high latency, low bandwidth, and short network lifespan due to the copious amounts of data generated by a wide variety of sensors. This study presents a deep reinforcement learning-based efficient routing method for IoT-enabled WSNs to combat latency as well as electricity consumption (DRL). The proposed strategy separates the network into unequal cluster according to the present data transmission existing in the sensors, hence preventing the network from collapsing prematurely. Extensive testing has been performed in ns3 using the recommended strategy. The results of the experiments are contrasted to the state-of-the-art methodologies to demonstrate that the proposed method is effective in the areas in received packets, connectivity latency, clean energy, and the amount of living nodes within a network.