Optimizing QoS‐Based Communication Protocol in Wireless Sensor Networks Using Genetic Algorithms With AI/ML Mechanisms for IoT Environment

ABSTRACT Wireless sensor networks (WSNs) designed for the Internet of Things (IoT) need to satisfy the service quality demands of real‐time applications. The communication and routing from WSN to IoT are adaptable to meet the service demands. To aid such a process, this article introduces a learning‐assisted genetic process for communication assistance (LGP‐CA) method. The proposed method identifies quality demands for the applications through intense requests and tenure. Based on the prolonged tenure, service degradations are identified to improve the back‐end application communication support. The prolonging tenures are identified using the genetic algorithm that generates time‐based remaining requests for the service intervals. In this case, the pending/prolonged service demands are serviced by allocating idle resources. This process is monitored using deep learning that executes the chances of request offloading or shared allocations. The communication assistance for prolonged service demands is handled by the WS nodes connected to the resources. Based on the learning suggestions, the node selection and discard processes are formulated. This process enhances the application service throughput, and service delivery, and reduces the service latency. The proposed method improves service throughput, delivery, and shared allocations by 9.83%, 8.99%, and 7.28%, respectively, for the maximum requests.