Fuzzy-Based Routing Scheme with Reliable Transmission Enhanced by Deep Learning for Wireless Sensor Networks

The advent of fifth-generation (5G) communication technologies is set to transform the global Internet of Everything (IoE) by linking numerous physical devices to the Internet. This extensive connectivity plays a vital role in IoT applications such as smart homes, city management, transportation, surveillance and the Internet of Medical Things (IoMT). Although wireless communication seeks to achieve broader coverage, lower latency, higher throughput, and increased reliability, challenges remain in optimizing limited communication resources and enhancing coding, modulation, and channel estimation methods. Managing transmission power is essential for mitigating interference and conserving energy. Our study introduces a fuzzy logic optimization routing (FLOR) method designed to extend the lifespan of Wireless Sensor Networks (WSNs) by making routing decisions based on nodes' energy levels and network distances. We evaluate FLOR's performance against current methods, showing the significant improvements in throughput, packet delivery ratio, average end-to-end delay and normalized routing overhead. In addition a deep learning-based strategy is proposed for sub-band assignment and power control in IoT cellular networks. Utilizing the convolutional neural networks (CNNs), we enhance the resource allocation, decrease latency, and improve energy efficiency. Our results demonstrate that deep learning methods outperform traditional techniques, offering faster speeds, reduced latency, and better energy efficiency. The integration of deep learning with edge computing and network slicing further benefits various IoT applications, boosting network performance, reliability, and energy efficiency.