Leveraging Edge-Fog Computing for Low-Latency and Resilient Medical IoT Communication

Abstract

Strong, low-latency communication frameworks are required for Medical Internet of Things (MIoT) applications due to the growing need for real-time data processing in the healthcare industry. The strict latency and reliability requirements of essential healthcare services are difficult for traditional cloud-based models to achieve. This study investigates how edge and fog computing can be combined to improve MIoT communication through increased resilience, scalability, and latency reduction. In order to ensure effective resource allocation, the suggested hybrid fuzzy-based reinforcement learning model optimizes task offloading among edge, fog, and cloud layers. According to experimental results, the model improves latency from 10.62 ms to 7.71 ms, a 27.4% reduction. Fault tolerance rises to 11.93%, guaranteeing system stability, while throughput rises by 26.1% to 15.20 Mbps. Scalability has also been improved to 10.05%, enabling dynamic response to rising device loads. These results demonstrate how edge-fog computing can facilitate safe, robust, and real-time MIoT communication networks.