Intelligent-Slicing: A Profitable Framework for 5G and Beyond Networks in Healthcare

Healthcare is being revolutionized by the development of 5G and other emerging technologies, which enable advanced services such as real-time patient monitoring, remote procedures, and telemedicine. However, maintaining excellent service quality while effectively managing network resources remains a major challenge. We address this challenge by introducing a profit-driven framework called Intelligent-Slicing, which leverages Multi-Agent Deep Reinforcement Learning (MADRL) to optimize the allocation of network resources. In the preparation and planning phase, our approach uses MADRL to dynamically anticipate resource requirements, configure network slices, and optimize resource leasing. In the scheduling phase, MADRL-based scheduling ensures real-time prioritization of critical healthcare services, such as remote surgeries, even under high traffic conditions. The proposed framework offers an intelligent, adaptive, and cost-efficient network slicing strategy that enhances the overall performance and reliability of healthcare services in 5G and beyond networks. To align network operator profitability with Quality of Service (QoS) compliance, the framework integrates a cost-revenue optimization model and Software-Defined Networking (SDN) for flexible slice control. Simulation results demonstrate that Intelligent-Slicing effectively prioritizes mission-critical healthcare services while improving resource efficiency, service reliability, and profitability in 5G and beyond networks.