Trustworthy AI-Driven 6G-IoT Architecture for Remote Healthcare: Reliable Resource Orchestration and Adaptive Network Intelligence
Yiya Sun, Bidare Divakarachari Parameshachari, Hao Wang · IEEE Internet of Things Journal · 2025
The convergence of sixth-generation (6G) communication networks with the Internet of Things (IoT) introduces a transformative paradigm for remote healthcare services, which enables ultra-reliable, low-latency, and intelligent medical communication in geographically challenged or underserved areas. However, the integration of Terrestrial Networks (TNs) and Non-Terrestrial Networks (NTNs) under a unified 6G framework presents unique challenges in resource management, signal robustness, and system trustworthiness—particularly for life-critical healthcare scenarios. In this paper, we present an AI-enhanced TN-NTN architecture designed to optimize the delivery of real-time healthcare applications such as telemedicine, remote diagnostics, and patient monitoring. Leveraging advanced intelligent models including proximal policy optimization, graph neural networks, multi-agent deep reinforcement learning, and federated learning, the proposed architecture dynamically manages network resources, mitigates interference, and adapts to varying traffic conditions. Simulation results demonstrate that the architecture achieves certain improvements in latency, throughput, and packet delivery rate compared to other models, even under high network loads. These findings highlight the potential of the proposed system to enable robust and cost-effective telemedicine services, bridging critical gaps in healthcare delivery for remote and resource-constrained environments. This research paves the way for AI-driven approaches in expanding the reach of healthcare services, with implications for next-generation 6G networks and beyond.