ABSTRACT In this paper, we propose a novel 6G‐enabled framework for enhancing fall risk detection in intelligent healthcare systems through the integration of communication, sensing, and computing. The method consists of three key components: (i) a Q‐learning‐driven adaptive modulation scheme that dynamically optimizes transmission parameters, (ii) a Long Short‐Term Memory (LSTM)‐based model for accurate estimation of channel parameters such as path loss and RMS delay spread, and (iii) a lightweight fall risk detection module that fuses wearable sensor signals (accelerometer and gyroscope) with channel quality indicators. This joint design ensures reliable data transmission, robust feature extraction, and accurate real‐time fall risk prediction under resource‐constrained conditions. Experimental evaluation on the MobiFall dataset demonstrates that the proposed framework achieves an accuracy of 98.5%, precision of 97.3%, and recall of 99.0%, outperforming state‐of‐the‐art baselines by 3.3–16.6 percentage points. Additionally, the system reduces energy consumption by 30% and achieves a throughput of 150 fps, compared to 80–100 fps for deep learning alternatives. These results highlight the framework's potential for practical deployment in resource‐constrained mobile healthcare environments.
Yan Zhang (2026) studied this question.