Intelligent Health Monitoring in 6G Networks: Machine Learning-Enhanced VLC-Based Medical Body Sensor Networks.

Area of Science:

• Wireless Communication
• Artificial Intelligence
• Medical Technology

Background:

• Sixth Generation (6G) wireless technology adoption is increasing in healthcare.
• Visible Light Communication (VLC) offers high data rates and mitigates electromagnetic interference (EMI) using existing lighting.
• Patient movement in hospitals creates dynamic channel conditions impacting VLC signal strength.

Purpose of the Study:

• To develop a novel channel modeling approach for VLC-enabled Medical Body Sensor Networks (MBSNs) in hospital settings.
• To integrate site-specific ray tracing with machine learning (ML) for accurate channel response prediction.
• To address challenges posed by dynamic hospital environments and patient mobility.

Main Methods:

• A Q-learning-based adaptive modulation scheme was introduced for real-time symbol error rate (SER) management.
• A Long Short-Term Memory (LSTM) network was developed to estimate path loss and Root Mean Square (RMS) delay spread.
• Ray tracing was used for channel impulse response (CIR) modeling combined with ML techniques.

Main Results:

• The Q-learning method achieved target SERs with spectral efficiency (SE) near the threshold.
• LSTM estimation demonstrated high accuracy in predicting path loss and RMS delay spread under dynamic conditions.
• The Intensive Care Unit (ICU) showed the highest Root Mean Square Error (RMSE) for path loss and delay spread with LSTM estimation (D1), while Family-Type Patient Rooms (FTPRs) showed the highest RMSE (D3).

Conclusions:

• This research presents the first combined ray-tracing and ML approach for VLC-MBSN channel modeling in hospitals.
• The proposed methods demonstrate high accuracy and adaptability to dynamic hospital environments.
• The findings support the integration of AI-driven VLC for reliable wireless communication in future healthcare.