Answers are grounded in this paper and cite their sources.
In remote healthcare systems, the efficient, secure, and real-time transmission of biomedical signals such as ECG is critical. Traditional RF-based communication often suffers from interference, limited bandwidth, and security concerns. Visible Light Communication (VLC), particularly when combined with Orthogonal Frequency Division Multiplexing (OFDM), presents a promising alternative due to its high bandwidth, electromagnetic immunity, and inherent data security. However, VLC systems are highly sensitive to environmental dynamics like ambient light variation and patient movement, limiting their reliability. Previous research has explored AI-assisted VLC systems and modulation schemes, yet many suffer from static configurations, limited adaptability, high energy consumption, and lack of real-time optimization. This work introduces a novel Q-learning-optimized OFDM-VLC system tailored for remote health monitoring. The system leverages reinforcement learning to dynamically adjust modulation schemes, transmission power, and encoding strategies in response to environmental conditions (e.g., SNR, ambient light, mobility), enabling energy-efficient and error-resilient data transfer. Using the MIT-BIH Arrhythmia dataset, ECG signals are preprocessed, digitized, modulated using adaptive QPSK or 16-QAM, and transmitted over a VLC channel. A Q-learning agent selects optimal actions in real time to minimize BER and energy use while maximizing throughput and SNR. MATLAB was employed for system design, simulation, and performance evaluation. Compared to static systems, the proposed method reduced BER from 0.078 to 0.015, improved SNR from 21.3 dB to 29.8 dB, increased throughput from 16.7 kbps to 22.4 kbps, and lowered latency from 14.6 ms to 9.0 ms. Energy consumption dropped from 1.35 J/bit to 0.89 J/bit, and ECG reconstruction accuracy rose from 85.3% to 96.7%. The integration of reinforcement learning with VLC-OFDM significantly enhances the reliability, efficiency, and adaptability of real-time biomedical data transmission in remote health monitoring.
No takes yet. Share an insight, caveat, or question.
Devikala et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: