Abstract

This research presents an optimized physiological monitoring system for COVID-19 patients, integrating Visible Light Communication (VLC)-based three-dimensional localization with Markov Chain Analysis. The proposed system enables real-time tracking of vital physiological indicators while balancing localization accuracy and computational efficiency, making it suitable for real-world healthcare applications. The VLC-based localization system was implemented using three LED beacons and optimized through Particle Swarm Optimization (PSO). The analysis revealed that optimal PSO parameters ( c(1)=1.9 , c(2)=2.1 , and w=0.8 ) significantly improved positioning accuracy, with 700 to 1100 particles providing the best trade-off between precision and computational cost. Additionally, the system successfully measured and calibrated cough frequency, respiratory rate, and oxygen saturation (SpO(2) ) using MEMS sensors. The results showed a cough frequency peak at 0.2 Hz, an average respiratory rate of 1.3 breaths per minute, and precise detection of hypoxia events through infrared and red light absorption. To assess disease progression, a Markov Chain Model was developed, analyzing heart rate, temperature, cough frequency, respiratory rate, and SpO(2) levels. The model identified four distinct patient states, ranging from mild to severe conditions, and provided probabilistic insights into symptom deterioration. A heat map analysis confirmed the reliability of state transition probabilities. The study underscores the critical trade-off between localization accuracy and computational efficiency, emphasizing the importance of careful parameter selection for real-time medical applications.