Computer vision-aided mmWave communications for indoor medical healthcare.

ABSTRACT

Comprehensive and exceedingly precise centralized patient monitoring has become essential to advance predictive, preventive, and efficient patient care in contemporary healthcare. Millimeter-wave (mmWave) technology, boasting high-frequency and high-speed wireless communication, holds promise as a viable solution to this challenge. This paper presents a new approach that combines mmWave communication and computer vision (CV) to achieve real-time patient monitoring and data transmission in indoor medical environments. The system comprises a transmitter, a reflective surface, and multiple communication targets, and utilizes the high-frequency, low-latency features of mmWave as well as CV-based target detection and depth estimation for precise localization and reliable data transmission. A machine learning algorithm analyses real-time images captured by an optical camera to identify target distance and direction and establish clear line-of-sight links. The system proactively adapts its transmission power and channel allocation based on the target's movements, guaranteeing complete coverage, even in potentially obstructive areas. This methodology tackles the escalating demand for high-speed, real-time data processing in modern healthcare, significantly enhancing its delivery.