Search

Software defined radio frequency (SDRF) sensing technology has revolutionized healthcare by enabling real-time monitoring and early diagnosis of patient health status with higher reliability, diagnostic accuracy, and enhanced healthcare services in a noncontact and noninvasive manner. However, RF sensing for breathing disorder diagnosis and monitoring is still an open research challenge. Further research is necessary to determine RF sensing accuracy and reliability for breathing disorders in different environments and applications. RF sensing is sensitive to environmental changes and shows nonlinear responses. Existing studies have explored RF sensing for breathing monitoring using fixed RF parameters to evaluate the system’s performance. However, several key parameters in RF sensing, such as operating frequency, sampling rate, bandwidth, gain, power, the height of antennas, and distance between transmitter and receiver, affect the system’s performance practicality. In this article, we used a reconfigurable SDRF sensing system to evaluate the RF parameters for monitoring breathing in order to understand their effects and enhance the performance of the sensing system. The correlation between RF sensing characteristics and wearable breathing sensors is evaluated using the correlation coefficient (CC) and mean square error (MSE). The findings reveal that a higher operating frequency of 4.8 GHz, a sampling rate of 300 samples/s, antennas on the line of sight, and distance up to 2 feet show the best performance, with an MSE of less than 0.1111 and a CC of 0.9943, indicating a significant correlation. The experimental study concludes that breathing monitoring performance using RF sensing heavily depends on RF parameters.

Millimeter-wave (mmWave) automotive radar can detect the surrounding environment using highly directional, high-frequency electromagnetic waves. With these advantages, mmWave radar has become an important component of autonomous driving and integrated sensing systems. Designing systems and sensing algorithms requires the application of realistic channel models and simulation. In this paper, the propagation channel is measured, characterized, and modeled for mmWave automotive radar with typical configurations in scenarios. Based on the channel measurements in urban street and expressway environments, ray-tracing (RT) technology is verified for modeling important objects regarding radar cross-section and echo power. A hybrid channel model is proposed by integrating RT with the stochastic modeling of targets and surrounding environments, which significantly improves simulation efficiency and provides more flexibility for virtual tests in various complex environments.