Low-cost UWB radar estimates breathing rates using CNN for remote healthcare

Researchers have developed a low-cost system using IEEE 802.15.4z compliant IR-UWB hardware to estimate human respiration rates for remote healthcare applications. A convolutional neural network (CNN) was trained to predict breathing rates from UWB channel impulse response data, achieving a mean absolute error of 1.73 breaths per minute (BPM) in unseen situations. This CNN model was optimized for embedded deployment, reducing memory requirements by 67% and inference time by 62% with a minimal increase in error, making it feasible for an nRF52840 system-on-chip. The system is highly energy-efficient, capable of operating for over 260 days on a single battery charge while continuously monitoring a room. AI

IMPACT Enables low-cost, continuous health monitoring, potentially improving early detection of respiratory issues.

RANK_REASON Academic paper detailing a novel application of UWB radar and CNN for health monitoring. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Anton Lambrecht
• breaths per minute
• CNN
• IEEE 802.15.4z
• nRF52840
• system on a chip
• ultra-wideband

• paper
• product
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →