Low-cost Embedded Breathing Rate Determination Using 802.15.4z IR-UWB Hardware for Remote Healthcare

2025-04-03

AI Summary

To address the high mortality rate associated with respiratory diseases, this paper proposes a contactless respiration rate monitoring system leveraging a low-cost IEEE 802.15.4z impulse-radio ultra-wideband (IR-UWB) radar. The method extracts human channel impulse responses and employs a lightweight convolutional neural network (CNN), which is the first to be deployed on the nRF52840 SoC with 8-bit weight quantization and tensor quantization—requiring only 46 KB of memory and 192 ms per inference. A cross-subject calibration strategy reduces the mean absolute error (MAE) in unseen scenarios to 0.84 BPM. Evaluated on a multi-environment dataset comprising 16 subjects, the system achieves an overall MAE of 1.73 BPM—improving upon conventional methods by 3.40 BPM. Quantization compresses model size by 67%, accelerates inference by 64%, and incurs only a +0.05 BPM accuracy degradation. With a battery lifetime of up to 268 days (extended to 313 days in bedside settings), the solution enables long-term, embedded, battery-powered respiratory monitoring.

Technology Category

• Machine Learning: Hardware-aware ML
• Application Domains: Internet of Things, Sensor Networks & Smart Cities
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Respiratory diseases account for a significant portion of global mortality. Affordable and early detection is an effective way of addressing these ailments. To this end, a low-cost commercial off-the-shelf (COTS), IEEE 802.15.4z standard compliant impulse-radio ultra-wideband (IR-UWB) radar system is exploited to estimate human respiration rates. We propose a convolutional neural network (CNN) to predict breathing rates from ultra-wideband (UWB) channel impulse response (CIR) data, and compare its performance with other rule-based algorithms. The study uses a diverse dataset of 16 individuals, incorporating various real-life environments to evaluate system robustness. Results show that the CNN achieves a mean absolute error (MAE) of 1.73 breaths per minute (BPM) in unseen situations, significantly outperforming rule-based methods (3.40 BPM). By incorporating calibration data from other individuals in the unseen situations, the error is further reduced to 0.84 BPM. In addition, this work evaluates the feasibility of running the pipeline on a low-cost embedded device. Applying 8-bit quantization to both the weights and input/ouput tensors, reduces memory requirements by 67% and inference time by 64% with only a 3% increase in MAE. As a result, we show it is feasible to deploy the algorithm on an nRF52840 system-on-chip (SoC) requiring only 46 KB of memory and operating with an inference time of only 192 ms. Once deployed, the system can last up to 268 days without recharging using a 20 000 mAh battery pack. For breathing monitoring in bed, the sampling rate can be lowered, extending battery life to 313 days, making the solution highly efficient for real-world, low-cost deployments.

Problem

• Develop low-cost IR-UWB system for breathing rate detection
• Compare CNN performance with rule-based algorithms
• Evaluate embedded feasibility for remote healthcare deployment

Innovation

• Uses IR-UWB radar for breathing rate detection
• Employs CNN to analyze UWB channel data
• Optimized for low-cost embedded devices