analiza aiqbench
This research demonstrates a practical application of machine learning by deploying an optimized convolutional neural network on ultra-low-power hardware to monitor breathing rates via IR-UWB radar. The study validates that 8-bit quantization enables efficient inference on embedded SoCs, significantly reducing memory and energy consumption while maintaining high accuracy in real-world environments. By releasing the dataset and proving feasibility for long-term battery operation, this work bridges advanced AI models with accessible IoT hardware for remote healthcare monitoring.
Low-cost Embedded Breathing Rate Determination Using 802.15.4z IR-UWB Hardware for Remote Healthcare
arXiv:2504.03772v3 Announce Type: replace-cross 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 used to estimate human respiration rates. We propose a convolutional neural network (CNN) specifically adapted to predict breathing rates from ultra-wideband (UWB) channel impulse response (CIR) data, and compare its performance with both other rule-based algorithms and model-based solutions. The study uses a diverse dataset, incorporating various real-life environments to evaluate system robustness. To facilitate future research, this dataset will be released as open source. 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/output tensors, reduces memory requirements by 67% and inference time by 62% 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 199 ms. Once deployed, an analytical energy model estimates that the system, while continuously monitoring the room, can operate for up to 268.2 days without recharging when powered by a 20 000 mAh battery pack. For breathing monitoring in bed, the sampling rate can be lowered, extending battery life to 313.8 days, making the solution highly efficient for real-world, low-cost deployments.
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 used to estimate human respiration rates. We propose a convolutional neural network (CNN) specifically adapted to predict breathing rates from ultra-wideband (UWB) channel impulse response (CIR) data, and compare its performance with both other rule-based algorithms and model-based solutions. The study uses a diverse dataset, incorporating various real-life environments to evaluate system robustness. To facilitate future research, this dataset will be released as open source. 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/output tensors, reduces memory requirements by 67% and inference time by 62% 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 199 ms. Once deployed, an analytical energy model estimates that the system, while continuously monitoring the room, can operate for up to 268.2 days without recharging when powered by a 20 000 mAh battery pack. For breathing monitoring in bed, the sampling rate can be lowered, extending battery life to 313.8 days, making the solution highly efficient for real-world, low-cost deployments.
Signal Processing (eess.SP); Machine Learning (cs.LG)
arXiv:2504.03772 [eess.SP]
(or arXiv:2504.03772v3 [eess.SP] for this version)
IEEE SENSORS JOURNAL, VOL. 26, NO. 14, 15 JULY 2026
https://doi.org/10.1109/JSEN.2026.3698502 DOI(s) linking to related resources
Text integral sub licența deschisă a publicației · 315 cuvinte
