Implementation and Analysis of a Low-Power Heart Rate Prediction Solution Using Hardware Acceleration on the ESP32-S3 Microcontroller
Resumo
This work details the implementation of an accelerometry-based heart-rate estimation system optimized for the ESP32-S3 microcontroller. It leverages PIE-based Single Instruction, Multiple Data (SIMD) extensions via ESP-DSP library to accelerate the pre-processing, training and inference of a continual-learning model. Under a modular FreeRTOS architecture, the system achieved 4.99 µs training latency, requiring less than 1 kB of RAM for the ML model itself. Results show a 30.6% computational performance gain and 8% energy consumption reduction with acceleration enabled. These efficiencies allow race-to-sleep strategies that extend the estimated battery life to 51 days on COTS hardware.Referências
Allen, J. (2007). Photoplethysmography and its application in clinical physiological measurement. Physiological Measurement, 28(3):R1–R39. PMID: 17322588.
Bieri, V. et al. (2023). Beliefppg: Uncertainty-aware heart rate estimation from ppg signals via belief propagation. In Uncertainty in Artificial Intelligence, pages 173–183. PMLR.
Del-Valle-Soto, C. et al. (2024). Unveiling wearables: exploring the global landscape of biometric applications and vital signs and behavioral impact. BioData Mining, 17(1):15.
Elsts, A. et al. (2018). On-board feature extraction from acceleration data for activity recognition. In Proceedings of the 2018 International Conference on Embedded Wireless Systems and Networks, EWSN ’18, page 163–168. International Conference on Embedded Wireless Systems and Networks (EWSN).
Espressif Systems (2025a). ESP32-S3 Technical Reference Manual. Espressif Systems (Shanghai) Co., Ltd., Shanghai, China. Version 1.7.
Espressif Systems (2025b). ESP32-S3-WROOM-1 & ESP32-S3-WROOM-1U Datasheet. Espressif Systems (Shanghai) Co., Ltd., Shanghai, China. Version 1.7.
Lee, H., Chung, H., and Lee, J. (2019). Motion artifact cancellation in wearable photo-plethysmography using gyroscope. IEEE Sensors Journal, 19(3):1166–1175.
Ni, J., Muhlstein, L., and McAuley, J. (2019). Modeling heart rate and activity data for personalized fitness recommendation. In WWW ’19: The World Wide Web Conference, pages 1343–1353.
Ogata, H. et al. (2024). Individually optimized estimation of energy expenditure in rescue workers using a tri-axial accelerometer and heart rate monitor. Frontiers in Physiology, Volume 15 - 2024.
Pacheco, A. G. C. et al. (2023a). Method and wearable electronic system for predicting the heart rate of a user during a physical activity, and, non-transitory computer readable storage medium. US Patent Application US 2023/0190120 A1. Published: Jun. 22, 2023. Filed: Feb. 24, 2022.
Pacheco, A. G. C. et al. (2023b). Towards low-power heart rate estimation based on user’s demographics and activity level for wearables. In ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1–5.
Pacheco, A. G. C. et al. (2024). Learning to estimate heart rate from accelerometer and user’s demographics during physical exercises. IEEE Journal of Biomedical and Health Informatics, 28(9):5092–5102.
Shin, J. and Cho, J. (2019). Noise-robust heart rate estimation algorithm from photo-plethysmography signal with low computational complexity. Journal of Healthcare Engineering, 2019:1–7.
Veluguri, S. P. (2025). Deep ppg: Improving heart rate estimates with activity prediction. In 2025 1st International Conference on AIML-Applications for Engineering & Technology (ICAET), pages 1–6.
Wilcoxon, F. (1945). Individual comparisons by ranking methods. Biometrics Bulletin, 1(6):80–83.
Zakynthinaki, M. S. (2015). Modelling heart rate kinetics. PLOS ONE, 10(4):1–26.
Zhang, Z., Pi, Z., and Liu, B. (2015). Troika: A general framework for heart rate monitoring using wrist-type photoplethysmographic signals during intensive physical exercise. IEEE Transactions on Biomedical Engineering, 62(2):522–531.
Bieri, V. et al. (2023). Beliefppg: Uncertainty-aware heart rate estimation from ppg signals via belief propagation. In Uncertainty in Artificial Intelligence, pages 173–183. PMLR.
Del-Valle-Soto, C. et al. (2024). Unveiling wearables: exploring the global landscape of biometric applications and vital signs and behavioral impact. BioData Mining, 17(1):15.
Elsts, A. et al. (2018). On-board feature extraction from acceleration data for activity recognition. In Proceedings of the 2018 International Conference on Embedded Wireless Systems and Networks, EWSN ’18, page 163–168. International Conference on Embedded Wireless Systems and Networks (EWSN).
Espressif Systems (2025a). ESP32-S3 Technical Reference Manual. Espressif Systems (Shanghai) Co., Ltd., Shanghai, China. Version 1.7.
Espressif Systems (2025b). ESP32-S3-WROOM-1 & ESP32-S3-WROOM-1U Datasheet. Espressif Systems (Shanghai) Co., Ltd., Shanghai, China. Version 1.7.
Lee, H., Chung, H., and Lee, J. (2019). Motion artifact cancellation in wearable photo-plethysmography using gyroscope. IEEE Sensors Journal, 19(3):1166–1175.
Ni, J., Muhlstein, L., and McAuley, J. (2019). Modeling heart rate and activity data for personalized fitness recommendation. In WWW ’19: The World Wide Web Conference, pages 1343–1353.
Ogata, H. et al. (2024). Individually optimized estimation of energy expenditure in rescue workers using a tri-axial accelerometer and heart rate monitor. Frontiers in Physiology, Volume 15 - 2024.
Pacheco, A. G. C. et al. (2023a). Method and wearable electronic system for predicting the heart rate of a user during a physical activity, and, non-transitory computer readable storage medium. US Patent Application US 2023/0190120 A1. Published: Jun. 22, 2023. Filed: Feb. 24, 2022.
Pacheco, A. G. C. et al. (2023b). Towards low-power heart rate estimation based on user’s demographics and activity level for wearables. In ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1–5.
Pacheco, A. G. C. et al. (2024). Learning to estimate heart rate from accelerometer and user’s demographics during physical exercises. IEEE Journal of Biomedical and Health Informatics, 28(9):5092–5102.
Shin, J. and Cho, J. (2019). Noise-robust heart rate estimation algorithm from photo-plethysmography signal with low computational complexity. Journal of Healthcare Engineering, 2019:1–7.
Veluguri, S. P. (2025). Deep ppg: Improving heart rate estimates with activity prediction. In 2025 1st International Conference on AIML-Applications for Engineering & Technology (ICAET), pages 1–6.
Wilcoxon, F. (1945). Individual comparisons by ranking methods. Biometrics Bulletin, 1(6):80–83.
Zakynthinaki, M. S. (2015). Modelling heart rate kinetics. PLOS ONE, 10(4):1–26.
Zhang, Z., Pi, Z., and Liu, B. (2015). Troika: A general framework for heart rate monitoring using wrist-type photoplethysmographic signals during intensive physical exercise. IEEE Transactions on Biomedical Engineering, 62(2):522–531.
Publicado
19/07/2026
Como Citar
CLEVELARES, João P. M.; OLIVEIRA, Higor D.; PACHECO, André G. C.; SOUZA JR, Luis A.; VILLAÇA, Rodolfo S..
Implementation and Analysis of a Low-Power Heart Rate Prediction Solution Using Hardware Acceleration on the ESP32-S3 Microcontroller. In: SEMINÁRIO INTEGRADO DE SOFTWARE E HARDWARE (SEMISH), 53. , 2026, Gramado/RS.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 388-399.
ISSN 2595-6205.
DOI: https://doi.org/10.5753/semish.2026.21854.
