Evaluating Stress Monitoring Pipelines at the Ultra-Edge: A Mobile Device-Based Study

• Vanessa Gamero USP
• Sergio T. Kofuji USP

DOI: https://doi.org/10.5753/sbcas.2026.21682

Abstract

Continuous stress monitoring using pervasive sensing generates multimodal data streams that are processed through pipelines, which may be implemented across distributed architectures. In this context, smartphones can operate as ultra-edge nodes, enabling local processing of these data while reducing latency and data transmission to upper layers. However, there is limited empirical analysis of how these pipelines operate under continuous execution at the ultra-edge. This study addresses this gap by evaluating the feasibility of executing a stress monitoring pipeline at the ultra-edge through the quantification of its data volume and computational requirements under continuous execution. For this purpose, each pipeline stage is analyzed in terms of generated data volume and the associated computational requirements in CPU, memory, and energy usage. A baseline configuration is compared with alternative strategies to reduce data volume and computational load. Preliminary results show that input data rates are low; however, continuous processing introduces sustained computational demands that must be considered for real-world use. These results provide a baseline for evaluating processing costs in subsequent pipeline stages.

References

Can, Y. S., Arnrich, B., and Ersoy, C. (2019). Stress detection in daily life scenarios using smart phones and wearable sensors: A survey. Journal of biomedical informatics, 92:103139.

Castaldo, R., Montesinos, L., Melillo, P., James, C., and Pecchia, L. (2019). Ultra-short term hrv features as surrogates of short term hrv: A case study on mental stress detection in real life. BMC medical informatics and decision making, 19(1):12.

Detweiler, C. A. and Hindriks, K. V. (2016). A survey of values, technologies and contexts in pervasive healthcare. Pervasive and Mobile Computing, 27:1–13.

Ferreira, D., Kostakos, V., and Dey, A. K. (2015). AWARE: Mobile Context Instrumentation Framework. Frontiers in ICT, 2.

Garcia-Ceja, E., Osmani, V., and Mayora, O. (2015). Automatic stress detection in working environments from smartphones’ accelerometer data: a first step. IEEE journal of biomedical and health informatics, 20(4):1053–1060.

Goodday, S. M. and Friend, S. (2019). Unlocking stress and forecasting its consequences with digital technology. npj Digital Medicine, 2(1):1–5.

Google for Developers (2026). Overview of google play services. Last updated 2026-03-02 UTC.

Jaiswal, D., Mukhopadhyay, S., and Sharma, V. (2024). Tinystressnet: On-device stress assessment with wearable sensors on edge devices. In 2024 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops), pages 166–171. IEEE.

Kang, S., Choi, W., Park, C. Y., Cha, N., Kim, A., Khandoker, A. H., Hadjileontiadis, L., Kim, H., Jeong, Y., and Lee, U. (2023). K-EmoPhone: A Mobile and Wearable Dataset with In-Situ Emotion, Stress, and Attention Labels. Scientific Data, 10:351.

Landreani, F., Faini, A., Martin-Yebra, A., Morri, M., Parati, G., and Caiani, E. G. (2019). Assessment of ultra-short heart variability indices derived by smartphone accelerometers for stress detection. Sensors, 19(17):3729.

Polar Electro Oy (2021). Polar ble sdk. [link]. GitHub repository, accessed 2026-03-07.

Rachakonda, L., Mohanty, S. P., Kougianos, E., and Sundaravadivel, P. (2019). Stresslysis: A dnn-integrated edge device for stress level detection in the iomt. IEEE Transactions on Consumer Electronics, 65(4):474–483.

Schaffarczyk, M., Rogers, B., Reer, R., and Gronwald, T. (2022). Validity of the polar h10 sensor for heart rate variability analysis during resting state and incremental exercise in recreational men and women. Sensors, 22(17):6536.

Shi, W., Cao, J., Zhang, Q., Li, Y., and Xu, L. (2016). Edge computing: Vision and challenges. IEEE internet of things journal, 3(5):637–646.

Smets, E., Rios Velazquez, E., Schiavone, G., Chakroun, I., D’Hondt, E., De Raedt, W., Cornelis, J., Janssens, O., Van Hoecke, S., Claes, S., Van Diest, I., and Van Hoof, C. (2018). Large-scale wearable data reveal digital phenotypes for daily-life stress detection. npj Digital Medicine, 1(1):1–10.

Van Gent, P., Farah, H., Van Nes, N., and Van Arem, B. (2019). Analysing noisy driver physiology real-time using off-the-shelf sensors: Heart rate analysis software from the taking the fast lane project. Journal of Open Research Software, 7(1).

Weiss, G. M. (2019). Wisdm smartphone and smartwatch activity and biometrics dataset. UCI Machine Learning Repository: WISDM Smartphone and Smartwatch Activity and Biometrics Dataset Data Set, 7(133190-133202):5.

Zhang, P., Jung, G., Alikhanov, J., Ahmed, U., and Lee, U. (2024). A reproducible stress prediction pipeline with mobile sensor data. Proceedings of the ACM on interactive, mobile, wearable and ubiquitous technologies, 8(3):1–35.