Abstract
Deploying deep learning (DL) models onto low-power devices for Human Activity Recognition (HAR) purposes is gaining momentum because of the pervasive adoption of wearable sensor devices. However, the outcome of such deployment needs exploration not only because the topic is still in its infancy, but also because of the wide combination between low-power devices, deep models, and available deployment strategies. We have investigated the outcome of the application of three compression techniques, namely lite conversion, dynamic quantization, and full-integer quantization, that allow the deployment of deep models on low-power devices. This paper describes how those three compression techniques impact accuracy and energy consumption on an ESP32 device. In terms of accuracy, the full-integer technique incurs an accuracy drop between 2% and 3%, whereas the dynamic quantization and the lite conversion result in a negligible accuracy drop. In terms of power efficiency, dynamic and full-integer quantization allow for saving almost 30% of energy. The adoption of one of those two quantization techniques is recommended to obtain an executable network model, and we advise the adoption of the dynamic quantization given the negligible accuracy drop (Chiara Contoli is a researcher co-funded by the European Union - PON Research and Innovation 2014-2020.).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Augasta, M., Kathirvalavakumar, T.: Pruning algorithms of neural networks-a comparative study. Open Comput. Sci. 3(3), 105–115 (2013)
Chaman, S.: Techniques for compressing deep convolutional neural network. In: 2020 International Conference on Computational Performance Evaluation (ComPE), pp. 048–053. IEEE (2020)
Choudhary, T., Mishra, V., Goswami, A., Sarangapani, J.: A comprehensive survey on model compression and acceleration. Artif. Intell. Rev. 53(7), 5113–5155 (2020). https://doi.org/10.1007/s10462-020-09816-7
Daghero, F., et al.: Human activity recognition on microcontrollers with quantized and adaptive deep neural networks. ACM Trans. Embed. Comput. Syst. (TECS) 21(4), 1–28 (2022)
Daghero, F., Pagliari, D.J., Poncino, M.: Two-stage human activity recognition on microcontrollers with decision trees and CNNs. In: 2022 17th Conference on Ph. D Research in Microelectronics and Electronics (PRIME), pp. 173–176. IEEE (2022)
Daghero, F., et al.: Ultra-compact binary neural networks for human activity recognition on RISC-V processors. In: Proceedings of the 18th ACM International Conference on Computing Frontiers, pp. 3–11 (2021)
Espressif: Esp32-c3-wroom-02 datasheet (2022). https://www.espressif.com/en/support/documents/technical-documents Accessed 07 Feb 2023
Ghibellini, A., Bononi, L., Di Felice, M.: Intelligence at the IoT edge: activity recognition with low-power microcontrollers and convolutional neural networks. In: 2022 IEEE 19th Annual Consumer Communications & Networking Conference (CCNC), pp. 707–710. IEEE (2022)
Gou, J., Yu, B., Maybank, S.J., Tao, D.: Knowledge distillation: a survey. Int. J. Comput. Vis. 129, 1789–1819 (2021)
InvenSense Inc.: Mpu-6050 product specification (2023). https://invensense.tdk.com/products/motion-tracking/6-axis/mpu-6050/ Accessed 07 Feb 2023
Khoram, S., Li, J.: Adaptive quantization of neural networks. In: International Conference on Learning Representations (2018)
Liang, T., Glossner, J., Wang, L., Shi, S., Zhang, X.: Pruning and quantization for deep neural network acceleration: A survey. Neurocomputing 461, 370–403 (2021)
National. Instruments: Pc-6251 datasheet (2020). http://www.ni.com/pdf/manuals/375213c.pdf Accessed 07 Feb 2023
Novac, P.E., Boukli Hacene, G., Pegatoquet, A., Miramond, B., Gripon, V.: Quantization and deployment of deep neural networks on microcontrollers. Sensors 21(9), 2984 (2021)
Novac, P.E., Castagnetti, A., Russo, A., Miramond, B., Pegatoquet, A., Verdier, F.: Toward unsupervised human activity recognition on microcontroller units. In: 2020 23rd Euromicro Conference on Digital System Design (DSD), pp. 542–550. IEEE (2020)
Pang, B., Nijkamp, E., Wu, Y.N.: Deep learning with tensorFlow: a review. J. Educ. Behav. Stat. 45(2), 227–248 (2020)
Reyes-Ortiz, J.L., Oneto, L., Samà, A., Parra, X., Anguita, D.: Transition-aware human activity recognition using smartphones. Neurocomputing 171, 754–767 (2016)
Rohde &Schwarz: Ngmo2 datasheet (2020). https://www.rohde-schwarz.com/it/brochure-scheda-tecnica/ngmo2/ Accessed 07 Feb 2023
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 IFIP International Federation for Information Processing
About this paper
Cite this paper
Contoli, C., Lattanzi, E. (2023). Energy Efficiency of Deep Learning Compression Techniques in Wearable Human Activity Recognition. In: Maglogiannis, I., Iliadis, L., MacIntyre, J., Dominguez, M. (eds) Artificial Intelligence Applications and Innovations. AIAI 2023. IFIP Advances in Information and Communication Technology, vol 675. Springer, Cham. https://doi.org/10.1007/978-3-031-34111-3_10
Download citation
DOI: https://doi.org/10.1007/978-3-031-34111-3_10
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-34110-6
Online ISBN: 978-3-031-34111-3
eBook Packages: Computer ScienceComputer Science (R0)