MLP-HAR: Boosting Performance and Efficiency of HAR Models on Edge Devices with Purely Fully Connected Layers
Open accessThe DOI matches its Crossref record; 29 of the 29 references it cites were found in a registry or an index.
Abstract
Neural network models have demonstrated exceptional performance in wearable human activity recognition (HAR) tasks. However, the increasing size or complexity of HAR models significantly impacts their deployment on wearable devices with limited computational power. In this study, we introduce a novel HAR model architecture named Multi-Layer Perceptron-HAR (MLP-HAR), which contains solely fully connected layers. This model is specifically designed to address the unique characteristics of HAR tasks, such as multi-modality interaction and global temporal information. The MLP-HAR model employs fully connected layers that alternately operate along the modality and temporal dimensions, enabling multiple fusions of information across these dimensions. Our proposed model demonstrates comparable performance with other state-of-the-art HAR models on six open-source datasets, while utilizing significantly fewer learnable parameters and exhibiting lower model complexity. Specifically, the complexity of our model is at least ten times smaller than that of the TinyHAR model and several hundred times smaller than the benchmark model DeepConvLSTM. Additionally, due to its purely fully connected layer-based architecture, MLP-HAR offers the advantage of ease of deployment. To substantiate these claims, we report the inference time performance of MLP-HAR on the Samsung Galaxy Watch 5 PRO and the Arduino Portenta H7 LITE, comparing it against other state-of-the-art HAR models.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“When resources are less constrained, efficient Deep Learning based classifiers can be employed, e.g., TinyHAR [222] and MLP-HAR [220].”
From Past, Present, and Future of Sensor-based Human Activity Recognition Using Wearables: A Surveying Tutorial on a Still Challenging Task (2025) · cites this paper as [220]
In the citing paper
“It refers to one of the state-of-the-art lightweight deep learning model designs for HAR applications.”
From Towards Customizable Foundation Models for Human Activity Recognition with Wearable Devices (2025) · cites this paper as [71]
Citations
Cites
Squeeze-and-Excitation Networks
Hu, J., Shen, L., & Sun, G. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00745 · Cited by 35,667
RepVGG: Making VGG-style ConvNets Great Again
Ding, X., Zhang, X., Ma, N., Han, J., Ding, G., & Sun, J. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.01352 · Cited by 2,907 · cited 1×
“These structures compromise memory utilization efficiency as they necessitate storing the outputs of each branch until combined through addition or concatenation [12], substantially increasing memory access costs and reducing inference speed.”
BackgroundDeep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition
Ordóñez, F., & Roggen, D. (2016) · Sensors · doi:10.3390/s16010115 · Cited by 2,761 · cited 4×
“Variants of RNNs, such as Gated Recurrent Units (GRU) [29], Long Short-Term Memory (LSTM) [25], and bidirectional-LSTM [38], are commonly used.”
BackgroundMethodInfluentialIntroducing a New Benchmarked Dataset for Activity Monitoring
Reiss, A., & Stricker, D. (2012) · 2012 16th International Symposium on Wearable Computers · doi:10.1109/iswc.2012.13 · Cited by 1,456
Transition-Aware Human Activity Recognition Using Smartphones
Reyes-Ortiz, J.-L., Oneto, L., Samà, A., Parra, X., & Anguita, D. (2016) · Neurocomputing · doi:10.1016/j.neucom.2015.07.085 · Cited by 740
Deep Learning for Sensor-based Human Activity Recognition
Chen, K., Zhang, D., Yao, L., Guo, B., Yu, Z., & Liu, Y. (2022) · ACM Computing Surveys · doi:10.1145/3447744 · Cited by 690 · cited 1×
“Wearable human activity recognition (HAR) has gained increasing attention due to its effectiveness in health monitoring, fitness tracking, and human-computer interaction [8].”
BackgroundDeepSense: A Unified Deep Learning Framework for Time-Series Mobile Sensing Data Processing
Yao, S., Hu, S., Zhao, Y., Zhang, A., & Abdelzaher, T. (2017) · Proceedings of the 26th International Conference on World Wide Web · doi:10.1145/3038912.3052577 · Cited by 667 · cited 2×
“For instance, DeepSense [35] and GlobalFusion [16] use spectrogram inputs obtained through FFT Transform.”
BackgroundMethodmHealthDroid: A Novel Framework for Agile Development of Mobile Health Applications
Banos, O., Garcia, R., Holgado-Terriza, J. A., Damas, M., Pomares, H., Rojas, I., Saez, A., & Villalonga, C. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-13105-4_14 · Cited by 583
A CNN-LSTM Approach to Human Activity Recognition
Mutegeki, R., & Han, D. S. (2020) · 2020 International Conference on Artificial Intelligence in Information and Communication (ICAIIC) · doi:10.1109/icaiic48513.2020.9065078 · Cited by 452 · cited 1×
“To capitalize on the strengths of both CNNs and RNNs, hybrid architectures [23, 25, 39] were developed.”
MethodDeep Residual Bidir-LSTM for Human Activity Recognition Using Wearable Sensors
Zhao, Y., Yang, R., Chevalier, G., Xu, X., & Zhang, Z. (2018) · Mathematical Problems in Engineering · doi:10.1155/2018/7316954 · Cited by 348
Research works printed but not linked: 13
- Nafees Ahmad and Ho-fung Leung. 2023. ALAE-TAE-CutMix: Beyond the State-of-the-Art for Human Activity Recognition Using Wearable Sensors. In 2023 IEEE International Conference on Pervasive Computing and Communications (PerCom). IEEE, 222--231.
- Jimmy Lei Ba, Jamie Ryan Kiros, and Geoffrey E Hinton. 2016. Layer normalization. arXiv preprint arXiv:1607.06450 (2016). · arXiv
- Si-An Chen, Chun-Liang Li, Nate Yoder, Sercan O Arik, and Tomas Pfister. 2023. Tsmixer: An all-mlp architecture for time series forecasting. arXiv preprint arXiv:2303.06053 (2023). · arXiv
- Sture Holm. 1979. A Simple Sequentially Rejective Multiple Test Procedure. Scandinavian journal of statistics (1979), 65--70.
- Diederik P Kingma and Jimmy Ba. 2014. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014). · arXiv
- Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. 2019. Pytorch: An imperative style, high-performance deep learning library. Advances in neural information processing systems, Vol. 32 (2019).
- Ilya O Tolstikhin, Neil Houlsby, Alexander Kolesnikov, Lucas Beyer, Xiaohua Zhai, Thomas Unterthiner, Jessica Yung, Andreas Steiner, Daniel Keysers, Jakob Uszkoreit, et al. 2021. Mlp-mixer: An all-mlp architecture for vision. Advances in neural information processing systems, Vol. 34 (2021), 24261--24272.
- Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. Advances in neural information processing systems, Vol. 30 (2017).
- Frank Wilcoxon. 1992. Individual comparisons by ranking methods. In Breakthroughs in Statistics: Methodology and Distribution. Springer, 196--202.
- Zhi-Qin John Xu, Yaoyu Zhang, and Tao Luo. 2022. Overview frequency principle/spectral bias in deep learning. arXiv preprint arXiv:2201.07395 (2022). · arXiv
- Jianbo Yang, Minh Nhut Nguyen, Phyo Phyo San, Xiaoli Li, and Shonali Krishnaswamy. 2015. Deep convolutional neural networks on multichannel time series for human activity recognition.. In Ijcai, Vol. 15. Buenos Aires, Argentina, 3995--4001.
- Haibin Zhao, Brojogopal Sapui, Michael Hefenbrock, Zhidong Yang, Michael Beigl, and Mehdi B Tahoori. 2023. Highly-bespoke robust printed neuromorphic circuits. In 2023 Design, Automation & Test in Europe Conference & Exhibition (DATE). IEEE, 1--6.
- Yexu Zhou, Tobias King, Yiran Huang, Haibin Zhao, Till Riedel, Tobias Röddiger, and Michael Beigl. 2024. Enhancing Efficiency in HAR Models: NAS Meets Pruning. In 2024 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops). IEEE, 33--38.
Cited by
Haresamudram, H., Tang, C. I., Suh, S., Lukowicz, P., & Plötz, T. (2025) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3729467 · Cited by 33
“For example, Contrastive Predictive Coding (CPC) [76, 136] utilizes a contrastive future prediction task, and it was found that predicting one or two timesteps in the future is (too) easy and results in poor performance, whereas predicting multiple timesteps into the future is much more complex,…”
Bian, S., Liu, M., Rey, V. F., Geißler, D., & Lukowicz, P. (2025) · ACM International Symposium on Wearable Computers (ISWC) · doi:10.1145/3715071.3750410 · Cited by 24
King, T., Zhou, Y., Röddiger, T., & Beigl, M. (2025) · Scientific Reports · doi:10.1038/s41598-025-90764-z · Cited by 24
“Recent methods, such as TinyLSTMs 24 , TinyHAR 19 and MLP-HAR 25 focus on designing efficient architectures that balance accuracy and resource constraints.”
Self-citationRey, V. F., Rezende, P. M. B., Zhou, B., Suh, S., & Lukowicz, P. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.129288 · Cited by 14
Long kernel distillation in human activity recognition
Yao, M., Cheng, D., Zhang, L., Li, W., Wu, H., & Song, A. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.113397 · Cited by 6
Metadata
- Authors
- Zhou, Y., King, T., Zhao, H., Huang, Y., Riedel, T., & Beigl, M.
- Affiliations
- Karlsruhe Institute of Technology (DE)
- Venue
- Proceedings of the 2024 ACM International Symposium on Wearable Computers
- Publisher
- ACM
- Published
- 2024-10-05
- Pages
- 133-139
- OpenAlex
- W4402828273
- Cited by
- 17
- References
- 27
- Citations by year
- 2026: 7 · 2025: 10
- FWCI
- 3.52
- Type
- conference paper
- Language
- en
- Open access
- Locations
- ACM International Symposium on Wearable Computers (ISWC) (publishedVersion)
- Repository KITopen (Karlsruhe Institute of Technology) (submittedVersion)
- KITopen (other-oa)
- Funders
- BMBF (01IS22095A)
- MWK BW (SDSC-BW)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Keywords
- human activity recognition, wearable devices, edge deployment, multilayer perceptron, fully connected layers, model compression, inference efficiency, multimodal fusion, temporal information
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: MLP-HAR: Boosting Performance and Efficiency of HAR Models on Edge Devices with Purely Fully Connected Layers. https://app.cochat.ai/papers/10.1145/3675095.3676624
- The paper
- Zhou, Y., King, T., Zhao, H., Huang, Y., Riedel, T., & Beigl, M. (2024). MLP-HAR: Boosting Performance and Efficiency of HAR Models on Edge Devices with Purely Fully Connected Layers. Proceedings of the 2024 ACM International Symposium on Wearable Computers. https://doi.org/10.1145/3675095.3676624
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.