Lightweight and Efficient CSI-Based Human Activity Recognition via Bayesian Optimization-Guided Architecture Search and Structured Pruning
This paper presents an integrated approach to developing lightweight, high-performance deep learning models for human activity recognition (HAR) using WiFi Channel State Information (CSI). Motivated by the need for accuracy and efficiency in resource-constrained environments, we combine Bayesian Opt...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/2/890 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832589179566948352 |
|---|---|
| author | Sungkwan Youm Sunghyun Go |
| author_facet | Sungkwan Youm Sunghyun Go |
| author_sort | Sungkwan Youm |
| collection | DOAJ |
| description | This paper presents an integrated approach to developing lightweight, high-performance deep learning models for human activity recognition (HAR) using WiFi Channel State Information (CSI). Motivated by the need for accuracy and efficiency in resource-constrained environments, we combine Bayesian Optimization-based Neural Architecture Search (NAS) with a structured pruning algorithm. NAS identifies optimal network configurations, while pruning systematically removes redundant parameters, preserving accuracy. This approach allows for robust activity recognition from diverse WiFi datasets under varying conditions. Experimental results across multiple benchmark datasets demonstrate that our method not only maintains but often improves accuracy after pruning, resulting in models that are both smaller and more accurate. This offers a scalable and adaptable solution for real-world deployments in IoT and mobile platforms, achieving an optimal balance of efficiency and accuracy in HAR using WiFi CSI. |
| format | Article |
| id | doaj-art-45a0241cf560410e88f6f2897be935b6 |
| institution | Kabale University |
| issn | 2076-3417 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-45a0241cf560410e88f6f2897be935b62025-01-24T13:21:13ZengMDPI AGApplied Sciences2076-34172025-01-0115289010.3390/app15020890Lightweight and Efficient CSI-Based Human Activity Recognition via Bayesian Optimization-Guided Architecture Search and Structured PruningSungkwan Youm0Sunghyun Go1Department of Information and Communication Engineering, Wonkwang University, Iksan 54538, Republic of KoreaDepartment of Computer Software Engineering, Wonkwang University, Iksan 54538, Republic of KoreaThis paper presents an integrated approach to developing lightweight, high-performance deep learning models for human activity recognition (HAR) using WiFi Channel State Information (CSI). Motivated by the need for accuracy and efficiency in resource-constrained environments, we combine Bayesian Optimization-based Neural Architecture Search (NAS) with a structured pruning algorithm. NAS identifies optimal network configurations, while pruning systematically removes redundant parameters, preserving accuracy. This approach allows for robust activity recognition from diverse WiFi datasets under varying conditions. Experimental results across multiple benchmark datasets demonstrate that our method not only maintains but often improves accuracy after pruning, resulting in models that are both smaller and more accurate. This offers a scalable and adaptable solution for real-world deployments in IoT and mobile platforms, achieving an optimal balance of efficiency and accuracy in HAR using WiFi CSI.https://www.mdpi.com/2076-3417/15/2/890WiFi sensinghuman activity recognitionBayesian optimizationneural architecture searchmodel compression |
| spellingShingle | Sungkwan Youm Sunghyun Go Lightweight and Efficient CSI-Based Human Activity Recognition via Bayesian Optimization-Guided Architecture Search and Structured Pruning Applied Sciences WiFi sensing human activity recognition Bayesian optimization neural architecture search model compression |
| title | Lightweight and Efficient CSI-Based Human Activity Recognition via Bayesian Optimization-Guided Architecture Search and Structured Pruning |
| title_full | Lightweight and Efficient CSI-Based Human Activity Recognition via Bayesian Optimization-Guided Architecture Search and Structured Pruning |
| title_fullStr | Lightweight and Efficient CSI-Based Human Activity Recognition via Bayesian Optimization-Guided Architecture Search and Structured Pruning |
| title_full_unstemmed | Lightweight and Efficient CSI-Based Human Activity Recognition via Bayesian Optimization-Guided Architecture Search and Structured Pruning |
| title_short | Lightweight and Efficient CSI-Based Human Activity Recognition via Bayesian Optimization-Guided Architecture Search and Structured Pruning |
| title_sort | lightweight and efficient csi based human activity recognition via bayesian optimization guided architecture search and structured pruning |
| topic | WiFi sensing human activity recognition Bayesian optimization neural architecture search model compression |
| url | https://www.mdpi.com/2076-3417/15/2/890 |
| work_keys_str_mv | AT sungkwanyoum lightweightandefficientcsibasedhumanactivityrecognitionviabayesianoptimizationguidedarchitecturesearchandstructuredpruning AT sunghyungo lightweightandefficientcsibasedhumanactivityrecognitionviabayesianoptimizationguidedarchitecturesearchandstructuredpruning |