Enhanced Human Activity Recognition Using Wi-Fi Sensing: Leveraging Phase and Amplitude with Attention Mechanisms

Wi-Fi-based human activity recognition (HAR) is a non-intrusive and privacy-preserving method that leverages Channel State Information (CSI) for identifying human activities. However, existing approaches often struggle with robust feature extraction, especially in dynamic and multi-environment scena...

وصف كامل

محفوظ في:

التفاصيل البيبلوغرافية
المؤلفون الرئيسيون:	Thai Duy Quy, Lin, Chih-Yang, Shih, Timothy K
التنسيق:	Journal article
اللغة:	English
منشور في:	2025
الموضوعات:	Wi-Fi sensing; channel state information (CSI); gate residual network; human activity recognition (HAR); multi-head attention; multi-scale convolutional neural networks; phase and amplitude
الوصول للمادة أونلاين:	https://scholar.dlu.edu.vn/handle/123456789/5040 https://www.mdpi.com/1424-8220/25/4/1038
الوسوم:	إضافة وسم لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!

Thư viện lưu trữ:	Thư viện Trường Đại học Đà Lạt

id	oai:scholar.dlu.edu.vn:123456789-5040
record_format	dspace
spelling	oai:scholar.dlu.edu.vn:123456789-50402025-08-20T09:28:01Z Enhanced Human Activity Recognition Using Wi-Fi Sensing: Leveraging Phase and Amplitude with Attention Mechanisms Thai Duy Quy Lin, Chih-Yang Shih, Timothy K Wi-Fi sensing; channel state information (CSI); gate residual network; human activity recognition (HAR); multi-head attention; multi-scale convolutional neural networks; phase and amplitude Wi-Fi-based human activity recognition (HAR) is a non-intrusive and privacy-preserving method that leverages Channel State Information (CSI) for identifying human activities. However, existing approaches often struggle with robust feature extraction, especially in dynamic and multi-environment scenarios, and fail to effectively integrate amplitude and phase features of CSI. This study proposes a novel model, the Phase-Amplitude Channel State Information Network (PA-CSI), to address these challenges. The model introduces two key innovations: (1) a dual-feature approach combining amplitude and phase features for enhanced robustness, and (2) an attention-enhanced feature fusion mechanism incorporating multi-scale convolutional layers and Gated Residual Networks (GRN) to optimize feature extraction. Experimental results demonstrate that the proposed model achieves state-of-the-art performance on three datasets, including StanWiFi (99.9%), MultiEnv (98.0%), and the MINE lab dataset (99.9%). These findings underscore the potential of the PA-CSI model to advance Wi-Fi-based HAR in real-world applications. 25 4 Khoa Công nghệ Thông tin 3 Thái Duy Quý 2025-07-14T07:14:35Z 2025-07-14T07:14:35Z 2025-02-09 Journal article Bài báo đăng trên tạp chí thuộc ISI, bao gồm book chapter https://scholar.dlu.edu.vn/handle/123456789/5040 10.3390/s25041038 https://www.mdpi.com/1424-8220/25/4/1038 en Sensors (Basel, Switzerland)
institution	Thư viện Trường Đại học Đà Lạt
collection	Thư viện số
language	English
topic	Wi-Fi sensing; channel state information (CSI); gate residual network; human activity recognition (HAR); multi-head attention; multi-scale convolutional neural networks; phase and amplitude
spellingShingle	Wi-Fi sensing; channel state information (CSI); gate residual network; human activity recognition (HAR); multi-head attention; multi-scale convolutional neural networks; phase and amplitude Thai Duy Quy Lin, Chih-Yang Shih, Timothy K Enhanced Human Activity Recognition Using Wi-Fi Sensing: Leveraging Phase and Amplitude with Attention Mechanisms
description	Wi-Fi-based human activity recognition (HAR) is a non-intrusive and privacy-preserving method that leverages Channel State Information (CSI) for identifying human activities. However, existing approaches often struggle with robust feature extraction, especially in dynamic and multi-environment scenarios, and fail to effectively integrate amplitude and phase features of CSI. This study proposes a novel model, the Phase-Amplitude Channel State Information Network (PA-CSI), to address these challenges. The model introduces two key innovations: (1) a dual-feature approach combining amplitude and phase features for enhanced robustness, and (2) an attention-enhanced feature fusion mechanism incorporating multi-scale convolutional layers and Gated Residual Networks (GRN) to optimize feature extraction. Experimental results demonstrate that the proposed model achieves state-of-the-art performance on three datasets, including StanWiFi (99.9%), MultiEnv (98.0%), and the MINE lab dataset (99.9%). These findings underscore the potential of the PA-CSI model to advance Wi-Fi-based HAR in real-world applications.
format	Journal article
author	Thai Duy Quy Lin, Chih-Yang Shih, Timothy K
author_facet	Thai Duy Quy Lin, Chih-Yang Shih, Timothy K
author_sort	Thai Duy Quy
title	Enhanced Human Activity Recognition Using Wi-Fi Sensing: Leveraging Phase and Amplitude with Attention Mechanisms
title_short	Enhanced Human Activity Recognition Using Wi-Fi Sensing: Leveraging Phase and Amplitude with Attention Mechanisms
title_full	Enhanced Human Activity Recognition Using Wi-Fi Sensing: Leveraging Phase and Amplitude with Attention Mechanisms
title_fullStr	Enhanced Human Activity Recognition Using Wi-Fi Sensing: Leveraging Phase and Amplitude with Attention Mechanisms
title_full_unstemmed	Enhanced Human Activity Recognition Using Wi-Fi Sensing: Leveraging Phase and Amplitude with Attention Mechanisms
title_sort	enhanced human activity recognition using wi-fi sensing: leveraging phase and amplitude with attention mechanisms
publishDate	2025
url	https://scholar.dlu.edu.vn/handle/123456789/5040 https://www.mdpi.com/1424-8220/25/4/1038
_version_	1845408551158153216

Enhanced Human Activity Recognition Using Wi-Fi Sensing: Leveraging Phase and Amplitude with Attention Mechanisms

مواد مشابهة