Aug 2026· Annals of Telecommunications· 0 citations· 22 references
TL;DR
This paper introduces a robust Transformer-based architecture designed to capture long-range temporal dependencies in wireless signals and validated using a comprehensive dataset from 86 volunteers, ensuring high generalization capabilities across diverse human motion patterns.
Abstract
Human activity recognition (HAR) plays a pivotal role in ambient assisted living, particularly for monitoring the elderly and patients with chronic conditions. However, traditional approaches relying on wearable sensors or video cameras face significant challenges regarding user compliance and privacy intrusion. To mitigate these issues, this paper proposes a device-free sensing (DFS) (
https://github.com/mestrelan/MDA-CSI
) framework utilizing Wi-Fi channel state information (CSI), named . We introduce a robust Transformer-based architecture designed to capture long-range temporal dependencies in wireless signals. was validated using a comprehensive dataset from 86 volunteers, ensuring high generalization capabilities across diverse human motion patterns.
This survey provides a systematic overview of cutting-edge research on Wi-Fi-enabled indoor human activity detection, classify mainstream technologies along three dimensions: signal processing pipelines, learning paradigms, and application granularity, and further dissect core challenges including environmental adaptab...
Zengqian Song, Jingming Li· Journal of Advances in Engin...· 0 citations
An innovative framework that integrates WiFi-based Channel State Information with the advanced object recognition power of YOLOv8 to enable robust, contactless activity classification and employs the deep learning capabilities of YOLOv8 for precise identification of diverse indoor actions.
Hicham Boudlal, Mohammed Serrhini, Ahmed Tahiri· Multimedia tools and applica...· 0 citations
A cross-environment transfer learning framework for CSI-based HAR that integrates CSI preprocessing, adaptive amplitude-phase fusion via TinyGate, an R(2+1)D backbone, and two temporal modeling strategies, namely Bidirectional Long Short-Term Memory (Bi-LSTM) and Transformer is proposed.
LipHS, a lightweight feature extraction framework capable of simultaneously capturing multi-level information from CSI signals, is presented, a lightweight feature extraction framework capable of simultaneously capturing multi-level information from CSI signals that achieves model lightweighting while maintaining robus...
ChunHao Xue· Poster Volume 0007 The 2026...· 0 citations
Radar-based human activity recognition (HAR) is a promising solution for elderly healthcare because it enables non-contact and privacy-preserving monitoring. However, existing lightweight CNN-based methods often have limited receptive fields, while many transformer-based approaches introduce high computational overhead...
Xiangbo Kong, Akari Takebayashi, Kenshi Saho et al.· IEEE International Conferenc...· 0 citations
A semi-supervised approach that integrates Contrastive Predictive Coding (CPC) with a hybrid BiGRU-Transformer architecture is introduced, thereby enabling comprehensive temporal modeling for human activity recognition in smart-home environments.
