Seminar on "A Deployment-Aware Methodology for CSI-Based Wi-Fi Sensing under Limited Labeled Data" by Miss He Wang, Hong Kong Polytechnic University

Date: 20 March 2026, Friday
Time: 4:30 pm
You are cordially invited to join our next seminar in online mode on March 20, Friday, 4:30pm (Hong Kong Time)
Zoom Meeting ID: 383 735 6917
Password: 270831

Speaker: Miss He Wang, Hong Kong Polytechnic University

Abstract: CSI-based Wi-Fi sensing enables device-free indoor perception, yet its robustness is fundamentally constrained by the quality of the collected CSI under deployment uncertainty and limited labeled data. This seminar presents a deployment-aware methodology that (i) models wall-reflected dynamic paths to explain how wall proximity can enhance or degrade sensing, (ii) develops a model-guided multi-sensor deployment framework to compare candidate topologies for stationary crowd counting, and (iii) leverages unlabeled CSI via semi-supervised learning and domain adaptation to improve generalization.  Extensive experiments across real-world scenarios validate that robust Wi-Fi sensing requires principled deployment design together with effective utilization of unlabeled data.

Speaker’s Bio: He Wang received the B.Eng. degree in Internet of Things Engineering from Beijing University of Technology, China, in 2020, and the M.Sc. degree in Electrical and Electronic Engineering from Imperial College London, London, U.K., in 2021. She is currently pursuing a Ph.D. degree in the Department of Electrical and Electronic Engineering at The Hong Kong Polytechnic University, Hong Kong. Her research interests include wireless sensing, multi-sensor fusion, machine learning, and wireless communications and networking.
 
WEBINAR WEBSITE:
https://www.ee.cityu.edu.hk/~cccn/
https://www.ee.cityu.edu.hk/~cccn/centre-seminars.htm