Performance Bound of Wi-Fi Sensing (Researching)
Wi-Fi sensing suffers from accuracy uncertainty, which obstructs the application in practice. This research aims to reveal the influence factor of sensing accuracy.
Precision Improvement Technology of Wi-Fi-Based Behavior Sensing System
In this research, we present WiRIM, a resolution improving mechanism for Wi-Fi based human sensing. We design a channel switching and aggregation algorithm to extend the effective bandwidth of commodityWi-Fi signals and improve the performance and efficiency of human sensing applications. With aggregated CSI, WiRIM constructs feature images which contain rich frequency, temporal and spatial characteristics, and then uses a deep learning method to process the extracted features.
We propose a cross-location human activity recognition (CLHAR) scenario as a case study. The CLHAR scenario requires a high enough resolution of the Wi-Fi signals to accurately recognize different activities under the interference of tiny changes in human location. The experiments demonstrate the generality and effectiveness of the proposed mechanism.
In this work, we address the issue of Wi-Fi-based human activity recognition (HAR) system in densely deployed Wi-Fi environments. With the benefit of sufficient information provided by Wi-Fi channel state information (CSI), HAR based on Wi-Fi has become an active research area in recent years. Traditional Wi-Fi CSI-based HAR applications usually focus on utilizing one Wi-Fi transmitter and one or several Wi-Fi receivers to extract the activity-related features, ignoring the communication among multiple Wi-Fi devices in the real world.
Besides, we present a novel Wi-Fi link selection model on the basis of continuous state decision-making process in which CSI is modeled as a part of the state. The model, referred to as WiAgent, takes an action of selecting one Wi-Fi link according to current state, and then updates the state for the choice of the next action. From extensive experiment results, our method performs better than other solutions in a given environment where multiple Wi-Fi transmitters exist.
 Xinbin Shen, Lingchao Guo, Zhaoming Lu, Xiangming Wen, Shuang Zhou. (2021, March). WiAgent: Link Selection for CSI-Based Activity Recognition in Densely Deployed Wi-Fi Environments. In 2021 IEEE Wireless Communications and Networking Conference (WCNC) (pp. 1-6). IEEE.
Links selection algorithm(Researching)
To reuse the densely deployed WiFi devices which are currently for communication, we present WiLink, the first 3D human pose estimation system for the large sensing range. Mainly contributions are as follows:
1、We propose a method to identify and remove bad WiFi links.
2、We propose indicators to measure the importance of links and the redundancy between links.
3、We propose an algorithm (DLS) based on maximum weights and minimum redundancy to dynamically select several effective WiFi links instead of all links.
4、We input CSI data of the selected links into neural network to extract features related to human pose, and convert these features into key point coordinates.
Experimental results show that compared with all WiFi links, using the links selected by the DLS algorithm as input of the neural network can enhance the estimation accuracy and reduce the computing time.