Recent studies have explored various object detection methods that use channel state information (CSI) in wireless local area networks (WLANs) such as IEEE 802.11ac. However, evaluations used with WLAN-based approaches by experimentation mainly target indoor object detection scenarios. Therefore, the effects of antenna placement of access points (AP) and stations (STA) on detection performance in outdoor environments have not been clarified. As described herein, by experimentation, we evaluate the performance of our developed WLAN-based object detection system with distributed antennas in outdoor and large-scale indoor scenarios. Then we clarify the effects of AP and STA placements on the achieved detection performance, where an off-the-shelf WLAN device is used to capture feedback CSI, including beamforming weight information. We leverage these data as feature information for machine learning (ML)-based object detection. Findings indicate that the object detection method can increase the detection probability effectively in specific locations by appropriate location of AP and STA, while reducing the necessary complexity for ML model training and detecting the target in outdoor and large-scale indoor scenarios.
Experimental Evaluation of WLAN-based Object Detection Using CSI in Outdoor and Large-scale Indoor Environments
2024-10-07
13157293 byte
Conference paper
Electronic Resource
English
MIMO-OFDM Channel Modelling for Outdoor-Indoor Environments
British Library Conference Proceedings | 2006
|DBSCAN-Based Mobile AP Detection for Indoor WLAN Localization
Springer Verlag | 2019
|