Transfer-Learning-Based Human Activity Recognition Using Antenna Array

• Published in: Remote sensing (Basel, Switzerland) Vol. 16; no. 5; p. 845
• Main Authors: Ye, Kun, Wu, Sheng, Cai, Yongbin, Zhou, Lang, Xiao, Lijun, Zhang, Xuebo, Zheng, Zheng, Lin, Jiaqing
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.03.2024
• Subjects: Accuracy; activity recognition; Antenna arrays; Antennas; Antennas (Electronics); Artificial neural networks; CSI; Fourier transforms; Human activity recognition; Learning; Low pass filters; Network interface cards; Neural networks; Noise control; Noise reduction; Noise threshold; Privacy, Right of; Receivers & amplifiers; ResNet18; Semiconductor industry; Training; Transfer learning; Transmitters; Wi-Fi; WiFi sensing; United States; China
• ISSN: 2072-4292; 2072-4292
• DOI: 10.3390/rs16050845

Due to its low cost and privacy protection, Channel-State-Information (CSI)-based activity detection has gained interest recently. However, to achieve high accuracy, which is challenging in practice, a significant number of training samples are required. To address the issues of the small sample size and cross-scenario in neural network training, this paper proposes a WiFi human activity-recognition system based on transfer learning using an antenna array: Wi-AR. First, the Intel5300 network card collects CSI signal measurements through an antenna array and processes them with a low-pass filter to reduce noise. Then, a threshold-based sliding window method is applied to extract the signal of independent activities, which is further transformed into time–frequency diagrams. Finally, the produced diagrams are used as input to a pretrained ResNet18 to recognize human activities. The proposed Wi-AR was evaluated using a dataset collected in three different room layouts. The testing results showed that the suggested Wi-AR recognizes human activities with a consistent accuracy of about 94%, outperforming the other conventional convolutional neural network approach.