Towards Anti-Interference Human Activity Recognition Based on WiFi Subcarrier Correlation Selection
Jinyang Huang, Bin Liu, Chao Chen, Hongxin Jin, Zhiqiang Liu, Chi Zhang, Nenghai Yu
Abstract
As an essential technology in the field of the Internet of Things, Human activity recognition (HAR) is a well-researched topic. Recently, some state-of-the-art WiFi-based HAR systems have been presented due to its characteristics of no-invasion, no privacy leakage, and high recognition accuracy rates (RARs). However, the commodity WiFi devices for identification are usually in a complex electromagnetic environment, where the interference caused by WiFi devices from channel overlap is common and severe. Furthermore, our extensive experiments show that the performance of these pioneer WiFi-based HAR systems may degrade significantly in co-channel interference (CCI) scenarios. To solve the above problem, we propose WiAnti, a WiFi-based HAR system that is robust to CCI. Two adaptive subcarrier selection algorithms, WiAnti-Pearson and WiAnti-DTW, are proposed to mitigate the impact of CCI and to improve the recognition performance in CCI scenarios. As demonstrated in the experimental results, WiAnti-Pearson yields a 95% RAR on average, which can improve up to a 14% RAR in the presence of constant CCI. Moreover, WiAnti-DTW achieves an 8% higher RAR in the varying CCI scenario, reaching 94%.