A high-noise CSI noise reduction method applied to wireless sensing

• 1、School of Computer Science, Beijing University of Posts and Telecommunications, Beijing, 100876

Abstract：At present, the Internet of Things technology is developing rapidly, and human gesture recognition and activity recognition are gradually becoming important promoters of the development of human-computer interaction technology, cameras and wearable technology. In recent years, WiFi wireless sensing technology has strong practical application value due to its simple deployment and wide coverage. The existing rerearch on gesture recognition based on WiFi perception focuses on solving the problem of accuracy, and it is necessary to use a laptop computer with a specific network card to perform gesture recognition at a specific location. For common smartphone devices in life, it also has the potential to realize WiFi wireless sensing, but due to the difficulty of collecting CSI and the large nose of CSI data collected at the same time, the application of WiFi wireless sensing on it needs to be studied. In order to solve the problem of high CSI noise collected by smartphones, this paper proposes an effective noise reduction method, which can ensure the effective extraction of action feature information from CSI data in the background of high noise and ensure the accuracy of its WiFi perception process.)

Keywords： wireless sensing smartphone CSI noise reduction method