A new method using covariance eigenvalues and time window in passive human motion detection based on CSI phases

Due to the rapid development of WLAN technology, device-free passive human detection utilizing the existing WLAN infrastructure has attracted the attention of the numerous scholars and holds more potential to ubiquitous smart applications. Despite of the prevalent signal feature Received Signal Strength (RSS), more reliable and robust feature Channel State Information (CSI) in the 802.11n standard has been proposed in recently years. In this paper, we explore to study the sensibility of amplitude and phase to human motion via CSI. We propose a novel eigenvalue based on phases to represent the full CSI information. With the dynamic time window algorithm, we develop CETW (covariance eigenvalues and time window) method to detect human motion. Based on the above research, we design a high precision device-free passive burst human motion detection system (HPMD). As demonstrated in the experiments, our system gained a high detection rate and a low false alarm rate.