Perceiving Accurate CSI Phases With Commodity WiFi Devices

WiFi technology has gained a wide prevalence for not only wireless communication but also pervasive wireless sensing.A wide variety of emerging applications leverage accurate measurements of the Channel State Information(CSI)obtained from commodity WiFi devices.Due to the hardware imperfection of commodity WiFi devices,the frequency response of the internal signal processing circuit is mixed with the real channel frequency response in passband,which makes deriving the accurate channel frequency response from CSI measurements a challenging work.In this paper,we identify a new non-negligible non-linear CSI phase error,and verify that IQ imbalance is the root source.We conduct intensive analysis on the characteristics of this non-linear phase error,and find that it's prevalent among various WiFi devices,rather stable along time and time-of-flight,but related to the received signal strength indication(RSSI),the carrier frequency and the specific transmission pair.After modeling on this non-linear phase error,we propose novel calibration methods to compensate both this non-linear phase error and the rotation phase error.We demonstrate the efficacy of the proposed methods by applying them in CSI splicing.Results of extensive real-world experiments indicate that our calibration methods work well and our enhanced Splicer significantly improves the performance of CSI splicing.Our work is of great importance to various CSI-based or PDP(power delay profile)-based applications.