RF signals can be used not only for data transmission, but also for environment sensing. In the indoor environment, signal generated by the transmitter propagates through multiple paths, such as direct radiation, reflection and scattering, and superimposes at the receiver, carring the information of the characteristics of the environment. It is not unusual to use radio for environment sensing. Typical applications include radar systems for detecting spacecraft and indoor localization with UWB. Compared with dedicated radar and even UWB signals, Wi-Fi has great advantages in ubiquity. However, Wi-Fi signals have narrow bandwidth, low time resolution and imperfect signal processing devices. Thus, it is necessitated to develop theory and technology for Wi-Fi based environment sensing, and achieve highly accurate environment sensing on commercial Wi-Fi devices. While both Wi-Fi and radar are not new, but for the first time, the combination of the two hit a spark of innovation in the field of mobile computing.

Wi-Fi based environment sensing technology has been widely applied. In security-sensitive scenarios, such as surveillance of confidential areas, disaster response, protection of important objects, etc., it needs to promptly detect the presence and activities of persons (staff or intruders) that do not carry any wireless devices in sensitive areas. Apart from security, passive human detection can be widely used in a variety of ubiquitous computing applications. For example, In museums, introduction of exhibits can be automatically displayed when visitors approach them; In supermarkets, items of most interest recently can be counted; Or in elevators and carriages, the number of passengers can be evaluated. In addition, Wi-Fi based environment sensing technology can be applied to detect fine-grained pose, gesture and breath, or tiny motions of human.


CSI can replace RSSI, for its finer granularity in frequency and higher stability in time. By applying CSI to model-based or fingerprint-based localization method, more accurate ranging result and richer fingerprint information can be obtained, leading to localization accuracy of submeter level.

LOS Recognition

Whether the line-of-sight path between the transmitter and the receiver exists, has great impact on communication performance. The line-of-sight path identification can be achieved using only signal amplitude and phase information on common WiFi devices, with recognition accuracy over 90%.

Human Sensing

In the scenarios of passive personnel detection, the personnel may not carry any electronic devices. By analyzing the impact of the personnel on the signal, the wireless sensing system detects the persence of the personnel, localizes the personnel and recognize different behaviors of the personnel.

Motion Recognition

With WiFi CSI, it is able to resolve multipath, and detect weak fluctuations of signals in the line-of-sight or non-line-of-sight pahts, which can improve the sensing sensitivity, expand the sensing area, enhance the sensing reliability, and enable recognition of human action, gesture, and even tiny motions such and breath.

Demo Videos

Widance,CHI'17,Human Activity Recognition
WiFiRadar,Mobile Sensing Platform based on Channel State Infomation
PADS,ICPADS'14,Indoor Human Detection
WiDar & WiDar2.0,MobiHoc'17 & MobiSys'18,Passive Human Tracking


Experimental Platform

Learn More

Real-Time Montoring System

Experimental Data Sample

  • Here are some data samples of experiments (such as WiDar, WiDar2.0, etc.) for researchers. Data samples are collected by WiSH.

Download WiDar Data Download WiDar2.0 Data


Pervasive Adoption

Xidian University, Tianjin University, Nanjing University of Posts and Telecommunications, PLA University of Science and Technology, etc.

High-Level Papers

In conferences like ACM CHI, IEEE INFOCOM and Journals like IEEE JSAC, IEEE TWC

Best Paper(Mention) Award

ACM CHI 2017, IEEE ICPADS Best Paper Honorable Mention Award; Best Paper Award of Tsinghua Science and Technology




Contact Us

Current Members

Past Members