New radar tech sees through walls by reading Doppler shifts in W-Fi and mobile communications

Smart phones actively connected to the Internet can pinpoint a troublemaker’s GPS coordinates, but the soft radio glow emitted by nearby wireless routers and mobile communications can now reveal their actual movement through 25 centimeter thick slabs of masonry. Ongoing research at the University College London (UCL) has conceived a method of detecting the Doppler shifts of Wi-Fi and mobile telephone signals to “see” people through obstacles. Useful in everything from hostage rescue to traffic control, the technique was awarded the Engineering Impact Award in RF and Communications at this summer’s National Instruments NI week meeting.

Through-wall radars have been previously constructed by MIT researchers Dina Katabi and Fadel Adib, but these transmit and receive, whereas UCL’s technique exclusively uses passive radiation from Wi-Fi routers, ambient GSM and LTE mobile signals, and more; meaning, there’s nothing to divulge that surveillance is taking place. Also, UCL’s “high Doppler resolution passive Wi-Fi radar” device is constructed from on two multi-frequency, software-defined, and an FPGA-based transceivers.

The system determines the target’s position by comparing the signal from a reference channel receiving the baseline signal from the Wi-Fi access point (or RF source), and a surveillance channel, which detects the Doppler-shifted waves reflecting from the movement subject. The minute frequency shifts between the two parameters are then interpreted and the target’s location and movement are revealed. It essentially leverages the Doppler Effect, which dictates that the change in the frequency of the wave is relative to an observer.

In its early prototype state, the device cannot reproduce anything as detailed as a photograph image, but can deliver a radar-style scatter plot with a splash of color that indicates the target. For a detailed rundown on how the system works, a written by UCL researchers Bo Tan, Qingchao Chen, Kevin Chetty, and Karl Woodbridge is available here .

Source: IEEE Spectrum