American and Chinese academics built a keystroke recognition system called WiKey consisting, at its simplest, of a standard router (sender) and laptop (receiver). WiKey can recognize typed keys in the middle of the system based on how the Wi-Fi signal lands on the receiver.
“In real-world experiments, WiKey can recognize keystrokes in a continuously typed sentence with an accuracy of 93.5 percent,” the researchers wrote.
However, they go on to say:
This is not something you should expect to see deployed in the real world tomorrow by spy agencies or hackers. Other distortions throw the entire thing off. If someone else is walking through—or simply in—the room, the current set up falters.
That’s somewhat reassuring but what they don’t say is that it tells us that our keystrokes, some our smallest movements, can be determined via radio waves. It doesn’t have to be Wi-Fi signals from access points already in the environment. It could be any type of custom built radio transmitters and receivers specifically brought to your location. By illuminating your environment with numerous transmitters/receivers one can imagine doing the equivalent of a CAT scan of your home/office, in real time, with centimeter resolution.
In this paper, we propose a novel approach for human identification, which leverages WIFI signals to enable non-intrusive human identification in domestic environments. It is based on the observation that each person has specific influence patterns to the surrounding WIFI signal while moving indoors, regarding their body shape characteristics and motion patterns. The influence can be captured by the Channel State Information (CSI) time series of WIFI. Specifically, a combination of Principal Component Analysis (PCA), Discrete Wavelet Transform (DWT) and Dynamic Time Warping (DTW) techniques is used for CSI waveform-based human identification. We implemented the system in a 6m*5m smart home environment and recruited 9 users for data collection and evaluation. Experimental results indicate that the identification accuracy is about 88.9% to 94.5% when the candidate user set changes from 6 to 2, showing that the proposed human identification method is effective in domestic environments.