the system cleverly measures the Doppler shifts created by human movement on regular Wi-Fi signals. That means that the system doesn't require line of sight for gesture detection and, the researchers claim, it could work with off-the-shelf Wi-Fi systems. If using Wi-Fi to detect gestures isn't wild enough, the researches claim that "The average accuracy is 94% with a standard deviation of 4.6% when classifying between our nine gestures."