Researchers at MIT’s Computer Science and Artificial Intelligence Lab have developed software that uses variations in Wi-Fi signals to recognize human silhouettes through walls. The researchers built a device, called RF-Capture, that transmits wireless signals and then analyzes the reflections of those signals to piece together a human form, according to a study published this morning.
The RF-Capture is sophisticated enough to determine subtle differences in body shapes, and, with 90 percent accuracy, distinguish between 15 different people through a wall. It can even determine a person’s breathing patterns and heart rate.
RF-Capture:
The RF-Capture is placed in a room, and a person walks in a neighboring room on the other side of the wall. The device emits wireless signals, which travel through the wall, and reflect off different parts of the human body as it moves. As various part of the body are reflected in the wireless signal, the RF-Capture takes snapshots.
Then, using an algorithm to identify body parts, it stitches the images together to create a silhouette of the moving figure. In some experiments, when the researchers focused the device on specific movement patterns they were able to trace a person’s handwriting as he wrote in the air.
