SideSwipe detects gestures based on disturbances to cell phone signals

Researchers at the University of Washington unveiled SideSwipe, a system that enables a cell phone to detect nearby hand gestures based upon how those movements disturb GSM signals.

Gesture-recognition is taking on growing importance as software developers seek new and better ways to enable interactions between people and devices, including smartphones, wearable gadgets and other consumer electronics.

In a research paper, the University of Washington scientists noted that input to current smartphones is hampered because they rely upon physical buttons, touchscreens, cameras or built-in sensors. But SideSwipe offers an alternative by leveraging actual, unmodified GSM signals to detect hand gestures around a handset, potentially enabling someone to handle incoming calls by waving their hand or navigate an onscreen map without touching the device.

"We developed an algorithm to convert the bursty reflected GSM pulses to a continuous signal that can be used for gesture recognition. Specifically, when a user waves their hand near the phone, the hand movement disturbs the signal propagation between the phone's transmitter and added receiving antennas. Our system captures this variation and uses it for gesture recognition," said the researchers.

In the incoming call management example, the researchers noted that a cell phone owner might use one of three standard hand gestures to respond to an incoming call. The user might enable silent mode with a downward gesture towards the phone, send a predefined text with a right swipe gesture or decline incoming calls by tapping on the phone.

Researchers validated their approach through a study with 10 participants, achieving average accuracy of 87.2 percent across 14 hand gestures at a distance of 25 to 30 centimeters from the handset. Though their initial research focused on GSM signals, the researchers said they believe SideSwipe can be easily adapted to other network standards, such as LTE, which they noted uses OFDMA/SC-FDMA in the uplink.

For more:
- see this research paper (PDF)
- see this MIT Technology Review article

Related articles:
MIT uses RF reflections to track people, bodily functions
Report: Microsoft to enable Kinect-like 3D gestures for its own Windows Phones
Samsung files patent for wearable device controlled by wrist gestures
Apple confirms acquisition of PrimeSense