Accelerometers and beyond. The most widely studied smartphone sensor is the accelerometer, which has been used in a vast
body of literature on activity recognition. uWave [84] uses the accelerometer for gesture recognition. TapPrints [85] shows
that the location of screen taps on smartphones and tablets can be identified based on accelerometer and gyroscope readings.
PhonePoint Pen uses the in-built accelerometer in mobile phones to recognize human writing [86]. SurroundSense [87]
combines the photo-acoustic input from a smartphone’s camera andmicrophone with accelerometer data to explore logical
localization by means of ambience fingerprinting. Accelerometer-based smartphone pedometry is studied in [88], and a
predictive model for activity recognition based on accelerometer data is developed in [89].
Sensory readings from accelerometers and gyrometers have been combined for indoor localization, for instance in
the Escort system [90], which guides a user to the vicinity of a desired person in a public place, and for driving style
recognition [91]. Nericell [92] uses the accelerometer, microphone, GSM radio, and/or GPS sensors in smartphones to detect
potholes, bumps, braking, and honking.