Smartphone Cover-State Classification via Acoustic Sensing for Smartphone Search in Indoor Environments

When we lose our smartphone, we often try to search the smartphone by listening to ringing sounds from the smartphone. This approach is inefficient because it requires individuals to rely on their sense of hearing and walk around their homes. We propose a smartphone searching method relying on the state of the smartphone’s surroundings. The state of the smartphone’s surroundings can be described by a cover state, contacting objects, and the room where the smartphone is located. In this paper, we show the feasibility of the cover-state classification via acoustic sensing. Reverberation data of a ringing sound was collected for different cover states with two different covering objects. We evaluated the cover-state classification performance using the collected data and confirmed that our method successfully classified the cover-state with a mean accuracy of 0.66.