Can two Emotiscopes on a network automatically identify if they are in the same room by comparing the similarity of averaged FFT frames?

[connornishijima]

FFT data would be averaged long enough to anonymize it effectively, then passed around over P2P to nearby Emotiscopes on the same network.

Two units could automatically identify if they're in the same space by having similar enough FFT data over time, then dynamically group themselves in the app.