[UweSchmidt]

Next level analysis could take into account which/whose phones are (untypically) switched off when the bombs went off. Next next level would take into account phones that are "left at home" - where the phone is not in use and not moving around according to usual movement patterns.

I would guess that there is already substantial research done on exactly this, and that it is possible to detect many deviations of normal behaviour.

[incompatible]

What are the typical movement patterns for a phone at home? Mine is usually sitting on a particular spot on my desk.

[spurgu]

In that case you should be safe to go out and do some mischief! Or... image recognition in CCTV's might recognize your face and couple it with the fact that you're not carrying your mobile.

[JohnFen]

Face recognition (even by humans), particularly with CCTV data, has a very high rate of false positives. You might be able to make the case that the fact that your phone wasn't there is evidence that the image isn't really you.

[spurgu]

Except that footage of someone recognized as you walking out of your own apartment might be hard to refute.

Plus I only see face recognition (and CCTV quality) getting better over time. Unfortunately.

[incompatible]

I don't usually carry it when I go out, actually. I may be unusual.

[ddalex]

Absence of a signal is a signal in itself.

[kingosticks]

They used to do that in the TV series 24.