Inferring car movement patterns from passive TPMS measurements

(dspace.networks.imdea.org)

41 points | by wisdomseaker 7 hours ago ago

5 comments

  • RyJones 5 hours ago

    I built a demo of this back when I worked at Qualcomm in Seattle; match this with WiFi beacons and you can trace a person fairly well. It's been over a decade, but at the time both iOS and Android would send pings fairly frequently to all known WiFi networks looking to see if they should switch to a faster one. With your device ID, list of SSIDs you know, and your TPMS data, a person can learn a lot about you.

    Like, where do you work? Where do you stay (Hotel SSIDs)? Who are your friends (other people's home SSIDs)?

    • CSSer 4 hours ago

      And this is what I exhaustively tell people who insist that [tech company] is listening. My reply boils down to, "Why would they need to when you already send them everything in writing?"

    • 0x3f 3 hours ago

      Phones randomize hardware addresses now, so this doesn't work. Although there are better, not-so-publicly-known, ways to do it anyway.

  • EvanAnderson 7 hours ago

    I have an RTL SDR setup in a retail business receiving temperatures from sensors around the property. Besides the neighbors' weather stations I see a lot of TPMS coming, presumably, from the parking lot. I see the same cars regularly. I could definitely correlate them with the POS terminals and identify individual customers.

    • 0x3f 5 hours ago

      Most places where this would be legal there are probably much more effective ways to do it at the POS, even non-biometrically. Although yes you might consider usng TPMS as part of an ensemble.