back to article Mobile location data identifies individuals

One of the arguments in favour of anonymous mobile location tracking, nanely that it doesn't provide enough information to identify individuals, has been slapped down by a US-Belgian study. An anonymous trace of one phone's movements, plus a small amount of external data, can pick out one person out of millions. An analysis of …

COMMENTS

This topic is closed for new posts.
  1. LarsG

    If you don't know this already you must be from another planet.

    1. Anonymous (Noel) Coward
      Big Brother

      Google is watching you!

      "Wherever you go, there you are!" ......... and we know it.

  2. frank ly

    location markers

    If you know that a person was at a particular private home, followed by a particular place of work; then of course it is very easy to identify them. For most people, daily activities involve being at one place for a certain amount of time, followed by being at another place for some time, often with an unchanging route between the two places. I could have figured that out.

    1. Danny 14
      Black Helicopters

      Re: location markers

      indeed. especially if you are in a car - for a fair few locations there may only be one particular road suitable so that will make you stand out.

    2. Pat 11

      Re: location markers

      They didn't know that someone was in a private home. They knew which cell towers they had connected to. The point is that the cell usage pattern is fairly unique, and if you also have a database of information about people including some locations, like their address, you can use the cell tower logs to match to an individual. I agree that this is intuitively obvious, but unless people actually, y'know, do the science, you never know which bits of obviousness are actually true.

      1. D Moss Esq

        Re: location markers

        "... unless people actually, y'know, do the science, you never know which bits of obviousness are actually true ..."

        Quite right.

        What's a phone? An IMEI? A phone no.? Sometimes you use someone else's phone. Sometimes someone else uses your phone.

        A discipline will develop stitching the patchy record together over the years, over the IMEIs, over the phone no.s, excluding the other people's use of one phone, including the use of other phones, into one person.

        It will have to.

        Because there will still be a job to do, creating new identities, e.g. for ambassadors travelling under a false identity or people in witness protection programmes. The location identity created for them will have to look realistically patchy.

        From my misspent youth, Dematerialised ID, May 2003, http://dematerialisedid.com/BCSL/29%20May%202003.pdf pp.31-3, §4.9.

      2. frank ly

        @Pat 11 Re: location markers

        "They knew which cell towers they had connected to. The point is that the cell usage pattern is fairly unique, ..."

        Only fairly unique? :)

        It's not just the cell you are connected to that is used. Data from attempts to connect to other cells are also used and the resulting signal strength data can be used to triangulate your position. Using just the mobile data signal in my phone (i.e. not GPS and not WiFi) Google Maps can place me within a few streets of where I am. The mobile phone companies have access to much better quality data from their tower records, especially the local signal strength maps around those towers and so can triangulate your position with greater accuracy.

        1. SImon Hobson Bronze badge

          Just an FYI ...

          >> The mobile phone companies have access to much better quality data from their tower records, especially the local signal strength maps around those towers and so can triangulate your position with greater accuracy.

          In general, signal strength is a poor measure to use. For GSM networks, there is a *very* accurate distance measure available. The way GSM networks operate, the cell tower keeps telling the mobile device to adjust it's timing so that it's signal arrives at the cell tower within the timeslot allocated to that device. By keeping track of the timing instructions, the cell tower can have a quite accurate measure of distance from the tower.

          Together with this, and the segment you are using, it can place you on a fairly narrow line marking part of a circumference of a circle round the tower.

          Add in another tower and it can do the same from that, and place you in one of two positions (where the lines intersect). Since one of those positions probably wouldn't result in you using the segment you are on, then that can be ruled out and your position is left.

          Much the same technique is used for aircraft navigation. The airborne station sends a request signal to the ground station, which sends a reply back. the round trip time allows the airborne unit to work out the distance. Some units can do this simultaneously with two ground stations and work out your position from that - and it's generally the most accurate radio navigation technique (apart from GPS).

  3. Disco Dance Donkey

    Anonymised mass, not units

    No one would argue that it takes little data to identify someone; what may be of use is averaged data showing streams/clouds of signals to calculate traffic.

    1. Anonymous Coward
      Anonymous Coward

      Re: Anonymised mass, not units

      You meant they won't just use it to spy on you 24/7? :P

      Forget traffic, they could sell you more stuff... they have no idea what it is they could sell you, but I'm sure they can find something in the mountain of spam they wish to send.

  4. Down not across

    Yeah this should all be pretty obvious to all of us here. Not necessarily so for your average Joe Bloggs who is likely to be hoodwinked into believing that anonymisation works. And even if they had some doubts it might still come as a shock how coarse the location data can be for them still to be identifiable.

    1. Vimes

      Except that it's not just the general public that has been hoodwinked where anonymisation is concerned - it's the very people that are supposed to be protecting our privacy that don't have a clue. Take for example the ICO's apparent view on Google Analytics and DNT:

      https://nodpi.org/2013/03/25/the-ico-are-google/

      Mind you, there are ex-ICO staffers that work at Google now - so they know who to lobby and how. Come to think of it, perhaps this lack of regulation is more down to plain corruption than mere incompetence? Or perhaps it's both?

  5. Anonymous Coward
    Anonymous Coward

    Important note:

    Make sure your mistress lives in the same cell area as work/home.

    1. Anonymous Coward
      Anonymous Coward

      Re: Important note:

      Haven't you heard of "don't sh*t in your own backyard"?

  6. Joe Harrison

    Google Latitude

    Google Latitude is interesting but scary. If you let it, it will produce a fascinating report with pie-chart showing how much of your time you spend at work, at home, other places (complete with addresses), your airline flights, plus detailed log of your movements day by day. All without you doing anything apart from install the Android app and letting it run.

  7. Anonymous Coward
    Anonymous Coward

    Well shucks

    Who wants to go tell the redhead from zero dark thirty that she didn't need to have a car full of geeks roaming around Pakistan for months? All she had to do was analyze some data and there you go!

  8. Wzrd1 Silver badge

    Funny thing.

    My work mobile stays home when I'm not at work.

    My home mobile stays home when I'm at work.

    Both stay home when my wife and I are going out for the evening.

    In that last case, should either one ring at an inopportune time, said unit would require replacement. It seems that our displeasure results in a massive ESD discharge.

    From a stun gun...

    Because, water damage is too obvious.

This topic is closed for new posts.

Other stories you might like