Slashdot Mirror


New Projects Use Phone Data To Track Big Cities' Mass Transit Use

An anonymous reader points out a New York Times article about a traffic analysis program that "'works by taking note of which cellphone tower a phone is communicating with. It then looks for disruptions in service followed by significant changes in location. If a phone located near Times Square suddenly loses service and reconnects at Prince Street and Broadway 15 minutes later, then it has almost certainly traveled there using the N or R trains.' In another interesting twist, the article briefly notes, 'The system will also include an experiment that uses phones' microphones to sense when riders are on buses.'" The article also mentions a similar project to track buses and trains in Los Angeles.

50 of 75 comments (clear)

  1. Not necessary/ for most modern systems by immakiku · · Score: 2

    Such as in any subway system in China. Where reception doesn't end at the subway entrance. People are making calls and surfing the web while riding the train.

    1. Re:Not necessary/ for most modern systems by icebraining · · Score: 1

      Same here in Portugal, since 2006.

    2. Re:Not necessary/ for most modern systems by mehrotra.akash · · Score: 1

      And India

    3. Re:Not necessary/ for most modern systems by DNS-and-BIND · · Score: 1

      But the main problem with the Chinese system. Users only can. Speaking in sentence fragments using the mobile phone.

      --
      Shutting down free speech with violence isn't fighting fascism. It IS fascism!
    4. Re:Not necessary/ for most modern systems by sych · · Score: 1

      My mobile phone works fine here in Beijing.

    5. Re:Not necessary/ for most modern systems by DNS-and-BIND · · Score: 1

      You: joke. Sound: whoosh.

      --
      Shutting down free speech with violence isn't fighting fascism. It IS fascism!
    6. Re:Not necessary/ for most modern systems by sych · · Score: 1

      I'm not sure what was supposed to be funny.

    7. Re:Not necessary/ for most modern systems by zig007 · · Score: 1

      And Sweden.

      --
      Baboons are cute.
    8. Re:Not necessary/ for most modern systems by zig007 · · Score: 1

      Me neither. I suppose that the sentence appear chopped up, but I can't decipher it.

      --
      Baboons are cute.
    9. Re:Not necessary/ for most modern systems by GodGell · · Score: 1

      Same here in Budapest, Hungary... ever since I've had a GSM phone. It's funny to hear people complaining about not hearing their phones because the subway cars are loud... :-)

      --
      [SHOW SOME LENIENCY TOWARDS ... I mean, FUCK BETA] Eat. Survive. Reproduce. GOTO 10
  2. Really? by cultiv8 · · Score: 3, Insightful

    So who all gets access to this information about which cellphone is connecting to which tower, and what rights did I give to this person/entity to use it for stuff like this?

    Should have RTFCA (Customer Agreement)

    --
    sysadmins and parents of newborns get the same amount of sleep.
    1. Re:Really? by camperdave · · Score: 3, Insightful

      More worryingly, how did they get access to the noise level from my cell phone's microphone? I don't buy the need for them to know where I am on the bus.

      --
      When our name is on the back of your car, we're behind you all the way!
    2. Re:Really? by Anonymous Coward · · Score: 1

      More importantly, I don't buy the reason why they need to hear anything at all from my phones microphone. I don't see why they need to know what I, or anyone else near me, is saying using this method. Wasn't there are book about something like this???

      It will help us fight the Terrorists!

    3. Re:Really? by Anonymous Coward · · Score: 2, Informative

      Or you could just RTFA which clearly states that its a pilot program that has willing participants. The feature has to be activated on an app.

    4. Re:Really? by xclr8r · · Score: 1

      or Movie.. the latest Batman used cell phone mics to pick up the location of his target. linky

      --
      Beware of those who profit off the docile and persecute the unbelievers.
    5. Re:Really? by wisnoskij · · Score: 2

      Way to ruin a perfectly good conspiracy theory with the facts.

      --
      Troll is not a replacement for I disagree.
    6. Re:Really? by TubeSteak · · Score: 1

      More worryingly, how did they get access to the noise level from my cell phone's microphone? I don't buy the need for them to know where I am on the bus.

      RTFA. It's something they plan to include (as an optional?) in an app that they're developing.
      It's entirely possible to do all the audio capture and processing on the phone, while only sending a "on the bus" or "not on the bus" packet to Densebrain.

      I can't imagine that they'd want to stream audio to their servers and chew through the data plans of their user base.

      --
      [Fuck Beta]
      o0t!
    7. Re:Really? by hedwards · · Score: 1

      That's definitely something to worry about. However some phones do have a secondary microphone for the purposes of noise reduction and that one would be the one which would presumably be used. But either way I'm not so sure I'd trust it not to be eavesdropping on me.

  3. Nextbus by doconnor · · Score: 1

    The location of all Los Angles buses by GPS is already publicly available, as well as several other transit systems. New York is piloting the same system for the B63 5th Avenue bus.

    GPS doesn't work underground, but I'm pretty sure the MTA already knows exactly where all its trains are. It's just a matter of making the data public rather then trying to interpolate it using cell phone signals.

    1. Re:Nextbus by pneum0nic · · Score: 1

      I think the point is to track ridership, not just where the trains are, which you are correct that the transit authority already knows about.

    2. Re:Nextbus by immakiku · · Score: 1

      I believe the point is to model how people are using the trains rather than the status of the trains. How many people/who/when are the questions they want to answer.

    3. Re:Nextbus by ganjadude · · Score: 1

      you mean they cant tell all that by you know.. the fact that you payed for a ticket for train X for Y time on Z day?

      --
      have you seen my sig? there are many others like it but none that are the same
    4. Re:Nextbus by scoobydoobiedoo · · Score: 1

      Not in the case of the subway train. Buy ticket, take train from X to R, connect to F and end at Z!

    5. Re:Nextbus by Marc+Madness · · Score: 1

      you mean they cant tell all that by you know.. the fact that you payed for a ticket for train X for Y time on Z day?

      That only tells you which train station you entered, not which train you boarded or transferred to.

    6. Re:Nextbus by mooingyak · · Score: 1

      Well since I and everybody else paid to ride the bus/subway you would think they would have that data already too.
      Sounds like a great way to track terrorists or sheep as the case my be.

      For the NYC subway, you pay at point of entry, and you walk through a turnstile on exit. There's no data on transfers, or which train you got on. There's also nothing to tie your exit to your entrance. Between knowing where each train is at any given moment and rates at which people are entering and exiting each station you could probably build a decent model of how many people are on any given train. Since in NYC pretty much everybody uses a metrocard now you can probably improve the model a bit by looking at repeated points of entry and making guesses about commuting patterns (if 49% of my entries are at point A, 49% are at point B, and 2% are random other stations, then it's likely that I regularly commute from A->B and B->A. Even more likely if the time of day is fairly consistent).

      Still, that's a very bitchy model. Tracking a population sample where you know exactly where and when they got on and off, then extrapolating from that sample sounds MUCH simpler.

      --
      William of Ockham had no beard. The most likely explanation is that it was chewed off by squirrels every morning.
    7. Re:Nextbus by mcmonkey · · Score: 1

      Because of course, anyone traveling without a cell phone isn't a real person.

    8. Re:Nextbus by Zcar · · Score: 1

      They know where you got on but not where you exited, right? Not quite the same data.

    9. Re:Nextbus by hedwards · · Score: 1

      You can do that with transit passes. It's not so easy with cash, but with the various cards that are in place they can definitely track those. The local transit service rolled theirs out a while back and they had to deal with the backlash. Apparently, the party paying for the card has access to all the information they have about where the cards are being swiped, which meant that people with company sponsored cards would have to depend upon company policy to guide tracking.

      But, all in all, it allowed them to more fairly split the money up from fairs amongst the various agencies handling the busing.

    10. Re:Nextbus by hedwards · · Score: 1

      Which works fine so long as the fare is the same no matter how far you travel. The local link light rail charges differing amounts of money depending upon how far you go, which requires you to swipe when you enter the station and when you leave the destination station. It's a simple enough system then if you want to transfer to a bus you'd have to swipe there again.

      Which can be a problem for people with privacy concerns because that information is then available to whomever it is that wants to subpoena it.

    11. Re:Nextbus by xaxa · · Score: 2

      For the NYC subway, you pay at point of entry, and you walk through a turnstile on exit.

      Is the fare the same, regardless of the length of the trip? Wikipedia suggests it is.

      In the largest European cities I've visited it's not, but I live in London so I'll describe that.

      In London there are 9 concentric fare zones (1 to 9, with 1 being the central zone, and few tourists venturing further than Zone 2. Property is often advertised as "5 minutes from a zone 3 station" etc). [PDF map]

      The fare depends on which zones you travel through, and the time of day (peak/off-peak). e.g.: (from this ridiculous table)
      zone 9 to zone 1 is £6.00 or £3.50 (peak, off-peak)
      zone 9 to zone 3 is £3.50 or £1.40 (peak, off-peak)
      zone 2 to zone 1 is £2.50 or £1.90 (peak, off-peak)

      Hence, if you use the electronic ticket (Oyster card) it needs to know what zone you start in, what zone you ended in, and whether you went through a more central zone (from zone 3 in the east to zone 3 in the west, through the centre, costs more than taking an orbital route avoiding the centre). The Oyster card must be used to open the ticket barrier at the start and the end of the journey. (Paper tickets must too -- they have the zones they are valid in encoded in a magnetic strip)

      Anyway, my point was that in London the transport network has the origin and destination data for the majority of tube or suburban train trips. It must be interesting to mine this -- you could see the effect of delays, how people respond to a planned closure, how many people don't make it home on Friday night, who goes to a political protest...

      There is some sample data here but I don't want to register at the moment.

    12. Re:Nextbus by mooingyak · · Score: 1

      NYC is a single fare for the subway, regardless of how far you're going (much like Paris, or at least, what I remember being told in 7th grade French class oh so many years ago. Our teacher made this sound exotic and unique despite the fact that we lived some 20 miles from the NYC border).

      There are also a variety of train systems that bring commuters from various suburbs into NYC, at least some of which operate on some kind of greater distance traveled == greater fare. LIRR for example splits its stations into zones and then your fare is determined by how many zones you cross. There are however no turnstiles or any other recording mechanism. All fares are checked by an on-board ticket collector who will if necessary sell you a ticket at your seat (they don't like this though, and you end up paying an extra fee in the territory of $5 or so)

      --
      William of Ockham had no beard. The most likely explanation is that it was chewed off by squirrels every morning.
    13. Re:Nextbus by xaxa · · Score: 1

      I changed my mind and registered for the Oyster data November 2009 "feed", they surely didn't verify my details as I was immediately given a confirmation (it's late evening).

      It's a CSV file, with 2.6M lines, but doing anything with it will have to wait until next month.

  4. Big Brother by Nkwe · · Score: 1

    The system will also include an experiment that uses phones' microphones to sense when riders are on buses.

    What else does it "sense" via the microphone?

    1. Re:Big Brother by vlpronj · · Score: 1

      That's why this, to me, deserves the "darkknight" tag - i.e. the "Cell Phone Sonar" system used. There may not be any altitude info, making it not quite a 3D map, but I'm sure the right algorithm would include tower distance > and height, and calculate accordingly.

    2. Re:Big Brother by creat3d · · Score: 1

      Stop asking questions and just feel safer! Isn't freedom great?

      --
      Grammar nazis are to this community what excrements are to gold.
    3. Re:Big Brother by icebraining · · Score: 1

      It was an app that you had to install voluntarily.

    4. Re:Big Brother by UncleTogie · · Score: 1

      It was an app that you had to install voluntarily.

      Voluntarily, for now....

      --
      Don't tell me to get a life. I'm a gamer; I have LOTS of lives!
  5. Any relation? by Zcar · · Score: 1

    Or just a funny coincidence?

    Or so Alex Morgan Bell hopes. Mr. Bell began designing the system last year, when he was studying electric engineering at Columbia. After trying to get the idea going by himself and luring only several hundred people as users, Mr. Bell joined Densebrain, a Web development company that makes NYCMate, a transit map app (and is perhaps best known for SitorSquat, an app that maps public restrooms).

    Alex Bell, figuring out new uses for the phone.

  6. Primitive by wsanders · · Score: 1

    This is going to be obsolete soon. We have 3G in most of the undergound SF Bay Area BART, and, sooner or later, WiFi (it is in trial now). I am sure other cities can claim the same.

    --
    Give a man a fish and you have fed him for today. Teach a man to fish, and he'll say "WHERE'S MY FISH, YOU IDIOT?"
    1. Re:Primitive by VortexCortex · · Score: 1

      Seconded -- The town of Corpus Christi, TX had decent WIFI coverage when I lived there. They got me hooked on using it for free before allowing PDQ, and other "providers" to charge me for the service -- I guess the city sponsored WIFI is "rented" to the "providers" that I then must pay to login. Not sure how that's working out for them, but it can't be too bad since the explosion of portable WIFI enabled Android and iOS devices happened shortly afterward (Making it possible to use these devices in WIFI mode sans 3G/Cellular data plans).

      My only beef was that the WIFI was unsecured, so I could play ARP games with others near me (even just requiring a WPA password of "CCTX" would prevent such things -- Take note Starbucks et al. Set a password (post it on a sign), or else; Open/Unsecured WIFI users -- use a VPN or else).

      I dearly miss the municipal WIFI -- All cities should have it, if for no other reason than to give police and EMS a WIFI Internet connection.

  7. Have you ever... by rwade · · Score: 1

    ...actually been on the New York Subway? See it's fair information.

    There's no such thing as "paying for a ticket for train X for Y time on Z Day"

  8. They already have and release ridership info by healyp · · Score: 1
    http://mta.info/persdashboard/performance14.html

    http://www.mta.info/developers/

    Though these are just aggregates of turnstile data, so they know that X people entered at Times square and Y people exited at prince within about a 4 hour resolution(the scheduled turnstile audits). The only new thing this scheme would add is to tie the specific entrances and exits together. I'm not sure how useful that actually is, you can extrapolate the most frequently ridden lines based on the aggregate entrances and exits from it. Plus to tie it to particular riders, anonymous or not is an AWFUL lot of data to process.

  9. Sure... by roc97007 · · Score: 1

    > The system will also include an experiment that uses phones' microphones to sense when riders are on buses.

    Yeah, that's what the experiment is for. To sense whether riders are on buses, to check mass transit ridership. Really.

    --
    Oliver's law of assumed responsibility: If you're seen fixing it, you will be blamed for breaking it.
  10. But not everyone has a phone... by Urban+Garlic · · Score: 1

    So phones are maybe a reasonable proxy for riders, but it seems to me that the MTA, being a closed system which already has entry controls for fare-collection purposes, just might have more direct ways of getting at ridership, including ones that don't have smartphones. They could instrument their exit gates, too, and correlate them with train arrival times, and figure out which trains are letting off lots of passengers at which stations.

    But maybe the phone thing still makes sense if they're not anonymizing it.

    --
    2*3*3*3*3*11*251
  11. You live in the People's Republic of San Francisco by Benfea · · Score: 1

    Other non-communist cities will not tolerate such a waste of taxpayer dolla... Oh. Rich people will also benefit from this? Then it might be a good idea! Is there any way we can prevent the peasants in the lower 98th percentile from using this system? [/conservolibertarian]

  12. Re:More Anthony Weiner weiner photos!!! by creat3d · · Score: 1

    Bush and Obama are simply two heads on the same hydra, get ove it.

    --
    Grammar nazis are to this community what excrements are to gold.
  13. No shit by creat3d · · Score: 1

    They could've at least tried to find a better excuse than that.

    --
    Grammar nazis are to this community what excrements are to gold.
  14. Chicago already has this by cslax · · Score: 1

    Chicago already has this for buses and trains.

  15. Why bother? by cusco · · Score: 2

    This is an utterly useless app. If transit authorities want to track riders' use of the system they already have a much better sensor network, the cameras that are on board most of their vehicles. Cameras can do a pretty good job of object counting, and if given enough CPU cycles of counting the number and direction of objects moving into and out of a motion detection zone (doorway). That wouldn't even be a hard app to write, and the only additional hardware needed would be for the app to report (wirelessly I'd assume) when the vehicle returns to base.

    --
    "Think about how stupid the average person is. Now, realise that half of them are dumber than that." - George Carlin