6 ms·

It's well-known that phones report our locations at all times, I'd like to see a phone that falsifies that data as well.

by sejje 12y ago

pavel_lishin 12y ago
Kind of hard to fool triangulation. Software doesn't beat the laws of physics.
- chongli 12y ago
  But hardware definitely could. With a manually aimed antenna and proper tools exposed to the user, a phone could be made to communicate with only one cell tower at a time. This would make triangulation impossible.
  - im3w1l 12y ago
    You could look at which tower you connect to with which strength at which time. This would not give as exact results as triangulation for sure, but you could probably still infer a rough trajectory.
- troymc 12y ago
  If the operating system provides an API for the actual/measured location, why can't that API just be changed to provide a fake location instead? It could be a user setting: location services - use fake location? yes/no
  - alexjeffrey 12y ago
    triangulation is performed by analysing response times from cell towers - the FBI doesn't just call yourphone.getLocation()
    - troymc 12y ago
      You're right, that's different, but it can also be foiled, at least in theory: you just have your cellphone add some time delay before responding. The delay could be different for each cell tower. I realize that's a big ask (and is probably something the operating system can't do), but it doesn't violate any laws of physics.
      - alexjeffrey 12y ago
        I'd be wary of how this would affect call quality - if you were adding delays to sending call data to the tower you might end up with a laggy/glitchy phone call.
        x1798DE 12y ago
        How often are you really on calls, though? Assuming this would be feasible, you could just have it turned off whenever you are not on calls or whenever you don't need your phone to tell you your location. I'm not really sure it's feasible, though. I don't know the details of how phones interact with towers, but I would think that what's happening is that your phone broadcasts a radio signal, and that same exact signal is picked up by more than one tower. If you want to triangulate the location of the radio signal, you don't need to know how long it took to get to the tower, you just need to know how much longer it took to get to one tower than the other, and the locations of the two towers. Additionally, I think you can make a similar calculation based on signal strength if you can account for anisotropies in the attenuation of the signal. This seems less reliable to me, but if this is how it is done you could potentially throw off your location by a little bit by putting your phone in a partially shielded case that increases the attenuation on one side relative to another.
      - davb 12y ago
        Generally, each base station within range receives all messages from your phone. Even if you were to address different messages (with different response delays) to each station, they would all still "see" the other messages. They then record that message (or an identifier - akin to a TCP sequence number) and compare the (synchronised) time at which this message was received. This is used to calculate the difference (thus delay) of each tower's reception of that message. I don't think what you're suggesting would actually work given the way mobile network infrastructure works (where multiple cell base stations may see each message at the same time).
- dublinben 12y ago
  Trilateration, actually. The towers aren't measuring your angle, but your distance.
  - colechristensen 12y ago
    Being pedantic, it's neither. Both signal strength (approximating linear distance) and angle (with many, directional antennas per tower location) are used to calculate approximate position. Triangulation, however, is an appropriate term with colloquial understanding.
  - kjs3 12y ago
    Towers do measure angle. See: http://en.wikipedia.org/wiki/Sector_antenna http://en.wikipedia.org/wiki/Sector_antenna
- x0054 12y ago
  You can fool the towers trying to triangulate you by adding a small random delay to the ping responses. Of course, this would require a custom modem firmware and/or hardware level access to the phone. The authority trying to triangulate you would have an ability to track your approximate location, with in several square miles, but it would make it much more difficult to pinpoint your exact location.
  - rasz_pl 12y ago
    you would also need highly directional antenna to make sure only one base station can hear your signal
  - kjs3 12y ago
    Unfortunately, fuzzing your location enough to really make it appreciably harder for LEO to find you would also interfere with your ability to reliably make calls.
mindslight 12y ago
The only way I've figured out how to do that involves a complete mobile network reengineering - anonymous bearer tokens (so not bitcoin) to pay for network access tower-by-tower, and then something like TOR hidden services for the presence registry. You could approximate it with a rotating prepaid data plan (can you actually obtain these without putting down an ID?) and actual TOR with heavily compensated VOIP (or forgo voice). You'd need to change the IMEI and any other identifying information of the network access device every time you refreshed the plan as well, or you could start a club of periodically exchanging them with likeminded people.

It's well-known that phones report our locations at all times, I'd like to see a phone that falsifies that data as well.

by sejje 12y ago

It's well-known that phones report our locations at all times, I'd like to see a phone that falsifies that data as well.

pavel_lishin 12y ago

Kind of hard to fool triangulation. Software doesn't beat the laws of physics.

chongli 12y ago
But hardware definitely could. With a manually aimed antenna and proper tools exposed to the user, a phone could be made to communicate with only one cell tower at a time. This would make triangulation impossible.
- im3w1l 12y ago
  You could look at which tower you connect to with which strength at which time. This would not give as exact results as triangulation for sure, but you could probably still infer a rough trajectory.
troymc 12y ago
If the operating system provides an API for the actual/measured location, why can't that API just be changed to provide a fake location instead? It could be a user setting: location services - use fake location? yes/no
- alexjeffrey 12y ago
  triangulation is performed by analysing response times from cell towers - the FBI doesn't just call yourphone.getLocation()
  - troymc 12y ago
    You're right, that's different, but it can also be foiled, at least in theory: you just have your cellphone add some time delay before responding. The delay could be different for each cell tower. I realize that's a big ask (and is probably something the operating system can't do), but it doesn't violate any laws of physics.
    - alexjeffrey 12y ago
      I'd be wary of how this would affect call quality - if you were adding delays to sending call data to the tower you might end up with a laggy/glitchy phone call.
      - x1798DE 12y ago
        How often are you really on calls, though? Assuming this would be feasible, you could just have it turned off whenever you are not on calls or whenever you don't need your phone to tell you your location. I'm not really sure it's feasible, though. I don't know the details of how phones interact with towers, but I would think that what's happening is that your phone broadcasts a radio signal, and that same exact signal is picked up by more than one tower. If you want to triangulate the location of the radio signal, you don't need to know how long it took to get to the tower, you just need to know how much longer it took to get to one tower than the other, and the locations of the two towers. Additionally, I think you can make a similar calculation based on signal strength if you can account for anisotropies in the attenuation of the signal. This seems less reliable to me, but if this is how it is done you could potentially throw off your location by a little bit by putting your phone in a partially shielded case that increases the attenuation on one side relative to another.
    - davb 12y ago
      Generally, each base station within range receives all messages from your phone. Even if you were to address different messages (with different response delays) to each station, they would all still "see" the other messages. They then record that message (or an identifier - akin to a TCP sequence number) and compare the (synchronised) time at which this message was received. This is used to calculate the difference (thus delay) of each tower's reception of that message. I don't think what you're suggesting would actually work given the way mobile network infrastructure works (where multiple cell base stations may see each message at the same time).
dublinben 12y ago
Trilateration, actually. The towers aren't measuring your angle, but your distance.
- colechristensen 12y ago
  Being pedantic, it's neither. Both signal strength (approximating linear distance) and angle (with many, directional antennas per tower location) are used to calculate approximate position. Triangulation, however, is an appropriate term with colloquial understanding.
- kjs3 12y ago
  Towers do measure angle. See: http://en.wikipedia.org/wiki/Sector_antenna http://en.wikipedia.org/wiki/Sector_antenna
x0054 12y ago
You can fool the towers trying to triangulate you by adding a small random delay to the ping responses. Of course, this would require a custom modem firmware and/or hardware level access to the phone. The authority trying to triangulate you would have an ability to track your approximate location, with in several square miles, but it would make it much more difficult to pinpoint your exact location.
- rasz_pl 12y ago
  you would also need highly directional antenna to make sure only one base station can hear your signal
- kjs3 12y ago
  Unfortunately, fuzzing your location enough to really make it appreciably harder for LEO to find you would also interfere with your ability to reliably make calls.

mindslight 12y ago

The only way I've figured out how to do that involves a complete mobile network reengineering - anonymous bearer tokens (so not bitcoin) to pay for network access tower-by-tower, and then something like TOR hidden services for the presence registry. You could approximate it with a rotating prepaid data plan (can you actually obtain these without putting down an ID?) and actual TOR with heavily compensated VOIP (or forgo voice). You'd need to change the IMEI and any other identifying information of the network access device every time you refreshed the plan as well, or you could start a club of periodically exchanging them with likeminded people.