Wireless Positioning

An anonymous reader writes "This Intel-written whitepaper introduces a way to determine location with the aid of freely accessible, nearby radio sources, such as fixed Bluetooth devices, 802.11 access points, and GSM cell towers. Basically, the device reads the IDs of these local 'radio beacons' (each of which has a unique or semi-unique ID), looks up their positions in a locally-cached database, and performs a computation akin to triangulation. Intel created Place Lab in an effort to satisfy the emerging requirement for location-awareness within mobile devices such as smartphones, PDAs, and laptops, or even moving vehicles. According to the whitepaper, over four million of the required radio beacons have already been mapped."

Comments

Time code reference?

[dada21]

Sounds interesting. As geeky teens we tried making our own positioning system using 3 transmitters, one receiver and a PC. It never worked well as we didn't know how to properly encode the current time into the 'pings' to calculate the transit time.

Do all these broadcast cells broadcast the time code? Are the clocks in sync or do they need to be? I'm guessing without a way to "time" pings received, there's no easy way to validate your position.

The "need" to find yourself seems sort of a waste for most. GPS is nice but I'm more interested in real time user voting on traffic (on their road, in their direction). GPS + realtime traffic heuristics could offer faster escape routes during evacuations, or better gas mileage by avoiding idle periods.

Huh?

[sdirrim]

Didn't we used to call this wardriving?

Re:Huh?

[sdirrim]

Oops, never mind. But couldn't this be hacked to determine where a given person is at any time? They better have tight security on this!

Re:Why not GPS

[thc69]

Because this is a neato satellite-free system. Also, because (maybe) it can be done entirely in software on existing hardware. From TFA:

Despite these efforts, building and deploying location-aware applications that are usable by a wide variety of people in everyday situations is arguably no easier now than it was ten years ago. First and foremost, current location systems do not work where people spend most of their time; coverage in current systems is either constrained to outdoor environments or limited to a particular building or campus with installed sensing infrastructure. Applications like location-aware instant messaging fall flat if they only work for a fraction of users or only during a fraction of a user's day.

Second, existing location technologies have a high cost of entry to both users and application developers. Many location systems require expensive infrastructure, time-consuming calibration, or special tags, beacons, and sensors. The privacy cost to the many stakeholders is also typically ignored or considered only after deployment.

Maybe it can be combined with GPS for better accuracy, too.

What's the deal with GPS on cell phones?

[swb]

OK, this is kind of off-topic, and I realize the idea is that cell phone companies want to charge you for everything, but...what's the deal with the GPS/location thing on my phone?

Why can't they tell me where I am on that thing using the same info they'd send to 911? I'm not even sure the "Get it Now" payware applications can access it.

It just seems like such an obvious extension of the cell phone, especially since they've already added the location technology.

Re:What's the deal with GPS on cell phones?

[iwsmith]

I am no expert, but I think the reason they have GPS is this: Some cell phones use a time modulation technique to fit multiple users on a small frequency range that was allotted to them. This technique involves shifting parts of your signal in the sub-microsecond range (I dont actually know how fast the shifts occur), and as such require very precise timing (if not the tower and phone would not be in sync I suppse). GPS is the most logical way to achieve this, as GPS sat's transmit the current time as well (each GPS sat has an atomic clock on board). So, they use GPS to get extremely accurate times, which are then used for what I can only describe as their 'ninja-foo' signal manipulation to fit many users in a rather limited frequency space.

Direct Download Links

[theGreater]

Sorry, I don't have a torrent hosting setup -- someone else want to grab these?

• Windows XP (and probably 2000)
• Linux
• Mac OS X
• Generic (no scripts)

  -theGreater.

Sounds similar to Psiloc's miniGPS

[bigtrike]

This sounds very similar to Psiloc's miniGPS, except with the addition of additional sources and a location database.

http://www.psiloc.com/index.html?action=ShowArticl eItem&ida=154

The resolution from a single GSM tower seems to be within a mile or two. You can use it to trigger actions on your phone when you get in a certain area. If phones were capable of tracking signal strengths of other towers (I assume they do in order to be able to handoff) you could do this much more accurately. Mapping that into coordinates is fairly tough however, which would make Intel's database very useful. It would probably be far easier to pay the cell carriers for their tower location/code databases though.

Prior Art

[ngr8]

Interesting article. Couple of observations: Triangulation doesn't require time, just imputed direction. http://www.loran.org/library.html has some interesting resources. Cellular location services at http://www.binspy.com/tech/lbsvs.html get a little further along. Also, whilst being able to ride on a lot of different "antennas", seems that one could get to an arbitrarily precise location in two (if not three) dimensions. (For example, the car is at (x1,y1) according to the FM stations, and the 802.x gets it down to a circular error probable of x1+/- 1 meter, y1+/- 1 meter.... ok mongo, throw the egg!)

Re:Malicious?

[sdaemon]

as I read it, this is allowing a receiver to listen to nearby signals and determine its own location, NOT for the transmitters to determine the location of the nearby receiver. In short, if you're not putting out a signal, you can't be tracked by it.

For those who WANT to be tracked, amateur radio has a neat little niche called APRS, but that's probably lost on this crowd :)

Herecast

[Anonymous Coward]

This sounds very similar to what was done by this student (http://www.herecast.com/) a few years ago.

Most useful in doors- factories, etc.

[cbelt3]

C'mon- take off the tinfoil hats already. This tech is already active in some places, primarily as a tracking tool for indoor industry. Here's an example:

Your company makes big widgets that get pushed around your factory floor on carts. You want your people to have the flexibility to push the carts where they need to go, but at the end of every shift carts are 'lost', the second shift guy has to go looking around for the half-assembled widget with the missing frannistan.

You can make everyone log their widget work into widget wherezit workstations, but the workers wont want to waste valuable beer time for that. So the widget wherezit workstation logging project fails.

So instead you put a wifi device on each cart. It reads where it is based on the location of access point antennas you've put up in your rafters. It then uses these AP's to periodically tell a server where it is. End results ? You know where your widgets are hiding all the time. Without anyone having to do anything.

I wish /. readers wouldn't be so anxious to find the 'evil government / corporate / wal-mart' "Threat" before they see the real world solution to real world problems.

Besides, the aliens who overthrew the gummint in the 50's already put chips in all your fool heads anyway...

Other Art: Cellular Location/LORAN/Miscellaneous

[ngr8]

Couple of observations:

Triangulation from fixed points does not require a time stamp, just directions.

Some other sources:
- Cellular Location Services (E911, drive by text ads...) some discussion at http://www.binspy.com/tech/lbsvs.html

- LORAN at http://www.loran.org/library.html

Arbitrary Precision
Having spent all of a minute to thing about this, wouldn't a multi-band/multi-protocol gizmo give the ability to find location in 2-space (if not 3-space) to an arbitrary level of precision? Example: the FM station signal locates the car in (x,y) with a circular error probable of 200 meters. AM station signals reduce it to a CEP of 10 meters (waves hands a lot now), and the radar leaks from airports reduce it to 2 meters....)

Made up gedanken example, but it does seem feasible to me, gentle /. readers.

Re:Done before?

[rpresser]

Why can't the beacon devices themselves use this method to locate themselves relative to each other? Add a protocol for exchanging this information, and whatever devices are in the neighborhood could quickly reach a consensus as to their relative positions. And if one or more of them are GPS-enabled, voila, we have automagical mapping.

This idea was used in at least one Vernor Vinge story, "Fast Times at Fairmont High". The protagonists dropped wireless routers as "breadcrumbs" and after about four were down, they could accurately identify their position (relative to the routers).

Re:Local Database?

[nietsch]

It sounds like a nice solution at first, but you have to rely on the AP or Cell tower to have a unique IP. That will not be the case, as AP (until IPv6 if fully accepted and implemented) usually act as NAT router too and have a local network adress. Can you locate 192.168.1.1 for me please?

The article mentions mac numbers or celltower ID's that have to be linked to a location. Maybe you could resolve those locations with a DNS-like system, but I am sure there are more eficient ways to do that.
A complicating factor is that the devices cannot rely on continuus net acces, so the lookups need to be queued until acces is available. And since they aim for handheld devices, you can assume the storagespace is restricted, making all the other fields that come with DNS not only useless but unwanted too.
I can see you recently had a DNS-hammer in your hands, but this does not look like a IP-lookup nail, sorry.