Forget GPS, Hello WPS

No France writes "A company known as skyhook wireless has announced the commercial availability of its Wi-Fi Positioning System, or WPS. The company has compiled a database of every wireless access point it can find in a given city. When a mobile user running th Skyhook client is in a recorded area, their position is calculated by selecting the surrounding signals and comparing them to the reference database. Currently there are 25 US cities mapped, including New York City, Boston, Chicago, Los Angeles and San Francisco. Apparently this device is accurate to within 20-40 meters, though one has to wonder how well it deals with people moving their wireless access points."

Comments

Too Simple

[nxtr]

They might as well give everybody a peice of paper with a huge X on it that says 'You are here'.

20 - 40 meters?

[Anonymous Coward]

20 - 40 meters? Who will be forgetting GPS with that kind of crappy accuracy?

Re:20 - 40 meters?

[ouzel]

Why, when you young whippersnappers were still in diapers we were feeling lucky to get 100m accuracy with SA enabled.

AND we had to walk uphill both ways in the snow to get it.

Re:20 - 40 meters?

[Tony Hoyle]

Sounds optimistic...

Let's say I'm within range of 50 access points all called 'Netgear'.

Where am I?

Re:20 - 40 meters?

[Anonymous Coward]

> Let's say I'm within range of 50 access points all called 'Netgear'.
> Where am I?

You do know that access points have MAC addresses, don't you? Every single MAC address is globally unique. They have a database of those, _not_ of the names.

Note that this database is going to get stale quickly, as people turn on new APs, move existing ones, or upgrade broken ones. Still, I think it's a great thing to keep in your arsenal of positioning tools.

Re:20 - 40 meters?

[wwest4]

Even better... the more sophisticated systems use some other data to improve accuracy: signal strength, and known mobility patterns. You may be able to get a better indication of where you are by examining the relative signal strength of nearby APs compared to the last known survey; the known mobility patterns will adjust if the algorithm places you somewhere you're not likely to be. e.g. if you're moving faster than 5 m/s you are probably not in a building (unless you've seriously taken a wrong turn).

Been there, done that!

[wintahmoot]

PlaceLab has been doing this for a while, and it's free.

Re:Been there, done that!

[Myself]

Strange how the world turns, I just mentioned PlaceLab to a friend before loading Slashdot. Spooky!

PlaceLab's big advantage is the ability to use multiple sources. Wardriving data is just one potential input. If you have a GPS receiver and a wi-fi card and a CDMA phone all connected, it'll use whichever is giving the most trustworthy results. So you can move smoothly between urban, rural, and indoor environments.

What absolutely makes me giggle is this: "Morgan adds that GPS typically only locates things within a few hundred meters, whereas the Wi-Fi location system can get within 20 to 40 meters of an object."

Re:Been there, done that!

[suineg]

Well think about the fact that you pretty much need to be within 20 to 40 meters of an AP to even pick the signal up then it would make absolute sense.

Re:Been there, done that!

[Myself]

But the original wardriving that put the APs into the PlaceLab database was using GPS as a reference.

I see what you're saying though, that a moving GPS on a single wardrive will have some error based on atmospheric effects, but repeated resurveys of an area on different days would tend to average these out, similar to the long-term averaging of a stationary GPS receiver.

My point was that the spokesdude in the article is either misquoted or misinformed about the accuracy of GPS, and that the neither Skyhook nor PlaceLab is likely to return better outdoor results than a consumer GPS receiver. Indoors is where this concept really shines.

Re:Been there, done that!

[Anonymous Coward]

The paper worth reading here is: Cheng et al's "Accuracy Characterization for Metropolitan-scale Wi-Fi Localization" which explores the issues related to AP density, churn, positioning algorithm, etc. See: http://www.cs.ucsd.edu/groups/sysnet/miscpapers/mo bisys05p124.pdf

What a joke...

[nemostultae]

Apparently this device is accurate to within 20-40 meters

Hell, I can guess where I am to that accuracy. I thought GPSs where accurate within 5-8 meters nowadays. And this sounds really useful out in the open ocean, you know, where all those rouge wireless access points hang out.

Re:What a joke...

[gardyloo]

And this sounds really useful out in the open ocean, you know, where all those rouge wireless access points hang out.

Ah, yes, the pink ones.

Re:What a joke...

[Andy Dodd]

Sub-5-meter accuracy: As another poster pointed out, SA was turned off a while ago. Pretty much any GPS unit will be accurate to within 5 meters if it has a decent signal. With tricks like code smoothing, most errors are probably less than 3-4 meters even for a moving receiver.

Sub-meter accuracy: A little bit of position averaging + basic DGPS makes this easy for a stationary receiver, even when SA was on DGPS could cure the intentionally added errors. Very difficult to use with a moving receiver unless combined with an inertial navigation system. (Rare except in modern airplane navigational systems)

Millimeter accuracy: Also possible before SA was turned off, but required the receiver to be stationary for a long period of time, and required significant postprocessing of the data using a variant of DGPS. It still requires stationary receivers for nonmilitary systems.

About the only thing that can't be done without a method for decrypting the P code is sub-centimeter positioning of a moving object. Even with the P code available it can't be done without combining a high-grade inertial navigation system with the GPS system.

Re:What a joke...

[Bogwood]

But vanilla GPS just isn't accurate in urban canyons. In London's docklands area (high buildings) GPS gives accuracy in the 200 to 400m region due to multipath effects. As well as obscuring satellites, tall buildings cause reflections of GPS signals which can cause large errors in the pseudo-range calculations making accurate position reporting very difficult. Places like NYC or Chicago are useless for GPS.

I would imagine that supplementing GPS with other position determining mechanisms (like WiFi) could be beneficial in these circumstances.

Dynamic data would be better

[photon317]

Rather than trying to maintain a static database of AP locations and signal strengths, they should just put some live wifi nodes out there with real GPS on them and track the AP map in realtime as it shifts. Or they could give free service to a select small percentage of customers in return for attaching a GPS device and helping recalibrate the map with some background software once a month or something.

Re:but there's really no point!

[JanneM]

Cities are the one place a positioning system is useless, so why develop it there?

You haven't tried to find a specific place in Tokyo or Osaka, have you?

Having a Gps is a life saver. You look up the place you want to go to on an online map, get the coordinates, and you're set. Without it, it's just too easy to miss the right building, mistake streets for each other or get lost in many other creative ways.

You could argue the other way round (and just as stupidly) - since there's one single highway or road in the entire area, why would you need a Gps to know which one you're on?

Re:but there's really no point!

[cryptoz]

Say what?!

Determine your location in a building with an accuracy of somewhere inside 40 meters. Oh yeah, that's useful. Which floor again? Which room? Eh?

Only 20 to 40 meters?

[meatflower]

I honestly would expect it to be even worse.
If this determines position by signal strength wouldn't it then be dependent on the type of antenna you were using with your WiFi card? Sometimes my signal moves around even in the same position or drops significantly lower in "dead spots". What if I'm using one of those crazy Pringles can antennas?

"Hey! 100% signal here, I'm here, over there and...yep, that a ways too!"

Anyhow, what an awesome idea, I mean, it's not like we have anything like this in existence, you know, that millions and millions of dollars were put in to launch satellites into orbit. No, nothing like that, nothing that has 10 feet or less accuracy. Guess we should all start usin this POS. No thanks.

You've never been to Japan I see!

[KNicolson]

Road naming is non-existant outside major thouroughfares; it works more by an irregular grid numbering system of blocks, not the roads in between them. House numbering is similarly vague, with no guarantee that house number 2 will be beside number 3 or 4. Block nameplates are usually pretty small and not in easy-to-predict places very often; GPS, even for pedestrians, is very useful in Japan. Even the taxi drivers haven't a clue where most places are!

Interesting. . . .

[Bagheera]

. . . but I suspect ultimately of little practical value. Having done quite a bit of RF scanning on the WiFi bands in one of their listed cities (San Francisco) I've seen first hand how signals behave in that dense urban environment.

GPS and WAAS operate on time signals and highly accurate positioning. Cell towers would be inherently more accurate since thier positions are accurately known and don't change (except under very unusual circumstances.

WiFi nodes come up and down constantly, and their position is rarely going to be accurately known by anyone but the person who installed it - and chances are they're not telling "you" exactly where the node is.

Given "walk around surveying" to map the nodes, it's not really a surprise they have accuracy that's no better than an early 2 channel GPS receiver.

And, as others have pointed out, if I'm in downtown San Francisco (or any other city) I don't need my GPS to tell me I'm at 5th and Townsend. For directions there's Mapquest, Google, Yahoo Maps, etc...

Interesting technology. But it sounds more like something a hobbiest would come up with than business.

Re:Interesting. . . .

[Dufffader]

But for typical outdoor GPS usage, depending on AGPS (Augmented GPS/Assisted GPS, etc) is a little redundant. You might as well stick to pure GPS for navigation.

What gets interesting is that AGPS is supposed to provide positioning indoors too, or in areas where GPS don't work so well. In a densely urban areas (think downtown NYC), acquiring the minimum 3 satellites for triangulation is not so easy. This is where using CellIDs and timing information from cellphone infrastructure to get your position is useful.

Now the question is... why isn't there a device to use both GPS and cell info to provide positioning info everywhere?

GPS SA off since may 1, 2000

[slew]

GPS selective availability has been turned off since May 1st 2000..Here's some more information.

Basically, the military figured out how to easily jam GPS in an area. But before then, there were GPS field units available that averaged out the error and got better than 2-3 meters so that it didn't really matter that much...

20-40 meters?

[Jurisenpai]

Don't these people realize how accurate GPS positioning has become?

MGIS-grade equipment can now give positions with sub-foot ( 30cm) postprocessed accuracy. Survey-grade equipment can get within 5-10 cm.

As neat as WPS sounds, I don't think that anyone will be giving up GPS soon if WPS can't get any more accurate than 20-40 meters.

Re:20-40 meters?

[Bastian]

Yeah, but most GPS equipment of that grade falls into the "If you have to ask how much it costs, you can't afford it" price range.

The system I work with can do sub-meter, sub centimeter with post-process. It retails for ~$40,000 plus a couple thou for a DGPS subscription and a few hundred to a couple thou for the DMI (odometer) equipment. And its precision falls off sharply (to as bad as 5 meters) in metropolitan areas where you get the GPS signal getting blocked by and bouncing off of tall buildings.

My guess would be that this system is sort of the opposite - a relatively cheap solution that works well when you don't need incredible accuracy , and it probably works best in dense urban areas where GPS tends to perform the worst.

That said, I'm not entirely sure how big their market is, because I imagine most people that need positioning equipment and work exclusively in areas with good wireless coverage are probably the type of customers who can afford expensive GPS equipment.

Re:20-40 meters?

[Jurisenpai]

I'm not sure what kind of system you're working with (or how old), but the system I linked to above is *very* roughly $3500. The software would add another $1500. So for about $5000, you have submeter accuracy. Multipath is still a problem, yes, but there have been great advances in solving that, too.

GPS has come down in price incredibly in the last few years. You don't even need a subscription to a DGPS service anymore.

Re:20-40 meters?

[invisigoth]

This can solve some problems associated with standard GPS. Namely, that satellite-based GPS systems fail when structures block line-of-sight to the satellites in the sky. This can include things like tall buildings in urban areas, or underground parking garages, etc. Since WPS is purely terrestrial, it can overcome many of these problems. Of course, satellite-based GPS is still preferable if you don't have these issues.

Re:20-40 meters?

[wramsdel]

What accuracy, exactly, does GPS give you when it can't see any satellites? Skyhook isn't trying to sell this as a replacement for GPS, rather as a complementary technology. In fact, they even have a nice USA Today-style graph on their "Technology" page showing that the overlap between GPS and WPS is fairly small. What marketing person in the world would look at this graph and say "Let's position our technology as a replacement for GPS!". That's not even Internet-bubble thinking.

That being said, keep in mind that the goal is locational awareness. Any information that you can feed into that equation has the potential to be helpful. It can also be harmful if used in an inappropriate manner. Take, for example, the case of an access point which has been moved. Your device could say "I see this access point, so I must be in some radius of its last known location." It could also say "I see this access point, but my last good GPS fix was 75 seconds ago, and the user can't have moved that far in 75 seconds, so I'll choose to disregard that information." If you consider all the potential inputs to the "locational awareness" equation: GPS, WLAN, WiMAX, fixed Bluetooth, RFID readers, etc., it becomes clear that you can start to do some pretty cool stuff with system redundancy.

Useless?

[knightPhlight]

I'm currently surrounded by the SSIDs, "linksys" and "default". Can someone tell me where I am?

Re:but there's really no point!

[JanneM]

u do have a road map in teh car dont u?

I'll get a road map in my car the moment I get a car. I'll put it right next to my dictionary.

Re:Oh man.

[nmb3000]

To be honest, I don't care how novel their ideas are. Using a system where position is located based on such arbitrary measurements is not only silly, but a waste of time. Not only can anyone move the access points around, but as they get shut off or more get added it will only make things worse. Also considering it can be influenced by minute things like weather and the position of the microwave in the apartment across the street make it a waste of time. You'd need to rescan it at least monthly to maintain even 20-40 feet (screw that) accuracy.

Let's see, a near-absolute positioning system based on immobile and unchanging (or extremely slowly) data, or something based on what could probably be described as a chaotic system? Not to be a jerk about it, but "Forget GPS"? More like "Ignore WPS".

Re:Oh man.

[Johnny Mnemonic]

Assuming people don't get swap-happy and trade access points all over the place, the reliability should be very high, too.

Well, that would be an issue, wouldn't it? And, unlike, say, a system you own, other folks own and control these datapoints that your system depends on.

I have to say that I regularly reconfigure my WAP becasuse, well, it's mine, and I chose to use it like a toy. I notice that a great many of my neighbors have WAPs of their own, but, not so surprisingly, I find that everytime I look the configuration is a little different--cause they're playing with theirs too.

You know, they lose power so vanish for awhile. Or are configured for security, but then the owner decides that's too much a pain in the neck so reverts. Or buys a different brand, hoping to get better signal. Or people move away--I hear that "rentals" are common in "urban centers", which tend to attract a transient demographic. Or the mix changes for any number of other reasons.

I hope for your sake that you guys took a snapshot and then took another 3 mos. later to determine average drift; I suspect that it'd be significant, enough that you couldn't triangulate off of it, at least.

Really, at this stage of your product cycle you shouldn't be guessing if this is feasible; you should be able to respond to this (obvious) criticism in the strong affirmative, without the guessing you displayed. How else can I be expected to trust it? And while it may be accurate for a month or so, it'll only be updated once a year? Gee, I sure hope I'm trying to get my position at the beginning of the year rather than at the end.

Sorry to be harsh, but really this is one of the stupider ideas I've seen posted here. You may as well give directions based on the make, model, and color of cars parked in driveways. Those don't change, much either. But over the course of a year, I guess they actually do, huh?

Re:Karma whore

[nametaken]

I'm worried less about privacy, and more about how I'm going to tape my wireless access point to my roomba just to mess with them.

But then, with such poor accuracy, it's not like anyone will be worried about 40 or 50 feet here and there.

Re:The Positioning Sledgehammer

[Technician]

Or, are any of my friends within 6 blocks of me?

They only way a free AP sponsor would be interested if any of your friends are within 6 blocks would be to suggest you all meet at Starbucks.

Not useful for typical GPS uses, but...

[SnprBoB86]

20-40 ft? This is totally useless for street navigation, surveying, etc.

What this is useful for is grander scale positioning without the need for a GPS device built into a portable device.

For example, timezones are far larger than 20-40ft. Laptops could be configured to automatically adjust the timezone setting to match the closest access points, no GPS device needed. A weather monitor utility could always automatically show the local weather. A star map could be configured to show the local sky. I'm sure many people can think of others.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Oh man.

[andy jenkins]

Yes, but my iPaq has Wi-Fi not GPS. And the screen can show a good map.

I could ensure I'm always carrying a GPS reciever or just a city map but you know what, I own both these and only carry them when I know I need them. Which in the case of GPS, is never.

Re:Open Source DGPS?

[DigitalRaptor]

I hadn't looked into this for a while, and it looks like some work has been done on it.

Several open source programs / projects are listed here.

I haven't tried whether they work the way I want them to, or whether they provide the accuracy I crave... :}

Re:but there's really no point!

[mrchaotica]

GPS is useful in the city... if you want your computer to know where you are. Imagine being able to broadcast "I am [here]" to your friends (and have them reciprocate) so that your computer tells you when you happen to be in the same place and you can meet.

Imagine being able to enter meetings in your PIM software by saying to someone "Let's meet here tomorrow at five" and having the computer know where "here" is (it would also know who you're talking to, using facial recognition, but that's a different topic...).

Imagine your computer reminding you of work-related tasks as you walk into the office, and home-related ones when you arrive at home.

Imagine walking into a place you've never been, and having your computer automatically display useful information about it.

Granted, all these scenarios depend on technology beyond what's described here (namely, wearable computers), but even so GPS (and WPS, for areas such as inside large buildings where the GPS signal can't reach) is (or will be) more useful than you think.

Re:Oh man.

[MoneyT]

Now the real question is, how often do access points get moved or shutdown?

Re:Not very accurate

[MoralHazard]

20 to 40 meters of accuracy? I work with various grades of GPS and even with low accuracy gps I can get within 10 feet no problem.

My FAVORITE kind of slashdot post to respond to is this one. It's a combination of "I'm missing the point entirely" AND "I think I'm a lot smarter than I am." All rolled up into four little sentences.

WHY IS WPS USEFUL: Because there are a lot of urban areas where you can't get GPS signals for shit. Try New York City, for one--you're lucky if you can get two or three satellites in Manhattan, most of the time. I've spent a LOT of time trying to make GPS work in this city, and it sucks--and I know there are probably other places where the same thing happens.

So if your GPS can't give you any kind of signal, 20-40 meters is pretty frickin' good, wouldn't you say? Sure, 2.4 and 5.8 GHz signals get attenuated by common building materials, but not so much that it throws off triangulation too badly. I can routinely do reasonbly accurate triangulations (+/- 10 meters) on access points (sort of a reverse of the WPS process) near the ground in all sorts of medium-heavy office buildings.

Now, add in the fact that there are probably at least 10 million wireless client devices in consumer hands in the US today (a number that grows as we speak!), and compare it to the distribution of GPS receivers. Chances are, most everybody with a laptop has WiFi, but there's a lot of people that don't have GPS. Cost isn't really such an issue--there are cheap-ass USB GPS receivers all over (running on Linux, even). I think it has more to do with the fact that GPS is less of a "need-to-have-it" thing, and it's rarely built into laptops and PDAs.

Point is: MANY more people have a WiFi receiver than have a GPS receiver, and it often works as well or better in urban areas or in buildings.

Do you see it, now? Sorry for the snark-attack, but man, how did you not get this?

Skyhook?

[kingofalaska]

That is a mean trick played on newbies in the Army, at least in my first Air Assault unit. An old-timer says "go over to supply and get a skyhook, and some frequency grease for the antenna." He has a rational sounding explanation that the skyhook is the thing under the helo that the slings hook into. The newbie goes over to supply, and the seargent, playing along, sends him to someplace else. Everyone passes the poor newbie around, as long as the game can play.

On a more serious note, this sounds suspiciously like a project I worked on years ago, only this one is not as deep. Or nearly as accurate.

KoA

Alaska men should hit the trail for breasts

Ekahau did this as well, but better.

[spiff42]

I have been working with a commercial localization system called Ehahau Positioning Engine which we have acquired a license for at the Technical University of Denmark. This system uses the signal strengths of available access points to determine the position. The client is a piece of software running either on a laptop or PDA (they also have some nifty tags that can be used).

The client software running on the tracked device measures the signal strength of the access points, forwards the data to the server which calculates the position. The big-brother scenario is avoided as long as you still have to install the client yourself.

The major drawback of the system is that it needs extensive calibration, since they are using not only the available access points, but also the signal strength of these. Normally they suggest calibration in a 5x5m (15x15ft) grid. More calibration points yield a more accurate result.

And now the piece of information you have all been waiting for: accuracy. With a good calibration this can yield accuracies of arround 1m. In my tests (indoor) the accuracies fluctuate a bit, but is at least better than 3m 95% of the time.

Just as the system described in the original post, Ekahau requires no extra hardware (we already have 2-300 APs on campus).

/Spiff

That's not a security measure...

[Otto]

Since a large number of people would check "Do not broadcast the SSID" following their manufacturer's manual on security,

What manual says to do this? Turning off SSID broadcast is *not* a security measure in any sense, at all.

For one thing, the SSID is included in every single packet that the access point sends out. Period. So getting it is easy with or without the SSID broadcast.

For another thing, turning off SSID doesn't prevent anybody from connecting to the network. It will prevent stupider displays like Windows's wireless page from showing that access point as available, assuming no other AP has the same SSID being broadcast, but if they select that SSID from another AP or if they put it in manually, then they'll connect to your network just fine.

If you want security, enable WPA. Turning off SSID, filtering by MAC address, these are not security related adjustments, and add precisely nothing to your overall security strategy. They might be a way to keep your idiot neighbor from connecting to your network by accident, but they won't keep anybody from connecting to your network on purpose.

Re:The Positioning Sledgehammer

[anthony_dipierro]

All this effort to develop technology to determine your location is great if the reason for finding your location is because you're lost. But, otherwise, it seems like another case of the technology industry developing a new market for devices of questionable usage.

Wow, I've gotta really disagree here. If GPS were cheap enough, I can think of a lot of legitimate uses for it other than when I'm lost. Basically, it's useful for tracking things (where did I park my car? where did the car thief take my car?), and tracking myself (how far did I travel for business purposes in 2004? what time did I get to work yesterday?).

Throw legitimate privacy issues into the mix - generally in the US at least and certainly elsewhere - the thought of some anonymous entity determining my location is positively horrifying.

If by "some anonymous entity" you mean the government or your phone company, well, they could have already done it anyway. I don't see it as horrifying. I don't think it's such a good idea, but I still have a cell phone.

Why is it so difficult to simply self-identify my location

Besides the fact that good voice recognition technology isn't that widespread, it'd be a big hassle to constantly identify ones location. ("OK computer, I've parked my car in section D-5. OK, now I'm arriving at work.")

rather than relying on the sketchy availability of GPS satellites or databases of WiFi APs and doing all that trigonometry

Sketchy availability? Availability of GPS is basically a question of "are you outside?" It doesn't work as well indoors, and that's a problem for a lot of uses we might dream up for it (tell 911 where I am), but it was built for guiding missles, not calling 911. This is one advantage that wifi access points may be able to give. Of course I'm skeptical as to how reliable such a scheme would be. It certainly will never be global like GPS. I can get a GPS reception in the middle of the pacific ocean.

Re:The Positioning Sledgehammer

[LocoMan]

Now it'll only have to specify on which starbuck you are to meet them out of the 10 within those 6 blocks.. :)

Re:but there's really no point!

[brontus3927]

Actually, this is right along the lines of what I was going to suggest. For me, I'm horrible at remembering appointments and other such tasks. Sure, I can set up reminders and what not to remind me at 5:00 that I'm meeting a friend after work, but what about stopping at the bank on my way to the groccery store. Can't tell you how many times I've forgotten to make that stop. So a *PS-enabled system to remind me to turn left at the light and go to the bank instead of right to the groccery store would be very helpful

They got it backwards

[Jahz]

This service is interesting, but not very useful. How often are you using a wireless connection but you dont know where you are?

It reminds me of a project a friend of mine was working on earlier in the year. The college of computer science here at my university in Boston has a brand-spanking new building with cisco wireless routers built in (through all 15 or so floors). My friend was working on using these routers to triangulate the position of all of the wireless users.

This was accomplished by maticulously mapping the average signal strength of each router at various distances (1 foot, 2 feet, etc...). Since all the routers overlap a bit (as they should), extrapolating the position is not that hard. You just need a minimum of 3 routers to see any user. The more routers seeing the user, the more accurate the location. Its strikingly similiar to how the epicenter of an earthquakes is determined.

The authentication server behind the wireless network could then deny access to connections outside the building. The aim was to stop leechers from the nearby dorms and apartments, while still providing free authentication-less access (http only without authenticating) to the inside of the building.

Anyway I need to check up and see how that project is going. It was progressing well last time I checked. I think network security is a far better application for WPS.

What about RPS?

[scorp1us]

Can't you triangulate from FM radio station signal strength?

It woud have the following advangates:
Way smaller database
Way more coverage

It should be easy to do with the adevnt of software radio.

The only down side, is that you wouldn't need this company anymore!