Traffic reports to derive from cellphone location data

Intelligent roadways and traffic monitoring systems have been available (albeit not always entirely accurate) for some time now, but if IntelliOne and AirSage have their way, finding out about real-time roadblocks (and voyeurism lawsuits) could become a more fleshed-out reality. The firms are looking to utilize that oh-so-telling "anonymous" location data from each traveler's cellphone to pinpoint locations and overlay that information with maps. If wireless companies open up that data at a rate of "twice per second" while users are conversing and "once every 30 seconds" when not on a call, the entrepreneurial duo hopes to offer more detailed information and pragmatic advice than "radar, helicopters or cameras" currently do. While keeping a keen eye on traffic developments certainly has its benefits, the real issue here is privacy (or the lack thereof); while government uses have already been in place, carriers are (understandably) more hesitant to turn over consumer data for locating purposes. While the service would be marketed free of charge to wireless carriers, interested customers not enraged by such intrusions could purchase the data for a monthly fee, and if all goes as planned, the Tampa pilot that is currently ongoing will lead to "40 other markets" being invaded by this time next year.

[Via The Wireless Report]

[Farris @ Nov 6th 2006 2:13PM]

So this assumes that every motorist has a cell phone. What if there is a large group of them without phones? Or some people that have two or even three (work, personal, small business) phones?

[Nick @ Nov 6th 2006 3:16PM]

no, it tracks individual phones. just looks at the speed of each one, so your concerns are irrelevant.

[Gil @ Nov 6th 2006 2:32PM]

Uhhhm:

http://www.engadget.com/2006/10/27/vodafone-tomtom-partner-to-create-real-time-traffic-data-networ/

[Kyle S @ Nov 6th 2006 2:35PM]

i couldn't care less if sprint knows where i talk on my cell phone (don't they know this anyway?). i think any privacy concerns here are overblown.

[Robbert @ Nov 6th 2006 2:41PM]

At http://actueleverkeersinformatie.brabant.nl/ (in Dutch, fill in the captcha at the left) you can see a realtime example in the Dutch province of Brabant, based on the information from Vodafone phones.

[Wonderboy @ Nov 6th 2006 3:18PM]

Oh no!!! Someone could look at a report like this and realize that a random person located at exit 32 has a cell phone... it's the end of privacy as we know it.

/sarcasm

As long as this information is provided without any identifying information it poses no risk... in other words Verizon sends info that it records 10 cell phones at the 32.3 mile marker and 2 cell phones at the 12.7 mile marker, maps auto-generated, and you have a relatively accurate, relatively accurate traffic estimate.

Farris, it's pure statistics... lets say 8 out of 10 motorists have a cell phone (rough guess), it would be pretty rare to have a significant back-up without a good portion of that back-up carrying cell phones. The invisible back-up you're expressing fears of is virtually impossible in todays cell-driven world.

[Farris @ Nov 6th 2006 4:27PM]

I realized your point after I wrote the first question of my comment, but was too lazy to erase it :-P

But my other question still stands. At the lab I work at, it seems like every other person here has two cell phones, including one guy that has three. In the mornings, there is terrible traffic coming out here, mainly because of the power plant that is down the road, and the hundreds of motorists driving there for plant-work and other assorted government work. Would this system be able to cross-reference phone numbers (or some other phone-specific ID#) and say, "Oh, car A has two phones, but it is only one phone" rather than "Two phones= Two cars."

And there are always commuter cars!

Anyways, just my curiosity wasting my time at work...

[M. Rod Von Hugenstein @ Nov 6th 2006 3:51PM]

Funny, how they plan to charge me for data that I am helping to provide.

I refuse to pay for this, simply because of that.

[Peter @ Nov 6th 2006 4:22PM]

M. Rod - You already know where you are. They're charging you to know where everyone else is. ;)

[Mike @ Nov 6th 2006 4:22PM]

How accurate can this possibly be?

[Wonderboy @ Nov 6th 2006 5:03PM]

Fair enough... I guess it would depend on how accurate the GPS is. If they can get it down to several feet then they can use some sort of basic estimation. For example, lets say they're getting a report of 32 cell phones in a 50 square foot section but their calculations show a maximum of 25 cars in that same area, they can assume multiple phones per person or multiple passengers per car. I agree, it's a tricky business with a lot of room for error.

Of course if they take a different approach and instead of reporting cell phone density on roadways they report cluster movement speed (since generally packs of cars in traffic stay together for a minute or two at a time, so those same 32 phones would create an ameba-like mass inching along the highway) they could pinpoint back-ups and bottle-necks much more accurately.

The third approach would be to identify individual phones and capture their speed and location... I find this doubtful because it would be far too labor intensive to keep track of the millions of phones. Of course if this is the case, I fight this as it borders on privacy issues since it's no longer a random signal, it's been given an identifier that tracks specific movements.

Bored at work too, though now it's time to go home... hope there's no traffic.

[scooter @ Nov 6th 2006 5:40PM]

its not what this does, but what it least to. Governments big busisness dont take your liberties all at once, thye slowly take it. Bit by bit your information is used (for your good) until it is abused in the name of terrorism and or whatever they feel it should be used for.

I do not and will not carry a gps equiped cellphone.

Just because i dont do something that deserves watching does not mean i want them to be able to.

[Jake @ Nov 6th 2006 5:43PM]

How are they going to know when you're driving and when you're not?

[Joel @ Nov 7th 2006 9:34AM]

So a bus full of people would count as 80 vehicles, then?