GPS signals are surprisingly easy to disrupt

http://www.economist.com/news/international/21582288-satellite-positioning-data-are-vitalbut-signal-surprisingly-easy-disrupt-out

by douglasfshearer • 13 years ago

218 119 13 years ago

economist.com

cschmidt• 13 years ago

That reminds me of another Economist article about GPS jamming from 2011:

http://www.economist.com/node/18304246

That article was inspired by a trucker who drove by Newark Airport every day, disrupting GPS service there.

bjcubsfan2• 13 years ago

I worked on that, but have not seen that particular article. There are actually a whole lot of them going by the Newark Airport even today.

VLM2• 13 years ago

"receivers cost £2,000 per vessel"

Hmm, yes kind of like how a new GPS costs $5K, because the first .mil model did cost that much. Or all computers currently cost millions of dollars because the first ones did.

hga2• 13 years ago

Errr, as I read it, the cost is for new eLoran receivers, intended to be an alternative backup to the Mark I eyeball if GPS flakes out for whatever reason. Not going to be widely deployed like GPS receivers, and big ships need reliability ... in fact, the only customers are those who really need this sort of thing, unless it becomes mandated.

snom380• 13 years ago

A new GPS made for commercial, certified use (eg. for use in aviation) will cost way more that $5K. Just a software update to it might cost as much. Technology is getting cheaper, certifying it isn't. In don't know about ships, but I assume there is some form of certification for nav equipment there as well.

D9u• 13 years ago

I don't know about presently, but in the past the GPS transmitters had an introduced error known as "Selective Availability," the purpose of which was to deny pin-point accuracy to unauthorized devices.

For a price, one could acquire the necessary hardware/software to mitigate Selective Availability in order to increase the accuracy of GPS readings, although I forget what those prices were/are.

ThatGeoGuy2• 13 years ago

As somebody who's primary field of study is GPS / Navigation / Positioning, I'd like to remark that survey grade receivers such as in the eLoran stations can cost much, much more than £2,000.

Their price isn't so much the hardware itself, in fact, most of the receivers seen in cell phones today cost less than $1USD. However, when dealing with more accurate receivers, the receivers typically need extensive calibration and testing. I mean, it's possible to get measurements from GPS accurate to 1mm, but you need to have both a careful setup and very good knowledge about the errors associated with your receiver. So yeah, these receivers are actually quite expensive, and most of it is because the calibration on repeatability and stability of these sensors is of the utmost importance.

On a secondary note, survey-grade receivers are typically not as prone to jamming as consumer-grade receivers, mostly because they can use multiple GPS frequencies / can collect over a wider bandwidth. There's a lot more to it than that, but effectively the point I'm making is that the receivers spoken of in the article are not the same as everyday GPS receivers.

If you want to see how crazy the prices on these receivers can get, look no further than http://www.surveyorsmart.com/product.sc?productId=548. I doubt many surveyors would go for that particular model by Leica, as it is definitely expensive, even for survey equipment standards, but it's not unreal in terms of pricing.

joosters3• 13 years ago

1mm?

lmm• 13 years ago

So going in the other direction, are there cheap eLoran receivers available? Would one be any use as a backup to, say, someone with a cheap second-hand yacht who occasionally goes sailing?

bkanber• 13 years ago

There are different grades of GPS receivers, all licensed differently. Consumer-grade receivers have a maximum accuracy of something like 3 meters and a slow response time. There are also aviation grade and military grade receivers, which cost more (build & calibration) but also require a special license.

ballard2• 13 years ago

I interned at Trimble in the radio group. Industrial GPS receivers could be purchased by more / less anyone, however they were built and priced for heavy-duty industrial consumption (mines, farms, ships, aircraft, etc.). The unassisted accuracy without kinematic corrections was around +- 1 m horizonal and ~10 m vertical, constantly changing as the constellation moves. This would be for a top-of-the-line 24 channel parallel receiver. WITH kinematic GPS, +- 10 mm (not a typo) horizonal and ~3 m vertical. It was so good that receivers could be installed on the left and right sides of a grader's blade that the angle and position could be known with great accuracy and precision. Kinematic means a nearby stationary reference ground base station sending updates over radio frequencies to another GPS receiver that could be in motion. This was also used to subtract SA pre-Clinton because the injected error would be the same in nearby locations.

randomdata• 13 years ago

$5K doesn't seem out of the question. Even commonly found used receivers can go into the thousands of dollars. Ex: http://www.agdealer.com/list/view_image.cfm?ID=401541

themstheones• 13 years ago

I love how it's just some dude running a delivery truck inadvertently doing this. It's like something out of a movie.

wmf2• 13 years ago

Yeah, I was hoping it was the ten minutes before market close every day and some evil genius HFT operator was trying to exploit lack of time sync at the exchange.

joosters3• 13 years ago

Same here. If they were really cunning, they could fabricate their own signal with a different timestamp. You could shut down rivals by making their computers believe that the trading day was over, or that their market feed was too slow to trust...

rmc• 13 years ago

It sorta drives home how vulnerable the systems are. If someone can do it accidentally, the what could a malicious, well resourced attacker do? How come a well resourced attacker hasn't done it?

zokier• 13 years ago

How come a well resourced attacker hasn't done it?

Umm.. the article mentions that NK has been jamming GPS, and Iran allegedly landed US drone via GPS hacking.

Well resourced attackers are doing it.

speeder• 13 years ago

There is a legendary tale of a how a random delivery truck took down a whole popular MMO.

The MMO company had a issue with their own physical logistics, and had to put their login server across the street in another building.

They put special antennas between the two buildings, to keep the thing running, and it worked great, until one day, it didn't.

After a while trying to figure what was going on, they noticed that a huge delivery truck parked illegally (the place where it parked was illegal to anyone park), and the place he choose made the bulk of the truck metal stay right between the two antennas...

EDIT: Also when internet was still dial-up, one delivery truck ignoring height limits managed to cut all phone cables near my town biggest ISP, leaving the entire city without internet. I did not appreciated it...

DrStalker• 13 years ago

I've heard of a similar issue with a wireless data link across a river that would cut out once or twice a day, but not everyday, and no one could figure out why. The answer only came after a few weeks of monitoring and an ametuer fisherman saying "hey, those are all high tide times!" at which point they realise that high tide was causing the mast of a docked boat to rise up enough to interfere, and that the boat was not always in its dock which is why some days were fine.

jjwiseman• 13 years ago

The article mentions North Korea's frequent GPS jamming, but doesn't mention that it's suspected as a cause of the fatal drone accident last May: http://lemondronor.com/blog/index.php/2013/3/gps-loss-kicked...

I haven't been able to find any more recent info about the incident.

jandrewrogers• 13 years ago

Systems that rely on GPS are defective in their design. US military systems have never trusted GPS, since inception, because the Soviets had the ability to destroy the satellites and tinker with the RF.

US military systems are inertially guided only accepting GPS corrections within the error bounds of their extraordinarily tight inertial measurements. If the GPS is outside the error bounds, it is presumed to be compromised. There is no such thing as an intrinsically GPS-guided weapon; they are always inertially guided with GPS corrections. Modern US inertial systems are similar to GPS in terms of precision so GPS is increasingly superfluous.

As a consequence, the US military has never been significantly vulnerable to loss of GPS. Their systems were designed, from Soviet days, to assume that GPS could be compromised or lost since the Soviets had that ability. Knock out GPS and you still have to deal with ultra-precise solid-state optical gyros.

Many commercial systems take a cheap shortcut and rely on GPS but, contrary to popular mythology, US military systems never have.

samspenc• 13 years ago

Bane?

jessriedel2• 13 years ago

In principle, GPS jammers are illegal, right? FCC regulations or some such?

wmf2• 13 years ago

In principle, you'd read the article before commenting.

michaelwww12• 13 years ago

What, you didn't read it for him?

DanBC• 13 years ago

(I didn't down vote you.)

This is in the UK, so it's probably OFCOM who do the regulation and prosecution. (http://stakeholders.ofcom.org.uk/enforcement/spectrum-enforc...)

EDIT: Whoops, yes, this is mentioned in the article.

jessriedel2• 13 years ago

Wow, I'll make sure never to comment again without reading the article three times over to make sure I didn't miss a sentence...

ChuckMcM• 13 years ago

As this becomes more annoying people will start watching for terrestrial GPS transmitters and start busting them.

drivingmenuts• 13 years ago

They seem to be taking it awfully casually, as if it's all some minor misunderstanding between gentlemen.

afterburner• 13 years ago

Well it is Britain.

eCa• 13 years ago

Should it not be possible to track the vehicle (if it is only one) by installing GPS receivers in the area and noticing when the disruption begins and ends for each receiver?

That, coupled with license plate scanners at those locations, ought to find the culprit.

(And then they can launch an armed drone to take him/her out...)

jessaustin• 13 years ago

(Presumably the drone will use some sort of inertial guidance rather than GPS.)

emiliobumachar• 13 years ago

The culprit is basically shining a very bright light (at invisible frequency) in all directions. Catching them should be even easier than your proposal.

(I suppose one could program a missile to follow the light...)

sbierwagen• 13 years ago

  basically shining a very bright light

GPS satellites broadcast 500 watts from a 13dbi antenna.

But they're far away. Really really far away, to the tune of 21,000km. (The Earth is about 12,000km wide.)

By the time the signal makes it to the surface of the planet, the inverse square law has punched it down to -130 dBi, which is well below the thermal noise floor of room temp electronics. GPS receivers pull a useful signal out of a storm of random noise using deep information theory magic, which is vulnerable to several clever low power attacks.

Even a dumb jammer doesn't need much power, though.

  one could program a missile to follow the light...

http://en.wikipedia.org/wiki/AGM-88_HARM

"The AGM-88 High-speed Anti-Radiation Missile (HARM) is a tactical, air-to-surface missile designed to home in on electronic transmissions coming from surface-to-air radar systems."

michaelwww12• 13 years ago

As someone who was recently a delivery driver, the recent addition of GPS tracking is one of the most annoying things about a delivery job. Far worse than drug testing. Driving can be difficult and sometimes you need to park and stretch your legs. I was called into the office for this and queried about each of my stops. GPS is going to become more prevalent as people like me are highly motivated to defeat the system, especially working in teams with other drivers. Of course there are work-arounds. Park at a legit spot on an approved break and leave all your electronic devices in the vehicle and walk to where you want to be but don't want them to know about, etc...

However, it is highly doubtful this was done by a delivery driver. There is no point in jamming the signal for 10 minutes. What could it accomplish? If the purpose is to take an illegal detour (going against a one way street) the company has logs of all your trips and will soon figure why you have a 10 minute gap everyday.

nfm• 13 years ago

I took it to mean that the 10 minute window was when the driver was driving past the LSE, where the detectors where.

michaelwww12• 13 years ago

It's even less likely he jams signals where ever he goes and LSE just happened to notice it. His company would be very curious why he was disappearing. They can pull him on a screen and watch when he goes in and out of range. It would happen only to him, in whatever vehicle he happens to check out that day. A driver would have to be far more clever than that.

The article states "timestamps on trades made in financial institutions can be affected" There's your likely culprit; someone who for whatever reason needs to affect timestamps.

tinco2• 13 years ago

I think the parent means that the person who is driving past the LSE with his jammer is less likely a delivery driver, as they are under close scrutiny but rather anyone else.

Might as well be anyone who'd rather not be tracked if they can help it. I can imagine perhaps a VIP or even an organized crime member who prefers their car free of GPS trackers when they drive around town visiting associates.

jonknee2• 13 years ago

Driving can be difficult and sometimes you need to park and stretch your legs.

If your employer doesn't allow legally required breaks you can use the GPS log in your suit against the company. Otherwise it sounds like you weren't doing your job.

tinco2• 13 years ago

Someone isn't doing their job when they are taking breaks more often than the legally requirement minimum?

Sounds a bit harsh to me.

toomuchtodo2• 13 years ago

People aren't machines and shouldn't be treated as such.

michaelwww12• 13 years ago

There were far worse reasons to file a lawsuit (such as the company encouraging out right fraud in record keeping) but not many drivers are going to do that. They'd rather just complain and play the game while staying gainfully employed. There are all kinds of subtle ways to get rid of problem drivers. You've heard of Three Felonies a Day? [1] A driver can easily break a lot of driving rules and if the company wants to be strict about it, you're gone.

[1] http://www.threefeloniesaday.com/Youtoo/tabid/86/Default.asp...

marincounty• 13 years ago

You should become a Manager. You would fit right in. Dems--Dees-and Doos. You would make a great middle management Dink.

polemic• 13 years ago

Courier companies are going to have to wise-up pretty quickly and realise that the operational benefit to accurate GPS tracking is far too important to risk using as an HR tool.

At the end of the day, route optimisation can make or break a courier company, and getting good metrics on driver routes is just as important as getting (for example) tracebacks on your errors when coding.

The smart companies will take a no-blame approach: a driver is responsible for his route with the assistance of the intelligence provided by the company. Any follow-up should be constructive, and the good drivers will feed information back to improve the system.

There is no way it should be used to check your break schedule, and any operations manager should be coming down like a ton of bricks on anyone who is undermining such an important tool in their business.

michaelwww12• 13 years ago

I didn't work for a top tier company. Both companies I drove for had incompetent management all the way up to V.P. level. They were petty, stupid, and they sure as hell will use GPS tracking as a punitive tool.

teyc2• 13 years ago

There's an Australian company who's got a land based positioning technology. It does not rely on satellites and I think it holds a lot of promises especially for airports.

dragonwriter• 13 years ago

Land-based, non-satellite radionavigation (particular for aircraft) is older than GPS (or satellites, or even jet engines.) (e.g., VOR, DME, TACAN, etc.)

adestefan• 13 years ago

Don't forget about LORAN.

goodcanadian• 13 years ago

I imagine they are not the only one, but are you thinking of Locata (http://locata.com/)?

teyc2• 13 years ago

yes, thanks!

scotty79• 13 years ago

I guess we should migrate towards non GPS position tracking faster: http://www.darpa.mil/NewsEvents/Releases/2013/04/10.aspx

ballard2• 13 years ago

A paper on GPS jamming countermeasures

http://servv89pn0aj.sn.sourcedns.com/~gbpprorg/mil/gps4/GPS-...

Ultimiately, the lack of MAC and encryption makes civilian GPS MITMable.

(Discliaimer: I worked at Trimble.)

ihsw• 13 years ago

Interestingly, Google's successor to BigTable -- Spanner -- utilizes GPS clocks and atomic clocks to maintain consistency of their data across globally-distributed data stores.

http://en.wikipedia.org/wiki/Spanner_(database)

jcurbo• 13 years ago

This is not unique to Google, GPS is widely used as a time source for telecommunications and as an input to NTP daemons.

sargun• 13 years ago

Also, Google uses accurate atomic clocks in each of their datacenters, so when their GPS connectivity gets cutoff, they can still keep time. Almost everyone does.

leif2• 13 years ago

I didn't know GPS hammers were a thing but now I want one. Thanks, LSE.

leif2• 13 years ago

Jammers. Thanks, iPhone.

skeoh• 13 years ago

Make sure it is legal to own/operate one in your jurisdiction before buying any sort of jammer.

stock_toaster• 13 years ago

When all you have are GPS hammers, everything looks like... GPS nails?

eli• 13 years ago

Isn't GPS clock sometimes used to set up high-speed network connections? Maybe someone is jamming GPS to mess with a particular HFT application.

rayiner• 13 years ago

GPS signals are unsurprisingly easy to disrupt. It's a location mechanism that depends on one-way communications and triangulating an extremely precisely-timed signal between at least three far away satellites. Position fixes don't rain down like mana from heaven.

csense• 13 years ago

surprisingly

GPS signals come from satellites. AFAIK there are about 20-40 GPS satellites. Let's say each covers at least 1/50 of the Earth's surface area, about 500 million km^2. That's 10 million km^2 per satellite.

What's the max power generation you can fit on reasonable-sized satellite that's going to be in orbit for decades? I'm guessing not more than 10 kW. So 10 kW / 10 million km^2 means you have .001 watt per km^2.

GPS signals being easy to jam isn't surprising. What's surprising is that you can detect them with equipment that's small and cheap enough to fit in a cellphone!

mikeash2• 13 years ago

Where it really gets amazing is when you discover that the GPS signal is around 20-30dB below the noise floor. That is to say, ambient noise drowns out the signal by a factor of 100-1000. Long-term correlation manages to pull out usable information from this extremely noisy signal.

An interesting consequence of this is that you don't need a very fancy receiver, hardware-wise, because your software already has to be extremely sophisticated. For example, you can use a 1-bit ADC rather than trying to be more sophisticated about exactly how strong the signal is at any given moment.

jordanthoms2• 13 years ago

That's crazy... I think I'll contain my frustration more now when my GPS takes a few minutes for a cold lock. It's even more crazy because with the latest GPS systems in phones, I'm able to get a fix with only a very small portion or even no sky visible.

bobdvb• 13 years ago

Many many years ago I was a specialist in satellites for a major European broadcasting company. We had a problem with couriers who used radar speed trap jamming devices. They had really noisy oscillators which leaked huge quantities of out of band noise into the dished which were located near the courier drop off point.

We initially used an expensive spectrum analyser attached to a spare LNB (pointed at the parking area) but that needed you to notice the noise. So I used a USB enabled RF power meter interfaced to Excel (so sue me, it was a rush job and we weren't equipped for development). A script ran which fetched values, when there was a spike in received power it played an alarm and flashed the screen.

The next step was to ask security to hold anyone in the carpark while a large angry engineer went to give the courier a piece of his mind. It worked, courier companies were informed, people were shouted at and I believe it stopped.

There are probably better ways of doing this, but it was effective and built from spare parts.

GPS signals are surprisingly easy to disrupt

Related Stories