Interesting article on the dangers of GPS jamming

[230RN]

Saglant, bring up the OP's article at

http://www.newscientist.com/article/dn20202-gps-chaos-how-a-30-box-can-jam-your-life.html?page=1

Go to page 3, and look at the last paragraph.

Terry, 230RN

[RoadKingLarry]

from 230rns link-

And eventually you'll be able to navigate without any external signals, thanks to devices called "inertial measurement units", which track your movements from a known start point. Today, these IMUs use gyroscopes to measure orientation, plus accelerometers to tell how fast it is accelerating. Using this information, plus time, the acceleration is converted into speed and distance to reveal relative location.

Not like that is new technology. subs have been using it for decades and it was accurate enough for SLBM targeting 30+ years ago, I can't imagine it has gotten worse in the intervening years.

My first boat we had a dual IMU system called DMINS, second boat we ran a dual ESGN (ElectroStatic Gyro Navigation) System again plenty accurate for targeting SLCMs.

[RevDisk]

$30 GPS jammers have a very small area of usefulness.  You're talking a building.  Best case, maybe a block if you know what you're doing.  It's extremely easy to pinpoint.   Honestly, the answer is going to be sideband logic circuits to tell if the signal is real or legit.  If the GPS signal strength is X, and suddenly you get 5 * X, well, obviously you're being jammed or spoofed.  Doesn't stop the jamming, but you can run lockdown and notification routines.  If the signal is the same strength as the legit signal, you're just going to interfere with the level of accuracy, which can also be compensated. 

Basically, the fix will mark up the cost of the units by maybe $20 in hardware (if that), and a couple dozen man-hours of coding.

This is basically what GPS guided weapons have incorporated, plus they have access to better algorithms than are publicly available.  The GPS system was set up to give an accuracy edge to the military, basically they can read the P(Y)-code transmitted on both L1 and L2 with significantly less intensive computation than those without the crypto key. 

All satellite navigation systems are vulnerable to this.   It's not that big of a deal aside from DoS, and it's pretty easy to track.  It's only a big deal if you didn't make a system with robust countermeasures.   

Here's basically what happens:

Security Geek:  Hey boss, can we add $20 to the price tag to make the thing more secure?
Marketing:  zOMG, where's my martini!   We can't raise the price, our competition will stomp us.  Besides, I'm making commission here, and fraud prevention isn't factored into that!
Boss: My annual bonus is based on how many units we sell!
Security Geek:  <facepalm>

(Few weeks/months later)

Security Geek:  Some teenagers figured out how to defraud our devices with about $2 worth of parts.
Boss:  zOMG!   The CEO threatened to fire me unless we can announce to the media and shareholders that we fixed the problem!
Security Geek:  It'll now cost a $100 in parts to fix, and $600 to pay people to install it.  PER DEVICE.  Instead of $20 per device, like I told you!
Boss:  Cool, go do that, and I'll suck up to the CEO saying it was your fault for not implementing the fix in the first place.

Repeat as necessary.   

This is how systems become more robust. 

[sanglant]

that's what i meant, 3 seconds doesn't require updating the time more then once a day. randomly at that.

[KD5NRH]

If the GPS signal strength is X, and suddenly you get 5 * X, well, obviously you're being jammed or spoofed.  Doesn't stop the jamming, but you can run lockdown and notification routines.

Even simpler; reality checks on the devices depending on the GPS.  If it claims your bicycle or aircraft carrier briefly fired up warp engines for a half-second round trip to Bolivia, drop that data and wait until something more realistic happens for 2-100 packets in a row.  The time-dependent ones are even easier; clocks accurate to milliseconds per week are cheap and tiny, so use two.  If they agree and the GPS is more than a second off, disregard it until it either comes back to a reasonable error margin or a user initiates a munal reset.

[roo_ster]

Everything RevDisk wrote is near enough to God's Own Truth and to several incidents where I was Security Analysis Geek y'all ought to read it twice. 

I do analysis at the behest of PHBs, customers, more-senior engineers, & such.  I also do parallel analysis for my own benefit to better understand the problem, risk mitigation, etc.  About 1/3 of the time, my risk mitigation "unauthorized why are you wasting your damn time on that s*** 'cause we have teh One Party Line Groupthink True Answer" analysis becomes the preferred alternative or highlights a risk that mgt finally takes seriously.  Yeah, only a ~33% average for my gut, but the risks it has addressed have amounted to enough $$$ saved/made that my gut is worth its weight in ruby-studded platinum slathered in truffle sauce and the time expended in the noise of any staffing budget.

"Yes, this shows that if your O&O for $uper-$ystem is as you have it currently, you'll get slaughtered by the T-55 driving, loincloth-wearing, illiterate, zebra-molesting locals.  Might want to address that, Mr. RetiredgeneralwehiredbasedonUSMAcontactsnotintelligence."