INS and GPS.

[Phalconphixer]

A moments levity regarding INS...
Originally published I believe in a BA Staff magazine...

Inertial Navigation (INS 101)

The following explanation, in simplified terms, is typical of all IN systems:


1. An aircraft knows where it is at all times; it knows this because it knows where it isn't.
By subtracting where it is from where it isn't, it obtains a difference, termed deviation.

2. The IN system uses deviation to generate a corrective command to drive the system from
a position where it is, to a position where it isn't, arriving at a position where, if it wasn't,
it is now. Consequently, the position where it wasn't follows the position where it was to
the position where it isn't.

3. Futhermore, the position where it is now is not the position where it wasn't, and the system
has therefore acquired a variation. A variation is difference between where the aircraft is
and where it wasn't, and is considered to be a significant factor which needs correction.


The logic of the system is as follows:

Because variation has modified some of the information obtained by the aircraft, the
aircraft is not really sure where it is; however, it is sure where it isn't (within reason),
and it knows where it was. The IN system now subtracts where it should be from where
it wasn't (or vice versa) and by differentiating this from the algebraic difference between
its deviation and variation, (which is called error), it computes the correct information to
compensate for all factors.

Mohit; hope this makes sense...from an avionics guy just up the road in Granada...

[keith smith]

Denti
I was thinking of the navigation situation in 10-15 years time when augmented versions of perhaps three constellations, plus ADS-B are available for all to use. With verything working (SESAR and NextGen) it will be possible to fill up a TMA to a considerably greater extent than at present. What "graceful degredation" plan will ensure safety, albeit with reduced traffic flow. Limited retention of things like DME and VOR is one possibility,but there are others, and the whole thing will eventually need to modelled with pilots and controllers in the loop.It is something that FAA and Eurocontrol will have to tackle in combination, but there is little sign of that yet.
Keith

[ExSp33db1rd]

Originally published I believe in a BA Staff magazine...

The Flt. Ops. one of course,tho' I confess I can't remember the title at the time.

being involved in early INS trials - the first was bolted to the side of a 707 freighter just inside the main freight door and was about the size of the side of my house - the article was invaluable to me !

I remember firing the thing up in the Central Area of LHR, and when we got to the threshold of 10 R ( now 09 R ) I was totally amazed to see that it had recorded that we had moved 1 nm West and 0.5 nm South.

Shades of things to come.

[hugel]

Inertial Navigation relies on knowledge of where you have been (relative). GPS gives an absolute position.

Input to either can be lost or fail. An integrated system uses the best of both, by optimising the availability and error probable.

re. SESAR/NEXTGen incorporating ADS-B. Very nice when critical mass of infrastructure available; mandatory equipage enforced; and procedures developed to capitalise on it.

hugel

[Denti]

ADS-B over Mode S is mandatory in large parts of europe allready, however i do not know how that will help with infrastructure problems in the future.

I do agree that working fallback plans have to be developed and in fact the regulators worldwide should try to get something working on the road which will be used worldwide and not the current localized thing we see. In europe it seems we are going the way of DME/DME as backup coverage, i do not know how it is handled in other parts of the world though.

[Mohit_C]

Thanks everyone for all the inputs! Another very informative thread.

[keith smith]

Denti
You are quite right to suggest that a SATNAV backup must be agreed world wide. The two most important authorities for doing this are FAA and Eurocontrol. As far as I can see neither of these bodies have seriously considered the matter,let alone discussed it jointly. There are several alternative solutions, each withits own advantages and disadvantages. there may seem to be no great hurry, because relying on SatNav for increased traffic is still some way off. But because there are multiple choices to be made it may take timeto come to a conclusion--better get working at it pretty soon.
You asked how easy is it to jam ILS. I have never heard of any attempt, but maybe someone else does.ILS has very narrow beams(aimed to minimise multipath effects) which also increses signal strength. There is ground monitoring, and it will still be possible to divert on receiving a warning. Much less worrying than SatNAV.
Keith

[AnthonyGA]

By the way, how hard is it to jam a LOC or GP? Really do not know, but would think they send with quite a bit of power.

A key difference between an ILS or VOR and GPS is that the former two are local radio navigation aids, whereas the latter is worldwide. To jam an ILS or VOR, then, you need local jamming transmitters, one for each VOR or ILS. To jam GPS, you just need a single powerful transmitter for a large area. The GPS signal is many times weaker than the ILS or VOR signal, which also helps to make it far easier to jam or spoof. GPS also uses a very limited range of frequencies and produces a signal that can only be distinguished from background noise using special detection techniques, both of which make it more vulnerable to interference.

Jamming the signal removes the ability to use GPS, which is quite a hazard in itself if you are depending on GPS (no ILS or VOR available, which is already the case where they never existed, and might be the case in the future if the authorities continue to recklessly decommission stations). Integrity monitoring may be able to detect a jammed signal, but it cannot replace it if the jamming is well done, so even if it detects the loss of the signal, you still have no GPS.

Spoofing is almost as easy and much more sinister. With spoofed GPS, it seems that everything is working and integrity monitoring may not notice anything strange, but you are no longer where your GPS says you are, and obviously on an approach in particular, this can be deadly. Spoofing GPS itself and WAAS augmentation is straightforward; spoofing LAAS is more difficult because of the local transmitters (as with ILS and VORs).

The cure to spoofing is encryption, and the military has used this for years, but it is not practical to implement it for civilian aviation.

For all of these reasons, having GPS as the sole means of navigation is very unwise. For the same reasons, decommissioning LORAN was a bad idea, as is any decommissioning of VORs or ILS.

[411A]

decommissioning LORAN was a bad idea...

Agree 100%.
LORAN was superb in its accuracy, even the old style LORAN A wasn't bad, except for the occasional lane skip.
My sources tell me that...the USCG, which operated LORAN, is having second thoughts.
I sure hope so.

NB. As I recall, LORAN was a British invention...good show

[tildeleb]

Great discussion.

What worries me about GBAS, is the fact that some teenager hacker seduced or cheated by terrorists or common criminals could somehow send wrong info to the belly GPS antennas. With SBAS, they can't do such thing as the antennas are on top of the fuselage.

As for the GPS signal jamming, I thought that it was difficult, as it is a military designed and the pseudorandom noise is suposed to be difficult to be jammed. I also heard that the soviets (or russians) had jammers for GPS guided missiles. So it is possible, but I don't know if it is easy.

1. Sending false GPS data is possible, but unlikely IMHO. It is true that GPS was designed to be difficult to forge and jam.
Transmitted GPS Signals
GPS explained: Error sources

2. Any signal can be jammed it's a matter of RF transmit power and occupied spectrum. I have seen a modified home cordless phone partially jam a wireless security camera from 50 feet. GPS is vulnerable to jamming because the received signal strength is very low by the time it reaches the receiver. RF is more difficult to jam over a large areas/volumes, easier to jam over smaller areas/volumes.

3. It's not clear that the antennae location makes that much difference. If you were directly underneath the plane it would make some difference. If you were at right angles to the side of the plane, it might may no difference at all.

4. We're headed into a period of heightened solar flares, that is probably my biggest concern with GPS and all satellite and RF based systems.
Solar Storms and You: Human Impacts

[keith smith]

1.I think GPS spoofing is much more difficult than jamming.You need to be able to the track the victim aircraft so that you can radiate it with a false position that convincingly deviates from the true track. That needs an independant track and a GPS simulator ---band of skilled engineers required and then only kill one target. Jamming you can do with a load of thugs.
2.Cancelling e-Loran was a disaster in my view,but it was a presidential directive (There are plenty of other things Obama could have cancelledor not even started)
3.The main jamming threat(by a load of thugs) will come in10-20 years time when the world is basing traffic capacity on functioning satnav and ADS-B, and by that time we must have a "get you out of trouble " back up.
Keith

[xyzzy]

LORAN was American, but it was a bigger, louder development of the British GEE.

[Loose rivets]

I'm astonished that we are investing in another satellite system. I would have thought that pouring every penny into INS research would have been the way forward.

Three systems was the minimum number to throw one into suspect mode, but now we could have multiple sensors and primary read-outs in a thing the size of a small radio. Make this a sensible size to accommodate military spec electronics, and then multiply by a dozen or so, and it could still fit in the skipper's flight bag.

The summing might have to be done on as many autonomous devices as there were sensors, but the only large cost would be the design and initial fabrication. Repeat manufacture would be a minute fraction of hurling very expensive devices into space.

This system would be very difficult to jam, and could only be done locally, and with vast amounts of energy.

I for one am concerned that GPS is almost too good to be true. Certainly, the satellite life-span has passed expectations, but as mentioned above, they are vulnerable to solar radiation. We've been lucky so far, but who knows what's round the solar corner?

GPS signal is pretty small when it's transmitted, but it diminishes - like all electromagnetic radiation - according to the inverse square law. What we receive is mind-bogglingly feint, and consequently vulnerable to simple swamping, with a relatively broad spectrum signal. i.e. Hard to deceive, but easy to kill.

[AnthonyGA]

1.I think GPS spoofing is much more difficult than jamming.You need to be able to the track the victim aircraft so that you can radiate it with a false position that convincingly deviates from the true track. That needs an independant track and a GPS simulator ---band of skilled engineers required and then only kill one target. Jamming you can do with a load of thugs.

Unfortunately, you don't need to track the victim aircraft.

GPS satellites simply broadcast their own positions and the time of day. A GPS receiver receives these broadcasts, which allows it to determine the position of all available GPS satellites in space. It then measures its own distance from each satellite, by comparing the time of day of each broadcast message with the time of day at which it was received by the GPS unit. The unit then determines its position through triangulation. This is a vast simplification, of course, but that's the basic idea.

The GPS satellite signal is extremely weak. In fact, it is actually lost in background noise. GPS units locate satellites by using known PRNs and shifting replicated broadcast signals in time until there is a slight jump in signal strength. It's delicate but, amazingly, it works well. However, the weak signal means that it's very easy to overpower the real-world signals with local, spoofed signals.

So spoofing simply requires overpowering the real signals and substituting fake signals that are correlated in such a way that they place the fake "satellites" in different positions. This will give any listening GPS unit an incorrect position for itself. Thus, spoofing moves all GPS units in space. You don't have to target an individual airplane … but then again, you don't have the option of targeting an individual airplane either—all spoofing affects all units in range, and the range depends mainly on line-of-sight and transmitter power. And remember that the transmitter need not be stationary on the ground.

RAIM cannot detect this type of spoofing. All the signals look good, and the GPS unit has no way of knowing that they've all been spoofed in a coordinated way. Everything fits, and yet the position given by the "satellites" is completely wrong.

The military gets around this by encrypting its precision code. Military GPS units have encryption keys and units that allow them to decrypt the encrypted signal. Civilian units can't use the signal. Since spoofers typically will not have the encryption keys, they cannot spoof the encrypted signal (that's why the military calls it "anti-spoofing," because that was the original goal). Unfortunately, encryption is not an option for civilian use, because every civilian user would have to have the keys, which would mean that they couldn't possibly be kept secret from potential spoofers.

Jamming is way easier. You just blast out noise on the GPS frequencies (easy to do because all satellites use the same frequencies). RAIM will detect this—but even if RAIM detects it, you're still deprived of GPS. Which means that if GPS is your sole or primary means of navigation (LORAN shut off, VORs decommissioned, ILS removed), you have a very big problem.

My worry is that, while paranoid government authorities pour all their time and energy into having people remove their shoes in airports, the bad guys are preparing GPS spoofing equipment. One day, they turn it on, and a hundred aircraft crash at once. I think we are protected to some extent in that it's much more difficult to spoof GPS than it is to put a bomb in a suitcase, but I don't think we should depend on that.

As for WAAS and LAAS, neither of which is part of GPS: WAAS can be spoofed in a way similar to that of GPS, but LAAS is much more difficult to spoof or jam, because it's local, like an ILS. You might be able to mess up one airport (and that would be pretty bad, and ironically LAAS would make it easier), but everyplace else would be unaffected unless you had multiple local spoofing transmitters.

[Denti]

You are right that GPS in general is easy to jam, and somewhat more difficult to spoof. However for LAAS jamming seems the only worthwhile method as for spoofing you need to not only spoof the GPS-correction values but also the geometrical approach info for each channel (instead of a frequency you dial in a 5 digit channel number on your NAV) separately as well as the checksums which are there to provide integrity monitoring, in fact you need a second correctly programmed LAAS with a higher power output than the original one. By the way, one LAAS can of course provide several airports at once with high precision approaches, although i do not know of any actual installation of that kind as even one per airport is much much cheaper than one ILS for each runway direction.

SBAS is easier to spoof as you only need to spoof the correction values and a very low power output from the WAAS or EGNOS satellite.

It will be interesting to see if jamming or spoofing will become a real thread anytime soon like it used to be with frequency jamming or spoofing for ATC frequencies some time ago. At the moment we still have the full backup of real ILS installations parallel to GLS plus other non precision approaches, not to mention that onboard monitoring of positions based on DME/DME and IRU information vs GPS information is done as well.

[keith smith]

I think that you do have to track the spoofing victim aircraft, at least at the point where you start the deception. The spoof signal must be the same as the real signal at the start point. Also spoofing should be detected by comparing satellite signal with INS over relatively short periods

[hugel]

It is obviously a major concern in a similar way to radar or IFF spoofing, but it's not a subject on which I would be giving tips and how-tos on a public forum.

If future surveillance relies primarily on autonomous position reporting such as ADS there had better be some kind of reversionary system, both for the ground surveillance to support traffic management following such a problem and for the aircraft itself...

hugel