[ryandrake]

> There's also a very real chance that commercial aviation may find itself operating in a GPS-denied environment, at least in various edge cases.

Yea, it's kind of terrifying that we are slowly putting all of our eggs in the GPS basket. I love GPS but when lives are at stake, you need a redundant backup navigation system that is robustly deployed and reliably works.

[ammar2]

The VOR Minimum Operation Network[1] in the US is basically supposed to be that. They're decommissioning a lot of the VORs but at least guaranteeing that you'll be 100NM away from a working VOR and an airport with an approach that can be accomplished with VORs for the initial fixes.

Still definitely feels like putting a lot of reliance on GPS but at least there's a backup for the worst case.

[1] https://www.faa.gov/about/office_org/headquarters_offices/at...

[tjohns]

There's also a DME Minimum Operational Network, for airliners that can use DME-DME RNAV. (That's too expensive for smaller aircraft to install, though.)

[cameldrv]

It's too bad that DME/DME RNAV isn't more widely available. The only real reason it's so expensive is that there isn't much demand for it since GPS (usually) works fine. Electronics-wise, it's not much more complicated than a transponder. Unlike a GPS, it does have to transmit, so it will always be somewhat more expensive than GPS.

The other problem is that there's a limit to how many aircraft a DME station can serve at a time (about 100), but I believe that could be greatly expanded if aircraft weren't pinging the DME so often. A position fix every second is generally fine, and it could be even more infrequent if you have a cheap inertial system to fuse with it that can fill in the track for a few seconds between pings.

[rlpb]

One thing I have yet to understand is why DME-DME is preferred over VOR-VOR. Because the latter can support unlimited aircraft, unlike DME.

[chipsa]

Probably the required accuracy. VOR is on the order of a degree for accuracy. DME is around 0.1nm. So if you’re 50nm from the VOR, then you may have a position fix error of 0.87nm across the radial, if I did my math right.

[rlpb]

That's a good point. But if you're 50nm from a DME, I think you're unlikely to be able to get a lock in practice.

[tjohns]

For altitudes above 12,900 ft AGL, the official service volume for a DME is 100-130nm.

Below that it's considered "line of sight"... and some quick math shows that you'd be able to get line of sight >50 nm for all altitudes above 1700 ft AGL (which is very low).

Source: https://www.faa.gov/air_traffic/publications/atpubs/aim_html...

[toomuchtodo]

> you need a redundant backup navigation system that is robustly deployed and reliably works.

Could existing cellular network base stations be this ground based backup navigation system?

https://people.engineering.osu.edu/sites/default/files/2022-...

[tjohns]

No. After about 5000 ft AGL (give or take) you can't pick up cell tower signals at all, since the antennas are directional and pointed towards the ground.

This is a deliberate design decision, because even a low-altitude aircraft would have hundreds of cell towers in sight and would overwhelm the network when handsets tried to register on all of them.

But also: Pilots like being able to have guarantees about system accuracy. We get notices anytime even a single GPS satellite is out of service (even though there are 31 of them), and have software tools in the aircraft to predict if there will be any signal degradation along our route (RAIM). I can't imagine having anything near that level of guarenteed safety with an ad-hoc system like described.

[someguydave]

Sure it could be done but good luck convincing cell phone vendors to adopt liability for aircraft safety

[nradov]

I agree. Starlink could potentially be used as an alternative to GPS (and other similar constellations). But it will probably be a long time, if ever, before it's certified for civil aviation navigation. And it's also vulnerable to jamming.

https://www.technologyreview.com/2022/10/21/1062001/spacex-s...

[mikewarot]

Starlink and GPS always face denial of service from exterior forces, or simply not being able to launch replacements up into orbit in a timely manner.

We should ALWAYS have full coverage from ground based navigation systems, and the pilots should be required to use them on a regular basis.