The Quiet Numbers Station: Decoding Nineteen Years of GPS Cryptography (opens in new tab)

You're assuming it requires specialized military gear, as opposed to consumer gear with a flashed firmware. I believe GPS L1 C/A subframe 4 is on the ordinary L1 C/A civil signal, which means commercial receivers can receive it. They just can't (ordinarily) decrypt it. But a few KB of extra code would change that. A pretty broad set of Android phones can receive this data, without even needing to reflash the GPS firmware: you can decrypt on the application processor, since this field is readable.

I think it's simply because of using a public channel for encrypted communication.

So, a number station sending a message 'detonate the bomb' isn't a number station, because most people don't have the bomb?

Yeah its not a number station at all.

I've worked with the guy credited in the article before, so I'll vouch for his general credibility and the underlying information likely being solid: there's good evidence for this field being some kind of encrypted data stream, probably key distribution, and the behaviour has changed over time. But the breathless LLM-tone really did make it hard to read.

The code is all available and every claim is traceable back to the statistical analysis. Results are reproducible from the original data which is archived on Zenodo. Further analysis would be very welcome. https://github.com/sjmurdoch/gps-special-messages

Paradoxically, in the blog post he speaks with his own voice, describing the evidence better amassed in ... the AI written article.

(This comment was posted when the article was https://lsc-pagepro.mydigitalpublication.com/publication/?i=...; we've since done a merge and changed the URL above)

So much AI. I stopped immediately. He might have something interesting to say, but apparently not important enough for him to write about it himself, so not important enough for me to read it either.

>It’s common knowledge that the military has access to a separate, encrypted, higher-precision GPS signal.

The most militarily-valuable aspect of the military GPS signals is actually the anti-spoofing qualities, rather than the higher precision. Survey-grade GPS gear has been able to achieve centimetre-level precision from the regular civilian signals for several years now, using RF fuckery like tracking the phase angle and other techniques.

To be sure, you want the precision too. NATO countries have M982 Excalibur GPS-guided artillery rounds that are precise enough that you can select not just the building you want to hit but the specific window you want the round to enter.

But the primary benefit of the encrypted signal is that it provides cryptographic assurance that the signal is not spoofed and one can be confident that one's GPS-guided cruise missile or other munition is not being diverted off-course.

Nowadays the military GPS signal has moved from transmitting the legacy "P(Y) code", which is a Cold War-era design, to the "M code" which incorporates several decades' worth of lessons learned in terms of spoofing resistance, cryptographic authentication, etc. It's actually a really neat rabbit hole to climb down.

> has access to a separate, encrypted, higher-precision GPS signal.

That's not it, though. This is available on the consumer L1 band, and you can even read that info using a $5 Ublox receiver (UBX-RXM-SFRBX command).

I don't think this qualifies as clickbait in the sense that the headline mismatches the contents. My experience with 404 Media is that they treat every article like they've just released the Pentagon Papers, so you just have to read with that in mind.

(This comment was originally posted to https://news.ycombinator.com/item?id=48414479, where the article was https://www.404media.co/the-u-s-military-quietly-turned-gps-.... We've since merged the threads.)

The going wisdom seems to be that the EU's Galileo is the most accurate system for civilian use. GPS has undergone frequent systematic update for almost a half century.

Indeed. i have some GPS receiver modules and had wondered about this data, I had assumed it was imprecision in my device or something to do with a satellite moving around. I'll have to plug it in and go back for another look.

(This comment was originally posted to https://news.ycombinator.com/item?id=48414479 - we've since merged the threads)

> What's interesting to me is how out of date US GPS system is compared to China's BeiDou

It’s significantly older though. What would you expect?

TLDW: Russia is jamming GPS and GNSS over Europe, purposefully, using a constellation of military satellites.

Theory is that Russia is constantly practicing to totally disrupt GPS and GNSS (and the Chinese system) across all of Europe.

Related thread https://news.ycombinator.com/item?id=48409664

> GPS was always a dual use system

It wasn't. It was going to be a military-only system, until KAL007 presented the obvious life-saving civilian case.

But yes, the title of this article might as well read "Satellite system developed for military use is being used for a military purpose."

It’s not surprising, but I find it interesting.

The dataset was 136GB (about 7GB per annum), and the Python implementation took 45 hours for each run. The Julia code that processed the whole dataset and built the database took 5 hours, which made iterative development much more pleasant. Of course, later stages in the pipeline had much less data to process and so were much faster. With metadata and indices, that was about 3GB. It's bigger than your estimate since there are multiple observations of the same satellite.

Though I take your point that it’s not big data by the conventional use (i.e. requiring a distributed computing to process). The phrasing in the original article was better: “To make iterative analysis practical, we wrote a Julia pipeline: NetCDF source files are converted to Apache Arrow, then thread-parallel bit extraction is performed into a DuckDB database.”

> May 26, 2011

> No publicly recorded NANU announces a fleet-wide event of this kind in the surrounding window.

I do remember living through this one in February 2011 which was very strange at the time: https://web.archive.org/web/20111015232120/http://navcen.usc...

“SOUTHEAST ATLANTIC COAST: GPS Testing Information THE GPS NAVIGATION SIGNALS MAY BE UNRELIABLE FROM 20 JAN 2011 - 22 FEB 2011 FROM 0000Z - 0245Z DUE TO TESTING ON GPS FREQUENCIES USED IN SHIPBOARD NAVIGATION AND HANDHELD SYSTEMS. GPS SYSTEMS THAT RELY ON GPS, SUCH AS E-911, AIS AND DSC, MAY BE AFFECTED WITHIN A 150 NM RADIUS OF POSITION 30 49.09N 80 28.18W. DURING THIS PERIOD GPS USERS ARE ENCOURAGED TO REPORT ANY GPS SERVICE OUTAGES THAT THEY MAY EXPERIENCE DURING THIS TESTING VIA THE NAVIGATION INFORMATION SERVICE (NIS) BY CALLING (703) 313-5900 OR BY USING THE NAVCEN WEB SITE'S GPS REPORT A PROBLEM WORKSHEET AT WWW.NAVCEN.USCG.GOV.”

I specifically remember it because I was trying to navigate to the Atlanta IKEA but my phone showed me as being, like, south of Macon; ~100mi of error. That timeframe could fit if they were testing something like key availability in a spoofing scenario before enabling real key material transmission.

(This comment was originally posted to https://news.ycombinator.com/item?id=48414479, where the article was https://www.404media.co/the-u-s-military-quietly-turned-gps-.... We've since merged the threads.)

So similar to the constellation discussed in the GNSS jamming article here? > https://news.ycombinator.com/item?id=48409664

Thank you! That first link does seem to be the best one (in terms of easiness to access and information explained). We'll use it above.

We'll merge the threads and put the relevant links in the toptext. Thanks!

That's not what a 0day exploit is. It doesn't allow you to take over arbitrary GPS receivers, for instance.