The implication from other reports, including USAF, is that all timing signals were out by 13 microseconds.
Peter wrote:
Why is that?
Read the article…. "some … would… it’s not clear… still chasing for further details… " and it degenerates even more after that.
What exactly was 13us is not clear to me, and presumably not to the BBC either. The 4km is just an extrapolation from 13us (*300m/us), but whether those 13us were relevant to the navigation task is not clear. It could simply have been an offset between GPS system time and UTC, which would only have affected timing functions, not the navigation function (your landing time would have been 13us off in your logbook, shock horror).
This is the original report by the finnish university.
Here is the NANU (which you’re all supposed to read before flight 
)
The Airforces 50th space wing is promising a more detailed report on what happened, but they claim the core navigation function had been unaffected, and that the error had been in the UTC timing signal. Which very much hints at an error in the UTC parameter block (page 32 of the GPS SPS Signal Spec)
If the navigation function had indeed been affected, it would have been all over the news; however, it seems that the BBC is just pissed their DAB network went haywire.
Dave_Phillips wrote:
that all timing signals were out by 13 microseconds.
The USAF press release does definitely not hint that “all timing signals were out by 13us”. It says that there was a problem with the relationship between GPS time and UTC time, not that there was a problem within the GPS time domain.
