Yes, but nobody is saying how it was actually done.
Also, as I said, I would be very surprised they would have been so precisely measuring and logging the carrier frequency of the plane’s transmissions. There is absolutely no reason to be doing that. If they really have that capability, I wonder why they have it. And if they don’t really have it then the story gets more interesting 
This article, out in the last hour or so, has more info.
My reading of it is that they basically did exactly what I suggest i.e. used the doppler shift which would obviously be different if the plane was flying to the north (away from the satellite) versus to the south (towards the satellite).
Had the satellite been positioned over Malaysia they would not have been able to do this, AFAICS.
I am still surprised Immarsat are logging that data. The reason for that capability must be for tracking the target, using multiple satellites – notwithstanding the fact that any actual data session (which needs a paying customer) requires a GPS position from the target. Probably not something they would have wanted to go public with… Exactly as with the routine tracking of GSM phones, and logging of text messages, with the tower-signal-trangulated coordinates logged for each call and each SMS, which was secret for years, till an American magazine wrote about it, maybe 10 years ago.
Satellite communication always has to compensate for relative motion and frequency shifts due to significant doppler usually much greater than this case in the situation of a non-geosynchronous satellite. Naturally in order to compensate you need to track the doppler shift. I don’t think it is about tracking – that is an accidental effect.
Yes – it was the logging (recording) of the data which is curious.
I know “storage is dirt cheap these days” but this stuff has to be downlinked eventually.
I would imagine it’s logging this data so if there are problems, it can aid engineers in debugging the problem. Certainly in IT systems is common to log all sorts of stuff that doesn’t look useful to keep a log of for nominal operations, but when something on the network breaks it suddenly becomes very useful into finding out what/why/when. I would have to imagine satellite communications are similar in that respect.
I think that all satellite providers keep logs in the event of needing to support the customer. I know this is common in shipping, and the system would also need to know the received power levels, S/N levels etc., and when to switch to another satellite (for example).
Yes – I posted that higher up.
But, on further thought, I still can’t see how straight doppler shift could help, because the doppler shift is symmetrical for the north arc and the south arc – because both arcs are symmetrical about the satellite. In both arcs the aircraft would have been flying away from the satellite and at the same speed. The explanation doesn’t add up.
Maybe there would be a small difference in the ground speed between the two arcs, due to the winds. The aircraft would have to be be assumed to be cruising at the same mach # in both cases. But the article doesn’t say anything like that…
The power levels i.e. S/N etc would be the same because the satellite is so far away, relative to the aircraft altitude etc. There might be 2nd order effects due to terrain or atmospheric attenuation, and that would make sense, if you compared the data with other 777s on the two tracks.
However there is probably another factor: one has to assume this was a criminal act of some sort, and you don’t want to tell the whole world you can track airliners with a satellite (whether it was the Immarsat one or some other one) in case there is more of this stuff being cooked up.
It turns out that the satellite actually moves up and down quite a bit, and the doppler shift resulting from that movement can be used to determine if the aircraft was on the north arc or the south arc.
If the satellite was perfectly geostationary, this would not be possible.
It turns out that the satellite actually moves up and down quite a bit
“up” and “down” are a bit ambiguous terms in this context. I think you mean “north” and “south”. In other words, the trajectory is not 100% aligned with the equator, but has a slight angle to it. So it’s longitude and altitude is more or less stable, but the latitude will vary a bit throughout the day.
