I have to take issue with the comments about the Satcom system. The delays in the system are very well understood and compensated for.
The initial reply transmission will be a deliberately shortened data packet to ensure that the last symbols in this packet do not collide with the next timeslot (it will arrive later by 3.3us/km), this will be during the setup phase. It might well take only one packet to determine this time delay, the next response packet will be correctly timed because the satellite end of the link will have determined the timing delay and told the aircraft terminal what timing advance to use to ensure that the packet arrives in the correct timing window. This process is exactly analogous to how 2G GSM mobile networks operate.
The doppler shift is calculated by the modem DSP software at the point where the demodulation is done, this is a digital correction inside the modem to de-rotate the symbol constellation points to establish the symbol timing. This doppler shift will be dependent on the relative velocity of the two ends of the link, it will be very accurate provided that the signal levels are adequate.
From the data a set of equations can determine the range of position, this of course is limited by the resolution of the timing advance and the possible speed variation of the aircraft.
Of all the information that has come out, the Inmarsat data will be absolutely solid and the least likely to be open to question.
Last edited by Feathers McGraw; 2nd Apr 2014 at 12:52.
Reason: Tidy up spelling and phraseology