Ranging and Pioneer

"Steve Willner" wrote in message
...
In article ,
George Dishman writes:
The transmitter can be switched off and on but
after switching on it can take a long time to
lock on.

Don't think about switching the transmitter on and off. Instead
command the spacecraft to continuously downlink the contents of some
memory location. Now send a command to change the contents of that
location, and see how long it takes for the new value to be received
in downlink. As Craig wrote, a more sophisticated equivalent of this
is how ranging to more modern spacecraft is done.

Apparently the Pioneer hardware and software don't allow this
procedure to give a useful range. Without knowing spacecraft
details, it's impossible to say exactly why, but I see no reason to
disbelieve the experts. I wouldn't be surprised, though, if signal
strength is the major issue: how many bits/s can Pioneer send back
now? That is certainly one fundamental limitation.

Data is transferred in blocks with forward error
correction. They have no control on the timing at
that level so when a block of e.g 256 bits [1]
arrives containing your memory area and gets decoded
there is an uncertainty in the time at which the
value was sampled. At extreme range I believe data
rate was of the order of 16 bps so a block of 256
bits plus error correction coding would have an
uncertainty of more than 16s and be pretty useless.

The normal ranging technique is done by frequency
(or possibly phase) modulating the uplink carrier.
If the craft is in transponding mode, the same
modulation appears on the downlink and a dedicated
hardware correlator was used to both generate and
lock on to that modulating waveform. Comparing the
time of a reference point in the waveform on the
Tx versus the Rx gives the time delay. The Pioneer
craft had transponders so were capable of supporting
this technique but all the attempts JPL made to use
the it resulted in loss of lock. This was even at
relatively short range just after the Jupiter
encounter.

George

[1] I am not sure of the actual block size they were
using and it may have depended on the range/data rate.