[Date Prev][Date Next][Thread Prev][Thread Next] - [Date Index][Thread Index][Author Index]

Re: Fade remover



Hi Brian and all,

> I was thinking about the (bad) situation with the AO40 telemetry data.
> It sounds like even with a really good antenna, the fading due to
> the spin rate kills any attempt to get good CRCed  data. 
> 
> Has anybody thoght about writing a program that could take two or more
> .wav files, recorded by separate stations during the same pass.  
> It would want to decode both files individually to see which frames
> they contained, then if it could line up both files at the point where
> they start to overlap, it could go through both (or all) bit by bit.
> most bits should agree, but when they don't the file with the higher
> signal strength could be taken (or majority vote if more than two files).
> If the two stations are far enough apart maybe only one will be faded
> at a time.  The effect should be similar to a post-detection-selection
> diversity recever.

I had about the same idea when we standarized the telemetry format.
To be more specific, I thought about an option where soft-decision
I/Q data at the symbol sampling points was aquisited and distributed into
the multicast backbone (although at that time nobody thought that it might
become useful since AO40 was broadcasting happily on 145.xxx).

I think it should be possible, and it should actually provide the same gain
as a normal antenna diversity solution.
We must, however, make sure that every "receiver" provides its own carrier,
symbol and block synchronization. This is not a problem at all, because the
loop bandwidths can be made small enough to survive some long deep fades
over 100 or so symbols without any problem.

I am not sure about the benefit we could expect from this, however. This
is due to the fact that I did not 100% understand how these fades are
generated by the spin. If the antenna is blocked by the thruster nozzle
or something during the fades, wouldn't this affect every ground station
on earth in the same way and at the same time?

Again, it would be nicer if we had some sort of convolutional FEC on the
signal. We could use some soft-output modified viterbi decoder (the BCJR
for example) to consider the reliablility of each symbol from each source
and weighting them apropriately before adding them up.

  -- Jens
----
Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org



AMSAT Top AMSAT Home