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

Re: Saudisat status

Thanks for your through explanation of the difference.
My response, written in haste between flights and passes, was an attempt to
let folks know if they are on the 9k6 digital birds these are the same.  
I was pretty sure I could count on someone to fill in the details :-)


At 08:56 AM 9/28/00 -0400, Robert McGwier wrote:
>Wellllllll,   I guess you could put it that way.
>On a good day, with strong signals, the G3RUH is
>indeed GMSK compatible since GMSK is a filtered
>FSK.   This tells less than half the story.
>MSK, or FFSK, is the version of 2 level FSK (two tones
>encoding zero or one), where the shift is exactly equal
>to one half the baud rate.  So, 9600 bps MSK would have
>a shift of 4800 Hz.    GMSK  means that before the pulse
>is allowed to modulate the FM transmitter, it has been
>filtered by a filter whose shape is  gaussian like and greatly
>reduces the energy that is present in the sidebands for
>this signal type.
>In both cases,  MSK and GMSK, the following signal condition is
>true.   From the center of one baud, to the center of the next
>baud, in a properly equalized signal,  the phase shift is
>exactly +90 degrees or -90 degrees.   This fact can be used
>in two ways.
>(a)  It is  pretty easy to coherently demodulate this signal.
>That means follow the center frequency and the phase from
>baud to baud.      This is a BIG improvement over the way
>G3RUH, K9NG, and most other demodulators do the job
>which is simply to measure the deviation from center frequency
>at the proper baud time.
>(b)   Let me use an example to demonstrate the other gain.
>         Suppose I transmit  000 in succession and 001 in succession.
>        From the beginning of the tribit,  I will have incremented phase
>        -90, -90, -90 for a total of -270.      In the latter I will have
>        incremented phase -90, -90, +90 for a total of -90.    Thus for
>        one bit difference, the place where we wind up on the circle
>        we are wrapping around in the farthest possible point.  With
>        knowledge of the filter applied by the transmitter, receiver,
>        and the channel,  we can make use of this kind of information
>        to recover the most likely path around the circle as we receive
>        the data.    This is a dynamic program and is commonly called
>        a viterbi decoder or equalizer.
>Both of these things taken together will allow a MUCH less capable
>station (hardware wise:  antenna, preamps, etc.) to receive the
>data from a GMSK transmission from a satellite at the same bit
>error rate as the G3RUH demod.
>Stephen Hodgart has been brilliant again, and given us a reasonably
>easy to code version the coherent, maximum likelihood path demodulator.
>I am not certain this will run on the DSP56002 experimenter boards but
>I believe it will.   I coded two G3RUH style modems to run in parallel
>with processing power to spare for the DSP2232 (one for each port).
>And added a fractionally spaced equalizer ON THE RECEIVE SIDE and should
>work for  GMSK until we can get a coherent demodulators (such as Hodgart's)
>BEWARE!!!    The coherent demodulator and the Hodgart demodulator
>require an  SSB receiver with sufficient bandwidth!!   An FM receiver
>will never work.  The only way to use an FM receiver is  the noncoherent
>demodulator (G3RUH, K9NG, or the one Pavel did for the Finnish DSP56000
>board, or the EVM, or lastly the N4HY DSP2232 demods).  Of  these
>the one Pavel did for the Finnish DSP 56000 board or the I did for the
>DSP2232 demods are theoretically the best since they have an equalizer in
>them but that needs testing to prove.  For the well equipped satellite
>the G3RUH will work fine as will the others.
>At 01:11 PM 09/27/2000 Wednesday, Jim White wrote:
>>GMSK is G3RUH compatible.
>>At 11:34 AM 09/27/2000 -0500, you wrote:
>>> > 9k6 FM GMSK.  The data will look much like the other 9k6 digital
>>>OK what is " GMSK " ???
>>Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
>>To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org
>Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
>To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org
Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org