[pskmail] Re: [pskmail] Re: Fwd: RE: [digitalradio] 16QPSK Modulation and Baud

  • From: Vojtěch Bubník <bubnikv@xxxxxxxxx>
  • To: pskmail@xxxxxxxxxxxxx
  • Date: Thu, 21 Sep 2006 22:58:59 +0200 (CEST)

Hi Rein.

> The modems Walt takes about must deliver lots more data per minute, but
> the throughput/bandwidth ratio is a lot worse than optimum.

I do not know any definition of optimum troughput/bandwidth. I know only one of 
the Shannon laws telling something about trading bandwidth for coding gain.

> Pskmail actually employs the most important tools mentioned in this text, 
> being
> automatically adaptive message length,
> and optimum baud rate for the HF channel. 

I like the adaptive message length you described.

> FEC is not wanted. FEC gives the RX a second chance to receive or reconstruct
> the data. Pskmail does this by repeating only the damaged segment, and does 
> not
> suffer the speed penalty of FEC.

I do not agree with that. I believe that adding reasonable amount of FEC is a 
good thing. It is better IMHO to make the frames 1/4 longer to not to have to 
repeat every third frame, because there is a single bit error in that frame, 
which may have been fixed by FEC.

 If the data stream would be compressed and sent as a raw 8 bit stream not 
varicode encoded, you would spare half of the bits IMHO, that may be used for 
FEC to improve SNR.

> There is a lot more to be said (and experimented) about this. Come and join 
> in!!

Thanks for the invitation. There is so many things I would like to try. Jim 
KC0HOS seems to be willing to port pskmail to desktop Windows using PocketDigi. 
I will extend PocketDigi for him with the same interface you added to gMFSK.

> Theory is nice,
> but we are hams so we can actualy try it out!!

I would rather get some insight first before I start coding something. If Jim 
ports pskmail to Windows, I hope to find time to experiment with it. I do 
restrain to maintain Windows and Linux on my PC. My life is too short for that. 
BTW my occupation is programming of custom CAD system on windows. I studied 
computer graphics and computational geometry.

It is much more fun to me to experiment with DSP and coding than to write a 
protocol, because DSP and coding techniques are new to me. I hope to find some 
time to modify the PSK low level coding for pskmail and do some tests with 
ionospheric simulator to convince you that FEC is actually not your enemy, but 
your friend.

73, Vojtech OK1IAK

Other related posts:

  • » [pskmail] Re: [pskmail] Re: Fwd: RE: [digitalradio] 16QPSK Modulation and Baud