Mark, When I visualize the MT-63 waveform, it seems like many little tones turning off and on, but all running at a very low baud rate. Thus the low baud rate for the mode even though it is a huge width of spectrum.
All these years we were told that low baud rates worked better under poor conditions and on HF (which often has poor conditions). And that ISI issues would cripple modes much over 100 to 200 baud. Kantronics proprietary G-TOR mode can switch between several baud rates but I bet it is rare that it gets up to the maximum 300 baud rate. I can think of some very amazing things that can be done with these higher speed modes that appear to be nearly magical compared to the science that we were basing everything on. The odd thing is that I am not hearing much from anyone else on this except for the ALE buffs. Can others comment on this too. If it is only the U.S. hams that are limited to 300 baud on HF, why is it that the NZ, and EU, and SA hams have not developed much faster sound card modes? The reason that Pactor 1 had a low baud rate of 100 and a high one of 200, was that their experiments (SCS) proved that 100 was about as high as you could get away with with significant ionospheric disturbances. During pristine times, you could move it up to 200. As Dr. Tom Rink said in 1995 in discussing how they came up with Pactor 2: "... signals with higher baud rates suffer from a significant loss of immunity a- gainst time smearing ... For these reasons, 200 baud is commonly con- sidered to be the maximum useful symbol rate of 2-tone FSK systems, operating over short wave links." also "DQPSK with 100 baud has proven to be a very good compromise between robustness against AWGN and time dispersion, especially if it is combined with powerful error control coding." Pactor 2 can switch between DBPSK (its most robust mode) to DQPSK for more speed and 8DPSK and even 16DPSK under excellent conditions. With Pactor 3, they basically added more tones, 18 of them compared with the 2 carriers in Pactor 2, and uses DBPSK and DQPSK but still running at 100 baud for each of the carriers. Thus the claim that it is 100 baud. This seems to mesh with what Walt was talking about earlier. If you added up the carriers and multiplied it times the baud rate, you would have 1800 baud. I am hoping other developers and experts here on this group will help us understand how the MIL modems can work at the much higher baud rates. As far as joining MARS again, I will pass. I was a NavyMARS member in college when I was around age 18 and my call was N0YUI which is now a reissued ham call. Many years later I joined AFMARS and was AFA3QH, and realized this organization was not for me even though I am an USAF veteran. This was a few decades ago and their structure and mission have changed quite a bit. But that doesn't mean that MARS and amateur radio can not share and collaborate when it is feasible to do so. And I strongly support that. 73, Rick, KV9U Mark Miller wrote: >>If I gave you some parameters of a waveform, what would you use to base >>your measurement of baud rate? >> >> > >I would look at the data, and see how it is modulated into an analog >waveform. For FSK we know that a 1 produces one symbol, and a 0 another >symbol. MFSK16 the symbols represent > >0000 >0001 >0010 >0011 >0100 >0101 >0110 >0111 >1000 >1001 >1010 >1011 >1100 >1101 >1110 >1111 > >4 bits per symbol. For MT63 there are 64 bits per symbol. All 64 PSK >signals combine to produce 1 waveform, just like a two tone, 3, tone or 4 >tone test produce a waveform. The complex voice signal produces a >waveform. PACTOR III uses the same logic...Up to 18 tones are used, spaced >at 120 HZ. I can take a picture of the MT63 waveform and put it on the >Internet if you like. > > > > >>Are you saying that the reason that packet performs so poorly is due the >>fact that it has no convolutional coding or interleaving? >> >> > >Yes, I would say that it is not as well suited for HF operation as other modes. > > > >>All along what Walt and I have pointed out was that ISI becomes >>intolerable with difficult propagation conditions (e.g., doppler, polar >>flutter, etc.) with short symbol lengths. The longest symbol length >>possible for 300 baud is 1000/baud or 1000/300 = 3.33 ms. That is a very >>short pulse for HF. That is why Pactor chose 100 baud = 10 ms minimum >>pulse length (assuming they are continuous with no gaps). That 10 ms >>length is about the right amount, particularly with some DSP enhancements. >> >> > >You can overcome those issues by interleaving, convolutional encoding, >redundancy, and spreading the signal. I would say the real reason why 100 >baud may be the limiting for PACTOR III is not only the RF medium, but the >radios that are using it. Amateur gear I am sure is not designed for low >group delay distortion. > > > > >>If the baud rate of a waveform was 2400 as Steve has often mentioned, >>wouldn't the longest possible symbol length be about 0.42 ms? If this >>really can work on HF, it is completely contrary to what I have learned >>over the past few decades, particularly when Pactor was first on the >>scene. Even with extensive DSP, can you overcome that large of an ISI >>issue? >> >> > >Apparently you can, however we will never know unless we join MARS, or get >the arcane 300 baud limit lifted. > >73, > >Mark N5RFX > > > >Need a Digital mode QSO? Connect to Telnet://cluster.dynalias.org > >Other areas of interest: > >The MixW Reflector : http://groups.yahoo.com/group/themixwgroup/ >DigiPol: http://groups.yahoo.com/group/Digipol (band plan policy discussion) > > >Yahoo! Groups Links > > > > > > > > > > > Need a Digital mode QSO? Connect to Telnet://cluster.dynalias.org Other areas of interest: The MixW Reflector : http://groups.yahoo.com/group/themixwgroup/ DigiPol: http://groups.yahoo.com/group/Digipol (band plan policy discussion) Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/digitalradio/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
