Hi, Channel coding will improve performance, depending on the type of code used. For example, in QRadioLink I use a rate 1/2 convolutional code, which leads to twice the symbol rate. This gives me an improvement of 3 dB over no code/half symbol rate mode. LDPC and puncturing would improve it even more with small increase in symbol rate or bandwidth.
Regards, Adrian YO8RZZ On 12/2/17, Nino Carrillo <kk4...@gmail.com> wrote: > Hi Sam, > > I’ll take a stab at your baud question. > > Seems to me that lower symbol rate equates to more energy per symbol, and > therefore higher probability of receiving said symbol at the distant > station. At the same received power level, a system with a lower symbol rate > will be more successful at synchronizing to the received symbol stream in > order to then (later in the receive chain) apply error correction to said > symbols. > > It’s outside the scope of this list, but I’ve been working on a related idea > using the ADF7021 transceiver chip. I’d be happy to chat with you about it, > feel free to email me. > > Nino Carrillo > KK4HEJ > kk4...@gmail.com > > From: Samuel Hunt > Sent: Saturday, December 2, 2017 10:35 AM > To: freetel-codec2@lists.sourceforge.net > Subject: [Freetel-codec2] Ultra-reliable sensitive comms > > I am hoping there will be some people on this list far more experienced > than me to be able to advise on this! > > Am playing with VHF voice, etc, and I am thinking what is the best way > to get ultra-reliable (super-sensitive) performance on a VHF modem. > > I am thinking for VHF, based on a radio where only a normal FM modulator > with a Class C PA is available, so it would have to be something like > 4FSK. Bandwidth isn't a huge problem (say 12.5khz is available which > gives loads of headroom for 1000 baud Codec2). > > Is it better to run 4FSK at a low baud rate (say using one of the low > baud rate versions of the codec), then have just a little error > correction, on the basis that low baud = narrow IF = sensitive, or is it > better to run at a high baud, really heavy on the error correction, such > as 4800 baud and then have loads of error correction such that the > actual throughput is only 1000 baud? > > Basically which would work better - 4800baud air with 1000 baud after > error correction, or 1200 baud air with 1000 baud after error correction? > > This would be assuming 4m band (70mhz) where Rayleigh fading isn't very > predominant. > > > Also, I know it becomes the law of diminishing returns between 4FSK, > 8FSK, 16FSK, etc, and it is generally felt that 4FSK is the most > reliable, but is there any advantage to spacing the tones further apart > except from a small "guard band" would help prevent interference between > the tones. Presumably if the tones start to get really spaced out then > there is absolutely no advantage because you just waste spectrum with > huge guard bands for the 4 filters on receive. > > I know that the work so far on Codec 2 has been based on low baud, low > error correction, but I am wondering if this equals best sensitivity or > if this is more about bandwidth efficiency. > > > I just wondering best way to get really sensitive comms, so a 5W HT > could go for miles and miles. > > Sam > M1FJB > > > > > > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > Freetel-codec2 mailing list > Freetel-codec2@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/freetel-codec2 > > ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Freetel-codec2 mailing list Freetel-codec2@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freetel-codec2
