p.s. Yes, I'm fishing a bit with this message. I stood around and looked stupid while a local good digital engineer (and RF guy) and another good RF engineer discussed this idea once, and they both expressed interest but no time to work on it right now.
Anyone who knows how to go about this level of DSP engineering to get to the raw bit level from a common FM discriminator (receiver) without resorting to buying the IMBE chipset (and I suppose it goes without saying that it'd have to be someone who also isn't under an NDA because they already work on such things) -- I'd be willing to put you in touch with both of them, put up a website, deal with the general "public" inquiries, or whatever administrivia would keep the world off your back while y'all are working on something that nifty. (I'm guessing I won't find any "takers" -- unless someone feels like doing a lot of hard work for free... but what the heck, it doesn't hurt to ask.) On Dec 28, 2007, at 4:31 AM, Nate Duehr wrote: > > On Dec 28, 2007, at 3:01 AM, [EMAIL PROTECTED] wrote: > >> In a message dated 12/28/2007 4:24:38 A.M. Eastern Standard Time, [EMAIL >> PROTECTED] >> writes: >> The disadvantages are many, but the main one is in that the digital >> signal is converted to analog and then back to digital, and the >> resulting "double-vocoded" audio which has been put through two lossy >> CODEC's will sound really bad. >> >> >> You cant double vocode using Maxtracs, they have no vocoder to >> begin with. This is a transparent repeater >> and will pass IMBE, VSELP, AEGIS and D-Star. Might even be wide >> enough for 12KB Securenet. What it lacks is error correction so >> digital errors present on the input will be passed right along. This >> results in reduced range. This is a good way for Hams to start using >> their P25 radios immediately while shopping for a Quantar. I am >> building one right now. >> >> Chris >> N9LLO > > > Oh, so this is just taking discriminator audio and passing it to > another exciter with no filtering? The drawing labels don't make that > clear, but not sure why you'd need Maxtrac rigs to do this. Virtually > any rig would do it. A lot of the repeaters built for 100% duty- > cycle at high power levels that folks are already doing will do it, if > modified appropriately. > > The disadvantage to this type of setup, is exactly what you mention -- > no bit-regeneration. Garbage in, garbage out -- probably with some > unintended additional bit errors added by audio shaping inside the rig > if its all not completely bypassed. > > Also has the disadvantage of being able to accept ultra-wide signals > and re-transmit them, even if coordinated for a much smaller occupied > bandwidth, if a hard limiter isn't inserted between the receiver and > transmitter in the audio path. > > Are you finding reasonable pricing on P25 radios in your area, Chris? > They're really not that reasonable out here, yet. I expect a lot of > P25 Phase I rigs will "drop out" of Public Safety service if/when > Phase II starts getting widely deployed. (That will be a while > yet...) > > There's been some minor discussion between the RF-heads around here > and some of the bit-jockeys (heh heh... just joking with the > nicknames) about building bit-regeneration devices to put in solid old > repeaters, but the problems seem to lie in detection of the raised- > cosign modulation type. It's not simple, except perhaps for DSP > engineers who are too busy building real products for their day jobs > to dive into writing the code needed to detect the P25 analog waveform > and convert it back to a bitstream. > > Then you have to go the other direction and create that same raised- > cosign waveform in the exciter for transmit. All pretty heavy duty > coding, even for good DSP engineers, unfortunately. > > Seems very "do-able" to create a real "blob" that would drop into > certain repeater-quality RF platforms to do this, but way beyond my > capabilities... and everyone I've talked to so far about it. > > It's similar but harder engineering than say, the old 9600 bps bit- > regenerative repeaters when Packet was popular. The modulation for > that was a quadrature signal (if I remember correctly) and much easier > to detect, even in discreet component electronics. This waveform that > P25 uses appears to be quite a different beast altogether. > > -- > Nate Duehr, WY0X > [EMAIL PROTECTED] > > > > > > > Yahoo! Groups Links > > > > -- Nate Duehr [EMAIL PROTECTED]
