Scott Zimmerman wrote: > Although what Joe says is true, the system is balanced more so than the math > would indicate. This is due to most users ears requiring more than .15uV > sensitivity to pick out a weak signal. realize that the SINADs of BOTH the > repeater and the mobile rig would be additive when the mobile is in a fringe > area. If a user is into the system and is, say 50% noise and the repeater's > transmitter is being received at an S1, say 30% noise, that adds to 80% > noise. That's pretty noisy!!
Yep, yep. Totally understand. Maybe for this type of analysis we should use the receiver's 20 dB quieting point instead of the 12 dB SINAD point. > For a typical user's DSP (the one between their ears) to be able to process > a weak signal into the repeater's receiver, they need to be able to hear the > repeater's transmitter almost full quieting. This is the reason that all > links between receivers strive to be absolutely full-quieting and have good > audio characteristics. If not, all of the system noise and distortions are > additive. Absolutely. More comments below as to why I find this all fascinating. > > Scott > > Scott Zimmerman > Amateur Radio Call N3XCC > 474 Barnett Rd > Boswell, PA 15531 I knew we'd "go there" to the "additive receiver sensitivity" part -- I just wanted to see the numbers. There are days I wish I had time to do this stuff for a living... but that'll never happen. Too busy playing telco at work. On the wireline side of things, VoIP is gettin' pretty darn good! We got "HD Audio" VoIP phones on our desks at work last week, and internal calls sound bloody AMAZING. I think they're using G.722.1 Annex C for a CODEC, but it may be full 64 Kb/s -- makes all the digital radios I'm playing with as a hobby sound shamefully bad, really. Hell it makes the analog rigs sound bad too, 'cause the phones have good quality mics and speakers. Even when going through the PSTN gateway, they sound great. 2-way radio tech seems to be in a "race to the bottom" in audio quality. Just look at cell phones to see who 2-way radio manufacturers are copying. My GSM phone sounds like utter CRAP compared to my old brick analog phone, and it's not getting better. Of course, we hams and repeater geeks all know the driving business reasons... they're having to save on RF on-air bandwidth. Most folks don't know that. Meanwhile the rest of the telecom world is going the other way... "Hey, we can make this phone sound BETTER!". Or more often than not, "Hey we can do full framerate VIDEO... why are we talking on a phone in 2008?" It's all kinda interesting, really -- if we could somehow get "Brain-DSP effectiveness" numbers (like MOS score, or other subjective tests done in digital two-way) for analog, and then apply them... it would be possible to state pretty confidently that a particular system, analog, digital, whatever... doesn't matter... if it's set up a particular way, we would be able to mathematically predict how many users would "enjoy" it, how many would find it "adequate" and how many would "dislike" it. Then it could be engineered from the start to sound "acceptable or better" to some percentage of users. Of course, wireless is going to fall further and further behind wireline... because of video adoption. But voice-only calls will never go away completely... they'll just become the "alternate" way to join a meeting in progress, over the long-haul. Interesting stuff to study, really -- because the voice quality issues learned in even an analog repeater applies to the human ear "interface" questions that are plaguing the digital repeater world today. DVSI's CODEC's truly do sound like crap, but they're what the world's using so far for the major systems. And I'm not knocking DVSI really, I'm knocking that "race to the bottom" bandwidth-wise. VARIABLE bitrates and TRUNKING *mixed together* seem to be the final solution. If only one user is active, they get "beautiful" audio. 100 users active, the quality goes down, but the communication still goes through. That seems long-term to me what will *eventually* happen... but it'll take years and years because infrastructure is expensive to upgrade/replace -- and I'm not seeing manufacturers make DSP code upgradeable in the field in repeater technology yet... maybe I've missed it, since I don't work in that industry. Just think, maybe in ten years we'll fully understand analog to the point where it can be mathematically proven like the old telco engineering levels were when it was "the Bell system", but at the same time, technology will have marched on, away from analog. There really was some good engineering going on at Bell Labs before divestiture, and it all got thrown in the dumpster... choice is nice, but well-engineered "always works" isn't something you get from three large heavily-competing telcos (which is what we'll eventually end up with), either. That only comes in that environment by regulations/laws... something that's not in vogue right now. (It'll come back... the pendulum swings.) Ironic, isn't it? Or maybe even fascinating, really. That it all intertwines as much as it does. Understanding a simple analog FM repeater better, leads to understanding the challenges in digital systems too. We really should be engineering and learning this stuff. I enjoy analyzing it with ham projects, since they're usually not under a deadline, and the owner can do whatever they want. (Like Scott... he wants to try out running 200W on his UHF system, he of course, can... no one says it's not in the budget but him. Ham Radio is amazingly fun that way.) Cool stuff. Wish I had more time to mess with it from a "thinking about it" point of view. Thanks for sharing the numbers, and the user impressions, both! Nate WY0X
