P25 Audio

[JRayfield]

I am enjoying this discussion. I enjoy good technical discussions that make me think.

There are no 'repeats' of data sent with DMR (at least not when sending voice packet). Forward error correction is used. With DMR, up to 10% packet loss can occur and the protocol will 'fill in the gaps'.

I completely agree that it would be very interesting to see test results showing how DMR compares to narrowband (12.5 khz) analog. I'll have to see what I can come up with. But, as I've mentioned before, in the 'real world', numerous on-the-air systems have proven to work much better, in terms of coverage, than the analog systems that they replaced (and that was not because of having more digital repeaters as compared to the analog systems).

John Rayfield, Jr. CETma

 

[Baylink]

I am enjoying this discussion. I enjoy good technical discussions that make me think.

There are no 'repeats' of data sent with DMR (at least not when sending voice packet). Forward error correction is used. With DMR, up to 10% packet loss can occur and the protocol will 'fill in the gaps'.

Well, that's "redundancy", whether you want to call it "repeats" or not. What, precisely, is the FEC fraction for the DMR air interface, do you know? The MPEG2 sat uplink where I work is 5/6.

I completely agree that it would be very interesting to see test results showing how DMR compares to narrowband (12.5 khz) analog. I'll have to see what I can come up with. But, as I've mentioned before, in the 'real world', numerous on-the-air systems have proven to work much better, in terms of coverage, than the analog systems that they replaced (and that was not because of having more digital repeaters as compared to the analog systems).

Well, ok, but once again: on what grounds do you assert that the issue wasn't the actual radios?

It is almost certainly the case that if you're doing before/after comparisons of a given system, that you're comparing 15 to 30 year old 'old' gear and installations, with Just Installed 'new' gear, which is in the best condition it will ever be in -- and the handhelds are also going to be 3 to 5 generations newer.

So while, in fact, a given 'after' may be much more tractable on the ground than the 'before' was, there are so many variables involved in the system design that you cannot reasonably point to that evaluation, and use it to say "the modulation on the 'new' system in inherently better than that of the 'old'," which is what it sounds to *me* like you're doing.

It may well be the case that DMR, or P25, or, hell, even OpenSky is in fact more communications-effective for a given received signal strength than analog 12.5...

but to use that as a selling point, you *really* need to do bench tests with microwatt meters, attenuators, defined characteristics...

and most importantly, analog gear of the identical vintage (and preferably model) to the digital gear.

Of *course* MotoTRBO is going to blow MastrPro gear out of the water.

 

[JRayfield]

"Repeats" is most definitely very different from Forward Error Correction. What you're calling 'repeats" would be more akin to an Ack/Nack protocol, which is not suitable for voice transmission. Forward Error Correct 'replaces' missing data packets - there is no 'repeating' of data involved.

You make a lot of major assumptions, such as assuming that all DMR systems that have replaced "15 to 30 year old 'old' gear and installations". How do you know that? I'm sure you don't. Likely, there have been some installations that did replace old equipment that was not operating at 'top efficiency', so to speak, but there have also been tests done using mixed-mode MOTOTRBO repeaters and MOTOTRBO radios, where DMR outperformed analog FM.

John Rayfield, Jr. CET-ma

 

[Baylink]

"Repeats" is most definitely very different from Forward Error Correction. What you're calling 'repeats" would be more akin to an Ack/Nack protocol, which is not suitable for voice transmission. Forward Error Correct 'replaces' missing data packets - there is no 'repeating' of data involved.

Just so we're clear; I never said the word "repeats", anywhere. At least not about the digital side. I *did* use it to describe the analog equivalent to FEC, cause that's how one *provides* forward error correction in marginal analog links.

I have carefully used the technical term, 'redundancy', everywhere I've made implications about FEC.

Here's what Wikipedia has to say about FEC, which checks with my understanding; they don't say "repeats", either:

https://secure.wikimedia.org/wikipedia/en/wiki/Forward_error_correction

You make a lot of major assumptions, such as assuming that all DMR systems that have replaced "15 to 30 year old 'old' gear and installations". How do you know that? I'm sure you don't. Likely, there have been some installations that did replace old equipment that was not operating at 'top efficiency', so to speak, but there have also been tests done using mixed-mode MOTOTRBO repeaters and MOTOTRBO radios, where DMR outperformed analog FM.

I didn't make any of those assumptions. I merely pointed out that the assumptions existed, and that in order to come to a conclusion about whether a given set of facts made DMR look better than analog, on an apples to apples basis, they were questions one had to ask.

Your phrasing of some of your assertions implied that *you* were making those assumptions; if not, my apologies, but you may wish to be clearer in your statements.

Now, on the last point, yes: using exactly the same towers, repeaters and rigs (oh, my!) are the test conditions I was suggesting. At that point, we ask the next question: what were your parameters for defining "works better"?

Were the tests against 12.5k (I assume a mixed-mode MT radio won't do 25k) or 6.25? What was the intelligibility limit set at for analog?

And, keeping in mind the use cases we're talking about, remember this:

The real question is: "is it at all possible to get the message out analog, no matter what you have to do (repeating words, spelling phonetically, sending morse -), etc, at a received signal strength lower than that at which the digital modulation falls off the cliff".

If I'm hanging from a tree off the edge of a cliff, I'm really not going to mind it i I have to tell the dispatcher where 5 or 6 times...

(And remember, I don't have a dog in this fight; I don't even use PS LMR, much less work in the field or sell it. I just believe in intellectual honesty and science over (let's say it softly) marketing. And I'm not calling you a liar, either, but I don't in fact know who buys your groceries; if it's a Motorola dealer, then you have a higher hill to climb, here.)

 

[JRayfield]

"It is almost certainly the case that if you're doing before/after comparisons of a given system, that you're comparing 15 to 30 year old 'old' gear and installations, with Just Installed 'new' gear, which is in the best condition it will ever be in -- and the handhelds are also going to be 3 to 5 generations newer."

That sure sounds like an assumption to me - you didn't 'ask' about this, you stated "It is almost certainly the case". You sure have an odd definition of "assumption".

"I assume a mixed-mode MT radio won't do 25k" - and yet another assumption...not a question, an assumption. Yes, MOTOTRBO radios will operate in the wideband FM mode (25 khz).

The comparisons that I've seen were done between wideband FM and MOTOTRBO and narrowband FM and MOTOTRBO. MOTOTRBO is going to have a definite 'edge' over narrowband FM in coverage performance, whereas it will probably be a lesser 'edge', or may be closer to 'equal', between wideband FM and MOTOTRBO. In general, there will be around a 3db difference (equal to doubling, or cutting in half, the transmitter power) between narrowband FM and wideband FM.

I've had this kind of discussion with one of the country's leading radio signal propagation engineers. His comments to me indicated that he expected to see a significant improvement of coverage area with MOTOTRBO as compared to analog FM (especially narrowband).

Keep in mind, that I have worked with electronic communications equipment and systems for over 32 years (from AM broadcast up to 900 mhz), not from the sales standpoint, but from the technical standpoint. I never approach anything from the standpoint that a salesperson does. In fact, I made the comment to one of my customers (who had just agreed to purchase a MOTOTRBO system from us), that "I'm not a salesman". His response was "I can tell". To me, that was one of the best compliments that I've ever received.

John Rayfield, Jr. - CETma

 

[Baylink]

"It is almost certainly the case that if you're doing before/after comparisons of a given system, that you're comparing 15 to 30 year old 'old' gear and installations, with Just Installed 'new' gear, which is in the best condition it will ever be in -- and the handhelds are also going to be 3 to 5 generations newer."

That sure sounds like an assumption to me - you didn't 'ask' about this, you stated "It is almost certainly the case". You sure have an odd definition of "assumption".

That wasn't an assumption. That was a response. To this, your earlier comment:

> Simply put, MOTOTRBO will perform at least equal to, and in many cases better than, analog FM. Analog is NOT 'better' than MOTOTRBO in performance. This has been proven time and again with on-the-air systems.

Particularly that last part. There's an unspoken, but nearly required assumption inherent in that, that the "on-the-air" analog systems are old; I'm fairly certain no one's installing big new trunked analog systems, and in order for you to be using old v new as a comparison, anyway, it effectively must be an old analog and new digital system, or the assertion doesn't support your argument.

"I assume a mixed-mode MT radio won't do 25k" - and yet another assumption...not a question, an assumption. Yes, MOTOTRBO radios will operate in the wideband FM mode (25 khz).

Really? Why did they... no, never mind; I can see a good reason for that, I suppose; systems they want to sell upgrades to that are still on 25k.

The comparisons that I've seen were done between wideband FM and MOTOTRBO and narrowband FM and MOTOTRBO. MOTOTRBO is going to have a definite 'edge' over narrowband FM in coverage performance, whereas it will probably be a lesser 'edge', or may be closer to 'equal', between wideband FM and MOTOTRBO. In general, there will be around a 3db difference (equal to doubling, or cutting in half, the transmitter power) between narrowband FM and wideband FM.

I've had this kind of discussion with one of the country's leading radio signal propagation engineers. His comments to me indicated that he expected to see a significant improvement of coverage area with MOTOTRBO as compared to analog FM (especially narrowband).

Perhaps. But what was the topic being discussed? Business LMR?

Cause that's not the conversation I'm having here. I'm having the "public safety; hanging from a tree" conversation. I have no trouble believing that digital modulations will be equal in performance to analog, especially wideband analog, in an LMR environment.

Keep in mind, that I have worked with electronic communications equipment and systems for over 32 years (from AM broadcast up to 900 mhz), not from the sales standpoint, but from the technical standpoint. I never approach anything from the standpoint that a salesperson does. In fact, I made the comment to one of my customers (who had just agreed to purchase a MOTOTRBO system from us), that "I'm not a salesman". His response was "I can tell". To me, that was one of the best compliments that I've ever received.

I would think so as well.

But I think these last 2 exchanges illustrate to everyone upthread why the thread went pear-shaped in the first place: there are *lots* of meters to read and terms to define properly to even have this conversation.

 

[JRayfield]

My discussions with the engineer had nothing to do with the 'type' of radio service using digital or analog communications. It was purely an engineering discussion, as to how the coverage area of MOTOTRBO and analog FM systems compare. The conclusion of our conversation was that MOTOTRBO will provide a usuable signal in areas that are unusable with analog FM (wideband or narrowband), when all other things are equal (antenna, antenna height, transmit power, receiver sensitivity, etc.).

"Cause that's not the conversation I'm having here. I'm having the "public safety; hanging from a tree" conversation. I have no trouble believing that digital modulations will be equal in performance to analog, especially wideband analog, in an LMR environment."

Now you've got me confused. You seem to be differentiating between "public safety" use of digital (specifically, MOTOTRBO, in this conversation) and "LMR" (Land Mobile Radio).

John Rayfield, Jr. - CETma

 

[Baylink]

My discussions with the engineer had nothing to do with the 'type' of radio service using digital or analog communications. It was purely an engineering discussion, as to how the coverage area of MOTOTRBO and analog FM systems compare. The conclusion of our conversation was that MOTOTRBO will provide a usuable signal in areas that are unusable with analog FM (wideband or narrowband), when all other things are equal (antenna, antenna height, transmit power, receiver sensitivity, etc.).

Yes, I've got that. The question is, "what's 'usable'?"

"Cause that's not the conversation I'm having here. I'm having the "public safety; hanging from a tree" conversation. I have no trouble believing that digital modulations will be equal in performance to analog, especially wideband analog, in an LMR environment."

Now you've got me confused. You seem to be differentiating between "public safety" use of digital (specifically, MOTOTRBO, in this conversation) and "LMR" (Land Mobile Radio).

You bet I am: public safety communications have *substantially* more stringent requirements than business land mobile does, at least in the world *I* live in. That "in extremis, in danger" situation, which is the driver for *caring* about the edges of the footprint, isn't pertinent to business LMR systems, so their performance isn't a factor in that conversation. Well, the conversation *I'm* having, which is "are digital LMR systems, specifically DMR, suitable for public safety communications?"

That seemed to be the argument everyone else was having, as I understood it, when *I* jumped in.

 

[JRayfield]

"Usable", as in being able to understand what is being said in a transmission, versus not being able to understand what is being said in that transmission.

LMR stands for Land Mobile Radio. That has nothing to do with the 'type' of use of Land Mobile Radio. LMR covers public safety use as well as business use.

John Rayfield, Jr. - CETma

 

[Baylink]

Yup, we're in violent agreement there: Land Mobile Radio equipment can be used in many different services.

That returns us to the debate which started this conversation: do different services have different definitions of "usable" (hint: yes , and what *are* those definitions?

And, by extension, does DMR's definition of "usable", for a given system build, meet or exceed *public safety's* definition of "usable". That was, as far as I can tell, the original point in dispute in this particular thread.

 

[Farscan]

I thought the initial post was he liked analog audio better than digital. With a nice external speaker digital can sound pretty good. The digital is trying to eliminate background noise, and improve readability which can lose a little of the quality. A true p25 radio sounds quite a bit different than how my 396xt makes it sound. My preference on audio is actually AM. While flying two people can double over each other and we
can hear both just fine at the same time. It also sounds quite good in the D clark headsets.

 

[Gezelle007]

I do prefer analog audio over digital...... But, I have received criticism because of the fact that Ive only heard digital audio from scanners. Not an actual RX/TX P25 radio. And I admit, it would be helpful to know how the real thing actually sounds up close in person before judging... Its just that I have heard so many samples of digital that seem garbled and electronic sounding... no fidelity as someone put it.

 

[jackj]

My son-in-law is an assistant fire chief with a department that has gone digital in the last couple of months. He carries a Motorola portable and it is on when he is at home. We were visiting New Year's Day and I heard it for the first time where I could pay close attention to how it sounds. To be honest, it isn't much, if any, better than my 996T. A few transmissions sounded like the guy had marbles in his mouth. Background noise would garble the mobile transmission. The system is P25 and the radios are brand new, not units from e-bay.

I'm sure that some of you folks are going to say that my old ears are analog only and not compatible with digital audio or that I'm just uneducated and, once I've been properly educated, I'll like P25 just like they do. But the truth is that it just doesn't sound good and is at time hard to understand. All this while in strong signal areas.

 

[hitechRadio]

I would agree that analog does sound better than digital. If the the analog is full quieting, if it is not full quieting, digital sounds much better, as far as audio.

 

[hitechRadio]

I started a thread in digital voice formats forum called.(Post your Digital Com Info Websites) If you guys have some good sites for digital information please post there.

 

[33kracing]

i think p25 audio sucks myself. I'd much rather listen to analog. P25 is very monotone, has no fidelity which is why it seems like everyone sounds the same.

x2.....

 

[Boatanchor]

OK, here's my 2c..
I use various Motorola, Tait and 'other' current model P25 radios. Some of these radios are worth $2500+. I also monitor various P25 networks with Uniden and GRE digital scanners.

IMHO, none of the expensive P25 radios sound as good as a $200 analogue FM radio when signals are > - 110dBm. This is not a fault of the radio, it is simply a fact that digital vocoders lose audio quality during the compression process. It is inherent with all digital radios and vocoders, not just IMBE P25.

P25 may have better low frequency audio response on paper, due to the 300Hz high pass filtering that is used in many analogue FM systems, but aside from that, I would have to agree that digital vocoders produce some pretty undesirable side effects.

Pro's of digital:
No white noise/hiss as signals get weak.
High level encryption.

Cons of IMBE/P25:
Obsolete and spectrally inefficient technology (compared to DMR and Phase 2)
No white noise as signals get weak (lack of user feedback when entering weak signal areas).
Reduced audio dynamic range.
Awful handling of background / ambient noise.
Vocoder artifacts and voice distortion.
Poor (non-existent) handling of audio warning/signaling tones.
Much more expensive fixed infrastructure requiring high quality, IP links with redundancy.
Difficult to roll out in remote / rural locations (due to above).

The bottom line is that Phase 1 P25 is no more spectrally efficient than NBFM in conventional or trunking formats and reduces the nuances of most human voices.

Phase 1 is the Betamax of the digital radio world and has obsolescence built into it. This will of course guarantee a bright and prosperous future for the Motorola's and the DVSI's of the world..
Shame about the poor taxpayers wallets though..