Results 1 to 19 of 19

Thread: MCC7500 and "P25" QCII tones

  1. #1
    Join Date
    Jun 09, 2012
    Location
    do20
    Posts
    70
    Thanks
    69
    Thanked 80 Times in 31 Posts
    Country: Canada

    Default MCC7500 and "P25" QCII tones

    Sorry if this has been asked (maybe even by me). Having a hell of a time getting two-tone paging over a P25 Phase 1 talkgroup.

    Before you start, I know, there are lots of other options like private call/call alert, etc., but I have at least one and maybe two agencies who are insistent on using tones.

    As I have said before I have a 7.14 system with MCC7500 consoles. The system the tones need to go to is an ISSI'd Harris P25 system (though that shouldn't matter, at least I hope it shouldn't). And just to make sure we throw another vendor in the mix, the radios receiving the tones are Kenwood NX5400 portables. (And of course there's an eventual plan to introduce Unication pagers.)

    I've learned fairly quickly that there are constraints on the tones for digital tone paging that there aren't on analog. I gather this is a known subject, but I can't find much literature past the help file for the Kenwood radios, where IIRC the tones are restricted to values between 281.25 and some value in the 3000 Hz range, at strict 31.25 steps. Obviously most of the Moto QCIIB tones don't fall into those slots.

    I have yet to be able to get my MCC7500 to successfully cause a Kenwood radio to alert over a talkgroup. I did try all the different QCII methods available in Provisioning Manager and/or the free form input paging box on the Elite software. I even tried the trick I read about in another thread here, of creating the tone sets in wav files and substituting them for alert tones on the toolbar alert buttons. Those didn't work despite me following all the steps (about the only thing I can think of is my volume may not have been -28dB).

    (Edit: Sorry - to clarify here, I am sending the "digital" tone values, i.e. 718.75/812.50, using the wav files, and have tried to find the closest approximations in the QCII tables, and neither work.)

    I was able to successfully get one Kenwood radio to trigger from another Kenwood radio sending the tone, so I know the system itself can support the tones, and the radios are able to decode over the TG. I'm toying with the idea of recording the Kenwood radio sending the tone, and trying to put that in the MCC as the .wav file again, but in case that doesn't work, I'm at a bit of a loss.

    Is it truly possible (I know, anything's possible) that Moto just doesn't support "digital" tones - only analog QCII - in the MCC7500? You would think there would be at least one or two agencies doing this somewhere in North America. Unfortunately, locally, even the agencies who are successfully 2-tone paging on this P25 system with Moto end-user radios are using non-Moto consoles. That's sadly not an option for me (besides the fact I am nowhere near purchasing authority).

    Is there maybe a module we have to buy/install from Moto to get those tones supported in the MCC7500? Would it be easier to just obtain a standalone encoder (and if so does anyone have suggestions)?


  2. #2
    Join Date
    Jan 28, 2018
    Location
    South East US
    Posts
    4
    Thanks
    0
    Thanked 3 Times in 1 Post
    Country: United States

    Default

    Have you looked in the digital profile section of D1 and worked the tones in that? I know it works with nxdn and have heard dmr, but have no positive experience or outcome.
    I do know kenwood added that with ver 2.50
    And maybe a minor fix with 2.52.

    Nice write up, but need to dig in a little more
    And see if I can duplicate it.

    Sent from my SM-G965U using Tapatalk

  3. #3
    Join Date
    Jun 09, 2012
    Location
    do20
    Posts
    70
    Thanks
    69
    Thanked 80 Times in 31 Posts
    Country: Canada

    Default

    I'm running 2.50 in my radios right now. I will see if I can get 2.52 and give it a try.

    I have successfully sent tones from one Kenwood radio to another, so I'm confident that I have digital two-tone receiving working in the radios themselves. The problem I seem to have is that the MCC7500 does not have the right "tone set" for P25 two-tone paging. Moto QCIIB (or C or D or E) doesn't work, of course, and I tried creating audio files with Audacity, setting them to the right properties (in terms of bitrate, sampling, mono, etc), but those didn't trigger the radios either.

  4. #4
    Join Date
    Dec 21, 2011
    Posts
    4,130
    Thanks
    2,772
    Thanked 4,921 Times in 1,468 Posts
    Country: Canada

    Default

    Respectfully, and I absolutely do not mean this in an inflammatory way, you need to put your radio technician pants on and tell the customer the way it is going to be. Youíre wasting your time dealing with some half-assed implementation that sounds like it will be completely unreliable.

    The problem is the vocoder. It does its best to imitate whatís coming into it, and the reproduce what it thinks should be on the output. Itís not meant to be high-fidelity, with resolution to 1 Hz accuracy.

    Iíll send you a note about this issue so we can discuss further.

  5. The Following 3 Users Say Thank You to Mars For This Useful Post:

    120Comm (Aug 06, 2018),rescue161 (Aug 25, 2018),Viper1-6 (Aug 07, 2018)

  6. #5
    Join Date
    Jan 28, 2018
    Location
    South East US
    Posts
    4
    Thanks
    0
    Thanked 3 Times in 1 Post
    Country: United States

    Default

    Quote Originally Posted by 120Comm View Post
    I'm running 2.50 in my radios right now. I will see if I can get 2.52 and give it a try.

    I have successfully sent tones from one Kenwood radio to another, so I'm confident that I have digital two-tone receiving working in the radios themselves. The problem I seem to have is that the MCC7500 does not have the right "tone set" for P25 two-tone paging. Moto QCIIB (or C or D or E) doesn't work, of course, and I tried creating audio files with Audacity, setting them to the right properties (in terms of bitrate, sampling, mono, etc), but those didn't trigger the radios either.
    Get the white paper on the dvsi ambe2+ I think it is, the tones are sent from the chip it self, not in the vocal stream. That's where I think the 7550 is messed up. I think. I have yet see where the mcc series has the actual codec and dvsi chip and regens the tone in digital format, it sends via vocal on the p25 data stream. 2 year ago that is.

    Nxdn, all tones are issued from the chip, even the two tone and dtmf.

    Sent from my SM-G965U using Tapatalk

  7. The Following 3 Users Say Thank You to kwd-guy For This Useful Post:

    120Comm (Aug 06, 2018),radioinstl (Aug 05, 2018),Viper1-6 (Aug 07, 2018)

  8. #6
    Join Date
    Feb 04, 2012
    Posts
    1,670
    Thanks
    78
    Thanked 366 Times in 188 Posts

    Default

    The AMBE vocoder has certain tones built into it (DTMF). When the vocoder sees what it takes to be a DTMF tone it sends a special character instead of the actual tone pair. This character is seen at the RX end and the proper DTMF audio pair is generated. All other signals (voice and tones) are a mathematical reconstruction of what the vocoder sees on the input.

    For paging type functions you are almost limited to using items like the call alert or status features. I have not specifically looked into QCII and APX as I have no need for ir BUT I am aware of one agency that uses a analog TG on a 'prefered/protected' channel for the type of paging features you are asking about.

  9. The Following User Says Thank You to Notarola For This Useful Post:

    120Comm (Aug 06, 2018)

  10. #7
    Join Date
    Dec 21, 2011
    Posts
    4,130
    Thanks
    2,772
    Thanked 4,921 Times in 1,468 Posts
    Country: Canada

    Default

    Just a reminder guys, the OP is discussing a P25 Phase I system, which uses the IMBE vocoder. It is extremely limited in terms of the reproduction of frequencies, outside of human speech. And even then, the vocoder codebook is limited. It also does not handle distortion (excessive input amplitudes) well at all.

    A QCII code pair of 410.8/539.0 for example, (moto codes 114/119) are not going to come out the same, when fed into the vocoder.

    As we are dealing with Public Safety entities, whatever implementation is conceived, must be 100% reliable. As weíre discussing P25 digital technology here, the only reliable method of paging is the P25 native implementation. That way, there is no loss of fidelity, distortion or tolerance (decoder) issues. With P25 paging, you get native support, forward error correction and reliable operation.

    I am not an expert in the refinements of the AMBE+2 vocoder (used in P25 Phase II), and perhaps there have been some enhancements to the way in which Quik-Call II frequencies are interpreted and reproduced by the AMBE+2 vocoder.

    IMBE is dated and very limited. Is not suitable to do Quik-Call II encoding over this vocoder.

  11. The Following 4 Users Say Thank You to Mars For This Useful Post:

    120Comm (Aug 06, 2018),MaxK98 (Oct 11, 2018),rescue161 (Aug 25, 2018),triptolemus (Aug 06, 2018)

  12. #8
    Join Date
    Dec 21, 2011
    Posts
    4,130
    Thanks
    2,772
    Thanked 4,921 Times in 1,468 Posts
    Country: Canada

    Default



    Demonstration of what I'm stating in this thread. I think the result speaks for itself.

    For the purpose of the test, the XTS3000 had AGC disabled on both analog/digital, and no tone-squelch methods were used on the analog test frequency.

    Non-embedded link: https://youtu.be/eiGpiXAjn44

  13. The Following 8 Users Say Thank You to Mars For This Useful Post:

    120Comm (Aug 06, 2018),Alpha (Aug 07, 2018),C.Cheesman (Sep 09, 2018),MaxK98 (Oct 11, 2018),p47r4ck (Aug 11, 2018),rescue161 (Aug 25, 2018),Viper1-6 (Aug 07, 2018),WB4JGI (Aug 09, 2018)

  14. #9
    Join Date
    Feb 04, 2012
    Posts
    1,670
    Thanks
    78
    Thanked 366 Times in 188 Posts

    Default

    The IMBE vocoder handles tones like DTMF the same way the AMBE+ does. I understand that a 1000hz tone is also generated by the vocoder in the same manner.

  15. #10
    Join Date
    Dec 21, 2011
    Posts
    4,130
    Thanks
    2,772
    Thanked 4,921 Times in 1,468 Posts
    Country: Canada

    Default

    Quote Originally Posted by Notarola View Post
    The IMBE vocoder handles tones like DTMF the same way the AMBE+ does. I understand that a 1000hz tone is also generated by the vocoder in the same manner.
    Respectfully, have you attempted to pass DTMF tones through the IMBE vocoder?

    Even DVSI themselves says DTMF sucks with IMBE:

    P25 uses a specific method of digitized voice (speech coding) called Improved Multi-Band Excitation (IMBETM). The IMBETM voice encoder- decoder (vocoder) listens to a sample of the audio input and only transmits certain characteristics that represent the sound. The receiver uses these basic characteristics to produce a synthetic equivalent of the input sound. IMBETM is heavily optimized for human speech and doesnít do very well in reproducing other types of sounds, including dual-tone multifrequency (DTMF) tones.
    Source: http://www.dvsinc.com/papers/p25_training_guide.pdf

  16. The Following User Says Thank You to Mars For This Useful Post:

    120Comm (Aug 06, 2018)

  17. #11
    Join Date
    Apr 09, 2012
    Location
    Australia
    Posts
    786
    Thanks
    161
    Thanked 375 Times in 160 Posts
    Country: Australia

    Default

    Quote Originally Posted by Notarola View Post
    The IMBE vocoder handles tones like DTMF the same way the AMBE+ does. I understand that a 1000hz tone is also generated by the vocoder in the same manner.
    No it doesnít. A 1000Hz tone passed through IMBE will end up quantised into a broken and distorted 1011Hz tone.

    On a slightly different point, xMBE codecs donít use codebooks, they make a pitch estimation that falls into a set of pre- defined audio frequency bands based on how much energy is detected in each band. The frequency bands are defined as a function of the internal xMBE sampling rate which is 8 kilohertz. This is where the quantisation is occurring.

    Mars analogy of matching paint colours with 16 predefined colours explains it well.

  18. The Following 3 Users Say Thank You to MattSR For This Useful Post:

    120Comm (Aug 06, 2018),hambone (Aug 06, 2018),Mars (Aug 06, 2018)

  19. #12
    Join Date
    Apr 09, 2012
    Location
    Australia
    Posts
    786
    Thanks
    161
    Thanked 375 Times in 160 Posts
    Country: Australia

    Default

    Apologies for the double post mess - Tapatalk barfed and the edit took longer than the timeout period of 10 mins.

  20. #13
    Join Date
    Nov 11, 2011
    Location
    /dev/null
    Posts
    233
    Thanks
    7
    Thanked 145 Times in 49 Posts
    Country: United States

    Default

    Unication pagers handle QC-II over IMBE quite well. They do this by covering a range of 37.25Hz or 62.5Hz. The tones have to be spaced fairly far apart. It works very reliably but it sounds like garbage.

    Mars's example would detect anything between 390.7Hz and 421.8Hz for group 114 and anything between 515.7Hz and 546.8 for group 119.

    Any tone within that range would activate the pager.

    Not necessarily a bad thing, it just takes a little more diligent planning when picking tones.

  21. The Following User Says Thank You to Pezking For This Useful Post:

    120Comm (Aug 06, 2018)

  22. #14
    Join Date
    Dec 21, 2011
    Posts
    4,130
    Thanks
    2,772
    Thanked 4,921 Times in 1,468 Posts
    Country: Canada

    Default MCC7500 and "P25" QCII tones

    Why not just use P25 digital paging? I donít understand. Itís whacker to use QC2 over P25. Just like MODAT people. Beeping sounds for no good reason. And itís unreliable, and thatís not acceptable for a Public Safety application.

    It might ďsort ofĒ work, but it is a complete waste of engineering effort. This is an example of morons dictating to system administrators, what they want, when a solution, more reliable and resilient, already exists. They just donít want to listen or accept facts.

    Itís like plugging a 12 VDC power supply into an AC inverter, to power a 12 VDC device, when you can connect a DC power lead to the batteries, which are powering the inverter.

    Analog encoder ó-> Digital (distortion) ó> Analog (unreliable + WTF)

    Digital ó> Device. Works perfect. FEC.

    Not to mention it doesnít tie up a voice channel with tone crap, and the page is pretty much instant.

    No observable logic here.

  23. The Following 2 Users Say Thank You to Mars For This Useful Post:

    120Comm (Aug 06, 2018),rescue161 (Aug 25, 2018)

  24. #15
    Join Date
    Jul 11, 2012
    Location
    Stinkadeener
    Posts
    817
    Thanks
    526
    Thanked 966 Times in 432 Posts
    Country: United States

    Default

    I have only done what you're looking to do with APX subscribers... and with them it works flawlessly. Phase I or II, encrypted or clear, it doesn't care as long as the tones originate from a wireline console or another APX subscriber. I can't speak to what goes on at the protocol layer but I can confirm 100% that the tones are reproduced locally in the subscriber, and match the tone frequency/capcode as entered into Provisioning Manager or CPS if coming from another subscriber. The reproduced tone fidelity is sufficiently clean to drive external decoders.

    When you start getting into third-party subscribers all bets are off and, as demonstrated above, legacy ASTRO and ASTRO25 subscribers puke trying to reproduce the tones. The only non-APX radio I've ever seen it work on is a Unication G5.
    "The Girl Scouts found several hungry REACT members at the finish line..."

  25. The Following 4 Users Say Thank You to motorola_otaku For This Useful Post:

    120Comm (Aug 06, 2018),Alpha (Aug 07, 2018),Mars (Aug 06, 2018),p47r4ck (Aug 11, 2018)

  26. #16
    Join Date
    Nov 11, 2011
    Location
    /dev/null
    Posts
    233
    Thanks
    7
    Thanked 145 Times in 49 Posts
    Country: United States

    Default

    Quote Originally Posted by Mars View Post
    Why not just use P25 digital paging?
    This uses resources as well. It doesn't use voice channels, but it requires two radio ids and the console expects an acknowledgement.

    I know QC-II sounds welfare over digital, but Unication found a way to make it work reliably.

    The only reason agencies are starting to go this route is to stop supporting two radio systems, Motorola never developed a P25 pager, and issuing every KD a several thousand dollar take-home daily carry APX 4000 is problematic for many reasons. A G4 makes sense over a Minitor VI.

  27. The Following User Says Thank You to Pezking For This Useful Post:

    120Comm (Aug 07, 2018)

  28. #17
    Join Date
    Jan 03, 2015
    Location
    Somewhere in the Lone Star Republic
    Posts
    242
    Thanks
    46
    Thanked 96 Times in 56 Posts
    Country: United States

    Default

    Quote Originally Posted by Pezking View Post
    This uses resources as well. It doesn't use voice channels, but it requires two radio ids and the console expects an acknowledgement.

    I know QC-II sounds welfare over digital, but Unication found a way to make it work reliably.

    The only reason agencies are starting to go this route is to stop supporting two radio systems, Motorola never developed a P25 pager, and issuing every KD a several thousand dollar take-home daily carry APX 4000 is problematic for many reasons. A G4 makes sense over a Minitor VI.
    If trunking, pages and status updates are control channel data. The major downside to P25's Page/Call Alert function is that in all the testing I've been able to do, it is only capable of sending pages on an individual level and not a group level. Thus, the major drawback is you are paging sequentially to affiliated units. So paging multiple users may take some time. Now, paging a specific station for station alert and/or using the page function to open automatic gatesÖit's quite practical if you can justify the subscriber cost.

    The way Unication initially setup P25 paging was essentially reserving a talk group for paging. When Unication approached some of the departments near me with their statement it was P25 paging capable, many of the departments were depressed to find that they would have to contact the system admins (it's a regional system) to have the paging talk group created and put forth in a manner it would page at all of the sites in the system that needed the page. When we actually tested functionality of the Unication G4/G5 pagers, we found it was great for simulcast systems but had many issues related to the design of the unit that really hobbled it in wide-area systems which after so much work from Unication to get an audience with the system adminsÖnot a single pager was sold on the system as agencies discovered it was more practical and cost effective to continue to utilize their existing two-tone systems or utilize the tone paging functionality built into APX SU's as Astro25 SU's where phased out.

  29. #18
    Join Date
    Feb 14, 2015
    Location
    Silicon Valley, CA
    Posts
    2
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Country: United States

    Default

    Okay, a bit late to the conversation. Some background, we're in the process of setting up a Telex C-Soft console on a DMR control station which I understand is also AMBE.
    We are trying to get the familiar sound of the Motorola Console alerts for consistency. So far the 1000hz 1/3 as well as the 700hz "Code 22 Marker" work just fine. The trouble is with the 1500hz to 800hz typical "Alert 2" hi-lo tones, as they sound like junk when transmitted in p25 or dmr. Does anyone have the exact Hz that the AMBE vocoder passes after the quantisation occurs? I've heard crystal clear alerts on P25 as well as DMR using the same (relative) hz. Apologies if this needed its own thread.

  30. #19
    Join Date
    Jul 11, 2012
    Location
    Stinkadeener
    Posts
    817
    Thanks
    526
    Thanked 966 Times in 432 Posts
    Country: United States

    Default

    You're not going to get clean tone fidelity through TRBO subscribers at all. Their implementation of the vocoder doesn't "know" what to do with single-tone audio and tries to pass it as voice, which results in distortion on the receiving end. Currently, the only product from Motorola that can handle single tone audio properly is an APX.
    "The Girl Scouts found several hungry REACT members at the finish line..."