Results 1 to 17 of 17

Thread: NXDN 12.5 vs. 6.25 Coverage

  1. #1
    Join Date
    May 13, 2012
    Posts
    623
    Thanks
    1,213
    Thanked 389 Times in 214 Posts

    Default NXDN 12.5 vs. 6.25 Coverage

    In the real world, which has better coverage and overall audio quality, 12.5 or 6.25 FDMA?

    Kenwood compares 6.25 FDMA to 12.5 TDMA, but no comparisons to 12.5 FDMA Digital.

    I find it interesting a local system is Licensed 12.5 on an NXDN system, but use 6.25 instead. (this is a type C, VHF Multisite system)

    It seems to me 12.5 would have been a better choice. But I am no NXDN expert.
    Radio Referenced...Those who think they know it all are very annoying to those of use who do.


  2. #2
    Alpha's Avatar
    Alpha is offline T S - Moderator
    CS Forums $upporter
    Join Date
    Feb 12, 2012
    Location
    Directly above the center of the earth.
    Posts
    2,780
    Thanks
    1,239
    Thanked 1,533 Times in 739 Posts
    Country: Christmas Island

    Default

    Generally anything TDMA vs. FDMA is going to be 1/2 the bit rate, so would sound correspondingly worse. A 12.5 FDMA is going to sound much better than 12.5 TDMA, which would allow 2 channels of 6.25, essentially.

    As far as coverage goes, from what I remember from some tests I did some years ago, the narrower modes are a little worse than wideband, but not very much. Perhaps someone with some more recent real-world experience can chime in, that was before NXDN featured any TDMA modes. I believe the loss of coverage was natural, mostly due to receiver physics and occupied bandwidth, IIRC.

  3. The Following 2 Users Say Thank You to Alpha For This Useful Post:

    Mars (Jan 11, 2020),moetorola (Nov 26, 2019)

  4. #3
    Join Date
    Feb 04, 2012
    Posts
    2,077
    Thanks
    281
    Thanked 1,009 Times in 448 Posts

    Default

    Alpha is correct. I have done some studies on the various modes/coverage. In many cases its hard to tell what is what because of the built in error correction for the various modes. The results I got basically distill down to this.

    FDMA vs TDMA = better coverage by 5-15 % this is because of the constant carrier vs the cycling carrier.
    25-12.5-6.25 modulation = the wider the better based on baudrate, there is nothing new here basically the higher the baudrate the better the audio. Just remember comparing oranges (IMBE) to apples (AMBE) introduces variables based on the vocoder.

    Another factor that we found affects the narrower bandwidth signals is the sensitivity to noise. Basically a 10db noise spike has a larger impact on the narrow signal than on the wide signal. This is strait proportional impact.

    Finally this was NOT a in depth study where all variables were logged and antenna types were compared. We simply took a APX 8500 and tried analog wide and narrow, P25 and TDMA all on VHF then switched radios to a TRBO XPR2500 radio did the same tests then switched to a Kenwood NX800 and tried NXDN. All testing was done on the same day with the same vehicle and antenna and location. My overall feelings on the results.

    RF
    Analog wide or narrow = good with wide a bit better especially with backround noise (stereo radio)
    P25 = exellent (the best)
    TRBO and NXDN = 'gooder' not as good as P25 but generally better than analog

    Audio
    P25 = best
    P25 TDMA - TRBO = good
    analog = good sometimes better than trbo sometimes not as good (static and flutter on moving /weak signals)
    NXDN = good but could tell the audio was having highs and lows filtered by the vocoder.
    NXDN = not quite as good

  5. The Following 6 Users Say Thank You to Notarola For This Useful Post:

    Alpha (Nov 27, 2019),moetorola (Nov 26, 2019),s2scotty (Jan 11, 2020),SPECIAL_EYE (Jan 14, 2020),SwissMoto (Jan 13, 2020),wa6jbd (Jan 10, 2020)

  6. #4
    Join Date
    Jun 21, 2019
    Posts
    1
    Thanks
    0
    Thanked 3 Times in 1 Post
    Country: Canada

    Default

    I did some rather unscientific tests about a decade ago... I had an NX700 radio as a base unit scanning the same frequency one as analog narrow-band, one as 12.5khz digital, and another as 6.25khz digital. I had a mobile set up the same way, I went for a drive and stopped periodically and did a test on each mode (I also had the base side connected to a recording device). In the end, I found out that the 6.25 digital worked better than the others at the fringe... maybe that's why iCom NXDN is only 6.25Khz capable?

  7. The Following 3 Users Say Thank You to TheWizard For This Useful Post:

    Alpha (Nov 27, 2019),moetorola (Nov 26, 2019),Viper1-6 (Jan 10, 2020)

  8. #5
    Join Date
    May 13, 2012
    Posts
    623
    Thanks
    1,213
    Thanked 389 Times in 214 Posts

    Default

    My bad, I was not really clear in my OP, I was really looking for comparison of NXDN FDMA 12.5 vs. 6.25.
    Or,Maybe I am missing something? Is NXDN Type C not capable of 12.5 FDMA?

    This department seems to have coverage issues and audio complaints . I was thinking of suggesting to them to test 12.5 FDMA (?9600?) instead of 6.25 FDMA (4800).

    Although I think most of there issues are system setup and also VHF interference.

    Not to get to far off topic and starting a new post, since I have two sharp guys in here,,,, is there a way to remove a channel from service on illegal carrier on a type C NXDN system? I can not find it, if there is.

    Disclaimer: I am used to P25 systems, so not awful familiar with NXDN, although through some research it seems to be awful familiar to SZ, just different names for features or settings.
    Radio Referenced...Those who think they know it all are very annoying to those of use who do.

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

    Alpha (Nov 27, 2019)

  10. #6
    Join Date
    Jul 11, 2012
    Posts
    1,134
    Thanks
    1,078
    Thanked 1,617 Times in 689 Posts
    Country: United States

    Default

    In my firsthand experience with conventional Very Narrow NXDN and Kenwood NX-5000 series radios both range and audio quality are fantastic. I'd even go as far as saying that NXDN on a NX-5000 series radio sounds better than DMR or P25. I can't speak to the older generation stuff or ICOM IDAS radios, but the NX-5000s sound awesome.

  11. The Following 5 Users Say Thank You to motorola_otaku For This Useful Post:

    Alpha (Nov 27, 2019),Bill_G (Nov 27, 2019),moetorola (Nov 28, 2019),PSEhub (Dec 01, 2019),slapshot0017 (Dec 01, 2019)

  12. #7
    Join Date
    Jan 05, 2018
    Location
    WTVLCA01DS0
    Posts
    20
    Thanks
    39
    Thanked 38 Times in 14 Posts
    Country: United States

    Default

    Quote Originally Posted by moetorola View Post
    My bad, I was not really clear in my OP, I was really looking for comparison of NXDN FDMA 12.5 vs. 6.25.
    Or,Maybe I am missing something? Is NXDN Type C not capable of 12.5 FDMA?
    It is. I'm running a NexEdge (NXDN Type C trunking) system in 12.5KHz mode. Originally set it up that way to give us more data capacity, but the data usage never materialized. Too much work to go back and change 450+ radios to 6.25.

    Quote Originally Posted by moetorola View Post
    This department seems to have coverage issues and audio complaints . I was thinking of suggesting to them to test 12.5 FDMA (?9600?) instead of 6.25 FDMA (4800).
    I never really took the time to play with the 6.25 vs. 12.5 thing when it comes to audio. I did play with the audio EQ settings quite a bit and got it sounding good. Finally my Kenwood rep sent me a document from K. with guidelines on setting up the radios based on usage. Found a setting that worked well for us and I've just about got all the radios updated.

    The slide deck from Kenwood is about 27 pages long, but here's the basics. Wish I could drop the whole file here:
    Environment
    Quiet Noisy (80-90dB) Very Noisy (100dB)
    RX Low Cut Unchecked (default) Unchecked (default) Unchecked (default)
    Auto Gain Control Off (default) Off (default) *1 Off (default) *1
    Audio Equalizer Flat (default) Flat (default) High Boost *2
    TX Noise Suppressor Checked (default) Checked (default) *1 Checked (default) *1
    Auto Gain Control On (default) On (default) On (default)
    Audio Equalizer Flat (default) Flat (default) Low Boost *2
    Mic Sense" 0dB (default) 0dB (default) -10~-20dB *2

    – *1 Noise Suppressor can be used to reduce background noise
    – *2 In a very noisy environment it is recommended to lower the
    microphone sensitivity and for the user to keep their mouth close to
    the microphone.
    – *2 It is also recommended to use the combination of RX Audio
    Equalizer “High Boost” and TX Audio Equalizer “ Low Boost”


    Quote Originally Posted by moetorola View Post
    Although I think most of there issues are system setup and also VHF interference.
    I'm fortunate enough to have my system on 5 800MHz NPSPAC channels, so interference hasn't been an issue.

    Quote Originally Posted by moetorola View Post
    Not to get to far off topic and starting a new post, since I have two sharp guys in here,,,, is there a way to remove a channel from service on illegal carrier on a type C NXDN system? I can not find it, if there is.
    I think it does it automatically, but I'd have to do some digging to confirm that.

    Quote Originally Posted by moetorola View Post
    Disclaimer: I am used to P25 systems, so not awful familiar with NXDN, although through some research it seems to be awful familiar to SZ, just different names for features or settings.
    My system replaced an ailing SmartZone system, and it was quite similar. Basically had to figure out the Motorola to Kenwood terminology translations.

  13. The Following 2 Users Say Thank You to mmckenna For This Useful Post:

    Mars (Jan 11, 2020),moetorola (Jan 13, 2020)

  14. #8
    Join Date
    Jan 05, 2018
    Location
    WTVLCA01DS0
    Posts
    20
    Thanks
    39
    Thanked 38 Times in 14 Posts
    Country: United States

    Default

    Well, that didn't format well. Anyway, figured out how to do the attachment. Try this on for size:

    NEXEDGE Audio Rev 02.pdf

  15. The Following 2 Users Say Thank You to mmckenna For This Useful Post:

    Mars (Jan 11, 2020),moetorola (Jan 13, 2020)

  16. #9
    Join Date
    Oct 28, 2019
    Location
    DM14ed
    Posts
    27
    Thanks
    15
    Thanked 51 Times in 18 Posts
    Country: United States

    Default

    Quote Originally Posted by Notarola View Post
    Alpha is correct. I have done some studies on the various modes/coverage. In many cases its hard to tell what is what because of the built in error correction for the various modes.
    I'm quite fascinated by the coverage claims made regarding 12.5 KHz vs 6.25 KHz digital, and even 5 KHz dev FM vs 2.5 KHz FM. For a given noise figure on a receiver front end, the "threshold" is dependent on the IF bandwidth. Owen Duffy has a "receiver sensitivity metric" calculator on his web page that allows one to plug in things like noise figure, IF bandwidth, and some definition of threshold to illustrate the effect of bandwidth on perceived sensitivity. In theory, a 6.25 KHz wide signal should have some measurable improvement over 12.5 KHz bandwidth signals. Most of us are aware of the fact that on HF, CW will get through when SSB won't. That's an artifact of the IF bandwidth vs. receiver noise figure. Many of the current crop of microwave radios take advantage of that with adaptive modulation and variable IF bandwidth. Take a path fade, the radio throughput slows down, but the link holds error free.

    But with these digital modes, that seems to be thrown out the window. I'd like to know precisely why, but the whole subject of the various codecs and their inner workings is somewhat beyond my grasp right now. The results observed are not in keeping with the theory. Even 2.5 KHz FM is reported to have less range than 5 KHz FM, in spite of using half the IF bandwidth. Yet, I can measure a several dB threshold improvement in NFM on the bench, in a radio programmed up on a channel in both wide and narrow FM.

    I expect the error correction and various algorithms used in the different CODECS all behave differently. A more definitive measure would be to stick with BER, as that would allow a straight on comparison between P25 FDMA, TDMA, NXDN, and so one. Audio quality is subjective. I prefer a slight hiss on an analog signal, vs a hiss-free mechanical sounding digital voice, but that's just me. Obviously, trying to compress voice into half the number of bits is going to cost some clarity. Intuition says it would gain something in rx threshold sensitivity, but it loses some there as well. A T1 channel bank took 64 kb to convert a human voice to digital. Getting it down to 6.25 KHz is a HUGE amount of compression.

    So, all of the above is rambling. All I really know is that we've traded our brain's ability (wetware) to decipher a weak noisy signal for a mathematical algorithm (software). I still think our brains do it better.

  17. The Following 2 Users Say Thank You to wa6jbd For This Useful Post:

    moetorola (Jan 13, 2020),Radioguy7268 (Jan 11, 2020)

  18. #10
    Join Date
    Nov 06, 2012
    Posts
    576
    Thanks
    621
    Thanked 586 Times in 309 Posts
    Country: United States

    Default

    Both first hand and anecdotal 3rd hand, the 6.25 NEXEDGE is preferred for coverage, and is the most popular in use. 12.5 is less popular. 6.25 allows the receiver to be more sensitive and provides more of a guard band from adjacent channel interference.

    Kenwood's receivers are not the hottest on the market, so its hard to do a fair Apples-Apples comparison for the NXDN protocol itself. However, knowing that Kenwood's receivers aren't the hottest (at least on the older models), and how good even old NX-300s perform, it must be one heck of a protocol.

    That being said, we do have to re-allign old NX300s pretty regularly, they drift worse than their Motorola XPR4000/6000 counterparts from the same time period.

    We have found DMR and P25 to handle interference significantly better than NXDN.

    Very Narrow NEXEDGE sounds better than DMR or P25 Phase I or II IMO.

    Only exception is NX-220/320/420. Complete bad apples in terms of audio quality. NX-300 sounds much better.

    The newer NX-1000/3000/5000 all sound as good or better than the NX-300.

    Kenwood radios not only have large speakers, but an "acoustic belly" to go with them.

    Thats why a CP200d sounds better than an XPR7550e. It has more of an acoustic belly, and speaker in CP200d is bigger and better than XPR6550, and at least as good as HT/Waris.

  19. The Following User Says Thank You to PSEhub For This Useful Post:

    moetorola (Jan 13, 2020)

  20. #11
    Join Date
    Nov 05, 2012
    Location
    True North Strong and Free
    Posts
    329
    Thanks
    2,351
    Thanked 623 Times in 270 Posts
    Country: Albania

    Default

    Quote Originally Posted by wa6jbd View Post

    So, all of the above is rambling. All I really know is that we've traded our brain's ability (wetware) to decipher a weak noisy signal for a mathematical algorithm (software). I still think our brains do it better.
    Well when you have an analogue signal at -119, -120dBm, you’re ear won’t be pulling much of anything out. My XPR7550 on the other hand will recover full audio at that signal strength in digital mode.

  21. The Following 3 Users Say Thank You to Viper1-6 For This Useful Post:

    AD0JA (Jan 16, 2020),moetorola (Jan 13, 2020),PSEhub (Jan 11, 2020)

  22. #12
    Join Date
    Oct 28, 2019
    Location
    DM14ed
    Posts
    27
    Thanks
    15
    Thanked 51 Times in 18 Posts
    Country: United States

    Default

    Quote Originally Posted by Viper1-6 View Post
    Well when you have an analogue signal at -119, -120dBm, youre ear wont be pulling much of anything out. My XPR7550 on the other hand will recover full audio at that signal strength in digital mode.
    That generates more questions...

    What's the 12 dB sensitivity of the analog receiver you refer to? I've seen -127, or 0.1 microvolt, and even better. In that receiver, -119 is quite readable, if unpleasantly noisy. What's the 5% BER threshold of your XPR7550? And the big question for me is, what BER reasonably corresponds to 12 dB SINAD, in terms of voice readability? This last question is real, not hypothetical - I really don't know the answer.

    Analog has a soft threshold where one set of ears might pick out what's being said, and another might not. Digital has a hard threshold where, once it goes garbled, it's just gone. For me, personally, the hardest adjustment has been trouble shooting system problems. Intermod, for example. With analog, if there's a crackling antenna, you hear crackling. If there's intermod, you hear the mixed signals. If it's a weak signal, you hear that the signal is weak. With digital, if there's a crackling antenna, you have a high BER. If there's intermod, you have a high BER (but I have some good stories!), if there's a weak signal, you have a high BER. Seeing a pattern there?

    For me, we just traded one set of problems for another. Adapt or die.

  23. The Following 2 Users Say Thank You to wa6jbd For This Useful Post:

    moetorola (Jan 13, 2020),PSEhub (Jan 11, 2020)

  24. #13
    Alpha's Avatar
    Alpha is offline T S - Moderator
    CS Forums $upporter
    Join Date
    Feb 12, 2012
    Location
    Directly above the center of the earth.
    Posts
    2,780
    Thanks
    1,239
    Thanked 1,533 Times in 739 Posts
    Country: Christmas Island

    Default

    Trying to remember the exact source, but IIRC, 3% is the target max BER for acceptability, or the "12dB SINAD/20dBQ" threshold point, if you will.

  25. The Following User Says Thank You to Alpha For This Useful Post:

    moetorola (Jan 13, 2020)

  26. #14
    Join Date
    Jan 31, 2015
    Location
    Portland OR
    Posts
    733
    Thanks
    1,234
    Thanked 1,192 Times in 539 Posts
    Country: United States

    Default

    Quote Originally Posted by Alpha View Post
    Trying to remember the exact source, but IIRC, 3% is the target max BER for acceptability, or the "12dB SINAD/20dBQ" threshold point, if you will.
    5% BER is the P25 standard threshold, and generally equivalent to the 12db SINAD in analog. It's a hard number test that can be recreated over and over. It also tends to coorespond with the DAQ 3.4 subjective test some customers conduct. But, one man's 3.4 is another man's 2.0 depending on their ear.
    Bow wow wow yippie yo yippie yay

  27. The Following 2 Users Say Thank You to Bill_G For This Useful Post:

    Alpha (Jan 12, 2020),moetorola (Jan 13, 2020)

  28. #15
    Join Date
    Mar 01, 2012
    Location
    Houston, TX
    Posts
    518
    Thanks
    89
    Thanked 173 Times in 75 Posts
    Country: United States

    Default

    i used to manage both a connect plus as well as a type C NXDN system. Both were on the same TX Combiners, RX multicouplers, antennnas, power out and RX sensitivity, both on the UHF band. Virtually as close to identical except for the air interface as you could get. What we discovered that the NX subscribers had noticeably better range than the TRBO subs on both system access and with in building coverage, both RX and TX.

  29. The Following 3 Users Say Thank You to xpr8300 For This Useful Post:

    moetorola (Jan 14, 2020),PSEhub (Jan 14, 2020),Viper1-6 (Jan 14, 2020)

  30. #16
    Join Date
    Nov 06, 2012
    Posts
    576
    Thanks
    621
    Thanked 586 Times in 309 Posts
    Country: United States

    Default

    Quote Originally Posted by xpr8300 View Post
    i used to manage both a connect plus as well as a type C NXDN system. Both were on the same TX Combiners, RX multicouplers, antennnas, power out and RX sensitivity, both on the UHF band. Virtually as close to identical except for the air interface as you could get. What we discovered that the NX subscribers had noticeably better range than the TRBO subs on both system access and with in building coverage, both RX and TX.
    Could you elaborate on models of portables/mobiles? Also curious if exclusive or shared freqs?

    How did de-sense seem with portables near eachother?

  31. The Following User Says Thank You to PSEhub For This Useful Post:

    moetorola (Jan 14, 2020)

  32. #17
    Join Date
    Mar 01, 2012
    Location
    Houston, TX
    Posts
    518
    Thanks
    89
    Thanked 173 Times in 75 Posts
    Country: United States

    Default

    XPR7550 handhelds on the Connect Plus, NX-300/320 Handhelds on the NXDN Type C. Exclusive FB2 channels on both systems. If i remember correctly, it seems i could key the NX radio and the XPR would lose the control channel, but when the TRBO was keyed i dont recall the NX radios going OOR. It may have ,but its been a few years.

  33. The Following User Says Thank You to xpr8300 For This Useful Post:

    PSEhub (Jan 15, 2020)