Dstar versus dmr - from wb8nut

Discussion in 'News' started by Andy G0VRM, 13 October 2013.

By Andy G0VRM on 13 October 2013 at 23:59
  1. Andy G0VRM

    Andy G0VRM Active Member
    Committee Member

    Joined:
    22 January 2011
    Messages:
    72
    Likes Received:
    13
    Recently I have noticed a few people promoting DMR as the DStar replacement. So I have been looking into DMR a bit. Some of you may have read about Mototrbo which is Motorola's marking name for the DMR technology.

    I intend to do a more in-depth comparison here between the various digital technologies. Not just DStar versus DMR, but also look at whatever Yaesu is coming out with, APCO25 and Nexedge/NXDN.

    So I signed up on Yahoo Groups with a few of the DMR and Mototrbo groups. Via radio reference website, there even is a link to listen to one of the DMR networks/repeaters.

    One of the first questions I asked on the groups, "can you build you own DMR/Mototrbo repeaters?" In other words, is it practical?

    The answer I got was basically "no" that it is not practical and they way to get a repeater is to "get some of my friends together to share the cost." Also, not practical.

    Now with DStar you can build a repeater yourself very easily. You need two transceivers with packet ports capable of supporting 9600 baud packet, a hotspot board and a computer. So here is what it cost to build a home-brew DStar repeater that a couple of folks in a local radio club are considering building.

    Two single band Alinco radios. $279 each. Moencomm GMSK Node Adapter (hotspot board) at $120 and a computer, in this case a Raspberry Pi with Ethernet Adapter ($35 for the RPi and $30 for the Ethernet adapter. Total cost of the DStar repeater is $743 using all new equipment except for the duplexers and antenna which the club already has for the old UHF FM repeater days.

    DMR well for a new repeater, since there are not a lot of used repeaters, about $3,000.

    The other issue I see as a problem is with DMR each radio has an ID that has to be registered and is tied to the callsign of the user. This appears to be a real issue with sharing equipment and callsign routing, that is not an issue with DStar.

    Also DMR programming is very complex. Mainly because DMR was meant for the commercial market. There are things that need to be programmed like "talkgroups" which do not exist in ham radio. If I read correctly, since all repeaters in a network are tied together, all transmissions go out over the entire network. There seems to be no way for an amateur operator to connect and disconnect the repeater to the network or network of choice. This is all easily done in DStar.

    So here is what it really comes down to. DMR was designed for the commercial user in mind. DStar was designed for amateur radio. One fits better than the other. Both are open protocols except for the Codec which in both cases is proprietary. Both sound equally good.

    DStar has over 1,000 repeaters world-wide. DMR less than 100. DStar has thousands of users world-wide while DMR has about 300.

    DStar current generation radios can be used without any programming except for entering in the callsign of the user. DMR has to be programmed.

    Clearly, DStar is way ahead. The question among many is will it remain there. DStar users say yes and DMR users say no. It's the 21st Century version of the code/no-code argument all over again.

    http://wb8nut.com/blog/files/March_13_2013.html
     
    #1
    Ron Young likes this.
Tags: this article has not been tagged

Comments

Discussion in 'News' started by Andy G0VRM, 13 October 2013.

    1. Ron Young
      Ron Young
      Yes an
      Yes and the problem the problem gets worse when Yeasu Finally puts its Repeaters on the market, Yet another incompatible system causing more devision in the DV ranks.
      Yeasu and Motorola are one but will they be compatible? not on your nelly..
      We will have 3 systems and all of them proprietory what a comercial hodge podge.
      If Radio Amateurs could have devised a single Non Proprietory system that all amateurs could subscribe to without the guilt of loosing the amateur radios inclusive principle.
      The expensive DV dongles and other responses are just eyewash, they just helps the likes of ICOM yeasu etc to up the Anty sorry Price for the HT and mobile rig.
      Maybe the codec is the key or 3 codecs, maybe thats the bit Radio Hams should have concentrated attention on...
      I am well in favour of DV but but not the total theft of the system which manufacturers are performing
    2. Andy G0VRM
      Andy G0VRM
      I think the way digital voice should go is to produce an add-on box which plugs into the 6-pin data socket present on most VHF/UHF radios. It would need it's own microphone and speaker socket on the front panel and would need to be controlled and programmed either via a PC or an LCD display (large enough for mobile use!). Most modern radios have a 9,600 baud packet mode built in which gives direct access to the modulation and demodulation stages.

      It would ideally run on open-source and hence easily reproducible hardware - something along the lines the existing DV-RPTR boards which are based on a PIC-32. Or, if more computer power is required, something along the lines of the Beaglebone Black (costs £35 +a widely available radio interface).

      Yes the CODEC is key. All current amateur digital voice systems to date use proprietary CODECS based on AMBE or it's derivatives that need the chip to do the encoding and decoding (hence the Dongles) - there is a freely available open source codec called CODEC 2 which does everything the commercial ones do but at a lower bit rate and higher audio quality. The differences between the proprietary codecs and the open source ones is the bits they decide to patent as most are based on freely available research!

      One thing which is essential is forward error correction - D-Star DV works down rough noise as it is able to piece together the signal from the information it has. The data stream embedded in D-Star doesn't have this correction and will only work with STRONG signals. The ability to use the whole data stream for data when voice isn't present - with the same level of forward error correction would be extremely useful.

      The modulation method is equally important, not just from the point of view of on-air performance, but from a compatibility with other systems. D-Star with its GMSK doesn't co-exist with other repeaters in the same way as the 4 level FM used in APCO 25 does, letting the APCO repeaters work on both FM and digital.

      I think I have probably summed up the ideal system Hardware, Modulation Method, FEC and CODEC. The next thing is presentation and routing. The callsign based routing on D-Star works well, but I do like the recent add on which gives the ability to dial someone up on DTMF pad and the repeater system finds them - and most of all, the ability to remember where other stations are, even the ones that aren't digital from a DTMF code they use to register on the system and transmit to them in the appropriate format...
      Ron Young likes this.
    3. phillip hardacre
      phillip hardacre
      Good post Andy you do have a good point, i have NXDN gear and as far as i know the audio quality is the same as trbo, the way i see it, it comes down to cost and as there are lots of mototrbo radios on ebay (without charges nudge nudge wink wink) then thats what they will buy i think.
      dealer prices for trbo are less the nxdn so the dealers sell that gear and then that finds its way on to ebay thats why theres no nxdn gear on but lots of trbo.
    4. phillip hardacre
      phillip hardacre
      Motorola only has the commercial Yeasu gear, Ham yeasu is Yeasu.
      Ron Young likes this.
    5. Ron Young
      Ron Young
      May be some on will develope an open source, public domain system that can co-operate with either TRBO or D-Star
      or both then we can believe that DV is an Amateur Radio option not just a way to bolster Icom / Yeasu profits
      or maybe no one has noticed that Icom/Yeasu pricing doubles the cost when we go digital, Lets get the Amateur bands back from these Pirate companies.
    6. phillip hardacre
      phillip hardacre
      Mototrbo and nexedge are open source
    7. Andy G0VRM
      Andy G0VRM
      The first task is to break the system down into layers... (using the ISO Open Systems Interconnection model)

      Physical - The point to point data link using whatever modulation method (GMSK, 4FM etc.)
      Data Link - The means of making this reliable, forward error correction etc.
      Network - The addressing or routing (translating callsigns to IP addresses)
      Transport - The means of delivering the data stream to the appropriate repeater (UDP)
      Session - Communications between nodes (keeping Callsign lists up to date)
      Presentation - The encoding and decoding of digital voice
      Application - The presentation of data via the radio's display / PC Software (talk groups / reflectors etc.)

      The concept being that each layer on one system talks to the corresponding layer in the other via the layers below it - by conversion to an intermediate format which preserves the data itself and metadata (data describing the data) and then into the appropriate format at the other end...

      A lot of work!
      Ron Young likes this.
    8. Ron Young
      Ron Young
      Yes I take your point, this is all done in software, who does all the work? a few committed amateurs /software writers mostly open source.
      Who locks us in and blows the whole open source principle? in D-Star its Icom with some sort of control via a codec chip. The dvnode and dvap conspiracy help to keep the prices sky high, my bet it will all happen again with Mototrbo and nexedge and whatever Yeasu conspires , Amateurs need to set the standards before manufactures own them all...
    9. Ron Young
      Ron Young
      I wonder if we implemented the Presentation layer in software and made it compatible with D-star would we be sanctioned by the Ambe codec patents. I hear the Peter Martinez G3PLX had some problems with his recent software Codec!
      Maybe Icom would implement the open source Software Codec in all future D-Star production hi hi, we can live in hope
      It would be Icom supporting the Amateur fraternity a nice reward for their customers loyalty.
    10. Andy G0VRM
      Andy G0VRM
      It would be theoretically possible to implement D-Star using another audio CODEC - except the current system doesn't have any provision for a data-only QSO and hence slotting in another CODEC without conflicts will always be impossible.

      However, it would be nice to use the existing infrastructure and repeaters...

      See http://www.rowetel.com/blog/?page_id=452 and http://codec2.org/

      Andy
      Ron Young likes this.
    11. Ron Young
      Ron Young
      Many Thanks for the links above, Its all very interesting, I have spent a few hours browsing and following. The preference for more open source is stronger than I thought, Sounds like the way to go.
      I do not think provision of data link will be a problem, looks like its under consideration already, even adding a PSK channel sounds viable, Peter Martinez is mentioned hope he gets involved with the MoDem implementation

      Thanks Again

      Ron

Share This Page