Homebrew Multimode Digital Voice Modem

There’s an old saying that the nice thing about standards is there are so many of them. For digital voice modes, hams have choices of D-Star, DMR, System Fusion, and others. An open source project, the Multimode Digital Voice Modem (MMDVM), allows you to use multiple modes with one set of hardware.

There are some kits available, but [flo_0_] couldn’t wait for his order to arrive. So he built his own version without using a PCB. Since it is a relatively complex circuit for perf board, [flo_0_] used Blackboard to plan the build before heating up a soldering iron. You can see the MMDVM in action below.

The build includes an Arduino, of course, and the neat perf board wiring makes for a good-looking project. We’ve covered digital voice that uses PCs before and even some digital ham modes that use an Arduino. Or check out the MMDVM project for more info.

20 thoughts on “Homebrew Multimode Digital Voice Modem

  1. I’ve actually met some of the principles involved in the MMDVM project; any ham that managed to get to the Dayton Hamvention two years ago had a chance to meet Jonathan Naylor, G4KLX in person (he was hanging out at the Northwest Digital Radio booth with John Hayes, K7VE and Bryan Hoyer K7UDR).

    1. And more sadly, digital voice is just not all that interesting. Don’t get me wrong, I appreciate the “hack” and the determination behind it. I guess I’m just disappointed that there’s no longer any real ham data experimentation. No, re-purposed Wi-Fi, which might as well be laser-based once you get beyond a city block, doesn’t count. HF modes with no error correction or session reliability don’t count. Yes, there are physical limits to the amount of data that can be transferred on VHF and UHF bands, but we (hams) need to be doing more there. There is some room for improvement in transmission modes, but much room for improvement in protocols and bandwidth conservation. The pinnacle of data communication is not 32 bit images and HD video over HTML 5. There are other, equally valid but more efficient ways to communicate. I hate that broadband killed packet radio. Had it survived long enough to evolve, it might be truly compelling today.

      1. These guys: http://www.rowetel.com/blog/?page_id=452 beg to differ. They’re making big strides in turning Codec 2, a free (as in speech) open-source digital voice codec, into a usable communication mode.

        Their latest is FreeDV 2400A, a 4FSK modem implementation of the codec readable to -132dBm, which is about 10Db below Analog FM.

        While we’re at it: http://physics.princeton.edu/pulsar/k1jt/ Joe Taylor of WSJT fame has done a lot with robust communications over HF-SHF frequencies. He started with developing modes for Moonbounce, but has others for weak-signal HF, Meteor scatter, etc. They’re robust against multi-path, Rayleigh fading, and other forms of interference thanks to strong FEC.

        There’s experimentation in amateur radio if you know where to look! I’m in the block diagram phase of implementing a TurboCode FEC that could be applied to HF Packet, VHF Codec2, etc. If you don’t think there’s enough experimentation, please, start doing some!

        A great place to start is this book: http://www.amazon.com/An-Introduction-Information-Theory-Mathematics/dp/0486240614?ie=UTF8&keywords=introduction%20to%20information%20theory&qid=1458788831&ref_=sr_1_1&sr=8-1

        It’s only $4 on Amazon.

        73,
        KC9SWV – Morgen

        1. Effectively, you’re doing FEC, modulation and analysis on RF signals… Sounds like you should take your block diagram, and make a GNU Radio flow graph out of it; given there’s *a lot* of useful blocks readily available, taking your idea to a working prototype might be less than an hour! Give the GNU Radio Guided Tutorials a read: https://gnuradio.org/redmine/projects/gnuradio/wiki/Guided_Tutorials They don’t cover every block available with GNU Radio (especially FEC isn’t properly covered by the tutorial), but you’ll quickly get a feeling for what’s happening there.

          Since Codec 2 is written in C, you can just let gr_modtool generate a minimal wrapper Block and use that in your GNU Radio DSP flow graph. Tadah, baseband signal, ready for transmission!

      2. Sounds like the posting of a person who’s not involved or experimenting. I’m excited and thinking about having a USRP type device in my pocket with all that cellular phones today offer, but with and SDR on board. I imagine connecting to an app store and downloading a new radio or new protocol. Take the computer from your smartphone and stuff that into an SDR. I’m excited. Where are you?

      3. To wit, I think it was the Internet itself that killed packet. That and a total and complete lack of improvements. Most people never used anything faster than the 1200 baud packet that existed in the early 80’s. Myself included. There were experimental 56k modems in the late 90s but they were so expensive that they didn’t really have any market penetration. Even 9600 baud that could be run through certain radios didn’t even hardly get used.

      4. You actually missed at least a third of the D-STAR protocol; the digital data part, but that’s partly Icom’s fault, since they limited the implementation to 1.2 GHz. You also apparently don’t follow the TAPR DCC stuff, since there have been some quite interesting developments from that in the last few years other than stuff about Wi-Fi.

      5. The problem is ham radio killed experimentation itself. It all about HF and doing ham radio like it 1970. Your comment for example. Packet was great when we were still using acoustic couplers at 300 baud than the first 1200 baud modems, but rest of the world move on to 4G and Ham radio dug in and stuck to the past. I got my ticket KG7CSS a few week before the “you know what ” storm over ARRL proposal of dong away with the symbol rate for a 2.8 kHz bandwidth. perhaps we need a new group other than Eham and QRZ.

  2. Is is common that people put the parts on the copper side of the PCB like this? In my 40 years of tinkering I hardly ever done that except for a few times when I wanted it to look more hi-tech. Maybe it’s because I used stripboard/veroboard for the first 20 years and then moved over to the more fancy donut and 3-hole boards…

    1. Actually… though that’s not what is going on here putting the components on the copper side is very common with amateur RF projects. Go to your favorite search engine and do a query for “Manhattan Construction” and check it out.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s