FreeDV – Digital Voice for HF bands
This weekend I had the time to try something new. FreeDV is software for digital voice on HF!
My buddy Jeff N9IZ first told me about this exciting, new, experimental mode.
FreeDV is pretty interesting to use. I had the chance to mess around with it for a few hours and here is a brief description of my experiences with it:
On Sunday morning I had some time to kill before the “big game” so I installed the software. FreeDV is opensource so it is completely free to use, and if your a developer you can take the code and build and improve upon it. They use the CSV system for submittals. Check out their website here for more information. The software is in BETA (version 0.91 at the time of this post), so it could be a challenge to get things working properly. You will need some computer know-how and a good deal of patience to get it to work properly. If you are a computer n00b, then I would suggest waiting for a bit until a production release of the software is available. BETA software is experimental and usually means that it is buggy, and you will have some issues working with it, so be forewarned.
To use FreeDV you will need the following hardware and or software:
- A computer.
- 2 computer sound cards (one for the radio, and one for FreeDV to use.
- A headset with boom microphone (or just a headset or speakers for receive only).
- Microsoft Windows XP or better (or Linux, but you will have to compile the code for your distro).
- You should have a dual core processor and at least 2GB of RAM I would think. This will be a poor experience on an old clunker Pentium 4 like PC, so if your computer is too old…don’t bother. But if you already use HRD+DM780 for digital modes your good to go.
- A small amount of HDD space. The extracted folder is only about 30MB.
- An HF transceiver that is connected to your computer with a sound card interface or direct USB type connection.
INSTALLATION (this applies to Windows type OS):
- Download the program at http://freedv.org/tiki-index.php. Scroll down and click on “Windows binary files (v0.91 beta)
” to download. - Once the .zip file is downloaded, extract the contents of the folder to your desktop.
- Make sure that your HF transceiver and sound card interface are connected properly so that you have the audio from your transceiver coming into your computer.
- Next make sure that your second sound card is functioning. I myself use a USB headset (the USB connection is a sound card itself – this is the easiest method I would think).
- Run the program from the extracted folder on your desktop named “freedv.exe”
- You should then see the main GUI for FreeDV.
- Next you need to configure the sound cards that FreeDV will use for connection between your headset and the sound card that is used for the sound card or radio interface. You can access the Audio Config from the Tools menu at the top left of the FreeDV GUI.
- Tune to 14.236 MHz, this is the main frequency that is used for FreeDV QSOs right now. There may be other frequencies that are used, but this is the most popular one in North America.
- Click the “start” button in the bottom right to start the decode.
That’s really the process in a nutshell. There is some issues with configuring PTT still. For example, my setup cannot be keyed with the PTT button in the GUI right now because the software does not yet support it. I use an Icom IC-7600 with it’s own USB sound card interface and CAT control using the Icom CI-V code set. Typically I use Ham Radio Deluxe and Digital Master 780 for digital modes and that software will support CAT control PTT. This is a feature that is currently being looked at by FreeDV developers to be incorporated into the next version of the software.
HOW IT WORKS:
Since I am not an expert, I will explain this in simple terms as I understand the operation. Basically FreeDV uses an audio codec (a piece of software that encodes and decodes audio like MP3’s use) as a software based modem to decode and encode the voice and text data from your computer. The data is encoded and then is transmitted over the HF radio using AF signals that can then in turn be decoded by the receiving station.
The codec that is used is called “codec2”. There is a lot of information about how codec2 is being used for digital HF and the information on it can be found here: http://codec2.org/
MY EXPERINCES WITH MAKING QSOs ON FREEDV:
When you just tune to the frequency using analog SSB you will hear a distinctive “warrrrrrrrrrrrrrr” growling sound of those transmitting using FreeDV. To me it sounds a bit like digital ASCII or MT63 if you know what those digital modes sound like. On an audio spectrum analyzer you will see several vertical lines that occupy the audio bandwidth of 1000 Hz to 2100 Hz if you are on center frequency.
FreeDV GUI with digital QSO carrier (note the audio spectum vertical lines from 1kHz to 2.1kHz)
Generally speaking this mode does not work well when the bands or noisy due to QSB or QRM. Making QSOs when the band is very busy with other stations 3kHz away using SSB and having them bleed over on you will cause your FreeDV QSO to be almost unintelligible. Typically it will sound garbled, almost like the transmitting station is trying to talk on a tin-can with a string while under water. Signal strength is important as too week a signal just will not be received. I have found generally that if you have an S-4 noise floor, the station needs to be S-6 or better to be properly decoded on FreeDV. This is definitely not a weak signal type digital mode.
One good piece of advice to start, is to uncheck the Squelch check mark box setting in the GUI so that you can hear what is going on. Once you get used to using FreeDV then you can use the squelch to block out the background noise. This just makes it easier to know that you can hear the audio from your HF rig to start. By default (after you run the program for the very first time) the squelch is set at 4.0 and is enabled (checked box).
Also, it is a good idea if you have a radio with an adjustable filter to narrow the filter bandwidth to 1.2kHz. This will eliminate most adjacent channel interference and extra noise. My Icom IC-7600 lets me adjust the filter bandwidth from 50Hz to 3.6kHz on SSB which is a really great feature that is now incorporated on most modern mid-range to high-end HF transceivers.
Once you do find a station that is transmitting a strong signal, the audio is really pretty good, especially for only using 1.1kHz of bandwidth. I made several QSOs and at times had trouble copying the transmitting station (and likewise they copying me as well due to noise) every now and then, but when it did work, it was really pretty neat.
There is also a text field where you can input your call sign, QTH, and other information. When you transmit that text data is also transmitted with your voice as well. Most people will put in their call sign, their first name and their QTH just so make it easier on the receiving op to know who is transmitting. I found this very helpful when the conditions were noisy and couldn’t copy the transmitting station’s call sign. With noise though this text field sometimes becomes garbled and unintelligible as well as the voice data.
MY SUMMARY OF FREEDV:
Here is my bottom line no B.S. assessment of FreeDV thus far.
THE CONS:
- Software is buggy, and can be difficult to configure if you don’t know much about computers.
- Noisy HF conditions are a problem and will cause unintelligible QSOs. NOT A WEAK SIGNAL, OR NOISE BUSTER HF DIGITAL MODE!!!
- Sound card-radio-headset configuration is not for newbies. But if you have successfully connected your HF rig to your PC for digital modes like PSK-31, RTTY, or OLIVIA, then you should know enough about what your doing to make this work.
- PTT can be a problem as there is not good support for different PTT schemes yet.
- WATCH YOUR SOUND LEVELS! You don’t want to damage your sound card because of a too strong audio level input.
THE PROS:
- Very cool idea. It’s about time we had some digital voice on the HF bands (besides DRM).
- FreeDV is completely open source and free. No proprietary protocols that only work with a certain brand of radio (um, yeah I am talking to you Icom :@ ).
- The 1.1kHz bandwidth makes for good audio quality when there is no QSB or QRM and the signal is strong.
- Nothing too special is required to make it work. If your a Ham that is already on PSK31 or using other digital modes, most likely you have everything you need. Just pickup a USB headset.
Here is a great demonstration video on YouTube from one of the author’s of FreeDV, Bruce Perens:
And here is an EXCELLENT video by sandydiesel on YouTube that is a quick start guide/tutorial! Very nice!
I hope you found this post informative and interesting.
Are you a FreeDV user? What have your experiences been like? Tell us about it in the comments section below this post. We would love to hear from you!!!
73! de Nick N9SJA
K9XN 
Good job,OM…and good write up. You’re like my personal beta tester. Now I’m itching to get it configured on the Flex. Linux compiling for the FT-897 was a dismal failure!
just started playing with freeDV… I’ve used HRD/DM780 for a year, so the setup required for freeDV was already done. Had to buy a USB headset w/ mic. Use digital VOX on my FT-897, but had to increase the VOX gain in order to key the rig. Didn’t use the FreeDV equalizer at first and the headset mic sounded terrible… adjusted the low freqs out of mic response and it sounds good now. Still working on my transmitted voice… very garbled as I hear it in sidetone… not sure what the problem is, but will continue to play with it. Here in Big Bear Lake, at 7,000ft, there isn’t anybody else using DV yet, so I’m talking/listening to myself for now. I am very optimistic that this new mode will evolve into something we will all like to play with… but certainly not a weak-signal mode.
73, Bill, W6BBL………..btu
Bill,
I have made several contacts with FreeDV here. Usually if you listen on 14.236, that’s mainly where the action is with this mode. Usually I do hear a station on, it’s just that I cannot copy the transmission because the signal is just to weak and the copy comes in garbled. I’m sure that this mode will pick up some traction with time. It is very interesting how it works and there needs to be a good open standard for digital voice. DSTAR is nice, but only on ICOM radios. It would be nice if this turns out to be another PSK-31 type situation where it just takes off. The biggest strike against it is that you just cannot decode anything weak like you can do on SSB. Keep listening, who knows maybe you will hear me on there calling CQ ;-). I do try it from time to time. Great to meet you Bill! 73 de Nick N9SJA
Version 0.96 performs a lot better in poor signal conditions. It’s worth revisiting. FreeDV will be at Hamvention booth HA-12, do please drop by if you’re going.