Does SDR really suck?

How does it come that such a promising technology like SDR just does not take off? SDR promises great transceiver performance, flexibility for an affordable price. So what is the problem? Read on!

This year during the World Radio Team Championship (WRTC 2010) not a single team out the 48 entered the arena with an SDR Radio. These guys are top notch contesters and most of their callsigns can be found regularly of the highest rankings throughout all major contests. Why didn’t anyone trust SDR with its superb characteristics?

Let’s have a look to the market: By today FlexRadio offers already SDR solutions designed and manufactured against industrial standards. These radios are at least as reliable as the Yeasu FT1000MP or Icom IC-756 Pro IIIs. So it can’t be the (hardware) reliability.

Before we go one let’s recall the differences between a classic radio and a software defined Radio.

Three fundamental building blocks

Any radio can be broken down into three fundamental components (also called building blocks):

  1. RF
  2. Signal Processing
  3. User Interface

In a classic radio (like your IC756 Pro III) all three components are inevitably located within the same black box which is standing on your shacks desk (schematic below).

On the other hand, the SDR approach is much more flexible. All three building blocks are completely independent of each other. There are no physical or logical dependencies. It is up to the radio designer to combine or separate these building blocks (see schematic below).

SDR Concept User Interface

A great example is the SDR-Cube. While the SDR-Cube takes care about the Signal Processing and the User Interface, the cube will work with any kind of Softrock like transceiver. Whatever you might have already purchased. Great reuse, isn’t it? This is how the logical view on the SDR-Cube looks like:

SDR Cube Logical View

Almost every day there are new SDR (RF) projects announced. Apart from Softrock you might have already heard of Genesis, FA-SDR-TRX, QS1R, PMSDR, Beaglebrick, LD-1, … and so on. If you don’t trust me, check out this list of SDR projects.

Also in the Signal processing domain various software packages and implementations have emerged over the last few years. Rocky, WebSDR, SDR-Radio, Quisk and PowerSDR are just a few ones.

But what about the third building block, the User Interface?

The User Interface

UI component Icon

This year I had a lot of conversations regarding the User Interface of Software Defined Radios. Somehow it is just amazing that the Ham community is paying so little attention to this aspect of an SDR. By today it is still widely accepted that the typical User Interface of an SDR consists of a Mouse, a Keyboard and a PC Screen. You don’t need a degree in ergonomics to know that it is very painful to operate your radio in a 24 hour or even 48 hour contest exclusively with the mouse.

Even FlexRadio, the leading supplier of Software Defined Radios seems to be ignoring the importance of a proper User Interface – and by the way, I do NOT consider the Griffin Powermate or Shuttlepro as an appropriate User Interface! By today these devices are just simulating mouse and keyboard-strokes (which requires the focus to permanently remain of the SDR application).

ShuttlePro 2Griffin Powermate

Today it is easier than ever to create a personalized User Interface. Microcontrollers are cheap and the  few buttons, knobs and adjustable potentiometers might already be waiting in your shack on their usage.

Modern Microcontrollers come with build-in USB protocol and easy adjustable PC drivers. There are plenty of beginner tutorials available on the web and since no RF is involved User Interface are actually the best suited building block to get started for beginners!

Who says that the radio controls must be on the front of a black box and located perpendicular to your shacks desk? SDR allows us for the first time to completely redefine and optimize the radios User Interface! Are you serious into contesting? If so, you want to pay special attention to ergonomics. Why not having the control console directly next to your keyboard? Or do you prefer portable, backpack operations? Then your focus will lie on the limited physical dimensions. N2ABPs SDR-Cube is a nice example on this.

Conclusions

So does SDR really suck? I think it is a promising technology, but for operational usage it still has some flaws. We still need to discover all the new possibilities which SDR enables us to do. RF and Signal Processing isn’t everything! The time is right for exploring the third SDR building block – the User Interface!

Do you have a similar opinion? Am I right or wrong? Please share your thoughts with me through a comment!

About Tobias (DH1TW)

Self-confessed Starbucks addict. Loves to travel around the globe. Enjoys the technical preparations of Amateur-Radio contests as much as the contests themselves. Engineer by nature. Entrepreneur. For more, follow him @DH1TW

Comments

  1. Vince Burolla says:

    I totally agree that a PC should not be the required as a UI for a radio. However, I absolutely do love the visual display of my MacBook when I’m running Web SDR, or DSP Radio with my Softrock Ensemble Receiver. What I dislike is using the keyboard and track pad instead of big, beautiful knobs. I purchased a Powermate, but haven’t yet figured out how to use it with SDR. So, in my opinion, a really slick, waterfall, visual display with real knobs, would be absolutely fantastic.

  2. peter says:

    Hi Tobias

    may be there are other elements, that influence the decision
    pro or contra SDR:

    Some hams might prefer spending their time with antenna building
    instead of comparing the performance of different SW packets on their
    bricks and rigging up even more cables between those.
    May be, as you suggest, even more progressive iplementations are needed
    to take full adavantage of the new technology, e.g. runing a station by
    sliding your finger over the touch sensitive display, continuously zooming in
    the area respectively the frequeny window which is just below the finger.
    Soert of a continuously sliding looking glass function. It would be nice to program
    such a function, but I am out of business meanwhile.

    73 es gud DX de Peter

  3. Mike Baker K7DD says:

    Hello Tobias.
    I agree completely that the bigest problem I have with SDR is the lack of a friendly user interface. The last thing I really want to do is park in front of the computer and work the mouse to death. It seems to me that a touch screen with menus would be one way to go and another would be to make the SDR transparent so the rig looks like any other rig. It just employes the SDR systems. If the rig has a USB connector out the back it could be plugged into your computer and all the sound card crap could go away. Data would go directly from rig to log and vice versa. Obviously there is a lot of work yet to do on the part of those trying to intergrate SDR systems into the Ham community but making it friendly and transparent to the user has barely scratched the surface.

  4. Hi Tobias,
    Thanks for raising this often neglected aspect. Apple’s success and appeal on all their platforms is based on clear consistent application of User Interface guidelines. Any website serious about audience members returning invests significantly in testing and improving the user experience, conducting detailed research into how users find their way about the site and carry out common tasks. Maybe SDR radio development teams need to seriously consider hiring usability experts and information architects.
    I suspect (and hope) that as the number of us using SDR sets reaches a critical mass, more energy will come to be focused on achieving the optimum match between the hardware/software and us. The most exciting thing about the communities that have coalesced around the different approaches (Softrock and Genesis are two I know) is that there’s a pool of expertise and imagination growing day by day.
    In the absence of a corporate sponsor like Apple the R&D into user interface may be more ad hoc, but I’m confident that the communities are capable of solving this.
    I agree that the approach of George N2APB & Juha, OH2NLT’s SDR-Cube is exciting from this perspective alone.
    73 Stephen VK2RH

  5. I’m relatively new to Amateur Radio, but my Dad’s been in it forever, and many of my coworkers are. I’ve tried to bounce the SDR thing off of both of them, and they find it fascinating, but mostly through the filter of me, and I’m more a computer person. I read about GNU radio for a long time, and it is amazing how the community has grown, and the endless plethora of links.

    But you bring up a great point. In my opinion, current SDR software is on par with some of the most advanced and rich interfaces I’ve come across, although the Large Hadron Collider public pages (and some of the screens I’ve seen on videos about it) can also achieve high information density with clarity, too, but this is a serious art.

    This is subject of Tufte and countless interaction designers and media pioneers. The SDR cube is appealing because of its polymorphism. Software is the same in this modular way. I’ve used Quisk, SDR Shell, and DSP Radio, and an Arduino with my setup. Browsing around the SDR Shell world, I saw an interesting project that parallels the simplicity of the cube, it has just your spectrum and a PTT button, so I know people out there are thinking about this too.

    I think we need more divergent thinking and collaboration. I would like to see more exploration of the TiVO aspect of things, I can scroll up and down the dial, but it’d be nice to scroll back and forward through time. Further, the web based SDR access is simply astounding to just send to someone as an introduction. I have friends that have no interest at all in radio be perplexed and amazed. You know, digital modes and their web records could be another thing an interface could have, and some of the cognitive radio ideas.

    Less is also more, too, maybe have a level of automation such that I come home and have a nice report of anything interesting throughout the day with audio clips in a interactive, magazine format.

    The community I think will work over time to make all of this software more modular and make sense to their users and you’ll hopefully see more plug and play components, and a spirit of minimalism, or just be open to some kind of exploration around interfaces? It is a great time to learn to code, it took me seconds to customize Quisk in Python to control a VFO.

    I also think there could be a lot of fun in making a drop-dead simple program that teaches even basic electronics through synthesis, sampling, and signal processing, too. Again, I’m a beginner, and I’m at home with a keyboard, so I can hear you on direct control. I also know computer interfaces can suck, and we just have to keep at it, and pick and choose what is right and what we really want to be able to do.

  6. Hi Tobias. I dont know anything about SDR, but it looks promising. Whats missing imo is the touch screen GUI where one can have a seperate screen from the usual computer one, dedicated to SDR.

  7. Kent, K6FQ says:

    Oh boy! Your way off the mark on this! I’ve been using my FLEX 3000 in a number of contests using N1MM and it IS WAY easier than using a “standard” radio. A computer logging program is pretty much required so the operator’s focus is on the computer screen or screens in case. So it seems that SDR running on the same screens would make more sense. There are some utilities that can be run to help keep the focus on N1MM but still allow you to change frequency with the mouse and not mess up N1MM. I have installed the Virtual Audio Cable (VAC) and Virtual Serial Cable (VSC) so my setup can do digital modes and N1MM can change/get frequency changes, change modes, etc. All seamless!

    My own observation is that folks who poo-poo SDR have not used one or taken the time if they did, to get familiar with it. There is a learning curve with operating SDR but it’s like any new radio, just have to do it! I haven’t mentioned that the FLEX radios have AWESOME receivers! Better than all of the big box Yeasu rigs (except the 5000, but then they are 10 kilobuck radios!).

    I use the latest version of PowerSDR and it looks like a real radio.

  8. Hi Tobias,

    I think anybody with a lot of experience with software radios will agree that things that can be done well analog should be done analog. I’m not a ham but I get the impression the contests being discussed above involve lots and lots of narrow band signals and if two or more happen to overlap the thing to do is tune to another frequency, more or less manually and randomly. In a that case maybe you don’t need the things software radios are good at: thousands of adaptive taps, baud synchronous sampling, Viterbi decoding, etc.

    What seems to be needed is what you might call SDFP: software defined front panels.

  9. John McKown says:

    Last night I was lying awake and started thinking about computer interfaces for ham contests. I don’t know why. This thread just stuck in my head somehow :-) The obvious thing to do if the mouse is the problem is replace the mouse, maybe with one of these

    http://www.belkin.com/IWCatProductPage.process?Product_Id=390404

    or a Microsoft Sidewinder Joystick. You could probably program some of those buttons to do what needs doing without modifying the radio software at all.

    The next car in this train of thought is to modify the clunky radio interface to take full advantage of the joystick. Automatic logging, whatever. I imagine contests involve a lot of standardized messages. Once you’re writing software, why not throw in a parser and some canned messages. Now it’s just point, click and on to the next one. You don’t even really need the operator any more. Take the man out of the loop and voila: a hambot that wins every contest.

    That, for me, is the train’s caboose. I personally lose interest in a game or sport as soon as I realize a machine could do it better than any human. To me chess doesn’t have much point anymore and car racing isn’t far behind. The invincible robot doesn’t even have to exist — it’s enough if I can see that one could assemble a team of people who could build it without needing to invent anything.

  10. Tim Pierce says:

    I agree with John McKown,
    The fun is quickly getting zapped right out of the promise of SDR.
    I got a ham lcense at the age of 13. I saved my money from my paper route
    and bought a few different Heathkit receivers and transmitters. I built them all
    and got tons of advice from hams in the local club (mostly WWII vets who loved radio).
    I understand the common SDR user couldnt write the software needed to interface these radios.
    Which is OK, that could just be considered a learning process. The user may or may not
    want to delve into coding. But building the Rx or TxRx units should stay a kit. Homebrewing
    was always a very large and satisfying part of ham radio. The radios are quickly turning into
    buisness’ for the designers, with completely built radios. I am a member of the Softrock group
    and that is all it is turning into. Getting complete info is like pulling teeth from other members.
    The same thing happened with the Quicksilver, and on and on.
    Also I am sorry to say to ham purists out there but HF is not that interesting or that fun.
    There is a lot more interesting stuff going on in UHF,VHF, and Microwave.
    I no longer use my ham ticket and have some pretty fancy radios in the basement gathering dust.

  11. Francis Pressland says:

    There are now several commercial radios that use sdr based technologies inside what to all intents and purposes is a classic transceiver! Just look at the Elecraft K3. Elecraft are also using an sdr based radio in their upcoming portable radio.

    I think a lot of radio amateurs when confronted with an “SDR” radio, think of a black box that plugs into their PC and uses an interface, Ham radio Deluxe style. I also think they often associate SDR radio with the “Panadapters” waterfall displays. Well, you don’t actually need an sdr radio to have that! Of course this sort of display is software based, it has to be to show on a pc screen, but my humble FT-817 can display a waterfall when I am looking for psk31 transmissions!

    The thing that excites me most about SDR is, that the rf signal can be manipulated with software versions of filters etc. allowing updates to be applied just like updating any software we use on a PC.

    The next big thing is to take the best of the software defined advantages and keep the physical user interface “control surfaces” to work the transceivers! Much as a recording studio will have huge mixing desks attached to numerous PCs running the digital processing back end. Elecraft seem to be one of the first out of the paddock with this, I expect the big Japanese companies will follow once they lose customers!

  12. Bill Tippett W4ZV says:

    “You don’t need a degree in ergonomics to know that it is very painful to operate your radio in a 24 hour or even 48 hour contest exclusively with the mouse.”

    Absolutely correct…yet after >10 years, Flex still doesn’t get it. At least they revised their original motto “Real radios don’t have knobs.” :-)

  13. Dennis Deaton KG7UY says:

    I have used a flex 1500 for a little over a year. Love the radio but still hate using the keyboard and mouse for control! did not like the other knob solutions out there but am about to order the dj board as I like the concept, By the way I AM a computer guy who has no problem using the keyboard and mouse,I just like a dedicated user interface that has knobs. Otherwise, even an inexpensive SDR has a superior receiver in almost every way to the traditional transceivers on the market. The transmit has been harder to get right on cw but it is getting there. As a point of reference I also own Elecraft K-2, Icom 7600 which I like very much, and have operated or owned most of the currently available mid priced (4000dollar) range transceivers as well as a bunch of the budget units. Forced retirement due to health has precluded the use of the real High end stuff!! The SDR is slow to catch on due to its computer based nature, coupled with the lack of knobs to adjust the unit. I think that a lot of hams today do not like the adventure of learning something new as we have been used to plug and play since the demise of the kit manufacturers . I could just use my old T-599A/R-599A twins for the knob experience but prefer the SDR Rx with a real interface. Thanks for making this possible.

  14. “You don’t need a degree in ergonomics to know that it is very painful to operate your radio in a 24 hour or even 48 hour contest exclusively with the mouse.”

    I didn’t think I would say this a year or two ago, but now when I operate contests I use the tuning knob less and less. For doing S&P I just click to tune in signals- this is much faster than a tuning knob, plus I can do it without listening, which is MUCH less tiring when operating SO2R. This is with a traditional radio (K3) with a SDR operating on its IF- so I can choose either knobs or a mouse :) Even when I am looking for a run frequency, I usually just let the computer find one rather than tuning with the knob- it is better at locating an open frequency than I am.

    I think the problem is not just “lack of knobs”, but lack of thinking out-of-the box regarding what is possible in UI interaction.

    • Hi Tor,
      thanks for your provocative comment. You are absolutely right. There still very view fellows trying to think out of the box. I would rather prefer a software searching me a free spot in the contest than rather having to search it on the spectrum view… or even wore tuning over the band in order to find a gap. Unfortunately, the only serious, professional, industrial manufactured SDR platform sold by today is the Flex-Family. I’ve been asking for a long time to provide a proper interface. But until today it’s just a dream. When ever I’m modifying something on the DJ-Console implementation, I have to recompile PowerSDR(-UI).

  15. Sid Boyce says:

    I can hardly describe commercial rigs as having desirable and friendly user interfaces.
    Even looking at a TS450S and an IC7200 sitting in the rack I am inclined to repeat what I wrote in a Yahoo group a couple of weeks ago – God the Father, God the Son, God the knobs and switches and these are simple compared to the high end rigs I see pictures of.
    Coming back to those rigs after a long time using quisk, qsdr, ghpsdr3-alex, etc. with Softrock, UHFSDR and HiQSDR radios I was lost performing various operations.
    The commercial stuff may be more familiar once you used them a while, pretty and impressive looking but certainly not elegant.

    They look more complex than I see sitting in the cockpit of an aeroplane I fly, just not as vital.

  16. glen worstell says:

    Hi Tobias,

    I am still very interested in adding an interface for the Hercules DJ-Control MP3 E2 to the powerSDR software for the hpsdr hermes board. I want to avoid re-inventing the wheel by doing all the code necessary to interface to the DJ USB.

    It would save me a lot of work if I could get a copy of the code for your modifications to powerSDR for the Flex radios – I used your code with the Flex 1500 and was very happy to have it.

    Is there anything I can do to get the source? I’d be pleased to send back to you any changes I make.

    cheers,
    g.

  17. re. top contest stations not using sdr based radios:

    IMHO, they have no need in doing so so far!
    Typically they are being heard so loud that they can
    run QSOs in ‘running mode’ most of the time on one
    frequency. Very little need to spoil their time with S&P
    traffic. S&P is very bad for high QSO rates per hour
    (i.e. total number of QSOs).

    If they want to grab ‘Multis’, they grab them from posts
    out of a DX-Cluster via Internet connection. Easy, just
    a mouse click away!

    Just imagine what happens to them if there was no
    DX Cluster info available (for them)! Totally different
    game then!

    We had used an SDR (Flex-3000 it was) 4 years back
    in time for the EU SSB-FD for the first time ever.
    50% of the Clubs OPs initially were quite reluctant to
    use the Flex, allthough there was a 2 hour introduction
    before the contest started.

    For this reason we had installed an XYZ brand ‘normal’
    transceiver in parallel for those not wanting the SDR
    radio for their contest shift.

    Tell you what: *NOBODY* of the whole group dropped
    the Flex-3000 and switched over to the XYZ brand
    ‘normal’ transceiver. They were all amazed by the
    flexibility which the SDR radio provided by the wide
    band scope feature.

    E.g.: Switch to 28.5 MHz, watch the spectrum for
    1 or 2 minutes and you know whether the band is
    open or not, and if open, to which parts of the world.

    Just try to duplicate that with a ‘normal’ transceiver.
    No way!

    Unless you consult the insane help of a DX cluster.

    My 2 cents.

    73, Klaus, DK3QN

  18. Dave says:

    I have to agree with much if not all of the above.

    SDR systems are seen as purely computer based. But, then there is the Elecraft KX3. It’s a single box physical rig, but internally, fully SDR based. A real work of art.
    The “UI” as it were, is much like it’s cousin the K3 (also Elecraft.)

    I have a KX3 and a Kenwood TS-870s. (That uses DSP to implement much of the filter and signal processing functions, but is not an SDR as such, it is still a very nice radio.)

    There is not a lot to choose between them, on RX. If anything, the KX3 has the slight edge. On TX, obviously the Kenwood has power (100W) to it’s advantage, but the KX3 is truly portable, and runs happily at QRP levels (3W or so) on internal batteries!

    The KX3 is also firmware upgradable, and has had many features improved or added over the last couple of years.

    Now… If someone was to come up with a “front panel” in hardware, that was much like a conventional rig, but was in fact the UI to a SDR suite of programs running on a PC hiding under the desk, that would probably be the best mix of technology.

    Again, Elecraft have a K3 lookalike, that is just the front panel, but communicates over the internet with a “real” K3 elsewhere, if that could be adapted to control a SDR program, that would probably be good too.

    You could use any of the Kenwood (and clone) rigs for that, as they can be setup to send out a stream of data, relating to their settings, as you work the controls. But its not a slick as one might like, however is good enough to parallel control/tune a second rig of the same model.

    The Touch Screen idea is valid, but they have their own problems, fine tuning something via a touch screen is often hit and miss sadly, plus they get covered in mucky finger marks. Good for rugged waterproof bits of equipment (Military and the like) but I suspect most Amateur rigs have too many needed functions to make that practical in the real world. Just my opinion, others may vary! :)

    There is nothing like having “real” controls to use, even if they in turn only send commands to software that is running on something (somewhere) else. But then, you need a suprisingly fast interface, to make it all “feel” real.

    The KX3 radio seems to do this with great effect.

    Cheers.

    Dave G0WBX/G8KBV.
    NO affiliation with Elecraft, other than as a happy KX3 owner.

Trackbacks

  1. [...] this question also applies to Software Defined Radio. I already discussed the importance of the User Interface in another blogpost – therefore I’ll come back now to the DJ [...]

Speak Your Mind

*