Software-Defined Radio (SDR) is a very big deal. This is one of the few areas about which I actually knew a little bit, as we were using some very expensive (complicated, confusing, poorly-interfaced, and worst of all, fragile) SDR equipment in Afghanistan in 2006.
The general idea for the layman (in which category I still and probably always will consider myself) is that because of awesome advances in technology over the last few decades, you no longer need purpose-built hardware to do very specific things in radio, but can build general purpose, flexible radio hardware and specify the parameters for your specific use in software, instead.
To use an analogy, playing video games used to require the purchase of arcade game cabinets. And then game consoles. And now you can play a video game on darned near anything. The hardware (e.g. your PC, your phone, your tablet) is general purpose, and the specific application is all done in software. (This is an apt analogy for another reason I’ll get to in a moment.)
Northwest Digital Radio had several representatives at TAPR DCC this year to present on various aspects of the field. John D. Hays, K7VE, spoke on considerations for designing dynamic web interfaces for digital 2-way radio. Once the radio is done up in software, there’s no reason not to remove the presentation layer (i.e. the buttons and knobs and speaker) from the box with the fans and the widgets and the magic blue smoke. He talked cogently about some of the general considerations involved, including the important need to consider your audience (or market). More and more, people expect to operate portions of their lives on portable devices, and the interface needs to reflect that.
Reminiscing for a second here.
I wish someone had thought in those terms for the equipment we had to use. The system (which will go unnamed, sorry) used a network interface exclusively; you talked to it with a Panasonic ToughBook. That was great so long as the network interface came up. The “military hardened” system got rained on once (not even hard, or for long) and it never reliably booted up again. You can’t talk to a system (even to troubleshoot) if the network stack never comes up.
The second issue was that ToughBook. It was tough, but this was Afghanistan. Over the course of the year, we utterly destroyed that screen. Touch-screen interfaces do not mix with sand and war. Note to designers.
Even without that consideration, the interface was clearly designed by and for someone using a large, high resolution monitor–not a 12″ laptop in direct sunlight, possibly wearing gloves and getting mortared. Bigger on-screen widgets, please.
Back to the story.
Anyway, I enjoyed it and am terribly glad that someone is really thinking about these things.
Bryan Hoyer, K7UDR, spoke on Friday afternoon about SDR Architecture for VHF/UHF. He was a fairly dynamic speaker and made a lot of “big picture” statements, not talking just specifics about NWDigitalRadio’s UDR56k transceiver project (which is not quite the Duke Nukem Forever of the SDR world in terms of timely shipping), but also where the state needs to go. He was very quotable.
The great thing about SDR is it represents a commitment by the company not to do anything. . . A purpose-built thing is always less expensive than a universal thing. I know two things about Swiss Army Knives: they are relatively expensive, and they aren’t very good knives.
This, too, is true in the video game analogy. Your tablet is a decent general purpose device that does games, but it’s not a Playstation 3. Or whatever number they’re up to now. Purpose-built hardware will always be better for a specific purpose–if you can afford to buy hardware for each purpose you have. I certainly can’t; my new HF rig (which I now realize I never talked about here) was specifically designed to do what I wanted most well, and everything else good enough.
On the proliferation of digital modes in fldigi:
It’s either a modem for each propagation [mode] or a modem for each person that uses it.
(Maybe you had to be there. I laughed.)
On not supporting certain legacy digital modes:
We’re not going to waste time with 1200 baud. Because I’m really trying to kill it.
His strongest, most passionate points were on the need not to develop “point solution” equivalents to other services (e.g. APRS vice Twitter, Winlink vice e-mail), but to build the framework for the network. It’s all about the network. I tend to agree: if you build the framework, people will figure out clever things to do with it. And we ought to be thinking bigger than 1200 baud.
…which is part of why the Yaesu product announcement was such a let down. More thoughts to follow on that.
Incidentally, before this week I’d never heard of NW Digital Radio or their UDR56k. Before yesterday, I didn’t really understand it. After yesterday, I feel like a complete failure as a human being for not already having pre-ordered one.