[hpsdr] SDR Architectures

Doug Adams douglas.adams at me.com
Sat May 24 11:17:56 PDT 2014


Phil & All,

Very good article, lots of great ideas.

As a relative newcomer to HPSDR and an Apple Mac-centric person it has become very evident to me that the software is dividing us. Since Sdr is so dependent on the "S" (software) it is a great shame that so much good work is done on the various software platforms (Windows, OS X, Linux, Ubuntu, etc.) and yet little of it is shareable. Many of us with programming skills and the urge to create something for SDR are unwilling to take the time to learn a new Language and/or Development Environment. In most cases that effort could take months or years.

My point is that your SDR Server approach could mitigate this issue. Its interface, ethernet (or whatever replaces it), is inherently multi-platform therefore a Mac developer should be able to use the server just as easily as a Windows developer. That's a first step and could make a great deal of functionality available to people who would otherwise never have access to it.

I would suggest that you could go one step further. Some of us are especially interested in a particular area of SDR; demodulation, filtering, audio processing, digital modes, etc.. The SDR server approach could accommodate multiple servers each focused on a specialized topic. They could even be physically different hardware based on different operating systems and developed with the tools that are in the skill set of its developer. Each server would be a "Black Box", much like buying an integrated circuit. Over time they could be upgraded or even replaced with newer, better, faster Black Boxes.

To push the thought again, suppose we developed an agreed upon protocol (like the Open HPSDR protocol) not only for servers but also for server components. If, for example, I wanted to try using the AM Demodulator from cuSDR64 in my Mac-based SDR project, today this would be very tedious if not impossible. If we could define a protocol for accessing the individual components of an SDR server (and if cuSDR64's AM Demodulator ran on a conforming server) it would be doable. All of the best algorithms/software would now be available to developers regardless of the platform they chose for their own development efforts.

I'm aware that I've just described tens or hundreds of man-years of work but we need to start or we'll never get there. If we make it possible to share our work, regardless of software platform, we might get there sooner rather than later.

73’s 
Doug - K3TZR

PS - I just took delivery of an ANAN-100 so could we get this done soon? :-)

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/attachments/20140524/947e63cc/attachment-0002.htm>


More information about the Hpsdr mailing list