[hpsdr] ozy/janus as soundcard for digital modes

[Steve Bunch]

Stephen,

On Nov 3, 2011, at 8:27 AM, Stephen West-Fisher wrote:

> ***** High Performance Software Defined Radio Discussion List *****
> 
> No need to shout, and actually I thought I was quite clear in my response to
> Phil that the TS-480 only has analog audio out and in. My understanding of
> the HPSDR project is that it is for experimentation and learning, so I'm
> doing both.
> 
> So you are saying that I/Q data can go into an analog audio input but an
> analog audio output cannot go into an I/Q data port?

The audio coming out of the TS-480 is "real" audio -- that is, it's simply the audio waveform.  When you sample it with an A/D converter at some sample rate S, you get a stream of digital samples at that rate, representing a sampled version of the waveform.  Because of the way digital sampling works, you can only represent signals in the original audio that are at the frequency S/2 or lower.

I/Q audio is two audio streams - one is the audio waveform, same as before, the other is the 90-degree (at the sample rate S) delayed version of it.  When you sample these two at the same instant, you get two digital samples, one of them delayed from the other by 90 degrees.  Because you now have more information than you got from sampling one real stream (think of it this way: after the sample of the real stream, the next sample tells you, in essence, whether the waveform went up or down after that, and how far), your digital samples can now represent signals up to frequency S.

If you connect a real audio signal to the I (In-phase) input of a digital radio that is expecting I/Q signals, and nothing to the Q (Quadrature) input, half your samples will still be meaningful.  The Quadrature half of the samples will be DC.  So it's as if the signal was quickly returning to zero immediately after each I-side sample. The side-effects of this will get suppressed when you low-pass filter the resulting digital samples, so you will indeed see signals from sampling your audio stream as the In-phase channel.  It won't be quite right, since the demodulation process will think it's able to reconstruct signals up to frequency S, but you've only sampled the real signal often enough to go up to S/2.  So you will have created some extra junk (at least one full image that isn't really there, for example), but it will do something semi-meaningful.  If the samples have been filtered before demodulation, the signals you're interested in may well have been reconstructed to the point of usability.  Transmitting is probably inadvisable, as you may be generating unwanted images that could be in the audio passband and not be filtered out.

So continue playing with it as you read up on digital signal processing, and have fun!

Steve, K9SRB

> Since I was able to
> decode PSK-32 doing just that I'm not sure I completely agree. I do know the
> panadapter display looks quite odd, and I have not done any comparison of
> how well it works compared to just hooking up to a standard soundcard.
> Performance may be quite marginal, but I did get a positive response.
> 
> So then, since Janus was designed as an A/D and D/A adapter (with the TI
> TLV320AIC23B) and has non I/Q inputs and outputs so it appears to me the
> hardware should be able to handle the appropriate audio to digital
> conversion. If the hardware is capable, then the software would be the
> problem and Phil's response to me leads to that conclusion (windows driver).
> If there is a problem with this statement, please let me know where I'm
> going wrong.
> 
> At this point the clue I am looking for is where the problem lies, Janus
> firmware or PowerSDR code?
> 
> --
> Stephen West-Fisher
> N4IK
> 
> 
> -----Original Message-----
> From: hpsdr-bounces at lists.openhpsdr.org
> [mailto:hpsdr-bounces at lists.openhpsdr.org] On Behalf Of george byrkit
> Sent: Thursday, November 03, 2011 8:14 AM
> To: hpsdr at lists.openhpsdr.org
> Subject: Re: [hpsdr] ozy/janus as soundcard for digital modes
> 
> ***** High Performance Software Defined Radio Discussion List *****
> 
> Dear Stephen,
> The 'normal sound card' is used to receive I/Q data from the SDR-1000, which
> is then processed by PowerSDR.
> 
> Janus has better resolution and more stable timing than most sound cards,
> and the USB conntection thru Ozy has much less latency (delay) than a normal
> sound card would.
> 
> So UNLESS YOUR TS480 OUTPUTS I/Q data (which you have not yet confirmed or
> denied), PowerSDR WILL NOT WORK WITH IT.
> 
> 73,
> George K9TRV
> 
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at openhpsdr.org
> Subscription help:
> http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
> HPSDR web page: http://openhpsdr.org
> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/
> 
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at openhpsdr.org
> Subscription help: http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
> HPSDR web page: http://openhpsdr.org
> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/


 1320332209.0