At 02:28 PM 1/23/01 -0000, you wrote:
Regarding I/Q sampling (is this the same as the old 11+
selection, hi?) and
filtering and smoothing to extract the phase, can we have a
program please,
so that we can join in the fun?
You will need the hardware, something like a 56002EVM that I use to do
all the high speed ( ! ) DSP
Sorry !
Unless someone can be persuaded to write something similar for the
soundcard.
If 8 bit sampling is good enough - it usually is if no strong signals
appear in the IF passband that exceed the noise by more than 30dB - one
of the sampling options on my PIC based interface serial A/D will I/Q
sample a 750 - 1250 bandwidth centred on 1kHz and a PC can do the rest.
A commercial version is available from HF-Instruments
WWW.HF-INST.CO.UK
NIAGARA can do it using a Sigma-Delta interface (available from me)
or it can use a sound card if it's based on the ES688 chip. At this
moment I haven't coded for SB16-type sound cards because the one I'm
using in my laptop is an ES688. I'll put the command into NIAGARA today
to output the same (.wjs) file format Andy uses. Is that OK with
you Andy? With NIAGARA using its 1000x magnification mode you get
an effective sampling rate of 8.2 samples per second - and bin widths
similar to what we've been playing with. And it now has a screen capture
to write a colour .BMP output file.
Saving raw data overcomes the problems of setting wrong colour
thresholds, FFT sizes and the like and means you can go back and replot,
changing any parameter at will. Saving two 16 bit numbers per sample
means 24 hours recording at 3.9Hz (8000Hz / 2048) generates a file of
just 1.35 Meg.
Again, NIAGARA now has the capability to change the threshold settings
and re-compute the entire screen so you don't have to wait another 6
hours for another screenful to come in before you see if you've chosen
the right setting, hi! And it can process pre-recorded raw audio.
I have several programs here that can save raw audio, you're welcome to
try any of them.
Bill VE2IQ
| 