domain signal to frequency domain can not carry that out since
critical phase information is missing when you do it that way. By
offsetting the signals by 15KHz and not using a Quadrature Detector
you are depending on the hardware to kill the images that are 30 KHz
in the opposite direction.
At 02:05 AM 4/23/2006, you wrote:
>The STAR generates it's I-Q in software and uses that for sideband
>rejection ... so does the RX320 which cannot do sideband in hardware
>due to a hardware 1st LO with 2.5Khz steps using a 10Khz filter (main
>purpose is image rejection not sideband rejection) bandwidth and fixed
>2nd and 3rd LOs it is impossible to to place the sidebands on the
>filter slope to filter out the opposite sideband, not to mention there
>is no way to select the sideband you want because the 1st, 2nd and 3rd
>LOs are the same whether you are in AM CW USB or LSB modes ... If you
>measure the opposite sideband rejection using the 320's 12Khz output
>and Alberto's program you'll find it's far less than 30 db but with
>the I-Q generation in the built-in software the 320's opposite
>sideband rejection exceeds 60 db ... no other way to explain that
>other than the 320 does it's I-Q generation and sideband rejection in
>software (Plus a couple of Ten-Tec engineers told me about the I-Q
>generation in software which they've been doing since the early 90's
>in their commercial DSP receivers) .... it's also why neither the
>STAR or the RX320 use a zero IF but actually use a 12Khz or 15Khz IF,
>I-Q generation in software won't work with a zero IF from what I've
>been told ... If I remember right Bob Larkin's DSP-10 2 Meter
>transceiver generated it's I-Q in software also
>
>You can see the STAR's software block diagram in the 2nd part of the
>series and it clearly shows I-Q generation in software .... the key is
>they mix to baseband in software not hardware and the generation of a
>15Khz I-Q lo and the mixers in software isn't that tough ... from
>there the software is essentially like the hardware I-Q versions
>
>You can join the STAR Yahoo group and all the details are in the files
>section, including the DSP source code
>
>JR
>
>
>--- In [email protected], "jabauzit" <[EMAIL PROTECTED]> wrote:
> >
> > --- In [email protected], "i2phd" <i2phd@> wrote:
> > >
> > > There is no way out, unless the
> > > external hardware operates in quadrature, no software magics can
> > > recreate those I/Q components that haven't been generated by the
> > > hardware, and the two sidebands are by now irremediably mixed
>together.
> > >
> > If I understand you correctly, a TRX like the Pic-A-Star which
> > extracts the I/Q components in the DSP works because the image has
> > been removed by the preceding stages and the I/Q channels are used
> > only for demodulation, not image rejection.
> >
> > Jean-Claude PJ2BVU
> >
>
>
>
>
>
>
>
>Yahoo! Groups Links
>
>
>
>
Cecil Bayona
KD5NWA
www.qrpradio.com
"Windows the worlds most successful software virus"
YAHOO! GROUPS LINKS
- Visit your group "soft_radio" on the web.
- To unsubscribe from this group, send an email to:
[EMAIL PROTECTED]
- Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
