Re: Fwd: Re: [Discuss-gnuradio] USB2 <-> fast ADC & DAC

[Henrique Miranda]

Alan,

>Further to this discussion, I've been wondering how difficult it is to deal
>with signals which are sampled in the even nyquist zones and thus have
>their phase reversed in the spectral sense.
>
>Is processing the signal going to be more difficult than a situation in
>which the signal is sampled in an odd zone, and is not reversed? Is there
>anything I should specifically look at to aid in processing?

>I ask because I have contacts within Panasonic here in Australia who are
>willing to supply me with Microtune 4702 DT5 tuners free of charge. These
>tuners are similar to the 4937 being used by others, but have an IF output
>at 36MHz and bandwidth of 8MHz. This signal is fully contained within the
>4th Nyquist zone when sampling at 20MSamples per second, so it should be
>viable to undersample it. My only concern is that it might be more
>difficult to process a complex signal with it's phase reversed. 
                                                                 
If the information conveyed in your IF signal is non-complex you have
nothing to worry about (for instance, FM and BPSK). Just program the NCO
of your digital downconverter to -4MHz (36-2*20 MHz) or +4 MHz. The
baseband signal will be the same in both cases.                           
                                                                 
                                                                 
On contrary, for complex-valued baseband signals (SSB and QPSK for example),
you can easily choose the intended spectral image by changing the sign
of the NCO frequency. In this case, setting +4MHz as the NCO frequency would
select the non-inverted spectrum. Equivalently, you can change the sign of
imaginary part of the baseband signal at the output of the downconverter
(conjugate operation).

So, like Matt stated, "it is just a matter of flipping a plus sign to a
minus sign."

Cheers

Henrique



On Tue, 3 Jun 2003, Alan Gray wrote:

> Further to this discussion, I've been wondering how difficult it is to deal 
> with signals which are sampled in the even nyquist zones and thus have 
> their phase reversed in the spectral sense.
> 
> Is processing the signal going to be more difficult than a situation in 
> which the signal is sampled in an odd zone, and is not reversed? Is there 
> anything I should specifically look at to aid in processing?
> 
> I ask because I have contacts within Panasonic here in Australia who are 
> willing to supply me with Microtune 4702 DT5 tuners free of charge. These 
> tuners are similar to the 4937 being used by others, but have an IF output 
> at 36MHz and bandwidth of 8MHz. This signal is fully contained within the 
> 4th Nyquist zone when sampling at 20MSamples per second, so it should be 
> viable to undersample it. My only concern is that it might be more 
> difficult to process a complex signal with it's phase reversed.
> 
> I have over 140 of these tuners sitting in a box at University, Panasonic 
> evidently bought a bunch of them and have no further use for the leftover 
> units and are looking to off load them to a good cause, so there is some 
> incentive for finding a good use for them. If these are in any why useful 
> for this project, let me know, apparently they've got another 500 or so 
> that they're looking to get rid of.
> 
> As always, any thoughts or suggestions are welcome
> Alan
> 
> 
> 
> >At 04:13 PM 30/05/03 -0600, you wrote:
> >
> >>At 3:07 AM +0100 5/30/2003, Henrique Miranda wrote:
> >> >Joseph,
> >> >
> >> >After some calculations I came with the following procedure:
> >> >
> >> >1: The Nyquist region can be obtained using n = floor(fc/B + 1/2)
> >> >   where floor(x) rounds x towards minus infinity, fc is the
> >> >   IF center frequency and B the signal bandwidth.
> >> >
> >> >2: The minimum fs can then be easily obtained by fs = (2*fc+B)/n
> >> >
> >> >3: If the optimum fs is intended use fs = 4*fc/(2*n-1) instead.
> >> >
> >> >Moreover, if (fc/B + 1/2) is an integer, then minimum fs equals
> >> >the optimum fs.
> >> >
> >> >If you find some inconsistency in this procedure please let me know.
> >> >Henrique
> >>
> >>Dear Henrique,
> >>
> >>No inconsistencies found. I had to brush up on my algebra of inequalities 
> >>(aarghh!) but I agree with your equations #1 & #2. Nicely done and thank 
> >>you again.
> >>
> >>However, I must admit that I still don't fully appreciate what qualities 
> >>make a particular fs "optimum".
> >>
> >>Best regards,
> >>Joseph
> >>--
> >>Joseph A. DiVerdi, Ph.D., M.B.A.
> >>http://xtrsystems.com/          970.980.5868 (voice)
> >>PGP Key ID: 0xD50A9E33
> >>
> >>
> >>_______________________________________________
> >>Discuss-gnuradio mailing list
> >>address@hidden
> >>http://mail.gnu.org/mailman/listinfo/discuss-gnuradio
>