At 08:02 AM 8/5/2008, Edwin Marzan wrote: >Greetings Gents, > >As I am not as tech saavy as most of the gents participating in this >discussion. I was wondering if someone could explain the benefits of >"dedicated I/Q settings per band". I operate my radio daily and it >seems to work fine without this feature. I'm all for making the >SDR1000 perform as best as is possible but I've never heard anyone >out there say that I would sound better or hear better "if only" I >had this feature. >
Potentially, you'd get better image and carrier rejection for both Tx and Rx. Right now, the radio easily gets better than 30dB rejection with a single cal point (and usually better than 40dB). A typical image rejection in the neighborhood of the cal frequency is probably 60-70 dB (which puts the image down in the noise, most of the time) However, because the radio has a nifty spectrum analyzer feature, you can SEE that -40dBc spurious signal (even if it had no practical user effect). And, the image is only 20 or so kHz away (as opposed to "out of band" for the usual ham receiver), so if you have a strong signal, you can tune the image to the signal and receive it. Other radios with low IF as their last IF (e.g. the IC7000) don't have this problem, because they have a conventional superhet architecture, and the image is tens of MHz away from the desired signal. >Are there any other radios besides the Flex 5000 that have this feature? > > For a whole raft of experimenter's SDRs, there are varying implementations of the I/Q balance thing. Inside commercial radios, especially those that have a conventional mixer based front end, I/Q is fairly rare. Instead, they do what's called IF sampling (i.e. you bring the signal down to a relatively low IF, and sample at, say, 4*Fif, then mathematically transform the samples into I/Q pairs. That way, you don't have any channel matching/balance issues. (actually, the IF usually winds up being close to an odd multiple of the Fsample/4). As long as you can buy an A/D with enough bandwidth and bit depth for your signal, then using a single converter is a good trade. On the transmit side, I/Q is popular in wideband/high speed radios (because there are inexpensive monolithic vector modulators), and for them, calibration is needed. The hassle of calibration is less than dealing with the need to do linear upconversion. You can do your carrier synthesis with saturated components(improving power efficiency), then put the modulation on at the end. As an example, you can google "Electra UHF" and find info on a SDR that is currently orbiting around Mars that uses these techniques. Jim, W6RMK http://trs-new.jpl.nasa.gov/dspace/bitstream/2014/7832/1/03-2150.pdf is a short article http://descanso.jpl.nasa.gov/Monograph/mono.cfm?force_external=0 is a whole book (look at the end for Volume 9) about SDR for deep space. Chapter 2 describes the radio http://descanso.jpl.nasa.gov/Monograph/series9/Descanso9_02.pdf _______________________________________________ FlexRadio Systems Mailing List FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archives: http://www.mail-archive.com/flexradio%40flex-radio.biz/ Knowledge Base: http://kb.flex-radio.com/ Homepage: http://www.flex-radio.com/
