It strikes me, when implementing algorithms to compute Periodograms using DFT/FFTs, there are tradeoffs in variance and spectral resolution. There are a number of factors, including window/FFT length, window function, amount of window overlap, degree of zero-padding, number of estimates to average, choice of normalization factors, etc. Then you've got a lot of considerations with respect to how that spectrogram is displayed on screen that can influence the visual detection problem. We haven't even addressed computational complexity / speed of computation.
I do not know how much flexibility the P3 allows for changing a number of these parameters, the manual indicates spectral resolution is changed with "span", so if everything is automatically selected for the end-user at this release of the software, perhaps some configuration options might be placed back into the hands of the user in a subsequent release, including a guide provided to help one understand these tradeoffs in PSD estimation and presentation and how to configure the display for optimal results depending on what you want to do. The manual is not very verbose in this regard. With sampling rate known, and these other parameters chosen, conceptually, the software can report an estimate of the spectral resolution which could be theoretically expected on a popup information window or other. This is one argument for having an "external", potentially more "open to adjustment" system for analysis of the IF information, rather than say a P3. KJ4VPI ________________________________ From: Alan Bloom <[email protected]> To: Larry K1UO <[email protected]> Cc: [email protected] Sent: Tue, September 14, 2010 1:00:30 PM Subject: Re: [Elecraft] LP Pan versus P3 Panadapters Hi Larry, It depends on the span. Unlike some other panadapters, the P3 automatically reduces the sample rate as you narrow the span, which improves the sensitivity. For example, I have my HP8656B signal generator set to 3640 kHz with the amplitude as low as it will go (-127 dBm). The K3 has the preamp off, the attenuator on, and default CW shift and width. Under those conditions I can just barely hear the carrier. It's weak enough that if it were a modulated CW signal I don't think I could copy it. However the carrier is clearly visible on the waterfall if I set the P3 span to 2 kHz. As I increase the span the signal gradually gets harder and harder to see. Above about 30 kHz or so the signal disappears into the noise. So that's the trick. Narrow the span to make those signals pop out of the noise. 73, Alan N1AL On Tue, 2010-09-14 at 06:03 -0700, Larry K1UO wrote: > Bill, > > I can tell you that just last night, on 160 meters, I copied 2 very weak > stations perfectly Q5 that did not show on the P3 waterfall. Now, Its > entirely possible its my adjustment skills but I dont think so. Yes..the K3 > IF Mod was factory installed. This has happened more than once. For > some reason I was expecting the P3 to show these very weak signals better > than I could find them myself while tuning but no amount of adjustment so > far will do this. I may be expecting too much? > Just so you know what anothers experience has been so far. > 73 > > > > > I too would like to see details of the improvement over LPan etc. I am a > long time user of SDR/computer panadapters for use in detecting very weak > signals at UHF/VHF/microwave. I use a Softrock on the K3 IF port since I got > my K3 a year ago. I have yet to see one report regarding weak sig detection > with the P3 after many requests to the initial testers of the P3. ANYONE out > there looking for weak weak sigs with a P3?? > > 73 Bill K0AWU EN37ed > > ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[email protected] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[email protected] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html
