This page is circa 2002. The first screen capture is of the W3CCX beacon here on a night with a lot of pulse noise, and the K2 Noise Blanker turned on. Just focus on the spectrum display at the top of the graph, not at the waterfall, below. The waterfall is essentially a history of all of the playing around I had done for the prior 10 minutes (time lines on the left). Note that the noise floor is about 2-3 dB on the scale and the W3CCX beacon at about 665 Hz reaches the 40 dB mark.
The second screen capture (immediately beneath this caption) was actually obtained just before the first one on this page, and is identical except that the K2 Noise Blanker has been turned off. Notice on the spectral display at the top of the display that the W3CCX beacon is still peaking at 40 dB, but the noise floor has come up to 20 dB, rising by about 17-18 dB. This is impressive! Now, if you flip back and forth between the first and second displays, and focus on the bottom of the waterfall displays, you will see how the waterfall is whited/yellowed out at the bottom of the second display, below, but that the signal is in the clear on a black background on the first display, above. There's no doubt here at W3SZ when the pulse noise starts! But its fun to play these electronic tricks and eliminate it...I'd actually been hoping for a night like last night when I took these pix to test everything out!
The next DSP-10 display (below) was taken the next day, of W5UN at 144.040 off the moon. The audio gains for the entire audio chain have been changed, so the displays are not directly comparable (Dave has a great signal, but its not within 10 dB of the local beacon, although it sometimes seems like it!!) The signal at 600 Hz is Dave, peaking at just under 30 dB (28.6 dB, to be exact) on the scale, or 15 dB above the noise. The other signals are birdies.
The next screen shot, below, shows the same beacon as the first DSP-10 screen shot. Note that the Spectran spectrum is missing details like dB markers that DSP-10 has. In addition, the spectrum cannot be scaled nicely like the DS10 spectra in terms of amplitude range (DSP-10 amplitude range can be set anywhere from 70 dB full scale range all the way down to 7 dB full scale range, and the baseline is independently adjustable). There is some scaling possible with Spectran, but it is rudimentary by comparison. One advantage of Spectran's spectral display is that it does a good job of automatically compensating for baseline/filter roll that might be present in the signal coming out of the analog rig. With DSP-10, this compensation can be done, but its not automatic and sometimes the software seems to get confused when one changes scales after normalizing in this manner. On the other hand, Spectran's waterfall display is far more to my liking than that of the DSP-10; details and nuances of weak signals are more easily seen, without the need for careful, sometimes tedious adjustments to optimize the waterfall as with DSP-10.
Below is SM5FRH via EME through the Elecraft K2 and LT2S transverter, as displayed by Spectran. The waterfall signal at 600 Hz is Tobbe. There is no corresponding signal on the spectrum because I captured the screen while he was listening. Look at the top of the waterfall and you'll see this. The other signals are birdies.
The screen below didn't do very well in the transition from bmp to jpg filetype, but you can see the essentials of a DSP Blaster screen captured as I was listening to F3VS off the moon. Across the top is the block diagram of the signal path in software. You put the mouse over a block and click to select it and adjust its parameters. Here I have the LMS Noise Reduction and the Coherent BandPass Filter activated. The cirsular display is of F3VS's signal in binaural mode, the 'pseudo-stereo' mentioned in the text. The text in the center is to set certain parameters, activate TSR mode, etc. The designations M1-M9 are to save user-defined filter combinations for future use. On the bottom left is a spectral display, with yellow realtime and blue averaged. On the right is a time domain realtime display. This program is the workhorse here at W3SZ for EME work. I keep trying new programs but always seem to end up back at DSP Blaster, because it just does a better job of processing the audio.
The Screen below is of SM5BSZ's Linux DSP work in progress, a receiver largely in software. The waterfall display at the top is of W3CCX. Below is the spectral display of the same signal. This was acquired using a homebrew computer-controlled receiver with 24 KHz-wide IF in front of SM5BSZ's DSP program, permitting spectral display of a 24 KHz slice of spectrum. This looks different from the DSP-10 and Spectran outputs because of this wide-band view, which I view as a GREAT advance. This program requires calibration with a pulse generator to normalize for the filter contour of the input signal, to minimize baseline roll due to filter and receiver characteristics which may vary over the 24 KHz bandwidth. In the waterfall the W3CCX beacon signal is the strong yellow/orange signal just below 15620 Hz. Below the waterfall in the main display is the green cursor line that was placed over the beacon's signal in the spectral display when I left clicked on it to 'zoom' it. Below the main display is a smaller high resolution or'baseband' display focused on the signal of interest, the W3CCX beacon. Beneath that is the DSP lilter/noise blanker display. NOTE that all parameters are mouse-click adjustable: the bandwidth of the waterfall and spectral displays, the zero level and signal level range of the spectral display, the bandwidth and signal level range of the 'baseband' or zoomed spectral display, the noise blanker characteristics, and the filter characteristics. To center this zoomed display you just left click with the mouse on the signal of interest on the main display. This promises to be a very nice package! Go to here for more info on this website!!
Finally, here (below) is another Screen Shot from Leif's program, this time with the same homebrew receiver plunked down in the CW portion of the 20 meter band. Again one signal has been selected by a left-click of the mouse and is therefore shown beneath a green vertical cursor line on the spectral display, and then this signal is zoomed in the smaller spectral displays below. I have switched the positioin of the high resolution and DSP filter / noise blanker displays on this screen, so I could expand the high resolution display to the maximum screen width. Note the tremendous amount of information available in the waterfall display at the top of the screen about signal activity over time in this 24 KHz slice of the 20 meter band. This program is INCREDIBLE! For more information go here
Brought to you by the folks at W3SZ