The effect of varying the smoothing data window in the corrected dimension

For optimium viewing please resize your browser window so that the two plots (full spectrum and fingerprint details) display side-by-side.

The raw data

For more info on the raw data, see the main page.

Facelift parameters: 32 2.0 32 5

Facelift parameters: 64 2.0 64 5

It would seem as if going to a window of 64 data points we retain more weak signals and peaks look a bit more natural. The artifact level stays pretty much put. The execution time does go up by ~60%, though. (13 min vs. 20 min on my Indy R4600PC 100 MHz)

More experiments:
• The main page
• The effect of varying the smoothing data window in the orthogonal dimension
• The effect of varying the standard deviation