Schulze2005

CREATED: 200906170310 LINK: url:~/Modules/Literature/Schulze2005.pdf $y = (s + b)*p + n$, where

• y is the observed signal.
• s is the true signal
• b is the background
• p is a blurring function (due to the instrument)
• n is the noise

The main problem is to compute $s = (y - n)*p^{-1} - b$

Surveys a number of different methods for baseline correction, categorize them according to how they deal with the unknowns $b, p, n$.