For my experiment which was related to picking signal from a sensitive load sensor I needed a way to remove the wandering baseline. The sensor was connected to small fish and the resulting signal suffers from natural wandering baseline. Simply taking a mean of buffered data could not give an acceptable result, therefore I wrote this piece of LabVIEW code.

This is a simple multi-step process. First the mean of the buffered data points are subtracted. Then and infinite impulse response (IIR) is used to remove the noise. Then a copy of buffer is filtered with lowpass filter and it is subtracted from the main denoised buffer. The result is stable background-corrected clean signal. This is not perfect but it did solve my problem.