Optimum Digital Filters for Baseline Restoration

We present a general method to directly calculate the digital zero-area filters to achieve the maximum possible signal-to-noise ratio (SNR), in presence of any additive stationary noise. This method is applied to two cases, and, can also be easily expanded to other situations. The synthesis results of a typical example are given and fully discussed, relations between SNR and other factors such as the filter length and noise spectrum are investigated. A priori knowledge of the noise power spectrum is not required, so it is suitable to be used in adaptive, self-calibrating digital spectroscopy systems