Digital filtering is the process of transforming a discretely sampled input signal into an output signal, such that certain spectral characteristics of the input signal are lost, while others are retained. In neuroscience, it is performed on time series that represent electrophysiological or hemodynamic signals measured over time. Whereas analog filters are applied online and implemented as electronic circuits, digital filters are applied off-line and implemented in software.
A digital filter is an important signal processing tool for the analysis of neuroscientific data. It is used to increase sensitivity to aspects of the signals that are of interest while suppressing noise. For the purpose of simplicity, we will focus on electrophysiological time series, e.g., temporal fluctuations of the electric potential measured at the scalp (for an extensive treatment of digital filters see Smith, 2003). Yet, a digital filter can be applied to any...
- Smith SW (2003) Digital signal processing: a practical guide for engineers and scientists. California Technical Publishing, San DiegoGoogle Scholar
- Mitra S (2010) Digital signal processing. McGraw-Hill Science/Engineering/Math, New York, NYGoogle Scholar
- Nitschke JB, Miller GA, Cook EW (1998) Digital filtering in EEG/ERP analysis: some technical and empirical comparisons. Behav Res Methods Instrum Comput 30(1):54–67Google Scholar
- Percival DB, Walden AT (1993) Spectral Analysis for Physical Applications. Cambridge University Press, CambridgeGoogle Scholar