Spectral analysis is used to discover and quantify signals that change periodically; this includes oscillations in EEG or LFP signals, but also the periodic BOLD signal changes that result from the regular alternation of preferred and non-preferred stimuli.
At the end of this two-week Module you will be able to:
- Explain the concepts amplitude spectrum and phase spectrum
- Calculate the spectrum of any time series
- Filter time series, determine how spectral content changes over time, and analyze coherence
- Study the
docs\spectralAnalysis.mlxfile. Make sure that you understand each line of code, play around with the various parameters, and run the code in the debugger so that you can step through the code while inspecting the contents of the various variables.
- Read this document for some background on aliasing.
- Learn about filtering, convolution, spectrograms, and coherence in
docs\spectralAnalysisII.mlx.