![]() ![]() "Number of tapers" is the number of DPSS tapers to be used to compute the spectrum. The bandwidth of the main lobe is also called the frequency resolution because it dictates the minimum difference in frequency that can be detected. taper_params: The time-half bandwidth product (TW) can be computed as N*(BW/2) where N is the length of the window (seconds) and BW is the bandwidth of the main lobe.frequency_range: Range of frequencies (Hz) across which to compute the spectrum.Fs: Frequency at which the data was sampled in Hz.The spectral parameters used in all implementations of the multitaper spectrogram are described here. See the R implementation folder for usage information and other details of the Matlab implementation.multitaper_spectrogram_R.R: baseline implementation in R with option for multiprocessing. ![]() See the python implementation folder for usage information and other details of the Matlab implementation.requirements.txt: contains names and versions of non-standard library Python packages required to run multitaper_spectrogram_python.py.multitaper_spectrogram_python.py: baseline implementation in Python with option for multiprocessing.See the matlab implementation folder for usage information and other details of the Matlab implementation.Data precision is reduced from double to single for major speed improvements. multitaper_spectrogram_mex.m: optimized implementation in C called from Matlab.multitaper_spectrogram.m: baseline parallelized implementation in Matlab.Sleep Neurophysiological Dynamics Through the Lens of Multitaper Spectral Analysis. Prerau MJ, Bianchi MT, Brown RE, Ellenbogen JM, Patrick PL. Because DPSS tapers are uncorrelated with each other, they can be averaged together as if they were independent trials of the same condition, producing a spectrum with reduced variance compared to periodogram and single-taper estimation.įind videos describing the theory of spectral estimation and demonstrating how multitaper spectral estimation works at on the Prerau Lab website. The innovation of the multitaper method is that, instead of using a single-taper function to compute the spectrum, it uses multiple taper functions called discrete prolate spheroidal sequences (DPSS). The multitaper method works by averaging together multiple independent spectra estimated from a single segment of data. Multitaper spectral estimation was developed in the early 1980s by David Thomson 2 and has been shown to have superior statistical properties compared with single-taper spectral estimates 3,4. This repository contains Matlab, Python, and R implementations of the multitaper spectrogram analysis described in the paper "Sleep Neurophysiological Dynamics Through the Lens of Multitaper Spectral Analysis" 1. Numerical Differences Between Implementations.Prerau Lab Multitaper Spectrogram Code Matlab, Python, and R implementations ![]()
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |