Main Content

Machine Learning and Deep Learning for Signals

Signal labeling, feature engineering, dataset generation

Signal Processing Toolbox™ provides functionality to perform signal labeling, feature engineering, and dataset generation for machine learning and deep learning workflows.

Apps

Signal AnalyzerVisualize and compare multiple signals and spectra
Signal LabelerLabel signal attributes, regions, and points of interest, and extract features
EDF File AnalyzerView EDF or EDF+ files

Functions

expand all

labeledSignalSetCreate labeled signal set
signalLabelDefinitionCreate signal label definition
countlabelsCount number of unique labels
filenames2labelsGet list of labels from filenames
folders2labelsGet list of labels from folder names
splitlabelsFind indices to split labels according to specified proportions
signalMaskModify and convert signal masks and extract signal regions of interest
binmask2sigroiConvert binary mask to matrix of ROI limits
extendsigroiExtend signal regions of interest to left and right
extractsigroiExtract signal regions of interest
mergesigroiMerge signal regions of interest
removesigroiRemove signal regions of interest
shortensigroiShorten signal regions of interest from left and right
sigroi2binmaskConvert matrix of ROI limits to binary mask
edfinfoGet information about EDF/EDF+ file
edfwriteCreate or modify EDF or EDF+ file
edfheaderCreate header structure for EDF or EDF+ file
edfreadRead data from EDF/EDF+ file
signalDatastoreDatastore for collection of signals
dlstftDeep learning short-time Fourier transform
stftLayerShort-time Fourier transform layer
findchangeptsFind abrupt changes in signal
findpeaksFind local maxima
findsignalFind signal location using similarity search
fsstFourier synchrosqueezed transform
instbwEstimate instantaneous bandwidth
instfreqEstimate instantaneous frequency
pentropySpectral entropy of signal
periodogramPeriodogram power spectral density estimate
pkurtosisSpectral kurtosis from signal or spectrogram
powerbwPower bandwidth
pspectrumAnalyze signals in the frequency and time-frequency domains
pwelchWelch’s power spectral density estimate
signalFrequencyFeatureExtractorStreamline signal frequency feature extraction
signalTimeFeatureExtractorStreamline signal time feature extraction
zerocrossrateZero-crossing rate

Topics

Related Information