Preprocessing and Feature Extraction

Extract signal features in time, frequency, and time-frequency domains

Extract meaningful signal features using command-line functions or the Signal Feature Extractor app to use in machine learning models.

Apps

EDF File Analyzer	View EDF or EDF+ files (Since R2021a)
Signal Analyzer	Visualize and compare multiple signals and spectra
Signal Labeler	Label signal attributes, regions, and points of interest
Signal Feature Extractor	Extract and analyze signal features (Since R2025a)

`detrend`	Remove polynomial trend
`envelope`	Signal envelope
`lowpass`	Lowpass-filter signals
`smoothdata`	Smooth noisy data
`cwtmag2sig`	Signal reconstruction from CWT magnitude (Since R2023b)
`stftmag2sig`	Signal reconstruction from STFT magnitude (Since R2020b)
`islocalmax`	Find local maxima
`findchangepts`	Find abrupt changes in signal
`findpeaks`	Find local maxima
`refinepeaks`	Refine peak value and location estimates (Since R2024b)
`zerocrossrate`	Zero-crossing rate (Since R2021b)

`audioDatastore`	Datastore for collection of audio files
`arrayDatastore`	Datastore for in-memory data (Since R2020b)
`imageDatastore`	Datastore for image data
`signalDatastore`	Datastore for collection of signals
`edfheader`	Create header structure for EDF or EDF+ file (Since R2021a)
`edfinfo`	Get information about EDF/EDF+ file (Since R2020b)
`edfread`	Read data from EDF/EDF+ file (Since R2020b)
`edfwrite`	Create or modify EDF or EDF+ file (Since R2021a)
`paddata`	Pad data by adding elements (Since R2023b)
`resize`	Resize data by adding or removing elements (Since R2023b)
`trimdata`	Trim data by removing elements (Since R2023b)

`mean`	Average or mean value of array
`rms`	Root mean square value
`std`	Standard deviation
`snr`	Signal-to-noise ratio
`thd`	Total harmonic distortion
`sinad`	Signal to noise and distortion ratio
`peak2rms`	Peak-magnitude-to-RMS ratio
`signalTimeFeatureExtractor`	Streamline signal time feature extraction (Since R2021a)
`timeScalarFeatureOptions`	Store information for converting time-domain feature vectors to scalar values (Since R2024b)

`meanfreq`	Mean frequency
`medfreq`	Median frequency
`bandpower`	Band power
`obw`	Occupied bandwidth
`powerbw`	Power bandwidth
`pwelch`	Welch’s power spectral density estimate
`signalFrequencyFeatureExtractor`	Streamline signal frequency feature extraction (Since R2021b)
`frequencyScalarFeatureOptions`	Store information for converting frequency-domain feature vectors to scalar values (Since R2024b)

`stft`	Short-time Fourier transform
`fsst`	Fourier synchrosqueezed transform
`emd`	Empirical mode decomposition
`vmd`	Variational mode decomposition
`hht`	Hilbert-Huang transform
`cwt`	Continuous 1-D wavelet transform
`wsst`	Wavelet synchrosqueezed transform
`modwt`	Maximal overlap discrete wavelet transform
`modwpt`	Maximal overlap discrete wavelet packet transform

`instbw`	Estimate instantaneous bandwidth (Since R2021a)
`instfreq`	Estimate instantaneous frequency
`spectralCrest`	Spectral crest for signals and spectrograms
`spectralEntropy`	Spectral entropy for signals and spectrograms
`spectralFlatness`	Spectral flatness for signals and spectrograms
`spectralKurtosis`	Spectral kurtosis for signals and spectrograms
`spectralSkewness`	Spectral skewness for signals and spectrograms
`tfridge`	Time-frequency ridges
`signalTimeFrequencyFeatureExtractor`	Streamline signal time-frequency feature extraction (Since R2024a)
`timeFrequencyScalarFeatureOptions`	Store information for converting time-frequency-domain feature vectors to scalar values (Since R2024a)
`timeFrequencyScattering`	Joint time-frequency scattering (Since R2024b)
`waveletScattering`	Wavelet time scattering
`wsstridge`	Time-frequency ridges from wavelet synchrosqueezing

`dlcwt`	Deep learning continuous wavelet transform (Since R2022b)
`dlicwt`	Deep learning inverse continuous 1-D wavelet transform (Since R2024b)
`dlmodwt`	Deep learning maximal overlap discrete wavelet transform and multiresolution analysis (Since R2022a)
`dlstft`	Deep learning short-time Fourier transform (Since R2021a)
`dlistft`	Deep learning inverse short-time Fourier transform (Since R2024a)
`cwtLayer`	Continuous wavelet transform layer (Since R2022b)
`icwtLayer`	Inverse continuous wavelet transform layer (Since R2024b)
`modwtLayer`	Maximal overlap discrete wavelet transform layer (Since R2022b)
`stftLayer`	Short-time Fourier transform layer (Since R2021b)
`istftLayer`	Inverse short-time Fourier transform layer (Since R2024a)

`framelbl`	Partition label sequence into frames (Since R2024a)
`framesig`	Partition signal into frames (Since R2024a)
`scalarFeatureOptions`	Store information for converting feature vectors to scalar values (Since R2024a)

Use Signal Feature Extractor App

Extract and rank features from signals interactively. Prepare signal datasets for classification tasks.
- STEP 1: Import Data into Signal Feature Extractor
- STEP 2: Feature Extraction Using Signal Feature Extractor
- STEP 3: Rank Features Using Signal Feature Extractor
- STEP 4: Export Labeled Signal Sets and Features
Manage Data Sets for Machine Learning and Deep Learning Workflows
Organize, access, and manage data sets for different AI applications.
Accelerate Signal Feature Extraction and Classification Using a Parallel Pool of Workers
Use parallel computing to extract signal multidomain features for bearing fault detection. (Since R2024b)
Accelerate Signal Feature Extraction and Classification Using a GPU
Use a graphical processing unit (GPU) to extract signal multidomain features for bearing fault detection. (Since R2024b)
View, Preprocess, and Write EDF File
View and preprocess data stored in an EDF file.

Use signal feature extraction objects and AI-based classification to identify faulty bearing signals in mechanical systems.

Use the locally interpretable model-agnostic explanation technique to interpret decision-making processes of a deep learning network.

Extract features from smartphone sensor signals and use them to classify human activity.

Construct wavelet-based feature embeddings to classify EEG signals from persons with and without epilepsy.

(Deep Learning Toolbox)

Classify electrocardiogram signals using features derived from wavelets and an autoregressive model.

(Wavelet Toolbox)

How useful was this information?

Unrated 1 star 2 stars 3 stars 4 stars 5 stars