Working with Signals
Wavelet scattering enables you to produce low-variance data representations that minimize differences within a class while preserving discriminability across classes. Wavelet scattering requires few user-specified parameters to produce compact representations of data which are robust against time shifts on a scale you define. You can use these representations in conjunction with machine learning algorithms for classification and regression.
You can use the continuous wavelet transform (CWT) to generate 2-D time-frequency maps of time series data, which can be used with 2-D convolutional networks. Generating time-frequency representations for use in deep CNNs is a powerful approach for signal classification. The ability of the CWT to simultaneously capture steady-state and transient behavior in time series data makes the wavelet-based time-frequency representation particularly robust when paired with deep CNNs.
With a Signal Processing Toolbox™ license you can include the short-time Fourier transform
into your machine learning and deep learning workflows. You can also use
Signal Labeler (Signal Processing Toolbox) to label signals for analysis or for use in
machine learning and deep learning applications. Signal
Labeler saves data as
labeledSignalSet objects. With a Audio Toolbox™ license you can Import and Play Audio File Data in Signal Labeler (Signal Processing Toolbox).
You can also use
melSpectrogram (Audio Toolbox) for feature extraction.
|Signal Labeler||Label signal attributes, regions, and points of interest, and extract features (Since R2019a)|
|Convert deep-learning CWT filter tensor to filter bank matrix (Since R2022b)|
|Continuous wavelet transform filter bank|
|Convert CWT filter bank to reduced-weight tensor for deep learning (Since R2022b)|
|Deep learning continuous wavelet transform (Since R2022b)|
|Deep learning maximal overlap discrete wavelet transform and multiresolution analysis (Since R2022a)|
|Deep learning short-time Fourier transform (Since R2021a)|
|1-D lifting wavelet transform (Since R2021a)|
|Mel spectrogram (Since R2019a)|
|Maximal overlap discrete wavelet packet transform|
|Maximal overlap discrete wavelet transform|
|Wavelet time scattering|
|Wigner-Ville distribution and smoothed pseudo Wigner-Ville distribution|
- Detect Air Compressor Sounds in Simulink Using Wavelet Scattering (DSP System Toolbox)
Use the Wavelet Scattering block and a pretrained deep learning network to classify audio signals.
- Detect Anomalies in ECG Time-Series Data Using Wavelet Scattering and LSTM Autoencoder in Simulink (DSP System Toolbox)
Use wavelet scattering and deep learning network to detect anomalies in ECG signals.
- Radar and Communications Waveform Classification Using Deep Learning (Phased Array System Toolbox)
Classify radar and communications waveforms using the Wigner-Ville distribution (WVD) and a deep convolutional neural network (CNN).
- Radar Target Classification Using Machine Learning and Deep Learning (Radar Toolbox)
Classify radar returns using machine and deep learning approaches.