To fully answer your question would require more info about your application and data. From what you wrote, perhaps you have a two-class classification problem where you want to classify a set of waves into one of two classes: "waves" and "inverse-waves". In that case, you will need training data consististing of a set of waves with class labels, and a separate test set of wavs with class labels. For each wave observation, you will need to extract a feature vector, or feature vectors, from the waves.
There are lots of ways to extract feature vectors from timeseries data. To make the feature extraction process interactive you can use an app such as Signal Labeler or Diagnostic Feature Designer. Once you have extracted features, you can send the extracted features to the Classification Learner app:
- Signal Labeler app (https://www.mathworks.com/help/signal/ref/signallabeler-app.html) See details at https://www.mathworks.com/help/signal/ug/extract-signal-features.html. After extraction, you can export features to the MATLAB® Workspace or to the Classification Learner app and save features as labels in your labeled signal set. For more information, see Export Features and Save Features as Labels.
- Diagnostic Feature Designer app (https://www.mathworks.com/help/predmaint/ref/diagnosticfeaturedesigner-app.html). Once you have defined your features, you can export your most effective features directly to Classification Learner for more insight into feature effectiveness and for algorithm training.
