This is a brief introduction to the code. Here we use hierarchical clustering to classify the seismic events, and then regroup them according to their peak frequency and frequency bandwidth.
The script require python and some commonly used packages.
The compressed package here include part of the data of 17 stations, which can be used to test the code after decompression.
You can run the test code as:
python Muti-station_volcanic_signals_classificaiton.py -P config_json
This will return the classification results.
The config_json shows the parameters you may need to modify and the files you need to perpare in advance.
The data is downloaded from iris and preprocessed (filter:1-15Hz, mean removal, instrument response correction, resample 100 Hz).