Superlet Transform (SLT) code repository.

Three versions are supplied:

1. Pure MATLAB implementation (aslt.m, faslt.m)

   • Can be run completely in MATLAB, no further steps required.
   • The matlab-pure folder also contains a script Superlet_Toy_Data.m which produces a test signal that proves the utility of SLT.

2. MATLAB interface for C++ backend (faslt.mex64)

   • The folder contains a Visual Studio 2019 (v16.0) solution with the necessary files to build the mex64 file
   • The mex64 file will only work with a 64-bit MATLAB (who uses 32-bit for data analysis anyway?)
   • The project file must be modified to include two sets of libraries:
     • the MATLAB C libs - usually in your MATLAB folder ($(MATLABPath)\extern\lib\win64\microsoft) - the last directory might be different for other operating systems
     • the FFTW libs - a .lib file is included in 'matlab-cdll/lib' folder. The required .dll file may be downloaded from the FFTW website. It must be put in the same folder with the .mex64 file in order for the script to work.

3. Pure Python implementation (superlet.py)

   • multiplicative and fractional adaptive SLT
   • needs only minimal dependencies as defined in environment.yml, available in any basic scientific Python environment
   • when run as a script from the command line via python superlet.py will produce the example output synthetic_example.png
   • can be imported as stand-alone module via from superlet import superlet