lmlib Documentation

Version: 1.0 beta, CFR 1

A Model-based signal processing library working with windowed linear state-space and polynomial signal models.

lmlib is open-source and for Python for fast model-based single- and multi-channel signal processing. To ensure either maximum portability or maximum processing speed, lmlib offers multiple backends:

  • Plain Python backend (highly portable)

  • Python with JIT–Just In Time backend (fairly protable, fairly fast)

  • Python with C/OpenBLAS backend (maximum processing speed)

The library is provided by Signal and Information Processing Laboratory (ISI), ETH Zurich and Bern University of Applied Sciences, Institute for Human Centered Engineering (HuCE), Switzerland.

For quick start, consider the application and tutorial examples.

Main Modules

module: statespace

cost functions based on discrete-time autonomous linear state-space models (ALSSMs)

module: polynomials

calculus for continuous-time polynomials

module: utils

utilities such as a (bio-)signal catalogs and test signal generators