QAMPy is a dsp chain for simulation and equalisation of signals from optical communication transmissions. It is written in Python, but has been designed for high performance and most performance critical functions are writen in Cython and run at C-speeds.
QAMPy can equalise BPSK, QPSK and higher-order QAM signals as well as simulate signal impairments.
For signal equalisation it contains:
- CMA and modified CMA equalisation algorithms
- Radius directed equalisers
- several decision directed equaliser implementations
- phase recovery using blind phase search (BPS) and ViterbiViterbi algorithms
- frequency offset compensation
It can simulate the following impairments:
- frequency offset
- SNR
- PMD
- phase noise
QAMpy is designed to make working with QAM signals easy and includes calculations for several performance metrics:
- Symbol Error Rate (SER)
- Bit Error Rate (BER)
- Error Vector Magnitude (EVM)
- Generalized Mututal Information (GMI)
We put a strong focus on documenting our functions and most public functions should be well documented. Use help in jupyter notebook to excess the documenation.
For examples of how to use QAMpy see the Scripts and the Notebooks subdirectory, note that not all files are up-to-date You should in particular look at the cma_equaliser.py and 64_qam_equalisation.py files.
QAMpy should work for both Python versions 2 and 3, however we mainly test on Python 3, so your mileage might vary when using Python 2.
QAMPy depends on the following python modules numpy, scipy, cython, bitarray and numba (as a fall-back option). You will also need to have a working c/c++ compiler with open-mp support installed to compile the cython modules.
On Linux installation works fine using the usual python3 setup.py build and python3 setup.py install.
On Windows, cython modules need to be compiled with the same compiler as python was compiled with, which is often a very old version of Visual Studio. If you do not use the correct Visual Studio version you will often encounter cryptic compile errors. We will try to include more specific instructions in the future.
QAMpy is still in alpha status, however we daily in our work. We will try to keep the basic API stable across releases, but implementation details under core might change without notice.
QAMpy was written by Mikael Mazur and Jochen Schröder from the Photonics Laboratory at Chalmers University of Technology and is licenced under GPLv3 or later.
If you use QAMpy in your work please cite us as Jochen Schröder and Mikael Mazur, "QAMPy a DSP chain for optical communications, DOI: 10.5281/zenodo.1195720".
The GPU graphics card used for part of this work was donated by NVIDIA Corporation