/libapsm

Primary LanguageCudaOtherNOASSERTION

NOMA APSM CUDA library

Compile library

To download submodules and compile please run the init.sh script.

./init.sh

Run example command line tool

cd build

bin/NOMA_detect -r dataset/offline/tx/NOMA_signals_qpsk_complex.bin -s dataset/offline/rx/time/rxData_QPSK_alltx_converted.bin -cs '{"algorithm": {"otype":"APSM", "windowSize": 80}, "modulation":"QPSK", "antennas": {"otype": "equidistant", "number": 8}, "user": 1}'
bin/NOMA_detect -r dataset/offline/tx/NOMA_signals_qpsk_complex.bin -s dataset/offline/rx/ofdm/rxData_QPSK_alltx_converted.bin -cs '{"algorithm": {"otype":"APSM", "windowSize": 80}, "modulation":"QPSK", "antennas": {"otype": "equidistant", "number": 8}, "user": 1}'

For more information about input file format and tool parameters please refer to tool overview file.

License and Citation

APSM library license is specified, as found in the LICENSE file.

If you use this software, please cite it as:

@article{libapsm,
    title={GPU-Accelerated Partially Linear Multiuser Detection for 5G and Beyond URLLC Systems}, 
    author={Mehlhose, Matthias and Marcus, Guillermo and Schäufele, Daniel and Awan, Daniyal Amir and Binder, Nikolaus and Kasparick, Martin and Cavalcante, Renato L. G. and Stañczak, Sławomir and Keller, Alexander},
    journal={IEEE Access}, 
    year={2022},
    volume={10},
    number={},
    pages={70937-70946},
    doi={10.1109/ACCESS.2022.3187040}
}