The decoding performance of polar codes strongly depends on the decoding algorithm used, while also the decoder throughput and its latency mainly depend on the decoding algorithm. In this work, we implement the powerful successive cancellation list (SCL) decoder on a GPU and identify the bottlenecks of this algorithm with respect to parallel computing and its difficulties. The inherent serial decoding property of the SCL algorithm naturally limits the achievable speed-up gains on GPUs when compared to CPU implementations. In order to increase the decoding throughput, we use a hybrid decoding scheme based on the belief propagation (BP) decoder, which can be intra- and inter-frame parallelized. The proposed scheme combines excellent decoding performance and high throughput within the signal-to-noise ratio (SNR) region of interest.
%0 Conference Paper
%1 cammerer2017combining
%A Cammerer, Sebastian
%A Leible, B.
%A Stahl, M.
%A Hoydis, Jakob
%A ten Brink, Stepahn
%B 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
%C Piscataway, NJ
%D 2017
%I IEEE
%K sent ubs_10005 ubs_20007 ubs_30073 ubs_40406 unibibliografie
%P 3664-3668
%R 10.1109/ICASSP.2017.7952840
%T Combining belief propagation and successive cancellation list decoding of polar codes on a GPU platform
%X The decoding performance of polar codes strongly depends on the decoding algorithm used, while also the decoder throughput and its latency mainly depend on the decoding algorithm. In this work, we implement the powerful successive cancellation list (SCL) decoder on a GPU and identify the bottlenecks of this algorithm with respect to parallel computing and its difficulties. The inherent serial decoding property of the SCL algorithm naturally limits the achievable speed-up gains on GPUs when compared to CPU implementations. In order to increase the decoding throughput, we use a hybrid decoding scheme based on the belief propagation (BP) decoder, which can be intra- and inter-frame parallelized. The proposed scheme combines excellent decoding performance and high throughput within the signal-to-noise ratio (SNR) region of interest.
%@ 978-1-5090-4117-6 and 978-1-5090-4116-9 and 978-1-5090-4118-3
@inproceedings{cammerer2017combining,
abstract = {The decoding performance of polar codes strongly depends on the decoding algorithm used, while also the decoder throughput and its latency mainly depend on the decoding algorithm. In this work, we implement the powerful successive cancellation list (SCL) decoder on a GPU and identify the bottlenecks of this algorithm with respect to parallel computing and its difficulties. The inherent serial decoding property of the SCL algorithm naturally limits the achievable speed-up gains on GPUs when compared to CPU implementations. In order to increase the decoding throughput, we use a hybrid decoding scheme based on the belief propagation (BP) decoder, which can be intra- and inter-frame parallelized. The proposed scheme combines excellent decoding performance and high throughput within the signal-to-noise ratio (SNR) region of interest.},
added-at = {2020-03-25T15:38:36.000+0100},
address = {Piscataway, NJ},
author = {Cammerer, Sebastian and Leible, B. and Stahl, M. and Hoydis, Jakob and ten Brink, Stepahn},
biburl = {https://puma.ub.uni-stuttgart.de/bibtex/2560158e0411a91c6b8575d09f0faa054/unibiblio},
booktitle = {2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
doi = {10.1109/ICASSP.2017.7952840},
eventdate = {2017-03-05/2017-03-09},
eventtitle = {2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
interhash = {d262b2021ebce10d351e72dd5cc1f538},
intrahash = {560158e0411a91c6b8575d09f0faa054},
isbn = {{978-1-5090-4117-6} and {978-1-5090-4116-9} and {978-1-5090-4118-3}},
keywords = {sent ubs_10005 ubs_20007 ubs_30073 ubs_40406 unibibliografie},
language = {eng},
pages = {3664-3668},
publisher = {IEEE},
timestamp = {2021-06-16T16:15:28.000+0200},
title = {Combining belief propagation and successive cancellation list decoding of polar codes on a GPU platform},
venue = {New Orleans, LA, USA},
year = 2017
}
