Combining belief propagation and successive cancellation list decoding of polar codes on a GPU platform

@article{Cammerer2017CombiningBP,
  title={Combining belief propagation and successive cancellation list decoding of polar codes on a GPU platform},
  author={Sebastian Cammerer and Benedikt Leible and Matthias Stahl and Jakob Hoydis and Stephan ten Brink},
  journal={2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  year={2017},
  pages={3664-3668}
}
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… 

Figures from this paper

An Efficient Parallel Successive Cancellation List Polar Decoder Based on GPUs
  • Xing Zhou, Rongchun Li, Shijie Li, Yuntao Liu, Y. Dou
  • Computer Science
    2019 IEEE Intl Conf on Parallel & Distributed Processing with Applications, Big Data & Cloud Computing, Sustainable Computing & Communications, Social Computing & Networking (ISPA/BDCloud/SocialCom/SustainCom)
  • 2019
TLDR
A parallel SCL decoder based on the graphic processing unit(GPU) is designed to reduce the latency and improve the decoding throughput and an efficient approach for sharing the intermediate values among different decoding paths is introduced.
GPU-based Implementation of Belief Propagation Decoding for Polar Codes
TLDR
An optimized software BP decoder for polar codes on graphics processing units (GPUs) that achieves throughput speedups from 2.59× to 131× and provides good tradeoff between error performance and throughput.
On iterative decoding of polar codes: Schedule-dependent performance and constructions
TLDR
To overcome performance impairments observed when using polar codes optimized for sequential scheduling under parallel schedules, this work presents a method to optimize codes for iterative decoders working with parallel scheduling.
Reducing Polar Decoding Latency by Neural Network-Based On-the-Fly Decoder Selection
TLDR
A neural network (NN)-based estimator is trained to pre-select — only based on the received noisy channel observation — the most efficient decoder (i.e., BP or SCL) before decoding, which enables the same (good) error-rate performance of the SCL decoder but with a higher decoding efficiency and, in particular, with an — on average — lower decoding latency.
Successive-cancellation list decoder of polar codes based on GPU
TLDR
By means of the parallel computing capabilities of GPUs, the proposed decoder achieves high throughput of 41Mbps on NVIDIA GTX 980 and 65Mbps on TITAN X while decoding the code with length of 1024 bits and 32 lists.
CRC-Aided Belief Propagation List Decoding of Polar Codes
TLDR
It is shown that iterative decoding of polar codes based on the belief propagation list (BPL) algorithm can approach the error-rate performance of CA-SCL decoding and, thus, can be efficiently used for decoding the standardized 5G polar codes.
Scaling Deep Learning-Based Decoding of Polar Codes via Partitioning
TLDR
This work partitions the encoding graph into smaller sub-blocks and train them individually, closely approaching maximum a posteriori (MAP) performance per sub-block, and shows the degradation through partitioning and compares the resulting decoder to state-of-the art polar decoders such as successive cancellation list and belief propagation decoding.
BP List Decoding of Polar Codes with Adaptive Bit Splitting over Critical Set
TLDR
A modified version of belief propagation (BP) decoding is resorts to, i.e., the so-called BS-BPL, and takes advantage of its parallel structure to address the issue of low decoding delay and block error rate.
A Novel Flip-List-Enabled Belief Propagation Decoder for Polar Codes
TLDR
A novel flip-list- (FL)-enabled belief propagation (BP) method to improve the error-correction performance of BP decoding for polar codes with low computational complexity.
Belief propagation decoding of polar codes on permuted factor graphs
TLDR
This work shows that the performance of iterative belief propagation decoding of polar codes can be enhanced by decoding over different carefully chosen factor graph realizations, and achieves the best performance of all iterative polar decoders presented thus far.
...
...

References

SHOWING 1-10 OF 26 REFERENCES
Algorithm and architecture for hybrid decoding of polar codes
  • Bo Yuan, K. Parhi
  • Computer Science
    2014 48th Asilomar Conference on Signals, Systems and Computers
  • 2014
TLDR
A hybrid BP-SC decoding scheme to improve the decoding performance of polar codes with relatively short latency is proposed and the low-complexity unified hardware architecture for the hybrid decoding scheme is proposed, which is able to implement SC and BP algorithms using same hardware.
A GPU implementation of belief propagation decoder for polar codes
TLDR
The BP decoding algorithm is implemented to utilize the parallel computing capability of the GPUs and can make use of parallelism both at the thread level and block level, and by utilizing the limited shared memory available on GPUs, a real time decoding performance is achieved.
List decoding of polar codes
  • I. Tal, A. Vardy
  • Computer Science
    2011 IEEE International Symposium on Information Theory Proceedings
  • 2011
TLDR
It appears that the proposed list decoder bridges the gap between successive-cancellation and maximum-likelihood decoding of polar codes, and devise an efficient, numerically stable, implementation taking only O(L · n log n) time and O( L · n) space.
Low-Latency Software Polar Decoders
TLDR
It is shown how adapting the algorithm at various levels can lead to significant improvements in latency and throughput, yielding polar decoders that are suitable for high-performance software-defined radio applications on modern desktop processors and embedded-platform processors.
Early Stopping Criteria for Energy-Efficient Low-Latency Belief-Propagation Polar Code Decoders
  • Bo Yuan, K. Parhi
  • Computer Science
    IEEE Transactions on Signal Processing
  • 2014
TLDR
Novel early stopping criteria for polar BP decoding to significantly reduce energy dissipation and decoding latency, and a novel channel condition estimation approach, which can help select different stopping criteria in different SNR regions are explored.
Polar codes: Primary concepts and practical decoding algorithms
TLDR
The butterfly structure of polar codes introduces correlation among source bits, justifying the use of the SC algorithm for efficient decoding, and state-of-the-art decoding algorithms, such as the BP and some generalized SC decoding, are explained in a broad framework.
An Adaptive Successive Cancellation List Decoder for Polar Codes with Cyclic Redundancy Check
TLDR
It is demonstrated that polar code with 24-bit CRC decoded by the proposed adaptive SC-List decoder with very large maximum list size can achieve a frame error rate FER ≤ 10-3{-3} at Eb/No = 1.1dB, which is about 0.25dB from the information theoretic limit at this block length.
Massively LDPC Decoding on Multicore Architectures
TLDR
Algorithms and data structures suitable for parallel computing are proposed in this paper to perform LDPC decoding on multicore architectures and achieve throughputs that in some cases approach very well those obtained with VLSI decoders.
How to Construct Polar Codes
  • I. Tal, A. Vardy
  • Computer Science
    IEEE Transactions on Information Theory
  • 2013
TLDR
A method for efficiently constructing polar codes is presented and analyzed, proving that for any fixed ε > 0 and all sufficiently large code lengths n, polar codes whose rate is within ε of channel capacity can be constructed in time and space that are both linear in n.
A Decision-Aided Parallel SC-List Decoder for Polar Codes
TLDR
A decision-aided scheme to reduce the number of split paths and thus reduce the sorting complexity at the parallel SC-List decoding of polar codes.
...
...