Neural Successive Cancellation Decoding of Polar Codes

Neural network (NN) based decoders have appeared as potential candidates to replace successive cancellation (SC) based and belief propagation (BP) decoders for polar codes, due to their one-shot-decoding property. Partitioned NN (PNN) decoder has provided a solution to make use of multiple NN decoders which are connected with BP decoding, with the presence of insufficient training data for practical-length polar codes. However, PNN decoder requires BP iterations that detrimentally affect the decoding latency as compared to noniterative approaches. In this paper, we propose a neural SC (NSC) decoder to overcome the issue associated with PNN. Unlike PNN, the NSC decoder is constructed by multiple NN decoders connected with SC decoding. Compared to a PNN decoder for a polar code of length 128 and rate 0.5, the proposed NSC decoder achieves the same decoding performance, while reducing the decoding latency by 42.5%.

[1]  Warren J. Gross,et al.  Neural offset min-sum decoding , 2017, 2017 IEEE International Symposium on Information Theory (ISIT).

[2]  Yair Be'ery,et al.  Learning to decode linear codes using deep learning , 2016, 2016 54th Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[3]  Xiaohu You,et al.  Improved polar decoder based on deep learning , 2017, 2017 IEEE International Workshop on Signal Processing Systems (SiPS).

[4]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[5]  Keshab K. Parhi,et al.  Early Stopping Criteria for Energy-Efficient Low-Latency Belief-Propagation Polar Code Decoders , 2014, IEEE Transactions on Signal Processing.

[6]  Erdal Arikan,et al.  Channel Polarization: A Method for Constructing Capacity-Achieving Codes for Symmetric Binary-Input Memoryless Channels , 2008, IEEE Transactions on Information Theory.

[7]  David Burshtein,et al.  Deep Learning Methods for Improved Decoding of Linear Codes , 2017, IEEE Journal of Selected Topics in Signal Processing.

[8]  Alexander Vardy,et al.  Fast Polar Decoders: Algorithm and Implementation , 2013, IEEE Journal on Selected Areas in Communications.

[9]  Alexander Vardy,et al.  List decoding of polar codes , 2011, 2011 IEEE International Symposium on Information Theory Proceedings.

[10]  Stephan ten Brink,et al.  Scaling Deep Learning-Based Decoding of Polar Codes via Partitioning , 2017, GLOBECOM 2017 - 2017 IEEE Global Communications Conference.

[11]  Alexios Balatsoukas-Stimming,et al.  LLR-Based Successive Cancellation List Decoding of Polar Codes , 2013, IEEE Transactions on Signal Processing.

[12]  Stephen B. Wicker,et al.  An Artificial Neural Net Viterbi Decoder , 1995, Proceedings of 1995 IEEE International Symposium on Information Theory.

[13]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[14]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[15]  Stephan ten Brink,et al.  On deep learning-based channel decoding , 2017, 2017 51st Annual Conference on Information Sciences and Systems (CISS).

[16]  Huazi Zhang,et al.  Performance Evaluation of Channel Decoding with Deep Neural Networks , 2017, 2018 IEEE International Conference on Communications (ICC).

[17]  Alexander Vardy,et al.  How to Construct Polar Codes , 2011, IEEE Transactions on Information Theory.

[18]  Furkan Ercan,et al.  Memory-Efficient Polar Decoders , 2017, IEEE Journal on Emerging and Selected Topics in Circuits and Systems.

[19]  Tomaso A. Poggio,et al.  Regularization Theory and Neural Networks Architectures , 1995, Neural Computation.

[20]  Warren J. Gross,et al.  A Semi-Parallel Successive-Cancellation Decoder for Polar Codes , 2013, IEEE Transactions on Signal Processing.

[21]  Warren J. Gross,et al.  Fast and Flexible Successive-Cancellation List Decoders for Polar Codes , 2017, IEEE Transactions on Signal Processing.