More Like this

1. Neural-Network-Optimized Degree-Specific Weights for LDPC MinSum Decoding

   https://doi.org/10.1109/ISTC49272.2021.9594227  

   Wang, Linfang ; Chen, Sean ; Nguyen, Jonathan ; Dariush, Divsalar ; Wesel, Richard ( August 2021 , 2021 11th International Symposium on Topics in Coding (ISTC))

   Neural Normalized MinSum (N-NMS) decoding delivers better frame error rate (FER) performance on linear block codes than conventional Normalized MinSum (NMS) by assigning dynamic multiplicative weights to each check-to-variable node message in each iteration. Previous N-NMS efforts primarily investigated short block codes (N < 1000), because the number of N-NMS parameters required to be trained scales proportionately to the number of edges in the parity check matrix and the number of iterations. This imposes an impractical memory requirement for conventional tools such as Pytorch and Tensorflow to create the neural network and store gradients. This paper provides efficient methods of training the parameters of N-NMS decoders that support longer block lengths. Specifically, this paper introduces a family of Neural 2-dimensional Normalized (N-2D-NMS) decoders with various reduced parameter sets and shows how performance varies with the parameter set selected. The N-2D-NMS decoders share weights with respect to check node and/or variable node degree. Simulation results justify a reduced parameter set, showing that the trained weights of N- NMS have a smaller value for the neurons corresponding to larger check/variable node degree. Further simulation results on a (3096,1032) Protograph-Based Raptor-Like (PBRL) code show that the N-2D-NMS decoder can achieve the same FER as N- NMS while also providing at least a 99.7% parameter reduction. Furthermore, the N-2D-NMS decoder for the (16200,7200) DVBS- 2 standard LDPC code shows a lower error floor than belief propagation. Finally, this paper proposes a hybrid decoder training structure that utilizes a neural network which combines a feedforward module with a recurrent module. The decoding performance and parameter reduction of the hybrid training depends on the length of recurrent module of the neural network. 

   more » « less
2. A Reconstruction-Computation-Quantization (RCQ) Approach to Node Operations in LDPC Decoding

   https://doi.org/10.1109/GLOBECOM42002.2020.9348139  

   Wang, Linfang ; Wesel, Richard D. ; Stark, Maximilian ; Bauch, Gerhard ( December 2020 , GLOBECOM 2020 - 2020 IEEE Global Communications Conference)

   null (Ed.)

   This paper proposes a finite-precision decoding method for low-density parity-check (LDPC) codes that features the three steps of Reconstruction, Computation, and Quantization (RCQ). Unlike Mutual-Information-Maximization Quantized Belief Propagation (MIM-QBP), RCQ can approximate either belief propagation or Min-Sum decoding. MIM-QBP decoders do not work well when the fraction of degree-2 variable nodes is large. However, sometimes a large fraction of degree-2 variable nodes is used to facilitate a fast encoding structure, as seen in the IEEE 802.11 standard and the DVB-S2 standard. In contrast to MIM-QBP, the proposed RCQ decoder may be applied to any off-the-shelf LDPC code, including those with a large fraction of degree-2 variable nodes. Simulations show that a 4-bit Min-Sum RCQ decoder delivers frame error rate (FER) performance within 0.1 dB of floating point belief propagation (BP) for the IEEE 802.11 standard LDPC code in the low SNR region. The RCQ decoder actually outperforms floating point BP and Min-Sum in the high SNR region were FER less than 10 −5 . This paper also introduces Hierarchical Dynamic Quantization (HDQ) to design the time-varying non-uniform quantizers required by RCQ decoders. HDQ is a low-complexity design technique that is slightly sub-optimal. Simulation results comparing HDQ and optimal quantization on the symmetric binary-input memoryless additive white Gaussian noise channel show a mutual information loss of less than 10 −6 bits, which is negligible in practice. 

   more » « less
3. Suspicion Distillation Gradient Descent Bit-Flipping Algorithm

   https://doi.org/10.3390/e24040558  

   Ivaniš, Predrag ; Brkić, Srdjan ; Vasić, Bane ( April 2022 , Entropy)

   We propose a novel variant of the gradient descent bit-flipping (GDBF) algorithm for decoding low-density parity-check (LDPC) codes over the binary symmetric channel. The new bit-flipping rule is based on the reliability information passed from neighboring nodes in the corresponding Tanner graph. The name SuspicionDistillation reflects the main feature of the algorithm—that in every iteration, we assign a level of suspicion to each variable node about its current bit value. The level of suspicion of a variable node is used to decide whether the corresponding bit will be flipped. In addition, in each iteration, we determine the number of satisfied and unsatisfied checks that connect a suspicious node with other suspicious variable nodes. In this way, in the course of iteration, we “distill” such suspicious bits and flip them. The deterministic nature of the proposed algorithm results in a low-complexity implementation, as the bit-flipping rule can be obtained by modifying the original GDBF rule by using basic logic gates, and the modification is not applied in all decoding iterations. Furthermore, we present a more general framework based on deterministic re-initialization of the decoder input. The performance of the resulting algorithm is analyzed for the codes with various code lengths, and significant performance improvements are observed compared to the state-of-the-art hard-decision-decoding algorithms. 

   more » « less
4. Information Bottleneck Decoding of Rate-Compatible 5G-LDPC Codes

   https://doi.org/10.1109/ICC40277.2020.9149304  

   Stark, Maximilian ; Bauch, Gerhard ; Wang, Linfang ; Wesel, Richard D. ( June 2020 , ICC 2020 - 2020 IEEE International Conference on Communications (ICC))

   The new 5G communications standard increases data rates and supports low-latency communication that places constraints on the computational complexity of channel decoders. 5G low-density parity-check (LDPC) codes have the so-called protograph-based raptor-like (PBRL) structure which offers inherent rate-compatibility and excellent performance. Practical LDPC decoder implementations use message-passing decoding with finite precision, which becomes coarse as complexity is more severely constrained. Performance degrades as the precision becomes more coarse. Recently, the information bottleneck (IB) method was used to design mutual-information-maximizing lookup tables that replace conventional finite-precision node computations. The IB approach exchanges messages represented by integers with very small bit width. This paper extends the IB principle to the flexible class of PBRL LDPC codes as standardized in 5G. The extensions include puncturing and rate-compatible IB decoder design. As an example of the new approach, a 4-bit information bottleneck decoder is evaluated for PBRL LDPC codes over a typical range of rates. Frame error rate simulations show that the proposed scheme outperforms offset min-sum decoding algorithms and operates very close to double-precision sum-product belief propagation decoding. 

   more » « less
5. CRC-Aided High-Rate Convolutional Codes With Short Blocklengths for List Decoding

   https://doi.org/10.1109/TCOMM.2023.3324370  

   Sui, Wenhui ; Towell, Brendan ; Asmani, Ava ; Yang, Hengjie ; Grissett, Holden ; Wesel, Richard D. ( January 2023 , IEEE Transactions on Communications)

   Tal, Ido (Ed.)

   Recently, rate-1/ n zero-terminated (ZT) and tail-biting (TB) convolutional codes (CCs) with cyclic redundancy check (CRC)-aided list decoding have been shown to closely approach the random-coding union (RCU) bound for short blocklengths. This paper designs CRC polynomials for rate-( n - 1)/ n ZT and TB CCs with short blocklengths. This paper considers both standard rate-( n -1)/ n CC polynomials and rate-( n - 1)/ n designs resulting from puncturing a rate-1/2 code. The CRC polynomials are chosen to maximize the minimum distance d min and minimize the number of nearest neighbors A dmin . For the standard rate-( n - 1)/ n codes, utilization of the dual trellis proposed by Yamada et al . lowers the complexity of CRC-aided serial list Viterbi decoding (SLVD). CRC-aided SLVD of the TBCCs closely approaches the RCU bound at a blocklength of 128. This paper compares the FER performance (gap to the RCU bound) and complexity of the CRC-aided standard and punctured ZTCCs and TBCCs. This paper also explores the complexity-performance trade-off for three TBCC decoders: a single-trellis approach, a multi-trellis approach, and a modified single-trellis approach with pre-processing using the wrap around Viterbi algorithm. 

   more » « less