Article
Engineering, Electrical & Electronic
Teng Lu, Xuan He, Peng Kang, Jiongyue Xing, Xiaohu Tang
Summary: This paper focuses on partitioning parity-check matrices (PCMs) to reduce hardware complexity and increase decoding throughput for row layered decoding of quasi-cyclic low-density parity-check (QC-LDPC) codes. The authors propose an optimization problem to minimize the maximum column weight of each layer while maintaining a block cyclic shift property among different layers. They present feasible solutions and a lower bound to evaluate an optimal solution. Additionally, they introduce a metric called layer distance to measure data dependency between consecutive layers and propose enumerative and greedy partition algorithms. The constructed codes demonstrate better error correction performance than the underlying 5G LDPC codes.
IEEE TRANSACTIONS ON COMMUNICATIONS
(2023)
Article
Engineering, Electrical & Electronic
Saleh Usman, Mohammad M. Mansour
Summary: A new fast column message-passing (FCMP) schedule for decoding LDPC codes is presented in this study. Compared to existing serial decoding schedules, FCMP converges in half the number of iterations and has lower computational complexity and power consumption. An FCMP decoder supporting IEEE 802.11ad (WiGig) LDPC codes is synthesized using TSMC 40 nm CMOS technology, achieving a throughput of 8.4 Gbps at a clock frequency of 200 MHz and energy efficiency of 8.6 pJ/bit, the best-reported energy-efficiency for a WiGig LDPC decoder.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS
(2021)
Article
Telecommunications
Qiaoqiao Xia, Xin Wang, Huajun Liu, Qing Lin Zhang
Summary: The study introduces a novel hybrid projection algorithm (HPA) that alternately uses approximate even-vertex projection algorithm (EVA) and other accurate projection algorithms to increase the percentage of unuseful projections, achieving better decoding performance while reducing projection time. Compared to the cut search algorithm (CSA), the proposed algorithm significantly saves average projection and decoding time.
IEEE COMMUNICATIONS LETTERS
(2021)
Article
Engineering, Electrical & Electronic
Seokju Han, Jieun Oh, Kyungmok Oh, Jeongseok Ha
Summary: The paper proposes a deep-learning based decoding algorithm tailored for breaking trapping sets in LDPC codes. By re-initializing the channel outputs for error variable nodes, the decoding failures are resolved, significantly improving the performance in the low error-rate regime.
IEEE TRANSACTIONS ON COMMUNICATIONS
(2022)
Article
Physics, Multidisciplinary
Pavel Rybin, Kirin Andreev, Victor Zyablov
Summary: This paper focuses on the specific construction and error performance of binary Low-Density Parity-Check (LDPC) codes transmitted over the binary symmetric channel (BSC). By proposing a low-complexity decoding algorithm, the existence of LDPC codes with exponentially decreasing error decoding probability as the code length increases is proven.
Article
Engineering, Electrical & Electronic
Qi-Yue Yu, Tang Li, Hong-Ru Lin, Feng-Feng Cao
Summary: This paper presents parity-check coding transmit diversity (PCTD) schemes based on half-rate invertible LDPC codes, which are capable of achieving both diversity and coding gains from the time-varying fading channels. By using a chase combining based message passing algorithm (CC-MPA), better bit-error-rate performance is achieved.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2022)
Article
Computer Science, Information Systems
Marco Bondaschi, Marco Dalai
Summary: In this paper, we revise the proofs of low-rate upper bounds on the reliability function of discrete memoryless channels for ordinary and list-decoding schemes. We present a more rigorous and accessible proof based on Blinovsky's idea of using a Ramsey-theoretic result and complement it with additional steps. Additionally, we demonstrate how these techniques can be applied to fix an error in Blahut's low-rate bound.
IEEE TRANSACTIONS ON INFORMATION THEORY
(2022)
Article
Computer Science, Information Systems
Sina Pourjabar, Gwan S. Choi
Summary: This paper introduces a novel WiMAX decoder design that achieves performance close to 5G at a lower cost. The improved decoder reduces initial decoding latency and power consumption while maintaining maximum throughput, with similar FER performance to 5G. The modified decoder also supports multi-rate codeword size and offers reduced latency, as well as decreased power consumption when decoding only shorter codes at SNR greater than 3 dB.
JOURNAL OF SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY
(2021)
Article
Engineering, Electrical & Electronic
Sirawit Khittiwitchayakul, Watid Phakphisut, Pornchai Supnithi
Summary: This paper introduces a new non-uniform schedule based on integer programming for spatially coupled low-density parity-check (SC-LDPC) codes. The design is more flexible and allows reduction of update numbers and performance losses. The authors evaluate the performance of their non-uniform schedules in additive white Gaussian noise (AWGN) and inter-symbol interference (ISI) channels.
IET COMMUNICATIONS
(2022)
Article
Chemistry, Multidisciplinary
Jakub Hyla, Wojciech Sulek, Weronika Izydorczyk, Leszek Dziczkowski, Wojciech Filipowski
Summary: This paper presents a research work that demonstrates the feasibility of applying LDPC coding scheme in a system with highly limited computational resources. By optimizing the algorithm, it is possible to achieve low-complexity encoding in IoT-type systems, while improving memory usage and decoding time.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Hardware & Architecture
Ryo Shibata, Hiroyuki Yashima
Summary: In this letter, the authors study the use of low-density parity-check (LDPC) codes for noisy channels with insertion and deletion (ID) errors. They propose a design method for irregular LDPC codes that can obtain degree distributions for different noise levels. They examine the decoding performances of the designed codes and compare them with the symmetric information rates of cascaded ID-noisy channels. Furthermore, they investigate the relationship between decoding performance and the code structure of irregular LDPC codes.
IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES
(2022)
Article
Engineering, Electrical & Electronic
Yong Ding, Zhiping Huang, Jing Zhou
Summary: This paper investigates the blind recognition method of sparse parity-check matrices of LDPC codes in noncooperative communication, which is crucial for reverse analysis of communication protocols. Two improvements are proposed to enhance the fault tolerance of Liu Qian et al.'s (2021) algorithm. Firstly, a Gaussian elimination method based on random column exchange and soft information is introduced. Secondly, a random extraction method is proposed to improve the fault tolerance of the algorithm based on the sparse property of the parity-check matrices. Simulation results demonstrate the superior performance of the proposed algorithm.
IET COMMUNICATIONS
(2023)
Article
Computer Science, Information Systems
Yu Qiu, Chao Liu, Jianrong Bao, Bin Jiang, Yanhai Shang
Summary: An efficient iterative timing recovery method is proposed, utilizing steepest descent of low-density parity-check (LDPC) decoding metrics to achieve accurate symbol timing synchronization without pilot symbols at low SNR. The algorithm is theoretically analyzed using Gaussian approximation to study the evolution trend of LDPC decoding metrics. Additionally, an approximate gradient descent algorithm is adopted for excellent timing recovery with low complexity and global convergence.
Article
Energy & Fuels
Jakub Hyla, Wojciech Sulek
Summary: This paper discusses the importance of inexpensive and energy-efficient terminals in IoT systems. It presents a design and implementation of error correction coding for energy-efficient operation, focusing on the use of the Raptor-like quasi-cyclic subclass of LDPC codes. The paper includes an encoding algorithm, decoding scheme, and a design procedure for the parity check matrices. Experimental results show that the short block QC-RL-LDPC code achieves higher energy efficiency per information bit compared to fixed-rate LDPC codes.
Article
Engineering, Electrical & Electronic
Haoyuan Wei, Amir H. Banihashemi
Summary: The paper introduces a novel ADMM check node (CN) penalized decoding algorithm that improves decoding performance by penalizing non-codeword solutions and examines the trade-offs between performance and complexity. Research indicates that the proposed CN-penalized decoder outperforms ADMM VN penalized decoders in both waterfall and error floor regions.
IEEE TRANSACTIONS ON COMMUNICATIONS
(2021)