International Symposium on Information Theory and its Applications

Optimization of memory utilization for partially parallel QC-LDPC decoder

Tsung-Che Wu, Yaowen Hu, Chang‐Ming Lee

Published Oct 1, 2010• 3 citations

Abstract

Quasi-cyclic (QC) low-density parity-check (LDPC) codes have advantages over other types of LDPC codes due to their cyclic shifting property. In this paper, a partially parallel QC-LDPC decoder with efficient memory design is proposed. To improve the memory utilization, the structures of efficient chunk and predictor are realized. The efficient chunk can store parts of check-to-variable messages. This design can effectively reduce the memory requirement. Furthermore, the predictor verifies variable-to-check messages and updates efficient chunks instantly. With less check-to-variable messages and variable-to-check messages stored in the memory, this decoding architecture only deposits few messages in several efficient chunks, such that the cost of memory is less. Eventually, the proposed approach dramatically reduces memory requirement about 65% compared to the traditional method while maintaining the same low-cost.

DOI: 10.1109/isita.2010.5650137

Related papers

Efficient Highly-Parallel Decoder Architecture for Quasi-Cyclic Low-Density Parity-Check Codes

Daesun Oh, Keshab K. Parhi

In this paper, the authors propose an efficient highly-parallel decoder architecture using partially overlapped decoding scheme for quasi-cyclic (QC) low-density parity-check (LDPC) codes, which leads to reduction in hardware complexity and power consumption. Generally, due to the regularly structured parity-check matrix H of QC LDPC codes, the message updating computations in the check node unit (CNU) and the variable node unit (VNU) can be efficiently overlapped, which increases the decoding throughput by maximizing the hardware utilization efficiency (HUE). However, the partially overlapped decoding scheme cannot be used to design a highly-parallel decoding architecture for high-throughput applications. For (3, 5)-regular QC LDPC codes, our proposed method could reduce the hardware complexity by approximately 33% for the CNU and 20% for the VNU in the highly-parallel decoder architecture without any performance degradation. In addition, the power consumption can be minimized by reducing the total number of memory accesses for updated messages.

IEEE International Symposium on Circuits and Systems proceedingsMay 1, 200714 citations

Low-complexity entirely-overlapped parallel decoder architecture for quasi-cyclic LDPC codes

Weizhi Lu, Piming Ma

In this paper, we propose a low-complexity entirely-overlapped partially-parallel decoder architecture for quasi-cyclic low-density parity-check (QC-LDPC) codes. For a (c, t)-regular QC-LDPC decoder, this decoder can achieve approximate 100% hardware utilization efficiency (HUE) and entirely overlapped decoding by using only one check node process unit group (CNUG) to process c groups of check node messages sequentially. Compared with traditional partially-parallel decoder, the quantities of check node process unit (CNU), barrel shifters and counters can be decreased prominently. Moreover, it is flexible in code rate and code length.

International Conference on Advanced Communication TechnologyFeb 15, 20090 citations

Low-Complexity High-Speed Decoder Design for Quasi-Cyclic LDPC Codes

Zhongfeng Wang, Zhiqiang Cui

This paper studies low-complexity high-speed decoder architectures for quasi-cyclic low density parity check (QC-LDPC) codes. Algorithmic transformation and architectural level optimization are incorporated to reduce the critical path. Enhanced partially parallel decoding architectures are proposed to linearly increase the throughput of conventional partially parallel decoders through introducing a small percentage of extra hardware. Based on the proposed architectures, a (8176, 7154) Euclidian geometry-based QC-LDPC code decoder is implemented on Xilinx field programmable gate array (FPGA) Virtex-II 6000, where an efficient nonuniform quantization scheme is employed to reduce the size of memories storing soft messages. FPGA implementation results show that the proposed decoder can achieve a maximum (source data) decoding throughput of 172 Mb/s at 15 iterations

IEEE Transactions on Very Large Scale Integration (VLSI) SystemsJan 1, 2007155 citations

Design and efficient hardware implementation schemes for non-Quasi-Cyclic LDPC codes

Baihong Lin, Yukui Pei, Liuguo Yin +1 more

The design of a high-speed decoder using traditional partly parallel architecture for Non-Quasi-Cyclic (NQC) Low-Density Parity-Check (LDPC) codes is a challenging problem due to its high memory-block cost and low hardware utilization efficiency. In this paper, we present efficient hardware implementation schemes for NQCLDPC codes. First, we propose an implementation-oriented construction scheme for NQC-LDPC codes to avoid memory-access conflict in the partly parallel decoder. Then, we propose a Modified Overlapped Message-Passing (MOMP) algorithm for the hardware implementation of NQC-LDPC codes. This algorithm doubles the hardware utilization efficiency and supports a higher degree of parallelism than that used in the Overlapped Message Passing (OMP) technique proposed in previous works. We also present single-core and multi-core decoder architectures in the proposed MOMP algorithm to reduce memory cost and improve circuit efficiency. Moreover, we introduce a technique called the cycle bus to further reduce the number of block RAMs in multi-core decoders. Using numerical examples, we show that, for a rate-2/3, length-15360 NQC-LDPC code with 8.43-dB coding gain for Binary PhaseShift Keying (BPSK) in an Additive White Gaussian Noise (AWGN) channel, the decoder with the proposed scheme achieves a 23.8%-52.6% reduction in logic utilization per Mbps and a 29.0%-90.0% reduction in message-memory bits per Mbps.

Tsinghua Science & TechnologyJan 26, 20175 citations

An Efficient General-Purpose Quasi-Cyclic LDPC Decoder Architecture

Shunliang Mei

This paper presents an efficient general-purpose QC-LDPC decoder architecture by combining hardware with software.The proposed architecture can be applied to regular and irregular QC-LDPC codes of various rates,lengths and parity-check matrix structures,and supports min-sum approximation decoding algorithms and realizes different message passing scheduling strategies.Some complicated message update is implemented by a hardware accelerator,thus the decoding throughput is improved.According to the quasi-cyclic structure of the parity-check matrix,the messages are stored and processed by blocks in decoding.Parallel message processing can also be realized with a small increase in implementation complexity.

Computer Engineering and ScienceJan 1, 20080 citations

A Memory Efficient Partially Parallel Decoder Architecture for QC-LDPC Codes

Zhongfeng Wang, Zhiqiang Cui

In this paper, a memory efficient partially parallel decoder architecture suited for high rate quasi-cyclic low-density parity-check codes using (modified) min-sum decoding algorithm is proposed. In general, more than thirty percent of memory can be saved over conventional partially parallel decoder architectures. To reduce the computation delay of check-node processing unit, an efficient architecture based on variants of rank order filter is presented. The optimized partially parallel decoder architecture can linearly increase the decoding throughput with small hardware overhead. Consequently, the proposed approach facilitates the applications of LDPC codes in area/power sensitive high speed communication systems

Asilomar Conference on Signals, Systems and Computers (ACSSC)Mar 22, 200622 citations