| 引用本文： |
-
杜立婵,黄绎珲,王文静,聂晶,黎相成.基于量化修正的低复杂度LDPC译码算法[J].广西科学,2022,29(2):293-300. [点击复制]
- DU Lichan,HUANG Yihui,WANG Wenjing,NIE Jing,LI Xiangcheng.Low Complexity LDPC Decoding Algorithm Based on Quantification Correction[J].Guangxi Sciences,2022,29(2):293-300. [点击复制]
|
|
| 摘要: |
| 为进一步降低基于可靠度的(Low Density Parity Check,LDPC)译码算法复杂度，提出一种基于量化修正的低复杂度LDPC译码算法，该算法在对信道信息预处理时引入量化信息修正处理策略，从而避免在译码迭代过程中进行译码信息修正处理操作，在保持译码性能的同时，较大幅度地降低译码复杂度。针对均匀和非均匀量化方案，本文实现了基于修正系数的均匀量化和基于列重修正的非均匀量化两种译码方案。仿真实验结果表明，所提出的两种译码方案在算法复杂度明显较低的情况下，仍然与MRBI-MLGD译码算法具有相同的译码性能与收敛速度。特别地，在基于列重修正的非均匀量化方案中，只需3-4 bits的低比特量化即可达到均匀量化中高比特量化(8 bits)相同的迭代次数和误比特率性能。 |
| 关键词: 低密度奇偶校验码 低复杂度 可靠度 量化修正 预处理 |
| DOI：10.13656/j.cnki.gxkx.20220526.009 |
| 投稿时间：2021-12-01 |
| 基金项目:国家自然科学基金项目(61761006，61961004)，广西自然科学基金项目(2017GXNSFAA198276，2018GXNSFAA138079)和广西高校中青年教师基础能力提升项目(2022KY1015)资助。 |
|
| Low Complexity LDPC Decoding Algorithm Based on Quantification Correction |
|
DU Lichan1, HUANG Yihui2, WANG Wenjing1, NIE Jing1, LI Xiangcheng2,3
|
| (1.School of Artificial Intelligence, Nanning College for Vocational Technology, Nanning, Guangxi, 530008, China;2.School of Computer, Electronics and Information, Guangxi University, Nanning, Guangxi, 530004, China;3.Guangxi Key Laboratory of Multimedia Communications and Network Technology, Nanning, Guangxi, 530004, China) |
| Abstract: |
| In order to further reduce the complexity of reliability-based LDPC decoding algorithm,a low complexity LDPC decoding algorithm based on quantization correction is proposed.The algorithm introduces quantization information correction strategy in channel information preprocessing,which avoids correcting decoding information during the decoding iteration.The decoding complexity is greatly reduced while maintaining the decoding performance.For the uniform and non-uniform quantization schemes,in this article,two decoding schemes of uniform quantization based on correction coefficient and non-uniform quantization based on column re-correction are realized.The simulation results show that the two proposed decoding schemes still have the same decoding performance and convergence speed as the MRBI-MLGD decoding algorithm under the condition of significantly lower algorithm complexity.In particular,in the non-uniform quantization scheme based on column re-correction,only 3-4 bits of low-bit quantization can achieve the same number of iterations and bit error rate performance as medium-high-bit quantization (8 bits) in uniform quantization. |
| Key words: low-density parity-check codes low complexity reliability quantization correction pre-processing |
