Low Power Decoder Architecture of Product Code for Storage Controller.
2022 19th International SoC Design Conference (ISOCC)(2022)
摘要
Error correction codes (ECCs) are essential for correcting the errors caused by the physical characteristics of 3D NAND flash in flash solution products. In this paper, we introduce low power decoder architecture for product code. Precisely, by using Hamming product code and Chase-Pyndiah algorithm, this work takes not only the same correction capability and area as low-density parity-check decoder, but also less power with relatively simple operation for decoding. Only the log likelihood ratios of error-suspicious bits are gathered in SRAM for the next iteration of decoding, so that the power consumption and area of SRAM read/write decrease. Compared with the conventional ECC decoder with the same correction capability and area, our decoder has achieved an average power reduction of 22.4%.
更多查看译文
关键词
Error correction code,Hamming product code,Chase-Pyndiah algorithm
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn