MLC型NAND闪存中Polar码的优化设计  

Optimal design of polar codes in MLC NAND flash memory

作　　者：卞建慧 赵生妹[1] 孔令军[1] BIAN Jianhui;ZHAO Shengmei;KONG Lingjun(College of Telecommunication ＆ Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210003, China)

机构地区：[1]南京邮电大学通信与信息工程学院,江苏南京210003

出　　处：《南京邮电大学学报（自然科学版）》2018年第3期40-46,共7页Journal of Nanjing University of Posts and Telecommunications：Natural Science Edition

基　　金：国家自然科学基金(61475075;61501250)资助项目

摘　　要：为了进一步提升多级存储单元的纠错性能,提出了一种基于多级存储单元阈值电压分布的Polar码优化设计方法。针对多级存储单元的固有特性,该方法迭代计算各存储单元比特的巴氏参数值优化设计Polar码。分析了不同构造方法对多级存储单元闪存性能的影响,并与文中所构造的Polar码和系统Polar码在多级存储单元信道中的性能进行了比较。仿真结果表明:在多级存储单元信道中,当误码率为10^(-5)时,本文所构造的Polar码与高斯信道下经典巴氏参数法构造的Polar码相比可获得约2 d B增益;当信噪比为21 d B时,与蒙特卡罗法构造的Polar码相比,文中设计的系统Polar码的误码率可提升2个数量级。To further improve the error correction performance of the multi-level cell （MLC） /lash memory, an optimal design method for Polar codes based on the threshold voltage distribution of MLC is proposed. Aimed at the inherent characteristics of MLC flash memory channel, the method can optimize polar codes by iteratively calculating the Bhattacharyya parameter of each memory cell. The influences of different construction methods on the performance of MLC flash memory are analyzed and compared with the Polar code and the systematic Polar code constructed in this paper, Simulation results show that the pro-posed Polar code outperforms the performance of the Polar code constructed by traditional Bhattacharyya parameter method in the Gaussian channel about 2 dB gain at the bit error rate （BER） of 10^-5 in MLC flash memory channel. The systematic Polar code can improve BER performance by 2 order of magnitudes compared with the Polar code constructed by Monte-Carlo method when the signal-to-noise rate （SNR） is 21 dB.

关 键 词：多级存储单元 Polar码 巴氏参数 

分 类 号：TN911.2[电子电信—通信与信息系统]