首页> 外文会议>2015 International Conference on VLSI Systems, Architecture, Technology and Applications >Design of any codeword length parallel long BCH encoders with the help of an efficient C-utility
【24h】

Design of any codeword length parallel long BCH encoders with the help of an efficient C-utility

机译:借助有效的C-utility设计任何码字长度的并行长BCH编码器

获取原文
获取原文并翻译 | 示例

摘要

Error correction has become crucial for NAND Flash based Solid State Drives. As we move towards MLC (Multi Level Cell) NAND Flash memories from SLC (Single Level Cell) NAND Flash memories to achieve higher data density at lower cost per storage unit, we see increased Bit Error Rate (BER) in MLC NAND memories. BCH (Bose-Chaudhuri- Hocquenghem) codes are popular to detect and correct such higher number of errors in many communication applications in addition to the storage domain. BCH encoders are conventionally implemented by linear feedback shift registers that can be directly derived from Generator Polynomial. This direct implementation of long BCH encoders is not suitable for high speed applications, demanding high speed parallel long BCH encoder architectures. One potential problem with such high speed parallel long BCH encoders is its routing complexity when realizing the encoders in hardware. In this paper, we present the design of a long parallel BCH encoder ((4928, 4096) BCH code) with the help of a simple and efficient C-utility. This C-utility takes Generator Polynomial, unfolding factor parameter as inputs, generates the Data Flow Graph in terms of nodes, connectivity matrix with delay elements; generates the resulting BCH encoder hardware code (Verilog RTL code). The proposed C-utility can be used to auto generate any codeword length BCH encoder hardware for configurable unfolding factor value, thus avoiding the RTL design and implementation of encoder, simplifying and speeding up the encoder design compared to the conventional encoder design approaches. The proposed C-utility also generates enough details for each design configuration to analyze the routing complexity of the design much in advance.
机译:对于基于NAND闪存的固态驱动器,纠错已变得至关重要。随着我们从SLC(单级单元)NAND闪存转向MLC(多级单元)NAND闪存,以较低的每存储单元成本实现更高的数据密度,我们看到MLC NAND存储器中的误码率(BER)有所提高。 BCH(Bose-Chaudhuri- Hocquenghem)代码在存储领域之外,在许多通信应用中也很常用来检测和纠正这种数量更多的错误。 BCH编码器通常由线性反馈移位寄存器实现,该寄存器可以直接从生成多项式导出。长BCH编码器的这种直接实现方式不适合高速应用,要求高速并行长BCH编码器架构。这种高速并行长BCH编码器的潜在问题是在硬件中实现编码器时其路由复杂性。在本文中,我们借助简单而有效的C-utility,介绍了一个长并行BCH编码器((4928,4096)BCH代码)的设计。该C实用程序将生成器多项式,展开因子参数作为输入,根据节点,具有延迟元素的连接矩阵生成数据流图;生成最终的BCH编码器硬件代码(Verilog RTL代码)。提出的C-utility可用于自动生成任何可配置展开因子值的码字长BCH编码器硬件,从而避免了RTL设计和实现编码器,与传统编码器设计方法相比,简化并加快了编码器设计。提出的C-utility还为每种设计配置生成了足够的详细信息,可以提前分析设计的布线复杂性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号