
For FullText PDF, please login, if you are a member of IEICE,
or go to Pay Per View on menu list, if you are a nonmember of IEICE.

A BitWriteReducing and ErrorCorrecting Code Generation Method by Clustering ECC Codewords for NonVolatile Memories
Tatsuro KOJO Masashi TAWADA Masao YANAGISAWA Nozomu TOGAWA
Publication
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Vol.E99A
No.12
pp.23982411 Publication Date: 2016/12/01
Online ISSN: 17451337
DOI: 10.1587/transfun.E99.A.2398
Type of Manuscript: Special Section PAPER (Special Section on VLSI Design and CAD Algorithms) Category: Keyword: nonvolatile memory, bitwritereducing, errorcorrecting codes, clustering conditions, Sbit flip conditions, Sbound graph, cluster graph, REC code,
Full Text: PDF(1.6MB)>>
Summary:
Nonvolatile memories are paid attention to as a promising alternative to memory design. Data stored in them still may be destructed due to crosstalk and radiation. We can restore the data by using errorcorrecting codes which require extra bits to correct bit errors. Further, nonvolatile memories consume ten to hundred times more energy than normal memories in bitwriting. When we configure them using errorcorrecting codes, it is quite necessary to reduce writing bits. In this paper, we propose a method to generate a bitwritereducing code with errorcorrecting ability. We first pick up an errorcorrecting code which can correct tbit errors. We cluster its codeswords and generate a cluster graph satisfying the Sbit flip conditions. We assign a data to be written to each cluster. In other words, we generate onetomany mapping from each data to the codewords in the cluster. We prove that, if the cluster graph is a complete graph, every data in a memory cell can be rewritten into another data by flipping at most S bits keeping errorcorrecting ability to t bits. We further propose an efficient method to cluster errorcorrecting codewords. Experimental results show that the bitwritereducing and errorcorrecting codes generated by our proposed method efficiently reduce energy consumption. This paper proposes the worldfirst theoretically nearoptimal bitwritereducing code with errorcorrecting ability based on the efficient coding theories.

