ECC and Data Coding for Non-Volatile Memories
The Non-volatile Systems Laboratory along with the CMRR are jointly working together on advanced coding schemes to improve the lifetime and BER of flash memories. Our investigations proposed Error Correcting Codes and coding schemes to be used in flash memories.
In our work, we propose a new error correcting scheme that take advantage on the classification of the error types in MLC flash in order to improve the BER. We also explore the potential use of Write-Once-Memory (WOM)-codes in flash memories. WOM-codes, first introduced in 1982 by Rivest and Shamir, allow writing data more than once in a so-called write-once storage medium. These media comprise of binary cells that can irreversibly change their state. The implementation of WOM-codes in flash memories can reduce the number of erasure by a factor of two or more. Our work also proposed new WOM-codes that improve the amount of data possible to write in the memory.
For more information, please contact Prof. Paul Siegel (psiegel@ucsd.edu), Prof. Lara Dolecek (dolecek@ee.ucla.edu), Eitan Yaakobi (eyaakobi@ucsd.edu), or Dr. Steven Swanson (swanson@cs.ucsd.edu).
Results
![]() |
![]() |
![]() |
|
Publications
-
Tackling Intracell Variability in TLC Flash Through Tensor Product Codes
International Symposium on Information Theory, 2012. -
Tackling Temporal Variability in Multilevel Flash: New Error-Control Code Design and Architectural Validation
Invited Talk, Forty-Ninth Annual Allerton Conference. -
Error Characterization and Coding Schemes for Flash Memories
IEEE Globecom 2010 Workshop on Application of Communication Theory to Emerging Memory Technologies (ACTEMT 2010). -
Efficient Two-Write WOM-Codes
ITW'10: Proceedings of IEEE Information Theory Workshop, 2010. -
Multiple-Write WOM-Codes
48-th Annual Allerton Conference on Communication, Control and Computing, 2010. -
Error Characterization and Coding Schemes for Flash Memories
Proceedings Workshop on the Application of Communication Theory to Emerging Memory Technologies, 2010.
Faculty
Collaborators
Students
Ryan Gabrys














