Abstract
The performance and reliability of nonvolatile NAND flash memories deteriorate as the number of program/erase cycles grows. The reliability also suffers from cell-to-cell interference, long data retention time, and read disturb. These processes effect the read threshold voltages. The aging of the cells causes voltage shifts which lead to high bit error rates (BER) with fixed predefined read thresholds. This work proposes two methods that aim on minimizing the BER by adjusting the read thresholds. Both methods utilize the number of errors detected in the codeword of an error correction code. It is demonstrated that the observed number of errors is a good measure for the voltage shifts and is utilized for the initial calibration of the read thresholds. The second approach is a gradual channel estimation method that utilizes the asymmetrical error probabilities for the one-to-zero and zero-to-one errors that are caused by threshold calibration errors. Both methods are investigated utilizing the mutual information between the optimal read voltage and the measured error values. Numerical results obtained from flash measurements show that these methods reduce the BER of NAND flash memories significantly.
Highlights
We propose a tracking method based on the asymmetry of 0- and 1-errors since the asymmetry has a higher mutual information for voltages near to the optimum, where the total number of error provides little information
The read threshold voltages of nonvolatile NAND flash memories depend on various factors such as the number of program/erase cycles, data retention times, and read disturb
We proposed two methods for adapting the read thresholds based on the number of errors observed in the codeword of an error correction code
Summary
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Many soft-input decoding approaches assume reading procedures that maximize the mutual information (MI) between the input and output of flash memory This requires an accurate adaptation of the read voltages and additional estimation of log-likelihood ratios (LLR) for the quantization intervals [15,22]. In the case of decoding failures, the flash controller can perform repeated reads with certain threshold voltages until all errors are corrected by the decoder Such read-retry mechanisms can be very time-consuming. The total number of errors is not suited to estimate small changes of the read voltages, which occur for different pages from the same block To cope with this issue, we propose a second approach which is a gradual channel estimation method.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
