US 11,704,178 B2
Estimating a bit error rate of data stored by a memory subsystem using machine learning
Saeed Sharifi Tehrani, San Diego, CA (US)
Assigned to MICRON TECHNOLOGY, INC., Boise, ID (US)
Filed by Micron Technology, Inc., Boise, ID (US)
Filed on May 14, 2019, as Appl. No. 16/412,325.
Prior Publication US 2020/0364103 A1, Nov. 19, 2020
Int. Cl. G06F 11/07 (2006.01); G11C 16/26 (2006.01); G06N 20/00 (2019.01); G06F 11/10 (2006.01); G11C 29/44 (2006.01); G11C 29/50 (2006.01); G06N 3/08 (2023.01); G06F 3/06 (2006.01); G06N 3/02 (2006.01); G06F 11/00 (2006.01)
CPC G06F 11/076 (2013.01) [G06F 11/0727 (2013.01); G06F 11/1076 (2013.01); G06N 20/00 (2019.01); G11C 16/26 (2013.01); G11C 29/44 (2013.01); G11C 29/50 (2013.01); G06F 3/0653 (2013.01); G06F 11/00 (2013.01); G06F 11/0793 (2013.01); G06N 3/02 (2013.01); G06N 3/08 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
reading encoded data from a group of memory cells of a memory subsystem;
obtaining population data for a number of memory cells in the group of memory cells having a read voltage within a range of read voltages the population data comprising:
a first number of memory cells with a read voltage less than a first threshold voltage; and
a second number of memory cells with a read voltage greater than the first threshold voltage and less than a second threshold voltage;
predicting, using a first trained machine learning model, that an estimated raw bit error rate of the encoded data satisfies a first threshold using the population data, wherein the estimated raw bit error rate is determined without performing an error correcting code decoding operation on the encoded data; and
initiating a first media management operation in response to determining that the estimated raw bit error rate satisfies the first threshold.