IP Library › Granted Patent US 12,158,808
Granted Patent B2
US 12,158,808 · App. 18/093,432 · Granted Dec 3, 2024

Method of characterizing errors in NAND flash memory, and error estimation method and storage system control method using the same

Inventors: Sung Gil Hong (Seoul, KR); Ha Young Lim (Seoul, KR); Ji Yoon Jung (Seoul, KR); Do Hee Kim (Seoul, KR)
Assignee: FADU Inc.
G06F11/1068G06F11/1008G11C29/52
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,158,808
App. No.
18/093,432
Granted
Dec 3, 2024
Kind
B2
Abstract

A method of characterizing an error in a NAND flash memory, and an error estimation method and a storage system control method using the same are provided. The method of characterizing an error in a NAND flash memory characterizes an error source of the NAND flash memory using a center distance between one pattern and another pattern among a plurality of patterns in a threshold voltage distribution of NAND flash memory cells as a parameter.

Claims (23)

1. A method of characterizing errors in a NAND flash memory, performed by a computing device, comprising:

step (a) of setting, as a parameter, a center distance between one pattern and another pattern among a plurality of patterns within a threshold voltage distribution of NAND flash memory cells; and

step (b) of characterizing an error source of the NAND flash memory by obtaining a value of the center distance for each program/erase (P/E) cycle by applying the parameter,

wherein the plurality of patterns corresponds to an erase state ER and first to N-th (N being a natural number of 3 or more) programmed states in order of increasing threshold voltage of programmed wordlines,

wherein the error source of the NAND flash memory is at least one of a retention, a read disturbance, and a cross temperature,

wherein, in the step (b), a first center distance value according to a retention period for each P/E cycle is obtained and converted into information when the error source is the retention,

wherein, in the step (b), a second center distance value according to a read count for each P/E cycle is obtained and converted into information when the error source is the read interruption, and

wherein, in the step (b), a third center distance value according to a program temperature and a read temperature for each P/E cycle is obtained and converted into information when the error source is the cross temperature.

2. The method of claim 1 , wherein, when the NAND flash memory cells are multi-level cells (MLCs) having an erase state ER and first to third programmed states P 1 to P 3 in order of increasing threshold voltage, the parameter is a center distance between a first pattern corresponding to the first programmed state and a second pattern corresponding to the third programmed state, and the error source of the NAND flash memory is the retention.

3. The method of claim 1 , wherein, when the NAND flash memory cells are triple level cells (TLCs) having an erase state ER and first to seventh programmed states P 1 to P 7 in order of increasing threshold voltage, the parameter is a center distance between a first pattern corresponding to the third programmed state and a second pattern corresponding to the seventh programmed state, and the error source of the NAND flash memory is the retention.

4. The method of claim 1 , wherein, when the NAND flash memory cells are quadruple level cells (QLCs) having an erase state ER and first to fifteenth programmed states P 1 to P 15 in order of increasing threshold voltage, the parameter is a center distance between a first pattern corresponding to the sixth programmed state and a second pattern corresponding to the fifteenth programmed state, and the error source of the NAND flash memory is the retention.

5. The method of claim 1 , wherein, when the NAND flash memory cells are triple level cells (TLCs) having an erase state ER and first to seventh programmed states P 1 to P 7 in order of increasing threshold voltage, the parameter is a center distance between a first pattern corresponding to the first programmed state and a second pattern corresponding to any one of the second to fourth programmed states, and the error source of the NAND flash memory is the read disturbance.

6. The method of claim 1 , wherein the parameter is calculated on the basis of an accumulated number of NAND flash memory cells.

7. A method of estimating an error in an estimation target NAND flash memory, performed by a storage device including a memory controller and a memory device, the method comprising:

step (a) of setting, as a parameter, a center distance between one pattern and another pattern among a plurality of patterns within a threshold voltage distribution of standard NAND flash memory cells; and

step (b) of characterizing a retention error source of the standard NAND flash memory by obtaining a value of the center distance for each program/erase (P/E) cycle by applying the parameter,

step (c) of calculating a predetermined center distance value by applying a parameter for the retention error source to the estimation target NAND flash memory, and obtaining predetermined P/E cycle information; and

step (d) of estimating a retention error of the estimation target NAND flash memory by comparing retention error characterization information generated in the step (b) with the predetermined center distance value and the predetermined P/E cycle information,

wherein the plurality of patterns corresponds to an erase state ER and first to N-th (N being a natural number of 3 or more) programmed states in order of increasing threshold voltage of programmed wordlines, and

wherein, in the step (b), the center distance value according to a retention period for each P/E cycle is obtained and converted into information.

8. A storage device control method performed in a storage device including a memory controller and a memory device, comprising:

(a) estimating, by the memory controller, the retention error by the method according to claim 7 ; and

(b) controlling, by the memory controller, at least one of refreshing, setting of a read bias, setting of a decoding method, or setting of a bad block with respect to the memory device on the basis of the estimated retention error.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 5, 2023
From: HONG, SUNG GIL; LIM, HA YOUNG; JUNG, JI YOON; KIM, DO HEE
To: FADU INC.
Reel/Frame 062281/0909 →
Priority Claims (1)
KR 10-2022-0032485 · Mar 16, 2022 · national
Continuity (2)
Provisional Application 63320291 · Mar 16, 2022
Related Publication 20230297471A1 · Sep 21, 2023