IP Library Granted Patent US 12,567,475
Granted Patent B2
US 12,567,475 · App. 18/415,722 · Granted Mar 3, 2026

Pre-characterizing weak bits for increased low density parity check (LDPC) speed

Inventors: Adam Jacobvitz (San Jose, CA); Piyush Dhotre (San Jose, CA); Niles Yang (San Jose, CA); Juan Carlos Lee (San Jose, CA); Eran Sharon (Rishon Lezion, IL); Idan Goldenberg (Rishon Lezion, IL); Zhenni Wan (San Jose, CA)
Assignee: Sandisk Technologies, Inc.
G11C29/1201G11C29/18G11C29/46G11C2029/1802
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 12,567,475
App. No.
18/415,722
Granted
Mar 3, 2026
Kind
B2
Abstract

A storage device may speed up error correction by pre-characterizing weak cell information in a memory device. The storage device includes a memory device with cells that may store multiple bits. A controller executes a pre-characterization operation on the memory device to identify a slow cell and/or a fast cell on the memory device. The controller retrieves weak cell information for the slow cell and/or the fast cell. The controller converts the weak cell information into values used by an error correction engine and provides the values to the error correction engine to be used in decoding information retrieved from the memory device.

Claims (29)

1 . A storage device pre-characterizes weak cell information in a memory device to speed up error correction, the storage device comprises:

a memory device including cells, wherein a cell in the memory device stores multiple bits; and

a controller to pre-characterize cells in the memory device to identify at least one of a slow cell and a fast cell based on a threshold voltage in the memory device, retrieve weak cell information for at least one of the slow cell and the fast cell, wherein the weak cell information is associated with a changed threshold voltage value in a weak bit voltage threshold window, convert the weak cell information into error correction values used by an error correction engine, and provide the error correction values to the error correction engine to be used to decode information retrieved from the memory device.

2 . The storage device of claim 1 , wherein the error correction engine is a low-density parity check (LDPC) engine, and the controller converts the weak cell information to Log-Likelihood Ratio (LLR) values that are inputted to the LDPC engine.

3 . The storage device of claim 1 , wherein the controller pre-characterizes the cells in the memory device by executing an initial test mode on the memory device.

4 . The storage device of claim 1 , wherein the controller pre-characterizes the cells in the memory device by executing a periodic test mode on the memory device.

5 . The storage device of claim 1 , wherein the weak cell information includes a first bit to indicate a bit location and a second bit to indicate when a cell is one of the slow cell and the fast cell.

6 . The storage device of claim 1 , wherein the weak cell information is ranked based on cell speed.

7 . The storage device of claim 1 , wherein the controller stores the weak cell information in a table, wherein the weak cell information is associated with a predefined number of cells per state.

8 . The storage device of claim 1 , wherein the controller stores the weak cell information in a table and the controller splits entries in the table based on at least one of a location read, cell behavior, temperature conditions, read disturb conditions, and data retention conditions.

9 . The storage device of claim 1 , wherein the controller obtains the weak cell information for corresponding wordlines and groups the corresponding wordlines.

10 . The storage device of claim 1 , wherein the controller identifies a slow cell as a cell that does not pass a program verify operation and the controller identifies a fast cell as a cell that passes a program verify operation a fewest number of pulses.

11 . A method for pre-characterizing weak cell information in a memory device to speed up error correction on a storage device, the storage device comprises a controller to execute the method comprising:

executing one of an initial pre-characterization operation and a periodic pre-characterization operation to pre-characterize cells in the memory device;

identifying at least one of a slow cell and a fast cell in the memory device based on a threshold voltage;

retrieving weak cell information for at least one of the slow cell and the fast cell, wherein the weak cell information is associated with a changed threshold voltage value in a weak bit voltage threshold window;

converting the weak cell information to error correction values; and

providing the error correction values to an error correction engine to be used to decode information retrieved from the memory device.

12 . The method of claim 11 , wherein executing the initial pre-characterization operation comprises executing an initial test mode on the memory device.

13 . The method of claim 11 , wherein executing the periodic pre-characterization operation comprises executing a periodic test mode on the memory device.

14 . The method of claim 11 , wherein the weak cell information includes a first bit to indicate a bit location and a second bit to indicate when a cell is one of the slow cell and the fast cell.

15 . The method of claim 11 , further comprising ranking the weak cell information based on cell speed.

16 . The method of claim 11 , further comprising storing the weak cell information in a table, wherein the weak cell information is associated with a predefined number of cells per state.

17 . The method of claim 11 , further comprising storing the weak cell information in a table and splitting entries in the table based on at least one of a location read, cell behavior and temperature conditions.

18 . The method of claim 11 , further comprising obtaining the weak cell information for corresponding wordlines and grouping the corresponding wordlines.

19 . The method of claim 11 , wherein the identifying comprises identifying a slow cell as a cell that does not pass a program verify operation and identifying s a fast cell as a cell that passes a program verify operation a fewest number of pulses.

20 . A storage device pre-characterizes weak cell information in a memory device to speed up error correction, the storage device comprises:

a memory device including cells, wherein a cell in the memory device stores multiple bits; and

a controller to pre-characterize cells in the memory device to identify at least one of a slow cell and a fast cell in the memory device based on a threshold voltage, retrieve weak cell information for at least one of the slow cell and the fast cell, wherein the weak cell information is associated-with a changed threshold voltage value in a weak bit voltage threshold window, convert the weak cell information to Log-Likelihood Ratio (LLR) values, and provide the LLR values to a low-density parity check (LDPC) engine to be used to decode a codeword retrieved from the memory device and return a logical value stored in a cell on the memory device.

Assignments (4)
PARTIAL RELEASE OF SECURITY INTERESTS Recorded Apr 25, 2025
From: JPMORGAN CHASE BANK, N.A., AS AGENT
To: SANDISK TECHNOLOGIES, INC.
Reel/Frame 071382/0001 →
SECURITY AGREEMENT Recorded Apr 25, 2025
From: SANDISK TECHNOLOGIES, INC.
To: JPMORGAN CHASE BANK, N.A., AS COLLATERAL AGENT
Reel/Frame 071050/0001 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 31, 2024
From: SANDISK TECHNOLOGIES LLC
To: SANDISK TECHNOLOGIES, INC.
Reel/Frame 069796/0423 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 31, 2024
From: JACOBVITZ, ADAM; DHOTRE, PIYUSH; YANG, NILES; LEE, JUAN; SHARON, ERAN; GOLDENBERG, IDAN; WAN, ZHENNI
To: SANDISK TECHNOLOGIES LLC
Reel/Frame 066311/0573 →
Continuity (1)
Related Publication 20250239317A1 · Jul 24, 2025
References Cited (6)
US 8832526B2 · Tseng · 2014 [cited by applicant]
US 9329928B2 · Fitzpatrick · 2016 [cited by applicant]
US 10366769B2 · Cho · 2019 [cited by applicant]
US 11361832B2 · Han · 2022 [cited by applicant]
US 20160077914A1 · Zeng · 2016 [cited by examiner]
US 20230094363A1 · Graumann · 2023 [cited by applicant]