IP Library Granted Patent US 12,451,187
Granted Patent B2
US 12,451,187 · App. 18/239,302 · Granted Oct 21, 2025

Data storage device and method for inferring a read threshold using a time tag determination

Inventors: David Avraham (San Jose, CA); Ariel Navon (Revava, IL); Alexander Bazarsky (Holon, IL); Eran Sharon (Rishon Lezion, IL)
Assignee: Sandisk Technologies, Inc.
G11C16/08G11C16/105
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 12,451,187
App. No.
18/239,302
Filed
Aug 29, 2023
Granted
Oct 21, 2025
Kind
B2
Art Unit
2824
USPC
365/185.23
Abstract

A data storage device has an inference engine that can infer a read threshold based on a non-linear function of inputs that reflect current memory and data conditions. The read threshold can be used in reading a wordline in the memory. Using a machine-learning-based approach to infer a read threshold can provide significant improvement in read threshold accuracy, which can reduce bit error rate and also improve latency, throughput, power consumption, and quality of service.

Claims (45)

1. In a data storage device comprising a memory, a method comprising:

using firmware in the data storage device to infer a read threshold to use in reading a wordline in the memory, wherein the read threshold is inferred:

prior to the data storage device receiving a read command from a host that causes the wordline to be read;

only during a time and temperature update; and

based only on one or more inputs available during the time and temperature update, wherein a wordline number of the wordline is not one of the one or more inputs; and

in response to receiving the read command from the host:

looking up the wordline number in a pre-calibrated wordline zoning table comprising a plurality of predefined correction values associated with a respective plurality of wordline numbers;

modifying the inferred read threshold based on a correction value associated with the wordline number in the pre-calibrated wordline zoning table; and

using the modified inferred read threshold to read the wordline.

2. The method of claim 1 , wherein the read threshold is inferred by using machine learning in a background and a static table in a foreground.

3. The method of claim 2 , wherein the machine learning is trained to take into account future static tables in the foreground.

4. The method of claim 1 , wherein the one or more inputs comprises programming temperature.

5. The method of claim 1 , wherein the one or more inputs comprises elapsed time since a prior time tag and temperature update.

6. The method of claim 1 , wherein the one or more inputs comprises bit error rate (BER) information.

7. The method of claim 1 , further comprising updating the read threshold periodically.

8. A data storage device comprising:

a memory; and

a processor configured to execute firmware to:

infer a read threshold to use to read a wordline in the memory, wherein the read threshold is inferred:

prior to the data storage device receiving a read command from a host that causes the wordline to be read;

only during a time and temperature update; and

based only on one or more inputs available during the time and temperature update, wherein a wordline number of the wordline is not one of the one or more inputs; and

in response to receiving the read command from the host:

look up the wordline number in a pre-calibrated wordline zoning table comprising a plurality of predefined correction values associated with a respective plurality of wordline numbers;

modify the inferred read threshold based on a correction value associated with the wordline number in the pre-calibrated wordline zoning table; and

use the modified inferred read threshold to read the wordline.

9. The data storage device of claim 8 , wherein the one or more inputs comprise programming temperature.

10. The data storage device of claim 8 , wherein the one or more inputs comprise elapsed time since a prior time tag and temperature update.

11. The data storage device of claim 8 , wherein the one or more inputs comprise bit error rate (BER) information.

12. The data storage device of claim 8 , wherein the processor is further configured to periodically infer read thresholds.

13. The data storage device of claim 12 , wherein the read thresholds are inferred when a time tag threshold has changed.

14. The data storage device of claim 12 , wherein the read thresholds are inferred when a temperature has changed.

15. The data storage device of claim 8 , wherein the memory comprises a three-dimensional memory.

16. A data storage device comprising:

a memory; and

means for:

using firmware in the data storage device to infer a read threshold to use in reading a wordline in the memory, wherein the read threshold is inferred:

prior to the data storage device receiving a read command from a host that causes the wordline to be read;

only during a time and temperature update; and

based only on one or more inputs available during the time and temperature update, wherein a wordline number of the wordline is not one of the one or more inputs; and

in response to receiving the read command from the host:

looking up the wordline number in a pre-calibrated wordline zoning table comprising a plurality of predefined correction values associated with a respective plurality of wordline numbers;

modifying the inferred read threshold based on a correction value associated with the wordline number in the pre-calibrated wordline zoning table; and

using the modified inferred read threshold to read the wordline.

17. The data storage device of claim 12 , wherein the read thresholds are inferred when a program-erase-cycle (PEC) count has changed.