IP Library Granted Patent US 12689392
Granted Patent B2
US 12689392 · App. 18/679,022 · Granted Jul 21, 2026

Associative computing for error correction

Inventors: Ameen D. Akel (Rancho Cordova, CA); Helena Caminal (Ithaca, NY); Sean S. Eilert (Penryn, CA)
Assignee: Micron Technology, Inc.
H03M13/1575G06F11/1068G11C15/04H03M13/43
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 12689392
App. No.
18/679,022
Granted
Jul 21, 2026
Kind
B2
Abstract

Methods, systems, and devices for associative computing for error correction are described. A device may receive first data representative of a first codeword of a size for error correction. The device may identify a set of content-addressable memory cells that stores data representative of a set of codewords each of which is the size of the first codeword. The device may identify second data representative of the first codeword in the set of content-addressable memory cells. Based on identifying the second data, the device may transmit an indication of a valid codeword that is mapped to the second data.

Claims (53)

1 . A method, comprising:

storing data representative of a set of valid codewords in a set of content-addressable memory cells;

receiving first data representative of a codeword of a size for error correction in accordance with storing the data representative of the set of valid codewords in the set of content-addressable memory cells;

identifying the set of content-addressable memory cells that stores the data representative of the set of valid codewords each of which is the size of the codeword;

storing, using the set of content-addressable memory cells, a first set of entries associated with a first table to perform an error correction operation based on identifying the set of content-addressable memory cells;

storing, using the set of content-addressable memory cells, a second set of entries associated with a second table to perform the error correction operation based on storing the first set of entries;

identifying second data in the set of content-addressable memory cells based on the first data; and

transmitting an indication of a valid codeword that is mapped to the second data based on identifying the second data.

2 . The method of claim 1 , wherein the second data is representative of the valid codeword that is within a threshold Hamming distance relative to the codeword.

3 . The method of claim 1 , wherein the first data further is representative of a set of sensed values for a set of memory cells.

4 . The method of claim 1 , further comprising:

identifying that the set of content-addressable memory cells stores data representative of sets of sensed values that are mapped to indices associated with different sets of bit values for the codeword based on identifying the set of content-addressable memory cells.

5 . The method of claim 1 , further comprising:

identifying, in the set of content-addressable memory cells, an index associated with the second data based on identifying the second data.

6 . The method of claim 5 , further comprising:

transmitting an indication of a set of bit values for the codeword based on identifying the index.

7 . A method, comprising:

storing data representative of a set of valid codewords in a set of content-addressable memory cells;

receiving first data representative of a codeword of a size for error correction in accordance with storing the data representative of the set of valid codewords in the set of content-addressable memory cells;

identifying the set of content-addressable memory cells based on receiving the first data;

storing, using the set of content-addressable memory cells, a first set of entries associated with a first table to perform an error correction operation based on identifying the set of content-addressable memory cells;

storing, using the set of content-addressable memory cells, a second set of entries associated with a second table to perform the error correction operation based on storing the first set of entries;

determining whether a second codeword of a size for error correction is identified in the set of content-addressable memory cells based on the first data; and

determining whether to transmit an indication of a valid codeword based on the second codeword.

8 . The method of claim 7 , further comprising:

determining that the valid codeword is mapped to the second codeword based on determining that the second codeword is identified in the set of content-addressable memory cells.

9 . The method of claim 7 , further comprising:

identifying an index associated with the second codeword based on determining that the second codeword is identified in the set of content-addressable memory cells; and

searching a second set of content-addressable memory cells for the valid codeword associated with the index.

10 . The method of claim 7 , further comprising:

transmitting a request for an updated portion of the first data based on determining that the second codeword is not identified in the set of content-addressable memory cells.

11 . The method of claim 7 , further comprising:

receiving an updated portion of the first data based on determining that the second codeword is not identified in the set of content-addressable memory cells.

12 . The method of claim 7 , further comprising:

transmitting the indication of the valid codeword based on determining that the second codeword is identified in the set of content-addressable memory cells.

13 . The method of claim 7 , further comprising:

identifying, in the set of content-addressable memory cells, second data representative of the valid codeword that is within a threshold Hamming distance relative to the codeword.

14 . A method, comprising:

storing data representative of a set of valid codewords in a set of content-addressable memory cells;

receiving first data representative of a set of sensed values for a set of memory cells in accordance with storing the data representative of the set of valid codewords in the set of content-addressable memory cells;

identifying the set of content-addressable memory cells based on receiving the first data;

storing, using the set of content-addressable memory cells, a first set of entries associated with a first table to perform an error correction operation based on identifying the set of content-addressable memory cells;

storing, using the set of content-addressable memory cells, a second set of entries associated with a second table to perform the error correction operation based on storing the first set of entries;

determining whether a second set of sensed values for the set of memory cells is identified in the set of content-addressable memory cells; and

determining whether to transmit an indication of a set of bit values based on the second set of sensed values.

15 . The method of claim 14 , further comprising:

transmitting a request for an updated set of sensed values based on determining that the second set of sensed values is not identified in the set of content-addressable memory cells.

16 . The method of claim 14 , further comprising:

receiving an updated portion of the set of sensed values based on determining that the second set of sensed values is not identified in the set of content-addressable memory cells.

17 . The method of claim 14 , further comprising:

determining whether a confidence value associated with the second set of sensed values satisfies a threshold based on determining that the second set of sensed values is identified in the set of content-addressable memory cells.

18 . The method of claim 17 , further comprising:

transmitting the indication of the set of bit values is based on determining that the confidence value associated with the second set of sensed values satisfies the threshold.