# US 2021/0281278 A1 · Improving performance of a bit flipping (bf) decoder of an error correction system

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0281278 A1
- Document ID: US20210281278A1
- Status: Published application
- Kind: A1
- Application number: 16/813,342
- Filed: 2020-03-09
- Published: 2021-09-09
- Earliest claimed: 2020-03-09
- Assignees: SK hynix Inc.
- Applicants: SK hynix Inc.
- Inventors: Xuanxuan Lu; Fan Zhang; Aman Bhatia; Meysam Asadi; Haobo Wang
- Web page: https://patentgrep.com/patent/US20210281278A1

## CPC

- H03M 13/1108: {Hard decision decoding, e.g. bit flipping, modified or weighted bit flipping}
- H03M 13/3715: {Adaptation to the number of estimated errors or to the channel state}
- H03M 13/015: {Simulation or testing of codes, e.g. bit error rate [BER] measurements}
- H03M 13/096: {Checksums}
- H03M 13/1111: {Soft-decision decoding, e.g. by means of message passing or belief propagation algorithms}
- H03M 13/1177: {Regular LDPC codes with parity-check matrices wherein all rows and columns have the same row weight and column weight, respectively}
- H03M 13/43: Majority logic or threshold decoding
- H03M 13/112: {with correction functions for the min-sum rule, e.g. using an offset or a scaling factor}
- H03M: CODING; DECODING; CODE CONVERSION IN GENERAL (using fluidic means F15C4/00; optical analogue/digital converters G02F7/00; coding, decoding or code conversion, specially adapted for particular applications, see the relevant subclasses, e.g. G01D, G01R, G06F, G06T, G09G, G10L, G11B, G11C, H04B, H04L, H04M, H04N; ciphering or deciphering for cryptography or other purposes involving the need for secrecy G09C)

## Abstract

Techniques are described for improving the decoding latency and throughput of an error correction system that includes a bit flipping (BF) decoder, where the BF decoder uses a bit flipping procedure. In an example, different decoding parameters are determined including any of a decoding number of a decoding iteration, a checksum of a codeword, a degree of a variable node, and a bit flipping threshold defined for the bit flipping procedure. Based on one or more of these decoding parameters, a decision can be generated to skip the bit flipping decoding procedure, thereby decreasing the decoding latency and increasing the decoding throughput. Otherwise, the bit flipping decoding procedure can be performed to compute a bit flipping energy and determine whether particular bits are to be flipped or not. Hence, the overall performance (e.g., bit error rate) is not significantly impacted.

## Claims

### Claim 1

A method for decoding a low density parity check (LDPC) codeword, the method implemented by an error correction system that comprises a bit flipping (BF) decoder using a bit flipping procedure, the method comprising:
during a decoding iteration and prior to performing the bit flipping procedure:
selecting a set of variable nodes;
determining a degree of a variable node of the set, wherein the degree is the same for the variable nodes of the set;
determining a decoding factor that comprises at least one of: an iteration number “m” of the decoding iteration or a checksum of the LDPC codeword, wherein the iteration number “m” is an integer equal to or greater than one;
determining a bit flipping threshold “T”, wherein the bit flipping threshold “T” is predefined for the bit flipping procedure based on the iteration number “m”; and
generating a decision indicating whether to perform or to skip the bit flipping procedure, wherein the decision is generated based on the degree, the decoding factor, and the bit flipping threshold “T”, and
wherein the LDPC codeword is decoded based on the decision.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
performing a first comparison between the degree of the variable node and a degree threshold, wherein the decision is generated based on an outcome of the first comparison.

### Claim 3 (depends on 2)

The method of claim 2, wherein the degree threshold is equal to or less than five.

### Claim 4 (depends on 2)

The method of claim 2, wherein the decoding factor comprises the iteration number “m”, the method further comprising:
performing a second comparison between the iteration number “m” and a number threshold; and
performing a third comparison between the bit flipping threshold “T” and one or more predefined values,
wherein the decision is generated based on an outcome of the second comparison and an outcome of the third comparison.

### Claim 5 (depends on 4)

The method of claim 4, wherein each one of the one or more predefined values is a different value within a range that is defined based on the degree threshold.

### Claim 6 (depends on 4)

The method of claim 4, wherein the decision is to skip performing the bit flipping procedure based on (i) the first comparison indicating that the degree is smaller than the degree threshold, (ii) the second comparison indicating that the iteration number “m” is smaller than the number threshold, and (iii) the third comparison indicating that the bit flipping threshold “T” is equal to any of the one or more predefined values.

### Claim 7 (depends on 4)

The method of claim 4, wherein the decision is to perform the bit flipping procedure based on any of: (i) the first comparison indicating that the degree is greater than the degree threshold, (ii) the second comparison indicating that the iteration number “m” is greater than the number threshold, or (iii) the third comparison indicating that the bit flipping threshold “T” is not equal to any of the one or more predefined values.

### Claim 8 (depends on 2)

The method of claim 2, wherein the decoding factor comprises the checksum, the method further comprising:
performing a second comparison between the checksum and a checksum threshold; and
performing a third comparison between the bit flipping threshold “T” and one or more predefined values,
wherein the decision is generated based on an outcome of the second comparison and an outcome of the third comparison.

### Claim 9 (depends on 8)

The method of claim 8, wherein the decision is to skip performing the bit flipping procedure based on (i) the first comparison indicating that the degree is smaller than the degree threshold, (ii) the second comparison indicating that the checksum is greater than the checksum threshold, and (iii) the third comparison indicating that the bit flipping threshold “T” is equal to any of the one or more predefined values.

### Claim 10 (depends on 8)

The method of claim 8, wherein the decision is to perform the bit flipping procedure based on any of: (i) the first comparison indicating that the degree is greater than the degree threshold, (ii) the second comparison indicating that the iteration number “m” is smaller than the checksum threshold, or (iii) the third comparison indicating that the bit flipping threshold “T” is not equal to any of the one or more predefined values.

### Claim 11 (depends on 1)

The method of claim 1, wherein the decision is to perform the bit flipping procedure, and wherein performing the bit flipping procedure comprises:
computing a flipping energy for the variable node;
performing a comparison of the flipping energy and the bit flipping threshold “T”; and
flipping one or more bits of the variable node based on the comparison indicating the flipping energy is greater than the bit flipping threshold “T”.

### Claim 12

An error correction system for decoding a low density parity check (LDPC) codeword, the error correction system comprising:
a bit flipping (BF) decoder configured to, during a decoding iteration and prior to performing a bit flipping procedure:
select a set of variable nodes;
determine a degree of a variable node of the set, wherein the degree is the same for the variable nodes of the set;
determine a decoding factor that comprises at least one of: an iteration number “m” of the decoding iteration or a checksum of the LDPC codeword, wherein the iteration number “m” is an integer equal to or greater than one;
determine a bit flipping threshold “T”, wherein the bit flipping threshold “T” is predefined for the bit flipping procedure based on the iteration number “m”;
generate a decision indicating whether to perform or to skip the bit flipping procedure, wherein the decision is generated based on the degree, the decoding factor, and the bit flipping threshold “T”; and
decode the LDPC codeword based on the decision.

### Claim 13 (depends on 12)

The error correction system of claim 12, wherein the BF decoder is further configured to:
perform a first comparison between the degree of the variable node and a degree threshold, wherein the decision is generated based on an outcome of the first comparison.

### Claim 14 (depends on 13)

The error correction system of claim 13, wherein the degree threshold is equal to or less than five.

### Claim 15 (depends on 13)

The error correction system of claim 13, wherein the decoding factor comprises the iteration number “m”, and wherein the BF decoder is further configured to:
perform a second comparison between the iteration number “m” and a number threshold; and
perform a third comparison between the bit flipping threshold “T” and one or more predefined values,
wherein the decision is generated based on an outcome of the second comparison and an outcome of the third comparison.

### Claim 16 (depends on 15)

The error correction system of claim 15, wherein each one of the one or more predefined values is a different value within a range that is defined based on the degree threshold.

### Claim 17 (depends on 15)

The error correction system of claim 15, wherein the decision is to skip performing the bit flipping procedure based on (i) the first comparison indicating that the degree is smaller than the degree threshold, (ii) the second comparison indicating that the iteration number “m” is smaller than the number threshold, and (iii) the third comparison indicating that the bit flipping threshold “T” is equal to any of the one or more predefined values.

### Claim 18

A memory device comprising an error correction system and storing computer-readable instructions, that upon execution by the error correction system that comprises a bit flipping (BF) decoder using a bit flipping procedure for decoding a low density parity check (LDPC) codeword, cause the error correction system to perform operations comprising:
during a decoding iteration and prior to performing the bit flipping procedure:
selecting a set of variable nodes;
determining a degree of a variable node of the set, wherein the degree is the same for the variable nodes of the set;
determining a decoding factor that comprises at least one of: an iteration number “m” of the decoding iteration or a checksum of the LDPC codeword, wherein the iteration number “m” is an integer equal to or greater than one;
determining a bit flipping threshold “T”, wherein the bit flipping threshold “T” is predefined for the bit flipping procedure based on the iteration number “m”; and
generating a decision indicating whether to perform or to skip the bit flipping procedure, wherein the decision is generated based on the degree, the decoding factor, and the bit flipping threshold “T”, and
wherein the LDPC codeword is decoded based on the decision.

### Claim 19 (depends on 18)

The error correction system of claim 18, wherein the operations further comprise:
performing a first comparison between the degree of the variable node and a degree threshold, wherein the decision is generated based on an outcome of the first comparison.

### Claim 20 (depends on 19)

The memory device of claim 19, wherein the decoding factor comprises the checksum, wherein the operations further comprise:
performing a second comparison between the checksum and a checksum threshold; and
performing a third comparison between the bit flipping threshold “T” and one or more predefined values,
wherein the decision is generated based on an outcome of the second comparison and an outcome of the third comparison.
