Soft Decoding LLR Switching for Early Failure Termination

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing error decoding processes in memory systems face inefficiencies due to high-reliability error rates (HRERs), leading to increased computational time and decreased speed when attempting to correct errors in codewords.

Innovation Solution

Implementing multiple iterative soft decoding processes with different log-likelihood ratio (LLR) sets, allowing for early termination based on the reliability values of bits during decoding, to adapt to varying HRERs and enhance computational efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If multiple iterative soft decoding processes with different LLR sets are implemented, then error correction efficiency is improved, but device complexity increases

Engineering Contradiction:

Improveerror correction efficiencyVSAvoiddecoding process complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The decoding process is segmented into multiple independent soft decoding processes, each using a different LLR set. Each process can be independently executed and terminated, allowing the system to divide the complex error correction task into manageable segments that can be tried in sequence until success.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system dynamically selects and switches between different LLR sets based on decoding performance. The controller monitors the decoding process and can terminate ineffective processes while switching to alternative LLR sets, creating a dynamic adaptation mechanism that responds to actual decoding needs.

Inventive Principle:

Principle #15Dynamics

2Loss of time

If early termination of decoding process is implemented, then decoding latency is reduced, but measurement precision of error detection may be affected

Engineering Contradiction:

Improvedecoding latencyVSAvoiderror detection accuracy

Core Design Contradiction:

Loss of timeVSMeasurement precision

Solution Approach 1:

The system performs preliminary actions by implementing multiple soft decoding processes with different LLR sets before committing to a final decoding result. By trying alternative LLR sets early in the process, the system can identify and correct errors before full decoding completion, reducing overall latency while maintaining accuracy.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The decoding process incorporates feedback mechanisms where the controller monitors decoding progress and performance metrics. Based on this feedback, the system can determine when to terminate a decoding process early or switch to alternative LLR sets, balancing speed and accuracy through continuous monitoring and adaptation.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20260031835A1Iterative error decoding with early termination of decoding process

Publication Date: 2026.01.29 MICRON TECHNOLOGY INC

AI summary

A method for decoding data is disclosed. The method includes determining that a codeword received from a memory device has errors and initiating a soft decoding process to decode embedded data from the codeword based on assigning respective LLR values to bits of the codeword and based on updating LLR values of the bits at each iteration of the soft decoding process. The method also includes incrementing an iteration counter at each iteration of the soft decoding process and determining that the soft decoding process will be unable to decode the embedded data before the iteration counter achieves a maximum iteration count based on the updated LLR values of the bits. The method further includes terminating the soft decoding process at an iteration count that is less than the maximum iteration count in response to the determination that the soft decoding process will be unable to decode the embedded data.