ECC Decoder Switching with Self-Generated LLR for Lower Latency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing error-correcting code (ECC) decoding techniques, such as LDPC codes, are computationally complex and require improvements in decoding performance, especially in data communication and storage systems where data corruption is common.

Innovation Solution

A decoder that generates reliability information for each code symbol through hard decision decoding, switching from hard decision to soft decision decoding based on a threshold number of flips or syndrome value changes, and back to hard decision if soft decision decoding does not improve the syndrome value, optimizing the decoding process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If soft decision decoding is used to improve decoding performance, then error correction capability is improved, but computational complexity increases

Engineering Contradiction:
Improveerror correction capabilityVSAvoidcomputational complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The decoder dynamically switches between hard decision decoding and soft decision decoding based on the syndrome value and iteration count. The switching rule adjusts the decoding mode in real-time: using hard decision for initial iterations when syndrome value is high, and transitioning to soft decision when syndrome value decreases below a threshold or after a predetermined number of iterations, thereby optimizing the balance between error correction capability and computational complexity

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the decoding parameter (decision type) based on the syndrome value and iteration progress. By monitoring these parameters and switching decoding modes accordingly, the system adapts the computational intensity to match the actual error correction needs, reducing unnecessary computational complexity while maintaining reliability

Inventive Principle:
Principle #35Parameter changes

2Productivity

If hard decision decoding is used to reduce computational complexity, then decoding speed is improved, but error correction performance deteriorates

Engineering Contradiction:
Improvedecoding speedVSAvoiderror correction performance
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The decoder employs periodic hard decision decoding interspersed with soft decision decoding. Instead of continuously using computationally intensive soft decision decoding, the system periodically switches to hard decision decoding at predetermined intervals or when syndrome value indicates sufficient correction, thereby maintaining decoding speed while periodically enhancing error correction performance through soft decision when needed

Inventive Principle:
Principle #19Periodic action

3Reliability

If soft data information is always used to improve decoding performance, then reliability is improved, but latency increases due to re-read requirements

Engineering Contradiction:
Improvedecoding performanceVSAvoiddecoding latency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies soft decision decoding partially rather than always. By using soft decision decoding only when the syndrome value falls below a threshold or after a predetermined number of hard decision iterations, the system achieves improved reliability only when necessary, avoiding the latency penalty of always using soft decision decoding or requiring re-reads for soft data information

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS10715182B2Systems and methods for decoding error correcting codes with self-generated LLR
Publication Date: 2020.07.14 INNOGRIT TECH CO LTD
  • US10715182B2 patent drawing
  • US10715182B2 patent drawing
  • US10715182B2 patent drawing

AI summary

Systems and methods are provided for decoding a codeword encoded by a linear block code. A method may comprise performing a hard decision decoding on a codeword, recording a number of flip(s) for each bit of the codeword, generating reliability information for each bit based on the number of flip(s) for each bit respectively, determining to switch to soft decision decoding according to a switching rule and performing a soft decision decoding on the codeword using the reliability information for each bit.