LDPC Decoding Parameter Tuning for Non-Volatile Memory Reads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing decoding methods for rewritable non-volatile memory modules face challenges in obtaining appropriate log likelihood ratios (LLR) for LDPC decoding, affecting the performance and efficiency of error correction and decoding operations.
Innovation Solution
A decoding method that calculates and adjusts decoding parameters, specifically the LLR, based on the distribution of bits read from memory cells, to improve decoding correction capability and probability of successful decoding, reducing decoding delay.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional fixed decoding parameters are used for LDPC decoding, then the decoding process is simple and fast, but the decoding correction capability and probability of successful decoding are insufficient
Solution Approach 1:
The patent applies dynamics by transforming the decoding parameter from a fixed value to a dynamically adjustable value. The system calculates the actual distribution of 0s and 1s in the received codeword and adjusts the LLR (log-likelihood ratio) initialization parameter accordingly. This dynamic adjustment allows the decoding process to adapt to different data distributions, improving correction capability while maintaining reasonable complexity through automated calculation.
Solution Approach 2:
The patent directly implements parameter changes by modifying the LLR initialization value based on the measured bit distribution. When the distribution deviates from the ideal 50-50 ratio, the system changes the LLR parameter to compensate for the imbalance. This parameter change approach resolves the contradiction by improving reliability through adaptive parameter tuning without requiring complex structural modifications to the decoder architecture.
2Reliability
If decoding parameters are adjusted to improve correction capability, then the probability of successful decoding increases, but the decoding delay increases
Solution Approach 1:
The patent applies preliminary action by calculating the bit distribution and determining the appropriate LLR adjustment before the main decoding process begins. The system performs a quick statistical analysis of the received codeword to determine the distribution deviation, then pre-adjusts the decoding parameter accordingly. This preliminary preparation ensures that the subsequent decoding operation can proceed efficiently with optimized parameters, reducing overall delay while improving success probability.
Solution Approach 2:
The patent implements feedback by using the actual bit distribution information from the received signal to adjust the decoding parameters. The system measures the distribution of 0s and 1s, compares it with the ideal distribution, and feeds this information back to modify the LLR parameter. This feedback mechanism ensures that parameter adjustment is based on actual channel conditions, improving decoding reliability while avoiding unnecessary adjustments that would increase delay.
3Reliability
If the distribution of bits is not even (deviation from 50-50 ratio), then errors occur during decoding, but adjusting parameters to handle uneven distribution increases processing complexity
Solution Approach 1:
The patent applies self-service by enabling the decoding system to automatically detect and correct its own parameter mismatches. The system independently calculates the bit distribution, identifies deviations from the ideal 50-50 ratio, and self-adjusts the LLR parameter without external intervention. This self-service capability maintains decoding accuracy under varying conditions while preserving operational simplicity through automated adaptation.
Solution Approach 2:
The patent resolves the contradiction through parameter changes by modifying the LLR initialization value based on the measured distribution. When bits are unevenly distributed, the system changes the parameter to compensate for the imbalance. This approach maintains decoding accuracy without requiring complex operational procedures, as the parameter adjustment is automatically calculated from the observed distribution statistics.
Data Source
AI summary
A decoding method, a memory storage device and a memory controlling circuit unit are provided. The method includes: receiving a read command sequence for reading a plurality of bits from the memory cells; calculating a first count value of a first value and a second count value of a second value in the bits; and adjusting a decoding parameter corresponding to the bits to a specific decoding parameter according to the first count value and the second count value, and performing a decoding operation according to the specific decoding parameter, where the adjusted decoding parameter affects a probability that the bits are considered as an error bit in the decoding operation.