Decoder Threshold Updating for Faster NAND Memory Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory devices, particularly NAND type memories, face challenges in improving decoding performance and reducing decoding time due to the monotonically decreasing flipping threshold during iterative decoding processes.
Innovation Solution
A decoder system with a first and second processing circuit, a processor, and a bit flipping circuit that dynamically updates the flipping threshold based on the changing state of the check expression weight and energy of the codeword in each iteration, allowing for more reasonable threshold settings and improved decoding performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a monotonically decreasing flipping threshold is used during iterative decoding, then the decoding process is simple to implement, but the decoding performance deteriorates and decoding time increases
Solution Approach 1:
The patent applies the dynamics principle by transforming the static monotonically decreasing flipping threshold into a dynamic threshold that adapts based on the changing state of the check expression weight. The processor monitors the check expression weight across iterations and adjusts the flipping threshold accordingly, making the threshold flexible rather than fixed. This resolves the contradiction by maintaining implementation simplicity while improving decoding performance through adaptive threshold adjustment.
Solution Approach 2:
The patent implements parameter changes by modifying the flipping threshold parameter based on the check expression weight state. Instead of using a fixed monotonically decreasing threshold, the system changes the threshold parameter dynamically according to the decoding progress and error patterns detected. This allows the system to optimize decoding performance by adjusting the threshold to match the actual decoding state, resolving the performance deterioration issue.
2Device complexity
If a monotonically decreasing flipping threshold is used during iterative decoding, then the threshold setting is simple, but the decoding time increases
Solution Approach 1:
The patent applies feedback by having the processor continuously monitor the check expression weight during decoding iterations and use this feedback to adjust the flipping threshold. The system feeds back the decoding state information to the threshold adjustment mechanism, creating a closed-loop control system. This resolves the contradiction by maintaining simple threshold setting procedures while reducing decoding time through real-time adaptive adjustments based on actual decoding progress.
3Reliability
If the flipping threshold is dynamically updated based on check expression weight, then decoding performance is improved, but the device complexity increases
Solution Approach 1:
The patent implements self-service by enabling the decoder to automatically adjust its own flipping threshold based on its internal state (check expression weight). The processor within the decoder performs the threshold adjustment autonomously using the decoding progress information already available in the system. This resolves the contradiction by improving decoding performance through dynamic threshold updates while minimizing additional device complexity, as the adjustment mechanism uses existing decoder components and information.
Data Source
AI summary
In an example, a decoder comprises a first processing circuit configured to: obtain a check expression and a check expression weight based on a codeword to be decoded in a current iteration and a check matrix; a second processing circuit configured to: obtain energy of the codeword to be decoded in the current iteration based on the check matrix, the check expression, and a flipping state of the codeword to be decoded in the current iteration; a processor configured to: determine a flipping threshold in the current iteration based on a changing state of the check expression weight; and a bit flipping circuit configured to: output a codeword to be decoded in a next iteration based on a comparison result of the energy of the codeword to be decoded in the current iteration and the flipping threshold in the current iteration determined by the processor.


