LDPC Decoder Threshold Sequencing for Faster NAND Memory Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing NAND type memory systems face challenges in efficiently decoding data due to hardware and software failures, leading to data errors, which current error correcting codes like LDPC struggle to address effectively, particularly in high-density memory environments.
Innovation Solution
A decoder system that selects a plurality of preset threshold sequences based on the check formula weight, assigning a flipping threshold in each iteration to a bit flipping circuit to determine bit flipping, improving flexibility and reducing decoding time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a fixed threshold sequence is used for bit flipping in LDPC decoding, then the decoding process is simple to implement, but the decoding performance deteriorates when error degrees vary
Solution Approach 1:
The patent implements dynamic threshold selection by maintaining multiple threshold sequences (first, second, third sequences) and dynamically switching between them based on the check formula weight in different iterations. The threshold changes from fixed to adaptive, allowing the decoding system to respond to varying error conditions and improve reliability without excessive complexity.
Solution Approach 2:
The patent changes the parameter of threshold values by providing multiple preset threshold sequences with different threshold levels. The processor selects appropriate threshold sequences based on iteration number and check formula weight, enabling the system to adapt to different error degrees and improve decoding performance through parameter variation.
2Reliability
If multiple threshold sequences are selected based on check formula weight, then decoding performance for varying error degrees improves, but decoding time increases
Solution Approach 1:
The patent performs preliminary action by presetting multiple threshold sequences before decoding begins. These sequences are pre-calculated and stored, so during actual decoding, the processor only needs to select from the preset options based on check formula weight, avoiding time-consuming real-time calculations and reducing overall decoding time.
Solution Approach 2:
The patent implements periodic action by using different threshold sequences in different iteration periods. The first threshold sequence is used in early iterations, the second in middle iterations, and the third in later iterations, creating a periodic pattern that balances decoding accuracy with time efficiency across the decoding process.
3Adaptability or versatility
If check formula weight is used to select threshold sequences, then adaptability to different error conditions improves, but processing complexity increases
Solution Approach 1:
The patent applies local quality by using the check formula weight (a local metric for each codeword) to determine which global threshold sequence to use. This allows the system to adapt to different error conditions locally based on the specific codeword's error characteristics while maintaining a manageable overall processing structure.
Solution Approach 2:
The patent reduces processing complexity through preliminary action by pre-organizing multiple threshold sequences with different characteristics. The processor only needs to compare the check formula weight against predetermined criteria and select from the preset sequences, avoiding the need for complex real-time threshold optimization algorithms.
Data Source
AI summary
According to one aspect of the present disclosure, a decoder is provided. The decoder may include a first processing circuit to: obtain a check formula and check formula weight based on a codeword to be decoded in a current iteration and a check matrix; a second processing circuit to: obtain energy of the codeword to be decoded based on the check formula, the check matrix and a flipping state of the codeword to be decoded; a processor to: in a first iteration, select a preset threshold sequence as an initial threshold sequence based on the check formula weight; and assign a flipping threshold in the initial threshold sequence to a bit flipping circuit; and the bit flipping circuit to: output a codeword to be decoded in a following iteration based on a result of a comparison between energy of the codeword to be decoded and the flipping threshold.