Storage Controller LLR Calibration for NAND Read Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data reading methods for rewritable non-volatile memory modules require additional storage space for verified data, reducing available space and operational efficiency due to the need for iterative decoding corrections.
Innovation Solution
A data reading method that calibrates the log-likelihood ratio table using soft information from a codeword stored in a specific physical page without preparing verified data, improving decoding accuracy and efficiency by updating the table with trust codewords and soft information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If verified data is stored in the rewritable non-volatile memory module for iterative decoding correction, then decoding performance is improved, but the remaining available space is reduced
Solution Approach 1:
The patent introduces a buffer memory as an intermediary component between the non-volatile memory module and the storage controller. The buffer temporarily stores read data during iterative decoding operations, eliminating the need to use spare memory space for verified data storage. This mediator allows the system to maintain high decoding performance while preserving 100% of the non-volatile memory's available space for actual data storage.
2Reliability
If verified data is prepared and stored for decoding correction, then correct rate of decoding operation is improved, but operational efficiency is reduced
Solution Approach 1:
The patent segments the memory system into distinct functional components: the non-volatile memory module for data storage, the buffer memory for temporary data handling during decoding, and the storage controller for control operations. This segmentation allows each component to perform its specialized function efficiently, with the buffer handling the time-critical verified data storage without impacting the operational efficiency of the main storage system.
Solution Approach 2:
The buffer memory acts as a mediator that handles the overhead of iterative decoding operations separately from the main storage operations. By offloading verified data storage to the buffer, the system maintains high operational efficiency in the non-volatile memory while still achieving high decoding correct rates through the buffer's temporary storage capabilities.
3Reliability
If spare space is used to store verified data, then iterative decoding correction can be performed, but the remaining available space for other operations is reduced
Solution Approach 1:
The buffer memory serves as a dedicated intermediary resource that provides verified data storage capability without consuming any space from the non-volatile memory module. This allows the system to maintain full adaptability and versatility in managing the non-volatile memory's available space for various storage operations, while the buffer handles all iterative decoding correction requirements independently.
Data Source
AI summary
A memory management method is provided. The method includes: using a preset read voltage, a left preset read voltage set and a right preset read voltage set corresponding to the preset read voltage to perform a read operation on a target codeword to obtain a hard bit codeword, a left bit codeword and a right bit codeword, respectively; performing iterative decoding operations on each of the hard bit codeword, the left bit codeword and the right bit codeword to identify a trust codeword having a smallest syndrome among the hard bit codeword, the left bit codeword and the right bit codeword; using the hard bit codeword, the left bit codeword, the right bit codeword and the trust codeword to perform a calibration on a log-likelihood ratio table of the iterative operations, so as to update the log-likelihood ratio table to a calibrated log-likelihood ratio table.