Compressed LLR Generation for Non-Volatile Memory Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing error correction codes (ECCs) in non-volatile memory devices require high computational resources and extra storage space for ECC parity bits, leading to inefficiencies in data reliability and complexity.
Innovation Solution
The method involves generating log-likelihood ratios (LLRs) in a compressed form, supporting flexible bit-widths and adapting to various bit error scenarios, using a combination of hard reads, quality metric computation, and compressed inconsistency memory to reduce memory usage and latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If strong error-correction codes (ECCs) are used to improve data reliability, then data reliability is improved, but storage space and computational resources increase
Solution Approach 1:
The patent segments the LLR generation process into multiple hard read operations with incremental updates. Instead of performing a single complex soft read, the system performs multiple simpler hard reads, each updating the LLR values incrementally. This segmentation reduces the computational complexity of each individual operation while maintaining the overall error correction capability.
Solution Approach 2:
The patent performs preliminary hard read operations to establish initial LLR values before the final decoding operation. By pre-computing quality metrics and updating LLR values through multiple hard reads, the system prepares the data in advance, reducing the computational burden during the actual decoding phase and improving overall efficiency.
2Reliability
If strong error-correction codes (ECCs) are used to improve data reliability, then data reliability is improved, but storage space increases
Solution Approach 1:
The patent dynamically changes the parameter representation of LLR values by using incremental updates and compressed formats. Instead of storing full-precision LLR values, the system stores differences from previous values and reconstructs them as needed. This parameter transformation significantly reduces the storage space required for ECC parity bits while maintaining the reliability benefits of strong error correction.
3Reliability
If multiple hard reads are performed to generate LLRs, then error correction performance is improved, but read latency increases
Solution Approach 1:
The patent maintains continuity of useful action by performing multiple hard reads in sequence, with each read contributing incrementally to the LLR generation. Rather than performing one lengthy soft read operation, the system continuously performs shorter hard reads, updating LLR values incrementally. This continuous incremental approach improves error correction performance while managing read latency through efficient parallel processing and incremental updates.
4Measurement precision
If LLR bit-width is increased to improve decoding precision, then measurement precision is improved, but storage requirements increase
Solution Approach 1:
The patent transitions from storing LLR values in a single dimension (fixed bit-width) to a multi-dimensional approach using incremental updates and compressed representations. By storing differences from previous values and using variable-length encodings, the system achieves high effective precision without requiring proportionally increased storage. This dimensional transformation allows flexible precision adaptation without linear storage growth.
Data Source
AI summary
Devices, systems and methods for improving a decoding operation in a non-volatile memory are described. An example method includes performing a first hard read to obtain a first set of values stored in a plurality of cells, storing the first set of values in a first buffer, performing a plurality of subsequent hard reads on the plurality of cells to obtain a plurality of subsequent sets of values, performing, for each subsequent set of values, the following operations: computing a quality metric, storing, in a second buffer, a difference between the subsequent set of values and the set of values stored in the first buffer, wherein the difference is stored in a compressed format, and storing, in response to the quality metric exceeding a threshold, the subsequent set of values in the first buffer, and generating, based on the first buffer and the second buffer, the log-likelihood ratio.