Memory Read Voltage Estimation for Soft Decoding Without Extra Reads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory systems face challenges in reducing latency and enhancing error correction capability, particularly in determining optimal read voltages to minimize read errors and improve data retrieval reliability in non-volatile memory devices.
Innovation Solution
A system and method that apply multiple read voltages to memory cells, process the resulting data to determine optimal read voltages using a non-negative regularization parameter, generate log-likelihood ratio (LLR) values, and perform soft decoding without additional reads, thereby improving latency and error correction capability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If additional reads are performed to generate LLR values for soft decoding, then error correction capability is improved, but latency increases
Solution Approach 1:
The patent performs preliminary reads at multiple candidate read voltages before decoding to collect cell count data. This preliminary action enables the system to determine the optimal read voltage and generate accurate LLR values without requiring additional reads during the decoding process, thus reducing latency while maintaining error correction capability
Solution Approach 2:
The patent collects cell count data from multiple candidate read voltages in advance to create a data set that cushions against voltage variations. This beforehand data collection allows the system to identify the optimal read voltage and generate robust LLR values that improve error correction while avoiding the need for additional corrective reads
2Reliability
If multiple candidate read voltages are applied to determine optimal read voltage, then data retrieval reliability is improved, but read operations complexity increases
Solution Approach 1:
The patent applies read voltages at multiple candidate levels (excessive action) to ensure the optimal read voltage is identified. By collecting cell count data from more candidate voltages than strictly necessary, the system guarantees reliable optimal voltage selection while managing complexity through efficient data processing algorithms
Solution Approach 2:
The patent varies the read voltage parameter across multiple candidate values to determine the optimal read voltage. By systematically changing this parameter and analyzing the resulting cell count data, the system identifies the optimal voltage that maximizes data retrieval reliability while using mathematical models to manage the complexity of multiple voltage applications
Data Source
AI summary
A controller of a memory system performs a soft decoding without additional reads. The controller applies each of read voltages to cells to obtain a corresponding cell count and corresponding data, stores the obtained data, and processes the stored data. The controller determines a set of parameters, based on (i) the read voltages, (ii) cell counts corresponding to the read voltages and (iii) a non-negative regularization parameter. The controller estimates an optimal read voltage based on the set of parameters, generates log-likelihood ratio (LLR) values using the processed data and the optimal read voltage and performs soft decoding using the LLR values.