Adaptive Read Detection for Multi-Level Solid-State Memory Drift
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multi-level solid-state storage devices face reliability issues due to signal level drift over time, which complicates the detection of codewords in memory cells, especially in flash and phase change memory technologies, leading to increased error rates and reduced storage capacity.
Innovation Solution
A method for detecting codewords in solid-state memory that involves reading memory cells to obtain signal components, ordering them by signal level, partitioning into segments based on frequency data, and determining reference signal levels to adaptively estimate current cell levels, thereby improving reliability and reducing errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multi-level cells are used to store more bits per cell, then storage capacity is improved, but reliability deteriorates due to signal level drift
Solution Approach 1:
The patent applies dynamics by making the reference signal levels adaptive rather than fixed. The reference levels are dynamically adjusted based on detected drift conditions, allowing the system to maintain accurate detection across varying operational states. This resolves the contradiction by enabling multi-level storage while compensating for drift-induced reliability issues through adaptive reference level adjustment.
Solution Approach 2:
The patent changes the parameter of reference signal levels from static to dynamic values. By estimating drift and adjusting reference levels accordingly, the system maintains detection accuracy despite physical quantity drift in the memory cells. This parameter change enables the system to achieve both high storage capacity through multi-level cells and reliable detection through adaptive reference adjustment.
2Reliability
If reference signal levels are varied to compensate for drift, then detection reliability is improved, but device complexity increases
Solution Approach 1:
The patent applies self-service by having the system estimate its own drift and adjust reference levels autonomously. The controller uses detected signal characteristics to estimate drift and automatically update reference levels without external intervention. This self-adjusting mechanism improves detection reliability while minimizing the complexity increase by eliminating the need for complex external calibration systems.
Solution Approach 2:
The patent implements feedback by continuously monitoring detected signal levels and using this information to estimate drift, which then feeds back into adjusting the reference levels. This closed-loop feedback mechanism maintains detection accuracy under varying conditions while keeping the controller complexity manageable through efficient use of detected information for automatic reference level adjustment.
3Measurement precision
If training data is used for drift estimation, then level estimation accuracy is improved, but storage capacity is reduced
Solution Approach 1:
The patent applies the extraction principle by removing the need for dedicated training data storage. Instead of reserving storage space for training data, the system extracts drift information from the actual user data during normal read operations. This allows the system to maintain full storage capacity for user data while still achieving accurate level estimation through drift detection and compensation using real data.
Solution Approach 2:
The patent makes the user data serve multiple functions: both storing information and providing training data for drift estimation. By using the same data for both purposes, the system eliminates the need for separate training data storage, thereby maintaining maximum storage capacity while achieving accurate level estimation through the multi-functional use of user data.
4Quantity of substance
If more levels are packed per memory cell, then storage density is improved, but measurement precision deteriorates due to increased susceptibility to drift
Solution Approach 1:
The patent applies dynamics by making reference signal levels adaptive rather than fixed. The reference levels are dynamically adjusted based on detected drift conditions, allowing the system to maintain accurate detection even when levels are closely spaced. This resolves the contradiction by enabling high storage density through multi-level cells while compensating for drift-induced measurement precision issues through adaptive reference level adjustment.
Solution Approach 2:
The patent changes the parameter of reference signal levels from static to dynamic values. By estimating drift and adjusting reference levels accordingly, the system maintains detection precision despite the reduced spacing between levels in high-density multi-level cells. This parameter change enables the system to achieve both high storage density and reliable measurement precision.
Data Source
AI summary
A method for detecting codewords of a length-N, qary-symbol code, the symbols of each codeword stored in respective q-level cells of solid-state memory, where q>2, includes reading from memory cells storing a group of codewords to obtain respective read signals each comprising N signal components corresponding to respective symbols of a codeword. The signal components of the read signals are ordered according to signal level to produce an ordered component sequence. The ordered component sequence is partitioned to obtain segments corresponding to respective memory cell levels. Each segment contains a number of components dependent on predetermined frequency data indicative of expected frequency of occurrence of the corresponding level in use of the code. A reference signal level corresponding to each q memory cell level is determined in dependence on the signal components. The codeword corresponding to each read signal is then detected in dependence on the reference signal levels.