Analog Memory Cell Reading via Multiple Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory devices face challenges in accurately reading data from analog memory cells due to distortion mechanisms such as electrical field coupling, disturb noise, and threshold voltage drift, leading to read errors and reduced data reliability.
Innovation Solution
A method involving multiple read operations using multiple thresholds positioned between adjacent nominal values, computing soft metrics from comparison results, and decoding data using Error Correction Codes (ECC) to improve error correction and data reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple read operations with multiple thresholds are performed, then data reading accuracy and error correction capability are improved, but device complexity and operation time increase
Solution Approach 1:
The reading process is segmented into multiple independent read operations, each using a different threshold value. Instead of attempting to read all data with a single threshold, the system divides the reading task into multiple stages, where each stage compares memory cell values against a specific threshold to generate partial results that are later combined for final decoding.
Solution Approach 2:
The system performs more read operations than the minimum single read would provide. By conducting multiple read operations with different thresholds, the system gathers excess information that goes beyond what a single read could provide, enabling robust error correction through soft metric decoding even though this increases operation count.
2Measurement precision
If multiple read operations with multiple thresholds are performed, then data reading accuracy and error correction capability are improved, but operation time increases
Solution Approach 1:
Multiple read operations are performed in advance to gather all necessary comparison results before the actual decoding process begins. By pre-fetching all the threshold comparison data needed for soft metric calculation, the system avoids repeated access delays during decoding and prepares complete information sets beforehand for more efficient processing.
3Measurement precision
If multiple thresholds are positioned between adjacent nominal values, then soft metric precision and decoding performance are improved, but device complexity increases
Solution Approach 1:
Different threshold values are strategically positioned at different locations between adjacent nominal values based on local characteristics of the memory cell value distributions. Rather than using uniform threshold spacing, the system places thresholds where they provide maximum discriminatory power for distinguishing between different stored values, optimizing local separation quality.
Solution Approach 2:
The system varies the threshold parameter across multiple read operations, using different threshold values to compare against memory cell contents. By changing this critical parameter (threshold value) across multiple operations, the system extracts multiple dimensions of information from the same memory cells, enabling more precise soft metric calculation.
4Reliability
If Error Correction Codes are used with multiple threshold decoding, then data reliability is improved, but processing complexity increases
Solution Approach 1:
The decoding process uses feedback from the multiple threshold comparison results to iteratively refine the soft metric values. The system feeds the comparison outcomes back into the decoding algorithm, allowing the decoder to adjust its interpretation based on accumulated evidence from multiple thresholds, thereby improving reliability through iterative refinement rather than single-pass decoding.
Data Source
AI summary
A method for operating a memory (28) includes storing data, which is encoded with an Error Correction Code (ECC), in analog memory cells (32) of the memory by writing respective analog input values selected from a set of nominal values to the analog memory cells. The stored data is read by performing multiple read operations that compare analog output values of the analog memory cells to different, respective read thresholds so as to produce multiple comparison results for each of the analog memory cells. At least two of the read thresholds are positioned between a pair of the nominal values that are adjacent to one another in the set of the nominal values. Soft metrics are computed responsively to the multiple comparison results. The ECC is decoded using the soft metrics, so as to extract the data stored in the analog memory cells.