Memory Charge Loss Tracking With Temperature Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Memory devices experience increased bit error rates due to slow charge loss, which causes threshold voltage distribution shifts over time and temperature variations, making it difficult to predict optimal threshold voltage offsets without compromising performance.
Innovation Solution
Implementing adaptive block family error avoidance (BFEA) by grouping blocks with similar programming times and temperatures, and using a table to adjust charge loss values based on operating temperatures relative to a reference temperature, ensuring accurate bin assignments and reducing errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If charge loss is tracked using fixed temperature assumptions, then device complexity is reduced, but measurement precision of charge loss deteriorates due to temperature variations
Solution Approach 1:
The patent implements dynamic temperature compensation by adjusting charge loss tracking based on actual temperature measurements. The system transitions from static fixed-temperature assumptions to dynamic temperature-adaptive tracking, where compensation values are modified according to measured temperature deviations from reference conditions.
Solution Approach 2:
The patent changes the parameter being tracked from fixed charge loss values to temperature-compensated charge loss values. By introducing temperature as a variable parameter and adjusting compensation accordingly, the system maintains measurement precision across varying thermal conditions without requiring completely new tracking infrastructure.
2Reliability
If threshold voltage offsets are adjusted frequently to account for charge loss, then bit error rate is reduced, but device complexity increases due to additional control mechanisms
Solution Approach 1:
The patent applies preliminary temperature compensation adjustments based on predicted temperature changes and historical charge loss patterns. By pre-calculating and applying compensation before actual threshold voltage shifts occur, the system reduces bit error rates without requiring complex real-time control mechanisms during critical operations.
Solution Approach 2:
The patent implements feedback mechanisms where actual temperature measurements and charge loss observations are continuously monitored and used to adjust threshold voltage offsets. This closed-loop feedback system maintains reliability by automatically compensating for drift while keeping control complexity manageable through established feedback control principles.
3Measurement precision
If temperature compensation is implemented for charge loss tracking, then charge loss measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent introduces temperature compensation values as an intermediary mechanism between raw charge loss measurements and corrected charge loss values. This intermediary layer allows precise temperature-based adjustment without requiring direct complex interaction between temperature sensing and charge loss tracking systems, simplifying the overall implementation.
Solution Approach 2:
The patent segments the temperature compensation process into distinct components: temperature measurement, reference value storage, compensation calculation, and application to charge loss tracking. By dividing the compensation mechanism into separate functional segments, the system achieves high measurement precision while managing complexity through modular design that allows independent optimization of each component.
Data Source
AI summary
A system includes a memory device and a processing device operatively coupled with the memory device to perform operations including determining an operating temperature value for the memory device; determining a charge loss adjustment value based on the operating temperature value of the memory device and a reference temperature value for the memory device; determining a charge loss value for a first block of a plurality of blocks of the memory device; obtaining a normalized charge loss value for the first block based on the charge loss value and the charge loss adjustment value; and performing a first memory operation on the first block based on the normalized charge loss value.