Adaptive Hard and Soft Bit Decoding for Non-Volatile Memory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-volatile memory systems face challenges in accurately reading and decoding data due to variations in threshold voltages over time, leading to errors that traditional hard bit decoding methods struggle to correct, especially as the memory ages and die-to-die variations occur.
Innovation Solution
Implementing a dynamic read table that adapts by tracking the most frequently successful adjustments to hard bit read reference voltages and incorporating soft bit decoding, allowing for dynamic hard bit plus soft bit read reference voltages to improve decoding accuracy and efficiency, while minimizing power and time consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional hard bit decoding is used, then decoding speed is fast, but decoding accuracy deteriorates due to threshold voltage variations and aging
Solution Approach 1:
The patent implements dynamic read tables that are updated based on historical decoding success rates. The system transitions from static hard bit reference voltages to dynamic soft bit reference voltages that adapt to aging conditions and threshold voltage variations, thereby maintaining high decoding accuracy without requiring exhaustive searches through all possible voltage combinations.
Solution Approach 2:
The system performs preliminary tracking of decoding success rates for different read reference voltage adjustments and stores these in dynamic read tables. When a decoding failure occurs, the system can quickly consult pre-computed soft bit reference voltages from the dynamic read table, avoiding time-consuming real-time calculations and enabling fast recovery from decoding failures.
2Measurement precision
If comprehensive adjustments to read reference voltages are attempted, then decoding accuracy improves, but power consumption and processing complexity increase
Solution Approach 1:
The patent applies different decoding strategies to different data groups based on their specific characteristics. Instead of uniformly applying complex soft bit decoding to all data, the system uses hard bit decoding for data groups that decode successfully and reserves soft bit decoding for specific cases where hard bit decoding fails. This localized application of decoding methods reduces overall processing complexity while maintaining high accuracy.
Solution Approach 2:
The system performs only the necessary adjustments to read reference voltages based on tracking historical success rates. Rather than exhaustively testing all possible voltage adjustments, the system identifies and applies only the most frequently successful adjustments from the dynamic read table, achieving high decoding accuracy with minimal processing overhead.
3Reliability
If soft bit decoding is always used, then decoding accuracy is high, but power consumption and processing time increase
Solution Approach 1:
The patent dynamically selects between hard bit and soft bit decoding based on real-time conditions and historical performance data. The system transitions from a static approach where soft bit decoding would be always used to a dynamic approach that adapts to current decoding needs, using soft bit decoding only when necessary to correct errors that hard bit decoding cannot handle, thereby reducing power consumption while maintaining high reliability.
Solution Approach 2:
The patent segments the decoding process into two distinct stages: first attempting hard bit decoding for all data groups, and then applying soft bit decoding only to specific data groups that fail the initial hard bit decoding attempt. This segmentation allows the system to achieve high decoding success rates while minimizing the power consumption and processing overhead associated with soft bit decoding by limiting its application to only when necessary.
Data Source
AI summary
Technology is described herein for operating non-volatile storage. In one embodiment, the memory system tracks which adjustments to default values for hard bit read reference voltages are most frequently successful to decode data in non-volatile memory cells. In response to a process that uses only hard bits failing to successfully decode data in a group of the non-volatile memory cells, the memory system attempts to decode the data in the group of non-volatile memory cells using dynamic hard bit read reference voltages and dynamic soft bit read reference voltages that correspond to only a subset of the most frequently successful adjustments to the default values for the hard bit read reference voltages. By only using a subset of the most frequently successful adjustments to the default values for the hard bit read reference voltages time and power is saved.