DRNVM Error Correction via Reed-Solomon Codes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Destructive read non-volatile memory (DRNVM) is prone to errors during power varying events, such as brownouts, leading to data corruption or loss, especially when values need to be written back, which can result in malfunctioning devices due to corrupted program code.
Innovation Solution
The implementation of error correction codes, specifically Reed-Solomon codes, are encoded in DRNVM cells, allowing for the correction of single symbol errors in each code word, thereby reconstructing original data without the need for continuous capacitor charging and large capacitor sizes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If error correction codes are implemented in DRNVM to correct data corruption during power varying events, then data reliability is improved, but memory structure complexity increases
Solution Approach 1:
The memory is segmented into code word symbols organized in columns, with each column containing error correction code symbols. This segmentation allows targeted error correction at the code word level while maintaining overall memory structure integrity.
Solution Approach 2:
Error correction codes are encoded into the memory structure in advance, before power varying events occur. The Reed-Solomon codes are pre-positioned in specific columns to enable immediate error detection and correction when data corruption occurs during brownouts or power loss events.
2Reliability
If traditional capacitor-based power protection is used to prevent data corruption during brownouts, then data reliability is improved, but device size and energy consumption increase
Solution Approach 1:
The patent replaces the mechanical/electrical capacitor-based power protection system with a software/code-based error correction system. Instead of using physical capacitors to maintain power during brownouts, Reed-Solomon error correction codes are used to detect and correct data corruption, eliminating the need for continuous capacitor charging and large capacitor sizes.
Solution Approach 2:
The approach changes the fundamental parameter from power maintenance (keeping voltage stable during brownouts) to error tolerance (allowing corruption and correcting it). This parameter change enables the system to function without large capacitors, reducing both size and energy consumption while maintaining data reliability.
3Reliability
If large capacitors are used to protect against power varying events, then data reliability is improved, but device area increases
Solution Approach 1:
The patent replaces the mechanical/electrical capacitor-based power protection system with a software/code-based error correction system. Instead of using physical capacitors to maintain power during brownouts, Reed-Solomon error correction codes are used to detect and correct data corruption, eliminating the need for large capacitor sizes.
Solution Approach 2:
The error correction approach treats data protection as a disposable, regenerative process rather than requiring permanent, large-scale physical protection infrastructure. The error correction codes enable the system to survive power varying events without needing continuous physical protection, reducing device area requirements.
Data Source
AI summary
In described examples, data are stored in a destructive read non-volatile memory (DRNVM). The DRNVM includes an array of DRNVM cells organized as rows of data. The rows of data are subdivided into columns of code word symbols. Each column of code word symbols is encoded to store an error correction code symbol for each column of code word symbols.