Memory Encoding with EDC-ECC Protection Against Tearing Errors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data corruption due to 'tearing' occurs when power is interrupted during writing or erasing operations in memory devices like RFID chips and IoT scenarios, leading to corrupted data.
Innovation Solution
A method involving encoding data with an error detection code (EDC) that detects up to 4-bit errors, followed by concatenation with inverted Hamming weights and error correction code (ECC) to correct 1-bit errors, ensuring data integrity during power interruptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is written to memory without error detection and correction codes, then memory operation speed and energy efficiency are improved, but data reliability deteriorates due to tearing errors during power interruptions
Solution Approach 1:
The error protection mechanism is segmented into two distinct components: an Error Detection Code (EDC) that detects up to 4-bit errors, and an Error Correction Code (ECC) that corrects 1-bit errors. This segmentation allows the system to handle different error types with appropriate mechanisms, improving reliability without requiring a single complex all-encompassing error protection system.
Solution Approach 2:
The patent applies error detection and correction codes preliminarily during the data encoding phase before memory writing operations. By pre-encoding data with EDC and ECC, and storing additional check bits (including inverted Hamming weights), the system prepares error protection in advance, enabling automatic error detection and correction during power interruption recovery without requiring complex real-time processing during critical write operations.
2Reliability
If error detection and correction codes are applied to all data, then data reliability is improved, but energy consumption increases
Solution Approach 1:
The patent implements partial error protection by applying EDC to detect up to 4-bit errors and ECC to correct 1-bit errors, rather than using more extensive error protection mechanisms. This partial action approach provides sufficient reliability for the intended application while minimizing energy consumption during encoding operations, balancing protection adequacy with energy efficiency.
Data Source
AI summary
Various embodiments relate to a method and system for encoding data to be stored in a memory, including: encoding the data to be stored in memory with an error detection code (EDC), that can detect up to 4 bit errors, as first encoded data; determining the Hamming weight of the first encoded data; inverting the determined Hamming weight; concatenating the first encoded data and three copies of the inverted Hamming weight as concatenated data; encoding the concatenated data with an error correcting code (ECC), that can correct 1 bit error, as second encoded data; and storing the second encoded data in the memory.