Memory Data Processing Structure for Multi-Bit Burst Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memories with channel error detecting and correcting functions can only detect and repair 1-bit data errors, which is insufficient for ensuring data accuracy during transmission and storage, necessitating a method to detect and correct multi-bit data errors.
Innovation Solution
A data processing method that involves obtaining raw data, grouping it into split data, encoding each split data with check data, reorganizing the encoded data to generate write data, storing it in a memory cell, and then decoding and checking the read data to generate corrected read data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing error detecting and correcting functions are used, then 1-bit data errors can be detected and repaired, but multi-bit data errors cannot be corrected
Solution Approach 1:
The raw data is divided into multiple first split data groups, and each group is independently encoded with check data. This segmentation allows the system to handle multi-bit errors by treating them as multiple 1-bit errors across different segments, thereby extending the error correction capability beyond the limitation of correcting only single-bit errors in traditional systems.
2Reliability
If data is directly stored without grouping and encoding, then storage process is simple, but multi-bit burst errors cannot be detected and corrected
Solution Approach 1:
The system performs preliminary grouping and encoding of raw data into first split data with check data before storage. This preliminary action embeds error detection and correction capabilities into the data structure itself, enabling the memory system to detect and correct multi-bit burst errors during the read and decode process without requiring complex post-processing circuits.
3Reliability
If raw data is grouped into split data with check data, then multi-bit errors can be corrected, but processing time and complexity increase
Solution Approach 1:
By segmenting the data into smaller first split data groups with individual check data, the encoding and decoding operations can be performed in parallel on multiple segments simultaneously. This segmentation reduces the overall processing time compared to encoding/decoding the entire data block as a single unit, thereby mitigating the time loss despite the added processing steps.
Data Source
AI summary
Embodiments relate to the field of semiconductor circuit design, and more particularly, to a data processing method, a data processing structure, and a memory. The data processing method includes: obtaining raw data to be stored, and grouping the raw data to obtain first split data, the first split data having equal number of code elements; encoding each of the first split data to generate first encoded data, where the first encoded data includes the first split data and check data corresponding to the first split data; reorganizing the first encoded data to generate write data; storing the write data into a memory cell; and obtaining read data in the memory cell, and decoding and checking the read data to generate corrected read data, to correct a multi-bit burst error occurring in the memory during data storage or data transmission.