One-Hot Encoding Error Protection for Storage Elements
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in effectively protecting against single event upsets (SEUs) and multi-bit upsets (MBUs) in storage elements, which can lead to errors due to particle strikes, requiring costly hardware redundancy and increased power consumption.
Innovation Solution
The use of one-hot or one-cold encoding combined with parity values to generate and store state indicating values in multiple portions of a storage element, allowing for error detection and correction without the need for redundant storage elements or voter logic, thereby reducing hardware overhead and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional redundancy methods (triplicated storage elements with voter logic) are used to protect against SEUs and MBUs, then reliability is improved, but device complexity and power consumption increase
Solution Approach 1:
The patent applies parameter changes by transforming the encoding format of stored data from conventional binary to one-hot or one-cold encoding schemes. This parameter change in data representation allows single-bit error detection and correction capabilities without requiring triplicated storage elements or voter logic, thereby maintaining reliability while reducing device complexity
Solution Approach 2:
The patent segments the storage element into multiple portions, with each portion storing a specific bit of the encoded data. This segmentation approach, combined with one-hot or one-cold encoding, enables the system to identify and correct single-bit errors by detecting which portion contains the erroneous bit, achieving error protection without complex redundancy structures
2Reliability
If traditional redundancy methods (triplicated storage elements with voter logic) are used to protect against SEUs and MBUs, then reliability is improved, but power consumption increases
Solution Approach 1:
By changing the encoding parameter from conventional binary to one-hot or one-cold encoding, the patent reduces the computational complexity of error detection and correction. This parameter change eliminates the need for continuous operation of voter logic in triplicated systems, thereby reducing power consumption while maintaining protection against SEUs and MBUs
Solution Approach 2:
The patent employs a simpler, less resource-intensive error protection mechanism that consumes less power compared to traditional redundancy methods. The one-hot or one-cold encoding scheme with single-bit error correction provides adequate protection without the continuous power demand of triplicated storage elements and voter logic
3Device complexity
If one-hot or one-cold encoding with parity values is used, then device complexity is reduced, but measurement precision of error detection may be compromised
Solution Approach 1:
The patent segments the storage element into multiple portions, each associated with specific parity bits in the one-hot or one-cold encoding scheme. This segmentation enables precise error detection by identifying which specific portion contains the erroneous bit, maintaining measurement precision while reducing device complexity compared to comprehensive redundancy methods
Data Source
AI summary
A state indicating value is encoded with a one-hot or one-cold encoding and each bit of the state indicating value is stored in a different portion of a storage element. Parity values are determined for each portion of the storage element and stored to a parity storage element. This allows errors caused by single event upsets or multi-bit upsets to be detected and corrected, with lower hardware cost compared to alternative approaches.