Global Error Variable for Propagating Coded Discrepancies
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for error propagation in safety-oriented computer programs executed on embedded hardware, such as PLCs, lack certainty and are vulnerable to systematic errors and compiler optimizations, making it difficult to reliably store and propagate error information.
Innovation Solution
A method using a globally valid error variable with a nonzero value and arithmetic sign to store and propagate error information, derived from discrepancies in coded variables, ensuring recognition and protection against systematic errors, with error propagation occurring through a combination of operations and operands in 'coded processing'.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If standard arithmetic operations are used for error propagation, then the error variable can be updated, but compiler optimizations may reduce propagation certainty
Solution Approach 1:
The error variable is segmented into multiple components: a global error variable for accumulation, local error variables for individual operations, and separate error detection mechanisms. This segmentation prevents compiler optimizations from affecting propagation certainty while maintaining clear error tracking paths.
Solution Approach 2:
An intermediary error detection mechanism is introduced between the coded processing operations and the error variable. This intermediary layer ensures that errors are properly detected and propagated without allowing compiler optimizations to bypass error detection, thereby maintaining propagation certainty.
2Reliability
If error information is stored in a single variable, then propagation is simplified, but systematic errors and reciprocal compensation cannot be prevented
Solution Approach 1:
The error storage is segmented into a global error variable and multiple local error variables associated with specific operations. This segmentation allows independent tracking of different error types, preventing reciprocal compensation while maintaining manageable complexity through structured organization.
Solution Approach 2:
The error propagation mechanism introduces asymmetry by treating different error types differently through stipulated arithmetic signs. Positive and negative errors are handled asymmetrically in the propagation process, preventing systematic errors from canceling each other out while maintaining a structured variable approach.
3Reliability
If coded processing is used for safety verification, then error detection capability is improved, but the complexity of error propagation mechanisms increases
Solution Approach 1:
The global error variable serves multiple functions: it accumulates errors from various operations, stores error information for propagation, and provides a centralized reference for error detection. This multi-functionality reduces the need for separate propagation mechanisms while maintaining high error detection capability through coded processing.
Data Source
AI summary
A method for storing and propagating error information in computer programs, in which a globally valid error variable is used for storing and propagating the error information, wherein for each recognized error a nonzero value for the error is added to the error variable as error information with a respective stipulated arithmetic sign, and wherein the value is formed from a discrepancy in the content of a coded variable from an expected value. This combination and integration of a separate global propagation variable with values derived from an error, particularly by virtue of detected discrepancies in the known error recognition and propagation paths using operations and operands in “coded processing”, achieves an increased propagation certainty.