Payload Data Verification via Meta Data Reference Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In safety-critical applications executed on non-safety-critical hardware, ensuring correct program flow and error detection is challenging due to complex data paths and dynamic reconfiguration in distributed systems, often requiring additional special hardware that is not always feasible.
Innovation Solution
A method that involves ascertaining reference result data for each sequence of data processing blocks, comparing payload and meta data result data to ensure adherence to the correct processing path, and triggering safety measures if discrepancies are found, allowing for flexible and correct data processing without additional hardware.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If safety-critical applications are executed on non-safety-critical hardware, then cost-effectiveness and processing power are improved, but reliability and error-proneness deteriorate
Solution Approach 1:
The patent segments the data processing into multiple independent sequences, each processed by different data processing blocks. This segmentation allows parallel verification of processing paths without requiring specialized safety hardware, thereby maintaining cost-effectiveness while improving reliability through distributed validation.
Solution Approach 2:
The patent implements a feedback mechanism where result data from multiple processing sequences are compared and verified. The system checks whether the result data agree across different sequences and uses this feedback to trigger safety measures if discrepancies are detected, thereby compensating for the lower inherent reliability of non-safety-critical hardware.
2Power
If multiple independent non-safety-critical systems are used to execute safety-critical calculations, then processing power and flexibility are improved, but device complexity and data path management worsen
Solution Approach 1:
The patent creates a universal verification mechanism that works across multiple different processing sequences and data paths. The same verification logic is applied regardless of which specific sequence is executed, allowing the system to leverage diverse processing paths while maintaining manageable complexity through standardized verification procedures.
Solution Approach 2:
The patent performs preliminary verification by comparing result data from multiple sequences before final output is generated. This preliminary action of cross-validation catches errors early in the processing pipeline, preventing the need for complex post-processing error management and simplifying overall system design.
3Reliability
If program flow control is implemented across multiple processing components, then correctness of processing path is improved, but system complexity and overhead increase
Solution Approach 1:
The patent uses copying by executing the same processing sequence multiple times in parallel and comparing the results. Instead of implementing complex control logic to track and verify each processing path, the system creates copies of the processing sequences and uses result comparison to verify correctness, thereby improving reliability while minimizing added complexity.
4Adaptability or versatility
If dynamic reconfiguration is enabled in distributed data processing systems, then adaptability and flexibility are improved, but measurement and monitoring of processing operations worsen
Solution Approach 1:
The patent implements self-service monitoring where each processing sequence independently generates and verifies its own result data. The system does not require external monitoring infrastructure to track dynamic reconfiguration; instead, the processing components themselves perform the verification through result data comparison, making monitoring as easy as the processing operations regardless of how dynamically the system is reconfigured.
Data Source
AI summary
A method for checking a processing of payload data. The method includes: ascertaining, for each one of multiple sequences of data processing blocks, reference result data that result when predetermined meta data are processed by the sequence of the data processing blocks; receiving payload data result data and meta data result data for a data processing of the payload data and of the meta data; checking whether the meta data result data for at least one sequence of the sequences admissible for the data processing of the payload data agree with reference result data ascertained for the sequence; and triggering a safety measure if the meta data result data for no sequence of the sequences admissible for the data processing agree with the reference result data ascertained for the sequence.


