Data Processing Device State Monitoring for Fault Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data processing devices face challenges in ensuring correct operation due to unforeseen faults or external attacks, which can lead to incorrect results or security gaps, as existing verification methods cannot cover all possible interactions between program parts and require specialized hardware for detection.
Innovation Solution
A method that temporarily changes a subsystem's state to unauthorized before executing a partial operating sequence, allowing detection of disruptions by evaluating the subsystem state and resetting it to authorized upon completion, using existing hardware mechanisms like memory management units to monitor correct operation without additional hardware.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If hardware measures like checkpoint registers, glitch sensors, frequency sensors, or light sensors are used to detect disruptions, then detection capability is improved, but device complexity and requirement for specialized hardware increases
Solution Approach 1:
The data processing device monitors its own operation by utilizing its existing subsystems and state evaluation capabilities. The processor automatically changes subsystem states, executes partial operating sequences, and evaluates states without requiring external specialized monitoring hardware. This self-service approach eliminates the need for additional checkpoint registers, glitch sensors, or other specialized detection devices.
Solution Approach 2:
Existing subsystems within the data processing device are made to serve dual purposes: performing their primary computational functions while simultaneously serving as monitoring mechanisms. The subsystems that normally process data are repurposed to detect disruptions by evaluating their own states after executing modified partial operating sequences, eliminating the need for separate specialized monitoring hardware.
2Reliability
If verification methods check all possible situations, then reliability is improved, but loss of time and productivity decreases
Solution Approach 1:
The system prepares for verification by pre-modifying partial operating sequences to include state changes and evaluation points. Instead of performing comprehensive verification after execution, the monitoring mechanism is built into the execution flow itself, with state evaluations occurring at predetermined points during normal operation. This preliminary structuring of verification points eliminates the need for separate, time-consuming verification passes.
Solution Approach 2:
The monitoring process is integrated into the normal operation of the data processing device, allowing verification to occur continuously during execution rather than requiring separate verification phases. The processor continues executing operating sequences while simultaneously monitoring subsystem states, ensuring that detection occurs without interrupting the useful computational work.
Data Source
AI summary
A method for monitoring the correct operations of a data processing device including changing a subsystem from an authorized state to an unauthorized state, executing the partial operating sequence, and resetting any subsystem state from the unauthorized state to the authorized state.