FSA-Based Biasing for Circuit Simulation Coverage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As digital circuit designs become increasingly complex, random input stimuli become less effective in covering important corner cases, and existing methods fail to provide a complete and automatic solution for biasing input stimuli to achieve coverage closure.
Innovation Solution
A system uses finite state automaton (FSA) instances to observe inputs and outputs of a circuit design, applying soft constraints to bias input stimuli, ensuring that FSA instances closer to an accepting state have a higher likelihood of influencing input selection, thereby guiding the simulation to satisfy temporal coverage properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If random input stimuli are applied to complex circuit designs, then the volume of input vectors increases, but the coverage of important corner cases decreases
Solution Approach 1:
The system continuously monitors coverage metrics during simulation and dynamically adjusts the biasing of random input stimuli based on what has been covered and what remains uncovered. This feedback loop ensures that important corner cases are targeted while maintaining overall coverage efficiency.
Solution Approach 2:
The system changes the parameters of input stimuli by applying dynamic biasing weights to different input combinations based on their importance and coverage status. This transforms uniform random stimulation into targeted random stimulation that prioritizes uncovered corner cases.
2Reliability
If existing methods are used to guide test selection, then some coverage improvement is achieved, but a complete and automatic solution for biasing input stimuli is not provided
Solution Approach 1:
The system automatically performs coverage analysis, identifies uncovered corner cases, and self-adjusts the input stimulus biasing without requiring manual intervention. This complete automation eliminates the need for manual test selection while achieving superior coverage.
Solution Approach 2:
The system pre-calculates coverage requirements and identifies critical corner cases before the main simulation begins, allowing the biasing mechanism to be pre-configured for maximum effectiveness throughout the verification process.
3Reliability
If FSA instances closer to accepting states are given higher biasing weight, then the likelihood of satisfying temporal coverage properties increases, but the complexity of managing multiple FSA instances increases
Solution Approach 1:
The system uses feedback from the current state of each FSA instance to dynamically adjust biasing weights, prioritizing FSA instances that are closer to satisfying their temporal coverage properties. This automated feedback mechanism manages the complexity of multiple FSA instances systematically.
Data Source
AI summary
One embodiment of the present invention provides a system which verifies a circuit design by biasing input stimuli for the circuit design to satisfy one or more temporal coverage properties to be verified for the circuit design. This system performs a simulation in which random input stimuli are applied to the circuit design. The system performs the simulation by using a finite state automaton (FSA) instance for a temporal coverage property to observe inputs and outputs of the circuit, and by using soft constraints associated with the FSA instance to bias the input stimuli for the circuit design so that the simulation is likely to progress through a sequence of states which satisfy the temporal coverage property.