Machine Learning Test Case Selection for IC Coverage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Integrated circuit testing is a resource-intensive process due to the large number of tests required to verify performance across various logic paths, timing constraints, and fabrication variations, often resulting in redundant test cases that waste computational resources.
Innovation Solution
A method using machine learning techniques to generate a coverage matrix and select a subset of test cases based on weights associated with each test case, ensuring each coverage point is hit at least once, thereby reducing redundancy and computational resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a large number of test cases are run to verify integrated circuit performance across various logic paths and fabrication variations, then testing coverage and reliability are improved, but computational resources and time are excessively consumed
Solution Approach 1:
The patent extracts and removes redundant test cases from the testing process by using machine learning to identify and eliminate duplicate coverage points. This extraction principle directly reduces the number of test cases executed while maintaining necessary testing coverage, thereby reducing computational resource consumption without sacrificing reliability
Solution Approach 2:
The patent changes the parameter of test case selection by using machine learning models to dynamically determine which test cases to execute based on coverage matrix analysis. This parameter change enables intelligent selection of test cases, optimizing the balance between testing coverage and computational resource usage
2Reliability
If multiple test cases are executed to cover different coverage points, then testing completeness is improved, but redundant tests increase and waste resources
Solution Approach 1:
The patent implements feedback through the machine learning model that analyzes the coverage matrix and learns from previous test executions. This feedback mechanism identifies redundant coverage points and prevents re-testing of already covered areas, thereby eliminating resource waste while maintaining testing completeness
Solution Approach 2:
The patent extracts redundant test cases by comparing coverage points across multiple test cases and removing duplicates. This extraction process ensures that each coverage point is tested only once, eliminating resource waste from redundant tests while preserving testing completeness
Data Source
AI summary
Certain aspects of the present disclosure provide techniques and apparatus for testing integrated circuit designs. An example method generally includes generating a coverage matrix associated with a plurality of test cases for an integrated circuit and coverage points associated with each test case of the plurality of test cases. A subset of the plurality of test cases is selected for execution based on weights associated with each test case of the plurality of test cases and a threshold weight value. Generally, the weights associated with each test case comprise weights in a machine learning model trained based on the coverage matrix. The integrated circuit may be tested based on the selected subset of test cases.