Constrained Random Circuit Simulation with Bayesian Test-Count Estimation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Constrained Random Verification (CRV) methodologies lack a systematic method to determine the optimal number of random test cases needed for achieving verification goals, leading to inefficiencies and resource wastage in hardware design verification.

Innovation Solution

Utilizing Bayesian estimation and machine learning to estimate the minimum number of test instances required for detecting failures in circuit designs by converting the Minimum Sufficient Test-Instance (MSTI) problem into a Bayesian estimation problem, employing random seed abstraction and conjugate prior distributions to improve estimation accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If constrained random verification is performed with a large number of test instances, then the reliability of detecting failures is improved, but the computational resources and time required increase

Engineering Contradiction:

Improvefailure detection reliabilityVSAvoidverification time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by performing Bayesian estimation before executing the full verification process. The system estimates the minimum sufficient test instances (MSTI) in advance using prior failure probabilities and likelihood functions, allowing verification engineers to determine the optimal test count before resource-intensive simulation begins, thus avoiding unnecessary computational waste while ensuring adequate failure detection coverage

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback through iterative Bayesian updating. As verification progresses and actual failure data becomes available, the system updates the posterior probability distribution, which then feeds back into refining the MSTI estimation. This closed-loop feedback mechanism allows the verification process to adapt dynamically, adjusting the number of test instances based on actual observed failures rather than relying solely on initial estimates

Inventive Principle:

Principle #23Feedback

2Loss of time

If constrained random verification is performed with insufficient test instances, then the computational resources and time are reduced, but the reliability of detecting failures deteriorates

Engineering Contradiction:

Improveverification timeVSAvoidfailure detection reliability

Core Design Contradiction:

Loss of timeVSReliability

Solution Approach 1:

The system performs preliminary Bayesian estimation to calculate the minimum sufficient test instances before verification begins. By using prior failure probabilities and likelihood functions, the system determines the optimal test count in advance, ensuring that verification achieves adequate failure detection reliability without performing excessively numerous tests that would waste computational resources

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent applies parameter changes by dynamically adjusting the number of test instances based on Bayesian estimation results. Rather than using a fixed or arbitrary test count, the system modifies the test instance parameter according to the calculated MSTI, which is derived from failure probabilities and verification goals, thus optimizing the balance between reliability and resource consumption

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If manual test case creation is used to cover specific scenarios, then the precision of verifying specific functions is improved, but the time and complexity of verification increase

Engineering Contradiction:

Improveverification precisionVSAvoidverification complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies self-service by enabling the verification system to automatically determine the optimal number of test instances without requiring manual intervention. The Bayesian estimation framework autonomously calculates MSTI based on failure probabilities and verification goals, eliminating the need for verification engineers to manually assess and decide on appropriate test counts, thus reducing verification complexity while maintaining precision

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent substitutes the mechanical process of manual test case creation and analysis with an automated Bayesian estimation system. Instead of relying on human engineers to manually determine verification parameters, the system uses probabilistic models and computational algorithms to automatically calculate optimal test instance numbers, replacing manual mechanical processes with automated intelligent systems

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12399219B1Constrained random simulation using machine learning and Bayesian estimation

Publication Date: 2025.08.26 SYNOPSYS INC

AI summary

Certain aspects of the present disclosure are directed towards a method for circuit testing. The method generally includes: determining a probability distribution indicating prior failure probabilities associated with a circuit design; determining a first likelihood associated with occurrence of at least one failure for the circuit design; determining a quantity of test instances to be performed using simulation to detect the at least one failure based on the probability distribution and the first likelihood; and outputting the quantity of test instances.