Circuit Variation Analysis via Cross-Scenario Information Sharing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of advanced integrated circuits exacerbates the impact of process variations, leading to high computational costs for variation characterization, especially when optimizing for performance, power, and area (PPA) at ultra-low voltage designs.
Innovation Solution
The approach involves performing simulations for multiple scenarios, grouping them into clusters based on similarity, and sharing information across scenarios to reduce redundant effort and accelerate characterization. This includes migrating full characterizations from reference scenarios to new scenarios, leveraging machine learning to reduce simulation requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full characterization simulations are performed for every scenario independently, then measurement precision is improved, but computational cost increases significantly
Solution Approach 1:
The patent performs preliminary simulations for all scenarios to obtain preliminary results before full characterization. These preliminary results are used to identify similar scenarios and select reference scenarios, so that full characterization simulations can be performed only on selected reference scenarios rather than all scenarios, thereby reducing computational cost while maintaining characterization accuracy through information migration from reference scenarios.
2Reliability
If the number of simulation scenarios is increased to cover more process variations, then reliability is improved, but productivity deteriorates due to increased computational burden
Solution Approach 1:
The patent makes the full characterization results from reference scenarios universally applicable to multiple similar scenarios through information migration. By identifying scenarios with similar characteristics and migrating characterization data from reference scenarios, the same simulation results serve multiple purposes across different scenarios, thereby improving productivity without compromising reliability in capturing process variations.
Solution Approach 2:
The patent performs preliminary simulations for all scenarios to enable subsequent identification of similar scenarios and selection of reference scenarios. This preliminary analysis allows the system to determine which scenarios can share characterization results, thereby reducing the total number of full characterization simulations needed while maintaining comprehensive coverage of process variations across all scenarios.
3Measurement precision
If comprehensive scenario evaluation is performed to capture extreme tails of probability distribution, then measurement precision is improved, but loss of time increases due to extensive simulations
Solution Approach 1:
The patent performs preliminary simulations for all scenarios to obtain preliminary results that capture basic statistical characteristics. These preliminary results are then used to identify similar scenarios and select reference scenarios for full characterization, including high sigma value analysis. This approach allows comprehensive probability distribution characterization to be performed only on selected reference scenarios while still capturing extreme tails for all scenarios through information migration, thereby reducing total simulation time.
Solution Approach 2:
The patent makes the comprehensive characterization results from reference scenarios universally applicable to multiple similar scenarios. By migrating characterization data including high sigma value information from reference scenarios to similar scenarios, the system achieves comprehensive probability distribution analysis for all scenarios without performing extensive simulations for each one, thereby reducing time loss while maintaining measurement precision for extreme tail characterization.
Data Source
AI summary
Simulations of a circuit are performed for many different scenarios. These simulations are subject to statistical variations and the simulations produce preliminary analyses of the circuit for the different scenarios. A full characterization of the circuit is estimated for a scenario of interest, by migrating a full characterization for a reference scenario from the reference scenario to the scenario of interest. The full characterization for the reference scenario was produced by additional simulations of the circuit under the reference scenario. The reference scenario may be identified by grouping the different scenarios into clusters.