Slew-Load Characterization Using Geometric Interpolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional circuit designs face challenges in accurately capturing timing and on-chip variation data, leading to increased simulation run-time overhead due to the need for extensive Monte-Carlo simulations across multiple slew/load points.
Innovation Solution
Implementing efficient slew-load characterization methodologies that simulate only an optimum number of points, using geometric progression interpolation and delta approaches to derive simulation values for a 7×7 timing table, reducing the number of required simulations from 49 to 3-7 diagonal reference points while maintaining desired accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If accurate timing and on-chip variation data are captured using conventional Monte-Carlo simulations across multiple slew/load points, then measurement precision is improved, but simulation run-time increases
Solution Approach 1:
The patent segments the 49 slew/load combination points into a smaller subset of representative points (e.g., minimum, maximum, and intermediate values). By dividing the full set into segments and selecting key representatives, the simulation workload is reduced from 49 points to a manageable subset while maintaining accuracy through targeted sampling of critical operating conditions.
Solution Approach 2:
The patent applies partial action by simulating only the necessary subset of slew/load points rather than all 49 combinations. By performing simulations on selected representative points (partial action) and using interpolation to derive remaining values, the approach achieves sufficient accuracy without the excessive computational burden of complete simulation coverage.
2Reliability
If all 49 slew/load combination points are simulated to ensure complete characterization, then reliability is improved, but productivity decreases
Solution Approach 1:
The patent performs preliminary action by pre-selecting representative slew/load points based on their significance to timing characterization. By identifying and prioritizing critical points (minimum, maximum, and key intermediate values) before simulation, the methodology ensures that the most impactful parameters are captured first, maintaining reliability while enabling faster completion through selective simulation.
Solution Approach 2:
The patent uses copying by deriving simulation results for non-simulated slew/load points through interpolation from the simulated representative points. Instead of performing actual simulations for all 49 points, the methodology creates copied/estimated values for remaining points based on the simulated subset, maintaining characterization completeness while significantly improving productivity.
3Measurement precision
If extensive Monte-Carlo simulations are performed for all timing arcs, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent extracts only the essential and critical slew/load combination points from the full set of 49 points for Monte-Carlo simulation. By taking out and focusing on the most significant points (minimum, maximum, and key intermediates) rather than processing all points, the methodology maintains measurement precision for critical parameters while reducing simulation process complexity and resource requirements.
Data Source
AI summary
Various implementations described herein are related to a method for constructing integrated circuitry and identifying input signal paths, internal signal paths and output signal paths associated with the integrated circuitry. The method may include generating a timing table for slew-load characterization of the input signal paths, the internal signal paths and the output signal paths. The method may include simulating corner points for the timing table, building diagonal points for the timing table based on the simulated corner points, and building remaining points for the timing table based on the simulated corner points and the diagonal points.