Load-Slew Index Generation via Intermediate Error Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining load-slew indices in circuit characterization, such as the NLDM model, often require a large number of pre-calculated data points for accuracy and may not yield the minimum number of sampling points, especially when dealing with 3D surfaces, leading to inefficiencies in data storage and processing.
Innovation Solution
A method and system for generating a subset of sampling points by iteratively calculating the error between actual and interpolated values, where points with peak deviations are not prioritized, but rather points with less than peak deviations are selected to achieve a reduced set of sampling points that meet a specified tolerance, allowing for efficient approximation of load-slew-delay surfaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods use peak deviation points for sampling, then accuracy is improved, but the number of sampling points increases leading to larger data storage requirements
Solution Approach 1:
The patent introduces intermediate value errors as a mediator between actual values and interpolated values. Instead of directly using peak deviations which occur at characteristic points, the method calculates errors at intermediate values between sampling points. This intermediary approach allows for accurate error assessment without requiring sampling at every peak deviation location, thus reducing the total number of sampling points needed while maintaining accuracy.
Solution Approach 2:
The patent applies partial action by selectively sampling only where necessary. Rather than uniformly sampling all peak deviation points, the method identifies regions where intermediate value errors exceed tolerance thresholds and samples only those specific areas. This partial sampling strategy maintains measurement accuracy in critical regions while avoiding redundant sampling in areas where the interpolation is already sufficiently accurate.
2Measurement precision
If more sampling points are used, then accuracy is improved, but data storage space and processing time increase
Solution Approach 1:
The method uses intermediate value calculations as a computational mediator that requires less processing than full peak deviation analysis. By evaluating errors at intermediate points between existing samples rather than identifying and processing all peak deviations, the algorithm reduces computational complexity and processing time while still achieving accurate determination of where additional sampling is needed.
Solution Approach 2:
The patent implements a self-service sampling strategy where the algorithm automatically identifies regions requiring additional sampling based on intermediate error calculations. The system serves itself by using its own interpolation and error assessment capabilities to determine where more data is needed, eliminating the need for external guidance or manual specification of sampling points.
3Quantity of substance
If conventional iterative methods are used to reduce sampling points, then data storage is reduced, but the method may not yield the minimum number of points required
Solution Approach 1:
The patent introduces intermediate value error assessment as a mediator that guides the sampling reduction process more efficiently than conventional methods. By calculating errors at intermediate points between samples, the algorithm can more accurately identify which sampling points are truly necessary and which can be removed, leading to a more optimal reduced set of sampling points that maintains accuracy with minimal data.
Solution Approach 2:
The method implements feedback by continuously evaluating intermediate value errors and using this information to guide subsequent sampling decisions. The algorithm calculates intermediate errors, identifies regions where errors exceed tolerance, and selectively adds or removes sampling points based on this feedback. This iterative feedback loop ensures that the final set of sampling points is minimized while still maintaining the required accuracy tolerance.
Data Source
AI summary
A method and system for generation of low-slew indices for circuit characterization are disclosed. In one embodiment, a method for automatically generating a subset of sampling points from a set of load and slew points for circuit characterization includes iteratively obtaining sampling points such that error between an actual value and an interpolated intermediate value is below or equal to a threshold error value. The subset of sampling points is then formed from the iteratively obtained sampling points.