Bonding Wire Circuit Parameter Estimation Using Trained Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately calculate circuit parameters of bonding wires in high-frequency circuits, especially when a large number of bonding wires are connected side by side, due to the difficulty in collecting and processing large amounts of data.
Innovation Solution
A program and calculation device utilizing machine learning to generate a trained model that estimates circuit parameters based on first and second information about the number of bonding wires, allowing for high-accuracy calculations using reduced data sets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional methods are used to calculate circuit parameters of bonding wires, then measurement precision may be maintained, but productivity deteriorates due to the difficulty in collecting and processing large amounts of data
Solution Approach 1:
The patent applies preliminary action by pre-collecting S-parameters for bonding wires with different numbers (1, N1, N2) and storing them in a database before actual design work. This pre-processing of data eliminates the need for time-consuming data collection during the design phase, significantly improving productivity while maintaining measurement precision through the use of pre-measured accurate parameters.
Solution Approach 2:
The patent uses copying by creating equivalent circuit models that replicate the electrical characteristics of actual bonding wires. Instead of measuring each physical bonding wire individually, the invention creates virtual copies through S-parameter data and equivalent circuits, allowing rapid analysis without repeated physical measurements, thus improving productivity while preserving measurement accuracy.
2Measurement precision
If the number of bonding wires increases, then the circuit parameter calculation becomes more complex, but the need for accurate parameter estimation increases
Solution Approach 1:
The patent applies parameter changes by utilizing S-parameters (scattering parameters) which are frequency-dependent complex parameters that fully characterize the bonding wire behavior. By changing from simple resistance/inductance models to comprehensive S-parameter models, the invention achieves accurate parameter estimation for any number of bonding wires while managing complexity through standardized parameter representations that can be stored and reused.
Solution Approach 2:
The patent reduces calculation complexity through preliminary action by pre-calculating and storing S-parameters for various bonding wire configurations (different numbers of wires) before actual design work. This pre-computation eliminates the need for complex real-time calculations during design, allowing engineers to quickly retrieve accurate parameters for any bonding wire configuration without performing complex simulations or measurements at the time of use.
Data Source
AI summary
A non-transitory computer-readable recording medium having stored therein a program for causing a computer to execute a process. The process includes acquiring first information about bonding wires other than a number of the bonding wires, and the second information about the number of the bonding wires, the bonding wires being connected between a first port and a second port, and estimating, from the first information based on a trained model, a first parameter, a second parameter, or a third parameter, and calculating a calculation parameter that is the circuit parameter for the number of the bonding wires, based on the parameter from the second information. The trained model is generated by performing machine learning on training data defining a relationship of the first information and the number of the bonding wires to the circuit parameter for the first information and the number of the bonding wires.