Optimal Input Data Generation for Semiconductor Design Simulators

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The increasing complexity of semiconductor devices makes it difficult and time-consuming to manually calibrate input data for design simulators, such as TCAD simulators, to accurately estimate semiconductor characteristics, leading to reduced accuracy and efficiency in obtaining target output data.

Innovation Solution

A computing device and method that generates optimal input data using a trained estimation model, selecting essential input parameters and generating training data with sample input and output data to automate the calibration process, thereby reducing manual intervention and improving accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual calibration procedures are performed to accurately estimate semiconductor characteristics, then measurement precision is improved, but loss of time increases

Engineering Contradiction:

Improveaccuracy of semiconductor characteristics estimationVSAvoidtime-consuming calibration process

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent creates a simplified copy or surrogate model that replicates the behavior of the complex design simulator. This surrogate model can be trained on sample data and then used to rapidly predict outcomes without running full simulations, thus maintaining accuracy while dramatically reducing calibration time.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system performs preliminary actions by pre-generating sample input data and corresponding output data before actual calibration is needed. This training data is prepared in advance to build the estimation model, so that when calibration is required, the model is already ready to provide rapid predictions without manual iteration.

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If the number of input parameters is increased to account for increasing complexity of semiconductor devices, then measurement precision is improved, but device complexity increases

Engineering Contradiction:

Improveaccuracy of semiconductor characteristics estimationVSAvoidcomplexity of calibration process

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the essential input parameters from the full set of possible parameters. By analyzing which parameters have the most significant impact on output characteristics, the system isolates and focuses on those critical factors, eliminating the need to manually manage and calibrate all possible parameters.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system transforms the complex multi-parameter calibration problem into a simpler form by changing how parameters are handled. Instead of manual adjustment of numerous parameters, the estimation model automatically processes multiple parameters simultaneously and identifies essential ones, effectively changing the parameter management approach from manual to automated.

Inventive Principle:

Principle #35Parameter changes

3Ease of manufacture

If manual calibration procedures are performed, then ease of operation is reduced, but ease of manufacture is maintained

Engineering Contradiction:

Improvefeasibility of calibration processVSAvoiddifficulty of manual calibration

Core Design Contradiction:

Ease of manufactureVSEase of operation

Solution Approach 1:

The estimation model performs self-service by automatically generating optimal input data without requiring manual calibration operations. The system trains the model on sample data and then uses it to autonomously determine appropriate input parameters, eliminating the need for operator intervention while maintaining the manufacturability of the process.

Inventive Principle:

Principle #25Self-service

4Productivity

If automated estimation model is used to generate optimal input data, then productivity is improved, but measurement precision may be reduced

Engineering Contradiction:

Improvespeed of calibration processVSAvoidaccuracy of output data

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system incorporates feedback mechanisms where the estimation model is trained on sample input-output data pairs. This training process allows the model to learn from actual simulation results and adjust its predictions accordingly, ensuring that automated predictions maintain high accuracy while providing rapid results.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20230169240A1Computing device and method generating optimal input data

Publication Date: 2023.06.01 SAMSUNG ELECTRONICS CO LTD

AI summary

A method of generating optimal input data for a design simulator providing output data related to output parameters in response to input data related to input parameters. The method includes; generating training data including sample input data and sample output data, selecting at least one essential input parameter affecting a plurality of output parameters from among the input parameters in accordance with an estimation model trained using the training data, and generating the optimal input data in accordance with essential input data corresponding to the at least one essential input parameter and the sample output data.