Analog Circuit Sizing via ML Structure Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In analog and mixed-signal electronic design, it is challenging to automatically adjust device sizes to satisfy electrical specifications due to the need for extensive manual effort and design knowledge, with existing solutions requiring extensive simulations and improper sizing constraints leading to optimization failures.

Innovation Solution

A computer-implemented method using machine learning models, such as decision tree and logistic regression, to analyze electronic design schematics, determine required and optional features, and automatically generate sizing constraints, reducing the need for manual input and improving optimization efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual methods are used to adjust device sizes and create sizing constraints, then design knowledge and control are improved, but productivity and time consumption deteriorate due to extensive manual effort and numerous simulations required

Engineering Contradiction:

Improvedesign accuracyVSAvoiddesign efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system enables automated structure prediction and sizing constraint generation that performs the design analysis task independently. The machine learning model automatically identifies circuit structures, predicts features, and generates sizing constraints without requiring manual intervention, thereby maintaining design accuracy while dramatically improving productivity

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the manual mechanical process of analyzing schematics and creating constraints with an automated computational system. The machine learning model substitutes the human designer's analytical work, using trained algorithms to predict circuit structures and generate sizing constraints automatically, reducing both time and manual effort

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If extensive simulations are performed to reach target specifications, then measurement precision is improved, but loss of time and productivity worsen due to the huge amount of simulations required

Engineering Contradiction:

Improvespecification accuracyVSAvoidsimulation time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary structure prediction and feature identification before the actual sizing optimization simulations. By using the machine learning model to pre-analyze the schematic and predict circuit structures, the system prepares the necessary information in advance, reducing the number of simulations needed to reach target specifications and thereby reducing time loss

Inventive Principle:

Principle #10Preliminary action

3Ease of operation

If improper sizing constraints are used, then ease of operation is improved by allowing automated optimization, but reliability worsens due to optimization failures

Engineering Contradiction:

Improveautomation levelVSAvoidoptimization success rate

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The system uses feedback from the predicted circuit structure and identified features to dynamically generate appropriate sizing constraints. The machine learning model analyzes the schematic, predicts structures, and uses this information to create constraints that are tailored to the specific circuit characteristics, ensuring both automation and high optimization success rates

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system changes the parameters of sizing constraints based on the predicted circuit structure and identified features. By adapting the constraint parameters to match the specific circuit configuration, the system maintains ease of automated operation while ensuring reliability through structure-appropriate constraint generation

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS11544574B1System, method, and computer program product for analog structure prediction associated with an electronic design

Publication Date: 2023.01.03 CADENCE DESIGN SYST INC

AI summary

The present disclosure relates to a computer-implemented method for electronic design. Embodiments may include receiving, using at least one processor, an electronic design schematic and optionally an electronic design layout. Embodiments may further include analyzing the electronic design schematic to determine if one or more required features of a particular circuit structure are present. If the one or more required features are present, embodiments may include analyzing, using a machine learning model, the electronic design schematic to determine if one or more optional features of the particular circuit structure are present.