# US 2016/0253445 A1 · Predictive multi-user client-server electronic circuit design system utilizing machine learning techniques

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2016/0253445 A1
- Document ID: US20160253445A1
- Status: Published application
- Kind: A1
- Application number: 15/055,279
- Filed: 2016-02-26
- Published: 2016-09-01
- Earliest claimed: 2015-02-26
- Assignees: Autodesk, Inc.
- Applicants: Autodesk, Inc.
- Inventors: Edward Sandor Pataky
- Web page: https://patentgrep.com/patent/US20160253445A1

## CPC

- G06F 30/3308: using simulation
- G06F 30/327: Logic synthesis; Behaviour synthesis, e.g. mapping logic, HDL to netlist, high-level language to RTL or netlist
- G06F 30/33: Design verification, e.g. functional simulation or model checking
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A predictive electronic circuit design system, method, and apparatus provide the ability to design an electronic circuit. An electronic computer aided design (CAD) environment server computing device is connected to the Internet and includes a machine learning module program. The machine learning module program has a netlist analyzer program, a database, and a classifier and predictor program. The netlist analyzer program receives a circuit netlist for a designed circuit from a user, characterizes the circuit netlist, and sends characterization data to the database. Characterization data from multiple users for multiple designed circuits are stored in the database. The classifier and predictor program uses design goal data, the characterization data for multiple designed circuits, and simulation results, to calculate and produce predictions and proposals for the user to make design changes to the designed circuit in order to better meet or exceed design goals.

## Claims

### Claim 1

A predictive electronic circuit design system comprising:
(a) an electronic computer aided design (CAD) environment server computing device connected to the Internet, wherein the server computing device comprises a machine learning module program that comprises:
(1) a netlist analyzer program that receives a circuit netlist for a designed circuit from a user, characterizes the circuit netlist, and sends characterization data to a database;
(2) the database, wherein characterization data from multiple users for multiple designed circuits are stored in the database; and
(3) a classifier and predictor program that uses, from the database, design goal data, the characterization data for multiple designed circuits, and simulation results, to calculate and produce predictions and proposals for the user to make design changes to the designed circuit in order to better meet or exceed design goals.

### Claim 2 (depends on 1)

The predictive electronic circuit design system of claim 1, wherein the netlist analyzer comprises:
an input characterization module that characterizes input sources in the designed circuit;
an output characterization module that characterizes properties of outputs from the designed circuit;
a component characterization module that characterizes circuit components of the designed circuit based on one or more parameters; and
a circuit topology characterization module that characterizes a circuit topology of the designed circuit.

### Claim 3 (depends on 1)

The predictive electronic circuit design system of claim 1, wherein the machine learning module further comprises:
a simulation results analyzer program that analyzes circuit simulation results and characterizes simulation output from the circuit simulation results and sends the simulation output characterization to the database.

### Claim 4 (depends on 1)

The predictive electronic circuit design system of claim 1, wherein the design goals comprise universal goals selected from a list consisting of:
power;
cost;
performance; and
size.

### Claim 5 (depends on 1)

The predictive electronic circuit design system of claim 1, wherein the design goals comprise user-inputted goals specific to a design comprising voltage gain and current limits.

### Claim 6 (depends on 1)

The predictive electronic circuit design system of claim 1, further comprising:
(b) an electronic CAD design environment client computing device that is connected to the Internet, wherein the client computing device comprises:
(1) a client CAD software application with a user interface that allows a user to design the designed circuit in the client CAD software application;
(2) a netlisting software application that converts circuit schematic information received from the designed circuit into the circuit netlist; and
(3) a circuit simulation software application that simulates the designed circuit represented by the netlist to produce the simulation results.

### Claim 7 (depends on 6)

The predictive electronic circuit design system of claim 6, wherein:
the client CAD software application receives proposed design changes from the classifier and predictor program;
the user interface visually presents the proposed design changes to the user by overlaying the proposed design changes over the designed circuit.

### Claim 8 (depends on 6)

The predictive electronic circuit design system of claim 6, wherein:
the client CAD software application receives proposed design changes from the classifier and predictor program;
the user interface visually presents the proposed design changes to the user in text form.

### Claim 9

A computer-implemented method designing an electronic circuit, comprising:
receiving, in a netlist analyzer program in a machine learning module program on an electronic computer aided design (CAD) environment server computing device connected to the Internet, a circuit netlist for a designed circuit from a user;
characterizing, in the netlist analyzer program, the circuit netlist;
storing the characterization data in a database, wherein characterization data from multiple users for multiple designed circuits are stored in the database; and
calculating and producing predictions and proposals for the user to make design changes to the designed circuit in order to better meet or exceed design goals, wherein the predictions and proposals are based on design goal data, the characterization data for multiple designed circuits, and simulation results, from the database.

### Claim 10 (depends on 9)

The computer-implemented method of claim 9, further comprising:
characterizing, in an input characterization module in the netlist analyzer, input sources in the designed circuit;
characterizing, in an output characterization module in the netlist analyzer, properties of outputs from the designed circuit;
characterizing, in a component characterization module in the netlist analyzer, circuit components of the designed circuit based on one or more parameters; and
characterizing, in a circuit topology characterization module in the netlist analyzer, a circuit topology of the designed circuit.

### Claim 11 (depends on 9)

The computer-implemented method of claim 9, further comprising:
analyzing, in a simulation results analyzer program in the machine learning module, circuit simulation results;
characterizing simulation output from the circuit simulation results; and
storing the simulation output characterization in the database.

### Claim 12 (depends on 9)

The computer-implemented method of claim 9, wherein the design goals comprise universal goals selected from a list consisting of:
power;
cost;
performance; and
size.

### Claim 13 (depends on 9)

The computer-implemented method of claim 9, wherein the design goals comprise user-inputted goals specific to a design comprising voltage gain and current limits.

### Claim 14 (depends on 9)

The computer-implemented method of claim 9, further comprising:
designing, in a user interface on a client CAD software application on an electronic CAD design environment client computing device that is connected to the Internet, the designed circuit;
converting, in a netlisting software application in the client CAD software application, circuit schematic information received from the designed circuit into the circuit netlist; and
simulating, in a circuit simulation software application in the client CAD software application, the designed circuit represented by the netlist to produce the simulation results.

### Claim 15 (depends on 14)

The computer-implemented method of claim 14, further comprising:
receiving, in the client CAD software application, proposed design changes from the classifier and predictor program;
visually presenting, via the user interface, the proposed design changes to the user by overlaying the proposed design changes over the designed circuit.

### Claim 16 (depends on 14)

The computer-implemented method of claim 14, further comprising:
receiving, in the client CAD software application, proposed design changes from the classifier and predictor program;
visually presents, via the user interface, the proposed design changes to the user in text form.
