# US 12,119,989 B2 · System and method for configuring network elements in a design network topology

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

- Number: US 12,119,989 B2
- Document ID: US12119989B2
- Status: Granted
- Kind: B2
- Application number: 17/513,957
- Filed: 2021-10-29
- Published: 2023-05-04
- Granted: 2024-10-15
- Earliest claimed: 2021-10-29
- Assignees: Keysight Technologies, Inc.
- Applicants: Keysight Technologies, Inc.
- Inventors: Vijay Anantha Murthy
- Web page: https://patentgrep.com/patent/US12119989B2

## CPC

- H04L 41/0816: the condition being an adaptation, e.g. in response to network events
- H04L 41/147: for predicting network behaviour
- G06N 3/04: Architecture, e.g. interconnection topology
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/045: Combinations of networks
- G06N 3/08: Learning methods
- G06N 3/088: Non-supervised learning, e.g. competitive learning
- G06N 5/022: Knowledge engineering; Knowledge acquisition
- H04L 41/0806: for initial configuration or provisioning, e.g. plug-and-play
- H04L 41/0866: Checking the configuration
- H04L 41/0886: {Fully automatic configuration}
- H04L 41/122: of virtualised topologies, e.g. software-defined networks [SDN] or network function virtualisation [NFV]
- G06N 20/00: Machine learning
- H04L 41/0226: Mapping or translating multiple network management protocols
- H04L 41/145: {involving simulating, designing, planning or modelling of a network}
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A method of configuring network elements in a design network topology includes receiving an image of the design network topology; attempting to retrieve design data from the received image corresponding to the design network topology; when the design data is retrieved, querying a topologies database using the design data to find a previously determined network topology that substantially matches the design network topology; when the design data is not retrieved, predicting a network topology using an unsupervised machine learning algorithm; identifying configurations for network elements in the matching network topology or in the predicted network topology in a configurations database; determining design configurations for the network elements of the design network topology from the identified configurations; translating the design configurations of the network elements to a standard format; and pushing the translated design configurations to actual network elements and/or virtual network elements corresponding to the network elements.

## Claims

### Claim 1

A system for configuring network elements in a design network topology, the system comprising:
a topologies database for storing a plurality of previously determined network topologies;
a configurations database for storing previously determined configurations of the network elements in the previously determined network topologies, respectively;
a processing unit; and
a non-transitory memory for storing instructions that, when executed by the processing unit, cause the processing unit to:
receive an image of a design network topology comprising a plurality of network elements, wherein the design network topology provides at least one of structural or logical arrangements of the plurality of network elements in a communication network;
attempt to retrieve design data from the received image corresponding to the design network topology using a constellation-based model and using a supervised machine learning algorithm when the design data cannot be retrieved using the constellation-based model, depending on a format of the received image;
when the design data can be retrieved, query the topologies database using the design data to find a previously determined network topology that substantially matches the design network topology, and identify configurations for network elements in the matching network topology in the configurations database;
when the design data cannot be retrieved, predict a network topology using an unsupervised machine learning algorithm, and identify configurations for network elements in the predicted network topology in the configurations database;
determine design configurations for the network elements of the design network topology from the identified configurations;
translate the design configurations of the network elements to a standard format; and
push the translated design configurations to at least one of actual network elements or virtual network elements corresponding to the network elements.

### Claim 2 (depends on 1)

The system of claim 1, wherein the instructions further cause the processing unit to:
store the translated design configurations of the network elements in the configurations database.

### Claim 3 (depends on 1)

The system of claim 1, wherein the instructions further cause the processing unit to:
translate the matching network topology or the predicted network topology to a standard format; and
store the translated network topology in the topologies database.

### Claim 4 (depends on 1)

The system of claim 1, wherein the standard format comprises JavaScript Object Notation (JSON) format and/or Yet Another Markup Language (YAML) format.

### Claim 5 (depends on 1)

The system of claim 1, wherein the received image of the design network topology comprises one of a JPEG, PNG, TIF or GIF image, or a Visio diagram.

### Claim 6 (depends on 1)

The system of claim 1, wherein the instructions further cause the processing unit to:
train the unsupervised machine learning algorithm by iteratively quantifying accuracies of intermediate predicted network topologies until attaining a desired confidence score for the predicted network topology.

### Claim 7 (depends on 1)

The system of claim 1, wherein the instructions further cause the processing unit to determine the design configurations for the network elements by optimizing the identified configurations of the network elements to provide the design configurations for the network elements.

### Claim 8 (depends on 1)

The system of claim 1, wherein the supervised machine learning algorithm comprises a Graph Neural Network (GNN) algorithm, and the unsupervised machine learning algorithm comprises a You-only-look-once version 3 (YOLOv3) algorithm.

### Claim 9

A method of configuring network elements in a design network topology, the method comprising:
receiving an image of a design network topology comprising a plurality of network elements, wherein the design network topology provides at least one of structural or logical arrangements of the plurality of network elements in a communication network;
attempt to retrieve design data from the received image corresponding to the design network topology using a constellation-based model and using a supervised machine learning algorithm when the design data cannot be retrieved using the constellation-based model, depending on a format of the received image;
when the design data can be retrieved, querying a topologies database, which stores a plurality of previously determined network topologies, using the design data to find a previously determined network topology that substantially matches the design network topology, and identifying configurations for network elements in the matching network topology in a configurations database;
when the design data cannot be retrieved, predicting a network topology using an unsupervised machine learning algorithm, and identifying configurations for network elements in the predicted network topology in the configurations database;
determining design configurations for the network elements of the design network topology from the identified configurations;
translating the design configurations of the network elements to a standard format; and
pushing the translated design configurations to at least one of actual network elements or virtual network elements corresponding to the network elements.

### Claim 10 (depends on 9)

The method of claim 9, further comprising:
storing the translated design configurations of the network elements in the configurations database.

### Claim 11 (depends on 10)

The method of claim 10, further comprising:
translating the matching network topology or the predicted network topology to a standard format; and
storing the translated network topology in the topologies database.

### Claim 12 (depends on 9)

The method of claim 9, further comprising:
training the unsupervised machine learning algorithm by iteratively quantifying accuracies of intermediate predicted network topologies until attaining a desired confidence score for the predicted network topology.

### Claim 13 (depends on 9)

The method of claim 9, wherein identifying the configurations for network elements in the matching network topology or in the predicted network topology comprises querying the configurations database, which stores network element configurations respectively corresponding to the plurality of previously determined network topologies, using the matching network topology or the predicted network topology to identify stored configurations of network elements.

### Claim 14 (depends on 9)

The method of claim 9, wherein determining the design configurations for the network elements comprises optimizing the identified configurations of the network elements to provide the design configurations for the network elements.

### Claim 15

A non-transitory computer readable medium storing instructions for configuring network elements in a design network topology, that when executed by at least one processor, cause the at least one processor to:
receive an image of a design network topology comprising a plurality of network elements, wherein the design network topology provides at least one of structural or logical arrangements of the plurality of network elements in a communication network;
attempt to retrieve design data from the received image corresponding to the design network topology using a constellation-based model and using a supervised machine learning algorithm when the design data cannot be retrieved using the constellation-based model, depending on a format of the received image;
when the design data can be extracted, query a topologies database, which stores a plurality of previously determined network topologies, using the design data to find a previously determined network topology that substantially matches the design network topology, and identify configurations for network elements in the matching network topology in a configurations database;
when the design data cannot be extracted from the received image, predict a network topology using an unsupervised machine learning algorithm, identify configurations for network elements in the predicted network topology in the configurations database;
optimize the identified configurations of the network elements;
translate the optimized configurations of the network elements to a standard format; and
push the translated optimized configurations to at least one of actual network elements or virtual network elements corresponding to the network elements.

### Claim 16 (depends on 15)

The non-transitory computer readable medium of claim 15, wherein the instructions further cause the at least one processor to:
store the translated optimized configurations of the network elements in the configurations database.

### Claim 17 (depends on 15)

The non-transitory computer readable medium of claim 15, wherein the instructions further cause the at least one processor to:
translate the matching network topology or the predicted network topology to a standard format; and
store the translated network topology in the topologies database.

### Claim 18 (depends on 15)

The non-transitory computer readable medium of claim 15, wherein the standard format comprises JavaScript Object Notation (JSON) format and/or Yet Another Markup Language (YAML) format.

### Claim 19 (depends on 15)

The non-transitory computer readable medium of claim 15, wherein the received image of the design network topology comprises one of a JPEG, PNG, TIF or GIF image, or a Visio diagram.

### Claim 20 (depends on 15)

The non-transitory computer readable medium of claim 15, wherein the instructions further cause the at least one processor to:
train the unsupervised machine learning algorithm by iteratively quantifying accuracies of intermediate predicted network topologies until attaining a desired confidence score for the predicted network topology.