# US 2024/0373337 A1 · Determining a machine-learning architecture for network slicing

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

- Number: US 2024/0373337 A1
- Document ID: US20240373337A1
- Status: Published application
- Kind: A1
- Application number: 18/774,224
- Filed: 2024-07-16
- Published: 2024-11-07
- Earliest claimed: 2022-03-15
- Assignees: Google Inc.
- Applicants: Google LLC
- Inventors: Jibing Wang; Erik Richard Stauffer
- Web page: https://patentgrep.com/patent/US20240373337A1

## CPC

- H04W 4/50: Service provisioning or reconfiguring
- H04W 48/18: Selecting a network or a communication service
- G06F 18/285: {Selection of pattern recognition techniques, e.g. of classifiers in a multi-classifier system}
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/09: Supervised learning
- G06N 3/092: Reinforcement learning
- G06N 20/00: Machine learning
- H04L 41/16: using machine learning or artificial intelligence
- H04L 41/5051: Service on demand, e.g. definition and deployment of services in real time
- H04W 8/08: Mobility data transfer
- H04W 24/10: Scheduling measurement reports {; Arrangements for measurement reports}
- H04W 64/003: {locating network equipment}
- H04Q 2213/13054: Expert system
- H04Q 2213/13098: Mobile subscriber
- H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- H04Q: SELECTING (switches, relays, selectors H01H; wireless communication networks H04W)

## Abstract

This document describes techniques and devices for determining a machine-learning architecture for network slicing. A user equipment (UE) executes a first application associated with a first requested quality-of-service level. The UE selects a first machine-learning architecture based on the first requested quality-of-service level. The UE transmits, to a network-slice manager of a wireless network, a first machine-learning architecture request message to request permission to use the first machine-learning architecture. The UE receives, from the network-slice manager, a first machine-learning architecture response message that grants permission to use the first machine-learning architecture based on a first network slice. The UE wirelessly communicates data for the first application using the first machine-learning architecture, the first machine-learning architecture being configured to compute an output based on an input using coefficients determined by the user equipment.

## Claims

### Claim 1

A method performed by a user equipment, the method comprising:
executing a first application associated with a first requested quality-of-service level;
selecting a first machine-learning architecture based on the first requested quality-of-service level;
transmitting, to a network-slice manager of a wireless network, a first machine-learning architecture request message to request permission to use the first machine-learning architecture;
receiving, from the network-slice manager, a first machine-learning architecture response message that grants permission to use the first machine-learning architecture based on a first network slice, the first machine-learning architecture forming a portion of at least one first end-to-end machine-learning architecture associated with the first network slice, the at least one first end-to-end machine-learning architecture being a distributed machine-learning architecture that is configured to process wireless communication signals and is formed by the first machine-learning architecture implemented by the user equipment, a machine-learning architecture implemented by a base station, and a machine-learning architecture implemented by an entity of a core network; and
wirelessly communicating data for the first application using the first machine-learning architecture, the first machine-learning architecture being configured to compute an output based on an input using coefficients determined by the user equipment.

### Claim 2 (depends on 1)

The method of claim 1, wherein the selecting of the first machine-learning architecture comprises:
receiving, from the network-slice manager, an available machine-learning architecture message that identifies available machine-learning architectures associated with the first network slice; and
choosing the first machine-learning architecture from the available machine-learning architectures.

### Claim 3 (depends on 1)

The method of claim 1, wherein the selecting of the first machine-learning architecture comprises:
receiving, from the network-slice manager, a machine-learning architecture selection message that comprises one or more selection rules; and
executing the one or more selection rules to determine the first machine-learning architecture.

### Claim 4 (depends on 1)

The method of claim 1, further comprising:
selecting a second machine-learning architecture;
transmitting, to the network-slice manager, a second machine-learning architecture request message to request permission to use the second machine-learning architecture; and
receiving, from the network-slice manager, a second machine-learning architecture response message that grants permission to use the second machine-learning architecture based on a second network slice.

### Claim 5 (depends on 4)

The method of claim 4, further comprising:
wirelessly communicating other data for the first application using the second machine-learning architecture, wherein:
the selecting of the second machine-learning architecture is based on the first requested quality-of-service level; and
the selecting of the second machine-learning architecture is responsive to at least one of the following:
the user equipment moving to a geographical location associated with a different base station or a different tracking area; and
the user equipment roaming between a home network and a visited network.

### Claim 6 (depends on 1)

The method of claim 1, further comprising:
executing a second application associated with a second requested quality-of-service level;
selecting a second machine-learning architecture based on the second requested quality-of-service level;
transmitting, to the network-slice manager, a second machine-learning architecture request message to request permission to use the second machine-learning architecture; and
receiving, from the network-slice manager, a second machine-learning architecture response message that grants permission to use the second machine-learning architecture based on the first network slice or a second network slice.

### Claim 7 (depends on 6)

The method of claim 6, further comprising:
responsive to receiving the second machine-learning architecture response message, wirelessly communicating other data for the second application using the second machine-learning architecture, wherein:
the selecting of the second machine-learning architecture is based on the second requested quality-of-service level;
the executing of the second application occurs during at least an overlapping portion of time that the first application is executed; and
the using of the first machine-learning architecture and the using of the second machine-learning architecture both occur during at least the overlapping portion of time.

### Claim 8 (depends on 1)

The method of claim 1, further comprising:
executing a third application associated with a third requested quality-of-service level;
selecting a third machine-learning architecture based on the third requested quality-of-service level;
transmitting, to the network-slice manager, a third machine-learning architecture request message to request permission to use the third machine-learning architecture;
receiving, from the network-slice manager, a third machine-learning architecture response message that does not grant permission to use the third machine-learning architecture;
responsive to receiving the third machine-learning architecture response message, selecting a default machine-learning architecture;
transmitting, to the network-slice manager, a fourth machine-learning architecture request message to request permission to use the default machine-learning architecture;
receiving, from the network-slice manager, a fourth machine-learning architecture response message that grants permission to use the default machine-learning architecture based on a third network slice; and
wirelessly communicating additional data for the third application using the default machine-learning architecture.

### Claim 9

A method performed by a network-slice manager, the method comprising:
creating a first network slice for supporting a first quality-of-service level;
associating the first network slice with at least one available end-to-end machine-learning architecture capable of meeting the first quality-of-service level, wherein each of the at least one available end-to-end machine-learning architecture is a distributed machine-learning architecture that is configured to process wireless communication signals and is formed by a machine-learning architecture implemented by a user equipment, a machine-learning architecture implemented by a base station, and a machine-learning architecture implemented by an entity of a core network;
receiving, from the user equipment, a first machine-learning architecture request message that requests permission to use a first machine-learning architecture, the first machine-learning architecture being configured to process the wireless communication signals;
determining whether the first machine-learning architecture forms a portion of the at least one available end-to-end machine-learning architecture associated with the first network slice; and
upon determining that the first machine-learning architecture forms a portion of the at least one available end-to-end machine-learning architecture, transmitting, to the user equipment, a first machine-learning architecture response message that grants permission to use the first machine-learning architecture.

### Claim 10 (depends on 9)

The method of claim 9, further comprising:
transmitting, to the user equipment, an available machine-learning architecture message that identifies the at least one available end-to-end machine-learning architecture.

### Claim 11 (depends on 9)

The method of claim 9, further comprising:
transmitting, to the user equipment, a machine-learning architecture selection message that comprises one or more selection rules enabling the user equipment to determine the first machine-learning architecture.

### Claim 12 (depends on 9)

The method of claim 9, further comprising:
creating a second network slice;
receiving, from the user equipment, a second machine-learning architecture request message to request permission to use a second machine-learning architecture;
determining that the second machine-learning architecture is not associated with any second end-to-end machine-learning architectures of the second network slice; and
transmitting, to the user equipment, a second machine-learning architecture response message that does not grant permission to use the second machine-learning architecture.

### Claim 13 (depends on 12)

The method of claim 12, wherein the second machine-learning architecture response message comprises a third machine-learning architecture associated with at least one of the second end-to-end machine-learning architectures.

### Claim 14

An apparatus for wireless communication at a user equipment, comprising:
a transceiver;
a memory; and
a processor coupled to the memory and the transceiver, the processor configured to:
execute a first application associated with a first requested quality-of-service level;
select a first machine-learning architecture based on the first requested quality-of-service level;
transmit, to a network-slice manager of a wireless network, a first machine-learning architecture request message to request permission to use the first machine-learning architecture;
receive, from the network-slice manager, a first machine-learning architecture response message that grants permission to use the first machine-learning architecture based on a first network slice, the first machine-learning architecture forming a portion of at least one first end-to-end machine-learning architecture associated with the first network slice, the at least one first end-to-end machine-learning architecture being a distributed machine-learning architecture that is configured to process wireless communication signals and is formed by the first machine-learning architecture implemented by the user equipment, a machine-learning architecture implemented by a base station, and a machine-learning architecture implemented by an entity of a core network; and
wirelessly communicate data for the first application using the first machine-learning architecture, the first machine-learning architecture being configured to compute an output based on an input using coefficients determined by the user equipment.

### Claim 15 (depends on 14)

The apparatus of claim 14, wherein the selection of the first machine-learning architecture configures the processor to:
receive, from the network-slice manager, an available machine-learning architecture message that identifies available machine-learning architectures associated with the first network slice; and
choose the first machine-learning architecture from the available machine-learning architectures.

### Claim 16 (depends on 14)

The apparatus of claim 14, wherein the selection of the first machine-learning architecture configures the processor to:
receive, from the network-slice manager, a machine-learning architecture selection message that comprises one or more selection rules; and
execute the one or more selection rules to determine the first machine-learning architecture.

### Claim 17 (depends on 14)

The apparatus of claim 14, the processor further configured to:
select a second machine-learning architecture;
transmit, to the network-slice manager, a second machine-learning architecture request message to request permission to use the second machine-learning architecture; and
receive, from the network-slice manager, a second machine-learning architecture response message that grants permission to use the second machine-learning architecture based on a second network slice.

### Claim 18 (depends on 17)

The apparatus of claim 17, the processor further configured to:
wirelessly communicate other data for the first application using the second machine-learning architecture, wherein:
the selection of the second machine-learning architecture is based on the first requested quality-of-service level; and
the selection of the second machine-learning architecture is responsive to at least one of the following:
the user equipment moving to a geographical location associated with a different base station or a different tracking area; and
the user equipment roaming between a home network and a visited network.

### Claim 19 (depends on 14)

The apparatus of claim 14, the processor further configured to:
execute a second application associated with a second requested quality-of-service level;
select a second machine-learning architecture based on the second requested quality-of-service level;
transmit, to the network-slice manager, a second machine-learning architecture request message to request permission to use the second machine-learning architecture; and
receive, from the network-slice manager, a second machine-learning architecture response message that grants permission to use the second machine-learning architecture based on the first network slice or a second network slice.

### Claim 20 (depends on 19)

The apparatus of claim 19, the processor further configured to:
responsive to receive the second machine-learning architecture response message, wirelessly communicate other data for the second application using the second machine-learning architecture, wherein:
the selection of the second machine-learning architecture is based on the second requested quality-of-service level;
the execution of the second application occurs during at least an overlapping portion of time that the first application is executed; and
the use of the first machine-learning architecture and the using of the second machine-learning architecture both occur during at least the overlapping portion of time.
