# US 2024/0205116 A1 · Method and apparatus for processing traffic by using artificial intelligence model in wireless communication system

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

- Number: US 2024/0205116 A1
- Document ID: US20240205116A1
- Status: Published application
- Kind: A1
- Application number: 18/523,006
- Filed: 2023-11-29
- Published: 2024-06-20
- Earliest claimed: 2022-12-19
- Assignees: Samsung Electronics Co., Ltd.
- Applicants: Samsung Electronics Co., Ltd.
- Inventors: Sunhyun KIM; Jiyoung CHA; Sangho LEE; Dongmyung KIM
- Web page: https://patentgrep.com/patent/US20240205116A1

## CPC

- H04L 43/026: using flow identification
- H04L 41/16: using machine learning or artificial intelligence
- H04W 16/18: Network planning tools
- H04W 24/02: Arrangements for optimising operational condition
- H04W 24/08: Testing, {supervising or monitoring} using real traffic
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72)

## Abstract

The present disclosure relates to a 5G communication system or a 6G communication system for supporting higher data rates beyond a 4G communication system such as long term evolution (LTE). A method performed by a user plane function (UPF) node in a wireless communication system includes receiving multiple traffic packets; identifying at least one encrypted traffic packet among the multiple traffic packets; identifying frame type information of the at least one traffic packet, based on an artificial intelligence (AI) model embedded in the UPF; identifying a quality of service (QoS) flow corresponding to at least one traffic packet, based on the identified frame type information; and transmitting the at least one traffic packet, based on the identified QoS flow.

## Claims

### Claim 1

A method performed by a user plane function (UPF) node in a wireless communication system, the method comprising:
receiving multiple traffic packets;
identifying at least one encrypted traffic packet among the multiple traffic packets;
identifying frame type information of the at least one encrypted traffic packet, based on an artificial intelligence (AI) model embedded in the UPF;
identifying a quality of service (QoS) flow corresponding to at least one encrypted traffic packet, based on the identified frame type information; and
transmitting the at least one traffic packet, based on the identified QoS flow.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
receiving information instructing activation of online training of the AI model from an artificial intelligence management function (AIMF); and
performing online training of the AI model, based on the identified frame type information.

### Claim 3 (depends on 2)

The method of claim 2, further comprising:
receiving a request for deployment of the AI model from the AIMF;
deploying a version of the AI model, based on the request; and
notifying the AIMF of completion of deployment of the AI model.

### Claim 4 (depends on 2)

The method of claim 2, further comprising:
receiving a request for backup of the AI model from the AIMF;
backing up a current version of the AI model; and
notifying the AIMF of completion of backup of the AI model.

### Claim 5 (depends on 2)

The method of claim 2, further comprising:
receiving a request for monitoring of the AI model from the AIMF;
assessing performance of the AI model; and
receiving an instruction determined based on the assessment result, from the AIMF,
wherein the instruction comprises at least one of an instruction to stop using the AI model, an instruction to roll back the AI model, or an instruction to back up the AI model.

### Claim 6 (depends on 5)

The method of claim 5, wherein assessing performance of the AI model is based on a classification ratio of the traffic packet, or is based on feedback information received from at least one node receiving the traffic packet.

### Claim 7 (depends on 6)

The method of claim 6, wherein, in case that an instruction to roll back the AI model is received from the AIMF, the method further comprises:
identifying a version of an AI model to be changed, based on at least one of a rollback condition included in the rollback instruction, or information regarding a backed-up AI model;
deactivating online training of the AI model;
changing the deactivated AI model to the identified version of an AI model to be changed;
activating online training of the AI model of the changed version;
identifying frame type information of the at least one encrypted traffic packet, based on the AI model of the changed version and the online training; and
notifying the AIMF of completion of an AI model rollback request.

### Claim 8 (depends on 1)

The method of claim 1, further comprising:
identifying an unencrypted traffic packet among the multiple traffic packets; and
identifying frame type information of the unencrypted traffic packet, based on header information of the unencrypted traffic packet and payload header information.

### Claim 9 (depends on 1)

The method of claim 1, wherein identifying frame type information of the at least one encrypted traffic packet comprises acquiring the frame type information by the AI model, by using header information of the at least one encrypted traffic packet as an input value, and
wherein the header information comprises at least one of a header size, a header reception interval, or an encoding algorithm.

### Claim 10 (depends on 1)

The method of claim 1, wherein identifying at least one encrypted traffic packet among the multiple traffic packets comprises identifying at least one encrypted traffic packet encrypted by a secure real-time transport protocol (SRTP), among the multiple traffic packets.

### Claim 11

A user plane function (UPF) node in a wireless communication system, the UPF node comprising:
at least one transceiver; and
at least one processor functionally coupled to the at least one transceiver,
wherein the at least one processor is configured to:
receive multiple traffic packets;
identify at least one encrypted traffic packet among the multiple traffic packets;
identify frame type information of the at least one encrypted traffic packet, based on an artificial intelligence (AI) model embedded in the UPF;
identify a quality of service (QoS) flow corresponding to at least one encrypted traffic packet, based on the identified frame type information; and
transmit the at least one traffic packet, based on the identified QoS flow.

### Claim 12 (depends on 11)

The UPF node of claim 11, wherein the at least one processor is further configured to:
receive information instructing activation of online training of the AI model from an artificial intelligence management function (AIMF); and
perform online training of the AI model, based on the identified frame type information.

### Claim 13 (depends on 12)

The UPF node of claim 12, wherein the at least one processor is further configured to:
receive a request for deployment of the AI model from the AIMF;
deploy a version of the AI model, based on the request; and
notify the AIMF of completion of deployment of the AI model.

### Claim 14 (depends on 12)

The UPF node of claim 12, wherein the at least one processor is further configured to:
receive a request for backup of the AI model from the AIMF;
back up a current version of the AI model; and
notify the AIMF of completion of backup of the AI model.

### Claim 15 (depends on 12)

The UPF node of claim 12, wherein the at least one processor is further configured to:
receive a request for monitoring of the AI model from the AIMF;
assess performance of the AI model; and
receive an instruction determined based on the assessment result, from the AIMF,
wherein the instruction comprises at least one of an instruction to stop using the AI model, an instruction to roll back the AI model, or an instruction to back up the AI model.

### Claim 16 (depends on 15)

The UPF node of claim 15, wherein, the at least one processor is configured to assess performance of the AI model based on a classification ratio of the traffic packet, or to assess performance of the AI model based on feedback information received from at least one node receiving the traffic packet.

### Claim 17 (depends on 16)

The UPF node of claim 16, wherein, in case that an instruction to roll back the AI model is received from the AIMF, the at least one processor is further configured to:
identify a version of an AI model to be changed, based on at least one of a rollback condition included in the rollback instruction, or based on information regarding a backed-up AI model;
deactivate online training of the AI model;
change the deactivated AI model to the identified version of an AI model to be changed;
activate online training of the AI model of the changed version;
identify frame type information of the at least one encrypted traffic packet, based on the AI model of the changed version and the online training; and
notify the AIMF of completion of an AI model rollback request.

### Claim 18 (depends on 11)

The UPF node of claim 11, wherein the at least one processor is further configured to:
identify an unencrypted traffic packet among the multiple traffic packets; and
identify frame type information of the unencrypted traffic packet, based on header information of the unencrypted traffic packet and payload header information.

### Claim 19 (depends on 11)

The UPF node of claim 11, wherein, in order to identify frame type information of the at least one encrypted traffic packet, the at least one processor is configured to acquire the frame type information by the AI model, by using header information of the at least one traffic packet as an input value, and
wherein the header information comprises at least one of a header size, a header reception interval, or an encoding algorithm.

### Claim 20 (depends on 11)

The UPF node of claim 11, wherein, in order to identify at least one encrypted traffic packet among the multiple traffic packets, the at least one processor is configured to identify at least one packet encrypted by a secure real-time transport protocol (SRTP), among the multiple traffic packets.
