Machine learning based traffic detection in a mobile system
AssigneeIntel Corporation
InventorsAlexandre Saso STOJANOVSKI, Zongrui DING, Meghashree Dattatri KEDALAGUDDE, Thomas LUETZENKIRCHEN, Qian LI
Life of this patent
Estimating the term…
Abstract
This disclosure describes systems, methods, and devices related to ML-based traffic detection. A device may receive a Policy Control and Charging (PCC) rule from a Policy Control Function (PCF), which includes an identifier (ID) for a standardized traffic pattern (STP) and a machine learning model identifier (ML model ID) of an ML model to be used for traffic detection. The device may send a Packet Detection Rule (PDR) to a User Plane Function (UPF) that includes an STP ID and the ML model ID to initiate an ML based detection of the STP.
Claims22
A device for a session management function (SMF) for managing traffic in a 5G system (5GS), the SMF comprising processing circuitry configured to:
- a processing circuitry configured to:
- receive a Policy Control and Charging (PCC) rule from a Policy Control Function (PCF), which includes an identifier (ID) for a standardized traffic pattern (STP) and a machine learning model identifier (ML model ID) of an ML model to be used for traffic detection;
- and
- send a Packet Detection Rule (PDR) to a User Plane Function (UPF) that includes an STP ID and the ML model ID to initiate an ML based detection of the STP;
- and
- a memory to store the PCC rule.
A method for Machine Learning (ML) based traffic detection in a 5G System (5GS), comprising:
- receiving a Policy Control and Charging (PCC) rule at a Session Management Function (SMF) from a Policy Control Function (PCF);
- processing the PCC rule to include identifiers for both a Standardized Traffic Profile (STP) and an ML model;
- instructing User Equipment (UE) via an N1 reference point to initiate detection of the STP using the identified ML model.
(canceled)
(canceled)
A non-transitory computer-readable medium storing computer-executable instructions which when executed by one or more processors for a session management function (SMF) for managing traffic in a 5G system (5GS), result in performing operations comprising:
- receive a Policy Control and Charging (PCC) rule from a Policy Control Function (PCF), which includes an identifier (ID) for a standardized traffic pattern (STP) and a machine learning model identifier (ML model ID) of an ML model to be used for traffic detection;
- and
- send a Packet Detection Rule (PDR) to a User Plane Function (UPF) that includes an STP ID and the ML model ID to initiate an ML based detection of the STP.
Dependencies are inferred from the claim text.