Packet data unit session for machine learning exploration for wireless communication network optimization
Devices, methods and computer programs for a packet data unit session for machine learning exploration for wireless communication network optimization are disclosed. A client device detects a trigger condition being satisfied for machine learning exploration for wireless communication network optimization. In response, the client device initiates establishment of a packet data unit (PDU) session for the machine learning exploration. The PDU session for the machine learning exploration is associated with a machine learning exploration indicator that indicates that the PDU session is reserved for machine learning exploration.
1. A client device, comprising:
at least one processor; and
at least one memory including computer program code;
the at least one memory and the computer program code configured to, with the at least one processor, cause the client device to at least perform:
receiving, from a network node device, a request to perform machine learning exploration comprising traffic steering optimization;
detecting a trigger condition being satisfied for the machine learning exploration for wireless communication network optimization, wherein the wireless communication network optimization comprises the traffic steering optimization to be performed by a session management function, SMF, entity by initiating at least one packet data unit session and applying a plurality of different filters, and wherein the wireless communication network optimization further comprises at least one of:
mobility optimization to be performed by an access and mobility management function, AMF, entity;
training of a time-sensitive networking, TSN, related functionality to be performed by the user plane function entity;
training of a radio specific procedure to be performed by a base station device;
protocol specific optimization to be performed by the base station device;
training of at least one of detection, decoding or power control to be performed by the client device; or
generating analytics related to radio access technology, RAT, or frequency selection to be performed by a network data analytics function, NWDAF, entity; and
in response, initiating establishment of a packet data unit, PDU, session for the machine learning exploration, the PDU session for the machine learning exploration being associated with a machine learning exploration indicator indicating that the PDU session is reserved for machine learning exploration.
2. The client device according to claim 1 , wherein the trigger condition comprises at least one of:
a need detected by the client device to perform the machine learning exploration; or
a request to perform the machine learning exploration received from a network node device.
3. The client device according to claim 1 , wherein the machine learning exploration indicator comprises at least one of:
a quality-of-service, QoS, identifier value indicating machine learning exploration;
a PDU session attribute indicating machine learning exploration;
a single network slice selection assistance information, S-NSSAI, identifier indicating machine learning exploration; or
a PDU session identifier indicating machine learning exploration.
4. The client device according to claim 1 , wherein exploration data for the machine learning exploration comprises at least one of:
randomly generated data;
data generated based on previously recorded data; or
data generated based on a parallel, non-machine learning exploration, PDU session.
5. The client device according to claim 1 , wherein the PDU session is to be established between the client device and a user plane function entity.
6. The client device according to claim 1 , wherein the wireless communication network optimization further comprises:
training of QoS handling to be performed by a user plane function entity.
7. A network node device, comprising:
at least one processor; and
at least one memory including computer program code;
the at least one memory and the computer program code configured to, with the at least one processor, cause the network node device to at least perform:
detecting a need to perform the machine learning exploration for wireless communication network optimization, wherein the wireless communication network optimization comprises traffic steering optimization to be performed by a session management function, SMF, entity by initiating at least one packet data unit session and applying a plurality of different filters, and wherein the wireless communication network optimization further comprises at least one of:
mobility optimization to be performed by an access and mobility management function, AMF, entity;
training of a time-sensitive networking, TSN, related functionality to be performed by the user plane function entity;
training of a radio specific procedure to be performed by a base station device;
protocol specific optimization to be performed by the base station device;
training of at least one of detection, decoding or power control to be performed by the client device; or
generating analytics related to radio access technology, RAT, or frequency selection to be performed by a network data analytics function, NWDAF, entity; and
transmitting a request to perform the machine learning exploration comprising the traffic steering optimization to one or more client devices.
8. The network node device according to claim 7 , wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the network node device to perform:
selecting the one or more client devices to which the request to perform the machine learning exploration is to be transmitted based on a selection criterion comprising:
client device capability to support a packet data unit, PDU, session for machine learning exploration;
a detected performance issue;
client device inactivity status; or
a client device related identifier, user category or subscribed service.
9. The network node device according to claim 7 , wherein the network node device comprises an operation, administration, and maintenance, OAM, entity or a session management function, SMF, entity.