Network node in communication system, user equipment and execution method thereof

Abstract

The present disclosure relates to a 5G communication system or a 6G communication system for supporting higher data rates beyond 4G communication systems such as Long Term Evolution (LTE). The present disclosure provides a method executed by a first network node, the method comprising: receiving a first message sent by a first node, the first message comprising information for indicating to deactivate a quality of service (QoS) flow of a user equipment (UE), sending the information for indicating to deactivate the QoS flow of the UE to a second node, and deactivating the QoS flow of the UE.

Description

Technical Field

[0001] The present disclosure relates to the field of communications, and more particularly, to a method performed by a first network node, a method performed by a second network node, a method performed by a user equipment, a first network node, a second network node, and a user equipment. Background Art

[0002] Considering the development of wireless communication from generation to generation, these technologies are mainly developed for services targeting people, such as voice calls, multimedia services, and data services. With the commercialization of the fifth-generation (5G) communication system, the number of connected devices is expected to grow exponentially. These will be increasingly connected to the communication network. Examples of the Internet of Things can include vehicles, robots, drones, home appliances, displays, smart sensors connected to various infrastructures, construction machinery, and factory equipment. Mobile devices are expected to develop in various forms such as augmented reality glasses, virtual reality headsets, and holographic devices. Efforts have been made to develop improved 6G communication systems in order to provide various services by connecting hundreds of billions of devices and things in the sixth-generation (6G) era. For these reasons, 6G communication systems are called super 5G systems.

[0003] The 6G communication system, which is expected to be commercialized around 2030, will have a peak data rate of tera (1,000 gigabits) bps and a radio latency of less than 100 μsec, thus being 50 times the data rate of the 5G communication system and having 1 / 10 of its radio latency.

[0004] In order to achieve such high data rates and ultra-low latency, the implementation of 6G communication systems in terahertz (e.g., 95GHz to 3THz bands) has been considered. It is expected that since the path loss and atmospheric absorption in the terahertz band are more serious than those in the millimeter wave (mmWave) band introduced in 5G, the technology that can ensure the signal transmission distance (i.e., coverage) will become more critical. As the main technology to ensure coverage, it is necessary to develop radio frequency (RF) elements, antennas, new waveforms with better coverage than orthogonal frequency division multiplexing (OFDM), beamforming and massive multiple input multiple output (MIMO), full dimensional multiple input multiple output (FD-MIMO), array antennas, and multi-antenna transmission technologies such as massive antennas. In addition, new technologies have been discussed to improve signal coverage in the terahertz band, such as metamaterial-based lenses and antennas, orbital angular momentum (OAM), and reconfigurable intelligence surface (RIS).

[0005] In addition, in order to improve spectrum efficiency and overall network performance, the following technologies have been developed for 6G communication systems: full-duplex technology for enabling uplink and downlink transmissions to use the same frequency resources simultaneously; network technology that utilizes satellites, high-altitude platform stations (HAPS), etc. in an integrated manner; improved network structure to support mobile base stations, etc., and to enable network operation optimization and automation, etc.; dynamic spectrum sharing technology with conflict avoidance based on spectrum usage prediction: using artificial intelligence (AI) in wireless communications to improve overall network operations by utilizing AI from the design stage of developing 6G and internalizing end-to-end AI support functions; and next-generation distributed computing technology that overcomes the computing power limitations of user equipment (UE) through ultra-high performance communication and computing resources achievable on the network (such as mobile edge computing (MEC), cloud, etc.). In addition, attempts are continuing to strengthen connectivity between devices, optimize networks, promote software-based network entities, and increase openness of wireless communications by designing new protocols to be used in 6G communication systems, developing mechanisms for implementing hardware-based security environments and secure use of data, and developing technologies for maintaining privacy.

[0006] It is expected that the research and development of 6G communication systems including hyperconnectivity of person to machine (P2M) and machine to machine (M2M) will bring about the next hyperconnectivity experience. In particular, services such as truly immersive extended reality (XR), high-fidelity mobile holograms, and digital replicas are expected to be provided through 6G communication systems. In addition, services such as remote surgery, industrial automation, and emergency response for enhanced safety and reliability will be provided through 6G communication systems, making the technology applicable to various fields such as industry, healthcare, automobiles, and home appliances. Summary of the invention

[0007] According to one aspect of the present disclosure, a method performed by a first network node in a communication system is provided, the method comprising: receiving a first message sent by the first node, wherein the first message comprises information for indicating deactivation of a quality of service QoS flow of a user equipment UE, sending the information for indicating deactivation of the QoS flow of the UE to a second node, and deactivating the QoS flow of the UE.

[0008] According to the method performed by the first network node in the communication system provided by the present disclosure, the method further includes: receiving a second message sent by the first node, wherein the second message includes information for indicating activation of the QoS flow of the UE, and activating the QoS flow of the UE, and sending the information for indicating activation of the QoS flow of the UE to the second node.

[0009] According to the method performed by the first network node in the communication system provided by the present disclosure, the method further includes: sending a third message to the second network node, wherein the third message includes information indicating whether deactivation of the QoS flow for the UE is successful.

[0010] According to the method performed by the first network node in the communication system provided by the present disclosure, the method further includes: sending a fourth message to the second network node, wherein the fourth message includes information indicating whether activation of the QoS flow for the UE is successful.

[0011] According to the method performed by the first network node in the communication system provided by the present disclosure, the method further includes: receiving a fifth message sent by the second network node, wherein the fifth message includes configuration information related to deactivating or activating a QoS flow.

[0012] According to the method performed by the first network node in the communication system provided by the present disclosure, the configuration information is received when a PDU session or a QoS flow is initially established or is received during one training iteration among multiple training iterations.

[0013] According to the method performed by the first network node in the communication system provided by the present disclosure, the configuration information includes at least one of the following: a protocol data unit PDU session identifier ID or a QoS flow identifier ID; in one iteration, the time when the QoS flow is activated for the first time, the time when the QoS flow is deactivated for the first time, the time when the QoS flow is activated for the second time, and the time when the QoS flow is deactivated for the second time; a model training iteration period; or the number of model training iterations.

[0014] According to the method performed by the first network node in the communication system provided by the present disclosure, the method further includes: sending a sixth message to the second network node, wherein the sixth message includes information for indicating whether the configuration of deactivating or activating the UE's QoS flow is successful.

[0015] According to the method performed by the first network node in the communication system provided by the present disclosure, the method further includes: receiving a seventh message sent by the first node, wherein the seventh message includes information indicating deactivation of the UE's QoS flow when the user equipment leaves federated learning.

[0016] According to the method performed by the first network node in the communication system provided by the present disclosure, the seventh message includes at least one of the following: a protocol data unit PDU session identifier ID or a QoS flow identifier ID; a PDU session or QoS flow status, which is set to be deactivated; a user equipment status of federated learning, which is set to indicate that the user equipment has left federated learning; a federated learning application identifier, which is used to identify a federated learning training task.

[0017] According to the method performed by the first network node in the communication system provided by the present disclosure, the method further includes: sending an eighth message to the second network node, wherein the eighth message includes information indicating whether deactivation of the QoS flow for the UE is successful when the user equipment leaves federated learning.

[0018] According to the method performed by the first network node in the communication system provided by the present disclosure, the method further includes: receiving a ninth message sent by the first node, wherein the ninth message includes information indicating activation of the UE's QoS flow when the user equipment rejoins the federated learning after leaving the federated learning.

[0019] According to the method performed by the first network node in the communication system provided by the present disclosure, the ninth message includes at least one of the following: a protocol data unit PDU session identifier ID or a QoS flow identifier ID; a PDU session or QoS flow status, which is set to activated; a user equipment status of federated learning, which is set to indicate that the user equipment joins federated learning; a federated learning application identifier, which is used to identify a federated learning training task.

[0020] According to the method performed by the first network node in the communication system provided by the present disclosure, the method further includes: sending a tenth message to the second network node, wherein the tenth message includes information for indicating whether the activation of the QoS flow is successful when the user equipment rejoins the federated learning after leaving the federated learning.

[0021] According to the method performed by the first network node in the communication system provided by the present disclosure, the user equipment leaves the federated learning, including at least one of the following: the user equipment leaves the federated learning in an iteration after joining the federated learning because the federated learning training requirements are not met; the user equipment leaves the federated learning during the federated learning model distribution phase or the training result reporting phase due to network congestion or deterioration of the wireless channel quality, or during the model distribution phase, the learning model related data is not received within a predetermined time window, or during the training result reporting phase, the first network node does not receive the training result data within a predetermined time window, resulting in the failure of model distribution or training result reporting; the user equipment leaves the federated learning due to training failure during the federated learning training phase; or the user equipment leaves the federated learning when the first network node detects that the service requirements of the QoS flow cannot be met.

[0022] According to the method performed by the first network node in the communication system provided by the present disclosure, the method further includes: sending an eleventh message to the first node and the second node, wherein the eleventh message includes information indicating deactivation of the QoS flow when the first network node monitors that the service requirements of the QoS flow cannot be met.

[0023] According to the method performed by the first network node in the communication system provided by the present disclosure, the first message and the eleventh message include at least one of the following: a protocol data unit PDU session identifier ID or a QoS flow identifier ID; or a PDU session or QoS flow status, which is set to be deactivated.

[0024] According to the method performed by the first network node in the communication system provided by the present disclosure, the second message includes at least one of the following: a protocol data unit PDU session identification ID or a QoS flow identification ID; or a PDU session or QoS flow status, which is set to activated.

[0025] According to the method performed by the first network node in the communication system provided by the present disclosure, the third message and the eighth message include at least one of the following: a protocol data unit PDU session identifier ID or a QoS flow identifier ID; a success response information indicating that deactivation of the corresponding PDU session or QoS flow is successful; or a rejection response information indicating that deactivation of the corresponding PDU session or QoS flow fails and the corresponding reason information.

[0026] According to the method performed by the first network node in the communication system provided by the present disclosure, the fourth message and the tenth message include at least one of the following: a protocol data unit PDU session identifier ID or a QoS flow identifier ID; a success response information indicating that the activation of the corresponding PDU session or QoS flow is successful; or a rejection response information indicating that the activation of the corresponding PDU session or QoS flow fails and the corresponding reason information.

[0027] According to the method performed by the first network node in the communication system provided by the present disclosure, the first network node is a base station, the first node includes one of the user equipment or the second network node, and the second node includes the other of the user equipment or the second network node, and the second network node is a core network device; or the first network node is a centralized unit control plane CU-CP of the base station, the first node is a core network device, and the second node is a centralized unit user plane CU-UP or a distributed unit DU of the base station; or the first network node is a main node MN, the first node is a core network device, and the second node is a secondary node SN.

[0028] According to a method performed by a first network node in a communication system provided by the present disclosure, the method further includes: receiving a handover request confirmation message sent by a third network node, and sending a radio resource control RRC reconfiguration message to a user equipment, wherein the handover request confirmation message and the RRC reconfiguration message include at least one of the following: a protocol data unit PDU session identifier ID or a QoS flow identifier ID; or a PDU session or QoS flow state, which is set to be deactivated.

[0029] According to the method performed by the first network node in the communication system provided by the present disclosure, the method further includes: receiving a switching command sent by the second network node, and sending a radio resource control RRC reconfiguration message to the user equipment, wherein the switching command and the RRC reconfiguration message include at least one of the following: a protocol data unit PDU session identifier ID or a QoS flow identifier ID; or a PDU session or QoS flow state, which is set to be deactivated.

[0030] According to a method performed by a first network node in a communication system provided by the present disclosure, the method includes: receiving a multicast broadcast service MBS session resource establishment request sent by a second network node, wherein the MBS session resource establishment request includes at least one of the following: MBS start and end time; a scheduled activation time and inactivation time sequence of an MBS session; and an MBS user identifier list for indicating a user equipment identifier to which the MBS will be applied.

[0031] According to a method performed by a first network node in a communication system provided by the present disclosure, the method includes: sending an MBS session resource establishment response message to a second network node, wherein the MBS session resource establishment response message includes at least one of the following: an MBS session identifier; an MBS QoS flow establishment list, including successfully established QoS flow information.

[0032] According to the method performed by the first network node in the communication system provided by the present disclosure, the method includes: sending a message related to MBS session resources to the second network node, wherein the message related to MBS session resources includes at least one of the following: MBS session identifier; cause information.

[0033] According to another aspect of the present disclosure, a method performed by a second network node in a communication system is provided, the method comprising: sending a first message to a first network node and deactivating a QoS flow of the UE, wherein the first message includes information for indicating deactivation of the quality of service QoS flow; and / or sending a second message to the first network node and activating the QoS flow of the UE, wherein the second message includes information for indicating activation of the QoS flow.

[0034] According to the method performed by the second network node in the communication system provided by the present disclosure, after receiving the twelfth message sent by the user equipment, the twelfth message includes information for indicating deactivation of the QoS flow, and sending the first message to the first network node.

[0035] According to the method performed by the second network node in the communication system provided by the present disclosure, after receiving the thirteenth message sent by the user equipment, the thirteenth message includes information for indicating activation of the QoS flow, and sending the second message to the first network node.

[0036] According to the method performed by the second network node in the communication system provided by the present disclosure, the method further includes: receiving a fourteenth message sent by a user equipment, wherein the fourteenth message includes information indicating deactivation of the QoS flow when the user equipment leaves the federated learning; and sending the information indicating deactivation of the QoS flow when the user equipment leaves the federated learning to the first network node.

[0037] According to the method performed by the second network node in the communication system provided by the present disclosure, the method further includes: receiving a fifteenth message sent by a user equipment, wherein the fifteenth message includes information for indicating activation of a QoS flow when the user equipment rejoins federated learning after leaving federated learning, and sending to the first network node the information for indicating activation of a QoS flow when the user equipment rejoins federated learning after leaving federated learning.

[0038] According to the method performed by the second network node in the communication system provided by the present disclosure, the method further includes: receiving a sixteenth message sent by a user equipment, wherein the sixteenth message includes information indicating deactivation of a QoS flow when the user equipment fails in federated learning training, and sending the information indicating deactivation of a QoS flow when the user equipment fails in federated learning training to the first network node.

[0039] According to the method performed by the second network node in the communication system provided by the present disclosure, the method further includes: receiving a path switching request message sent by a third network node, wherein the path switching request message includes at least one of the following: a protocol data unit PDU session identifier ID or a QoS flow identifier ID; or a PDU session or QoS flow state, which is set to be deactivated.

[0040] According to the method performed by the second network node in the communication system provided by the present disclosure, the method further includes: receiving a switching request confirmation sent by a third network node, wherein the switching request confirmation includes at least one of the following: a protocol data unit PDU session identifier ID or a QoS flow identifier ID; or a PDU session or QoS flow state, which is set to be deactivated.

[0041] According to another aspect of the present disclosure, a method performed by a user equipment UE in a communication system is provided, the method comprising: determining that a condition for deactivating and / or activating a quality of service QoS flow of the UE is met, sending a first message to a first network node or a second network node, the first message including information for indicating deactivation of the QoS flow, and / or sending a second message to the first network node or the second network node, the second message including information for indicating activation of the QoS flow.

[0042] According to the method performed by the user equipment UE in the communication system provided by the present disclosure, the method further includes: receiving a seventeenth message sent by the server, wherein the seventeenth message includes information for indicating that the server decides that the user equipment leaves federated learning.

[0043] According to the method performed by the user equipment UE in the communication system provided by the present disclosure, the method further includes: receiving an eighteenth message sent by the server, the eighteenth message including information for indicating that the server decides that the user equipment will rejoin the federated learning after leaving the federated learning.

[0044] According to a method performed by a user equipment UE in a communication system provided by the present disclosure, wherein, the method further comprises: sending a nineteenth message to a server, wherein the nineteenth message comprises information for indicating that federated learning training has failed. According to another aspect of the present disclosure, a first network node is provided, wherein the first network node comprises: a transceiver configured to send and receive signals to and from an external device; and a controller configured to control the transceiver to perform the method performed by the first network node.

[0045] According to another aspect of the present disclosure, a second network node is provided, comprising: a transceiver configured to send and receive signals to and from the outside; and a controller configured to control the transceiver to execute the method executed by the second network node.

[0046] According to another aspect of the present disclosure, a user equipment is provided, the user equipment comprising: a transceiver configured to send and receive signals to and from the outside; and a controller configured to control the transceiver to execute the above method executed by the user equipment.

[0047] According to another aspect of the present disclosure, a non-transitory computer-readable recording medium is provided, on which a program for executing any of the above methods when executed by a computer is stored. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] For a more complete understanding of the present disclosure and its advantages, reference is now made to the following description taken in conjunction with the accompanying drawings, wherein like reference numerals represent like parts:

[0049] Figure 1 An example wireless network according to an embodiment of the present disclosure is shown;

[0050] Figure 2 An example base station according to an embodiment of the present disclosure is shown;

[0051] Figure 3 An example user device according to an embodiment of the present disclosure is shown;

[0052] Figure 4 shows a process of example federated learning according to an embodiment of the present disclosure;

[0053] Figure 5 An example process of an activation and deactivation mechanism based on GBR QoS Flow of a UE based on federated learning training according to an embodiment of the present disclosure is shown;

[0054] Figure 6 Another example process of an activation and deactivation mechanism based on GBR QoS Flow of a UE based on federated learning training according to an embodiment of the present disclosure is shown;

[0055] Figure 7 An example process of a network-triggered QoS flow release or establishment process according to an embodiment of the present disclosure is shown;

[0056] Figure 8 An example process of a network-triggered PDU session release or establishment process according to an embodiment of the present disclosure is shown;

[0057] Fig. 9 An example process of GBRQoS processing of a UE trained by federated learning in an abnormal scenario according to an embodiment of the present disclosure is shown;

[0058] Fig.10 Another example process of GBRQoS processing of a UE trained by federated learning in an abnormal scenario according to an embodiment of the present disclosure is shown;

[0059] Fig.11 Another example process of GBRQoS processing of a UE trained by federated learning in an abnormal scenario according to an embodiment of the present disclosure is shown;

[0060] Fig.12 Another example process of GBRQoS processing when the federated learning trains the UE to switch based on the Xn interface according to an embodiment of the present disclosure is shown;

[0061] Fig.13 Another example process of GBRQoS processing when the federated learning trains the UE to switch based on the Ng interface according to an embodiment of the present disclosure is shown;

[0062] Fig.14 An example process of UE-triggered federated learning training for UE joining a multicast MBS for model distribution according to an embodiment of the present disclosure is shown;

[0063] Fig.15 An example process of a federated learning training UE joining a multicast MBS for model distribution triggered by an FL server according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0064] Before carrying out the following specific embodiments, it may be advantageous to set forth the definitions of certain words and phrases used throughout the patent document. The term "connection" and its derivatives refer to any direct or indirect communication between two or more elements, regardless of whether those elements are in physical contact with each other. The terms "send", "receive" and "communicate" and their derivatives include both direct and indirect communication. The terms "include" and "include" and their derivatives mean unrestricted inclusion. The term "or" is compatible, meaning and / or. The phrase "associated with" and its derivatives mean including, included, interconnected with it, including, included, connected to or connected with it, coupled to or coupled with it, can communicate with it, collaborate with it, interweave, juxtapose, approach, be bound to or bound with it, have, have the property of, have to ... or have a relationship with ... etc. The term "controller" means any device, system or part thereof that controls at least one operation. Such a controller can be implemented in a hardware manner or in a combination of hardware and software and / or firmware. The functionality associated with any particular controller, whether local or remote, may be centralized or distributed. The phrase "at least one of" when used with a list of items means that different combinations of one or more of the listed items may be used, and only one item in the list may be required. For example, "at least one of A, B, and C" includes any of the following combinations: A, B, C, A and B, A and C, B and C, and only A, only B, only C. Likewise, the term "set" means one or more. Thus, a set of items may be a single item or a set of two or more items.

[0065] Moreover, various functions as described below can be implemented or supported by one or more computer programs, each of which is formed by a computer-readable program code and embodied in a computer-readable medium. The terms "application" and "program" refer to one or more computer programs, software components, instruction sets, processes, functions, objects, categories, instances, related data, or a portion thereof suitable for implementation in a suitable computer-readable program code. The phrase "computer-readable program code" includes any type of computer code, including source code, object code, and executable code. The phrase "computer-readable medium" includes any type of medium that can be accessed by a computer, such as a read-only memory (ROM), a random access memory (RAM), a hard drive, a compact disc (CD), a digital video disc (DVD), or any other type of memory. "Non-transitory" computer-readable media excludes wired, wireless, optical, or other communication links that convey instantaneous electrical signals or other signals. Non-transitory computer-readable media include media in which data can be permanently stored and media in which data can be stored and rewritten later, such as rewritable optical discs or erasable memory devices.

[0066] Definitions for certain other words and phrases are provided throughout this patent document. Those of skill in the art should understand that in many, if not most instances, such definitions apply to prior, as well as future uses of such defined words and phrases.

[0067] The figures and various embodiments used to describe the principles of the present disclosure are included herein for illustration only and should not be construed in any way as limiting the scope of the present disclosure. In addition, those skilled in the art will appreciate that the principles of the present disclosure can be implemented in any appropriately arranged wireless communication system.

[0068] The following Figures 1 to 3 Various embodiments of the present disclosure are described as being implemented in a wireless communication system. Figures 1 to 3 The description is not meant to imply physical or architectural limitations to the manner in which different embodiments may be implemented. Different embodiments of the present disclosure may be implemented in any suitably arranged communications system.

[0069] Figure 1 An example wireless network according to an embodiment of the present disclosure is shown. Figure 1 The embodiment of the wireless network shown in FIG. 1 is for illustration only. Other embodiments of the wireless network 100 may be used without departing from the scope of the present disclosure.

[0070] like Figure 1As shown, the wireless network includes a base station (next generation nodeB, gNB or gNodeB) 101, gNB 102, and gNB 103. gNB 101 communicates with gNB 102 and gNB 103. gNB 101 also communicates with at least one network 130 such as the Internet, a proprietary Internet Protocol (IP) network, or other data network.

[0071] The gNB 102 provides wireless broadband access to a network 130 for a plurality of first user equipments (UEs) within a coverage area 120 of the gNB 102. The plurality of first UEs include a UE 111 that may be located at a small business (SB); a UE 112 that may be located at an enterprise (E); a UE 113 that may be located at a WiFi hot spot (HS); a UE 114 that may be located at a first residence (R1); a UE 115 that may be located at a second residence (R2); and a UE 116 that may be a mobile device (M) such as a cellular phone, a wireless laptop, a wireless personal digital assistant (PDA), etc. The gNB 103 provides wireless broadband access to the network 130 for a plurality of second UEs within a coverage area 125 of the gNB 103. The plurality of second UEs include UE 115 and UE 116 and subscriber stations (SS, e.g., UEs) 117, 118, and 119. In some embodiments, one or more of gNBs 101-103 may communicate with each other and UEs 111-116 using existing wireless communication technologies, and one or more of UEs 111-119 may communicate directly with each other (e.g., UEs 117-119) using other existing or proposed wireless communication technologies.

[0072] Depending on the type of network, the term "base station" or "BS" may refer to any component (or collection of components) configured to provide wireless access to a network, such as a transmit point (TP), a transmit-receive point (TRP), an enhanced (or "evolved") base station (eNodeB or eNB), a 5G base station (gNB), a macro cell, a femto cell, a wireless fidelity (WiFi) access point (AP), or other wireless capable devices. A base station may provide wireless access according to one or more wireless communication protocols, such as 3GPP 5G new radio (NR), Long Term Evolution (LTE), LTE-Advanced (LTE-A), high speed packet access (HSPA), Wi-Fi 802.11a / b / g / n / ac, etc. For convenience, various names of base station type devices and functions are used interchangeably in this patent document to refer to network infrastructure components that provide wireless access to remote terminals. Also, depending on the network type, the term "user equipment" (UE) may refer to any component such as a mobile station (MS), a subscriber station (SS), a remote terminal, a wireless terminal, a reception point, or a user device, etc. For convenience, various names of user equipment type devices and functions are used interchangeably in this patent document to refer to a remote wireless device that wirelessly accesses a BS regardless of whether the UE is a mobile device (such as a mobile phone or a smart phone) or a remote wireless device that is generally considered to be a fixed device (such as a desktop computer or a vending machine).

[0073] Dashed lines illustrate the approximate extents of coverage areas 120 and 125, which are shown as generally circular for purposes of illustration and explanation only. It should be clearly understood that coverage areas associated with a gNB, such as coverage areas 120 and 125, may have other shapes, including irregular shapes, depending on the configuration of the gNB and variations in the wireless environment associated with natural and artificial obstacles.

[0074] As described in more detail below, one or more of UEs 111-119 include circuitry, programming, or a combination thereof. In certain embodiments, one or more of gNBs 101-103 include circuitry, programming, or a combination thereof.

[0075] although Figure 1 An example of a wireless network is shown, but Figure 1Various changes may be made. For example, wireless network 100 may include any number of gNBs and any number of UEs in any suitable arrangement. Moreover, gNB 101 may communicate directly with any number of UEs and provide those UEs with wireless broadband access to network 130. Similarly, each gNB 102-103 may communicate directly with network 130 and provide UEs with direct wireless broadband access to network 130. In addition, gNBs 101, 102, and / or 103 may provide access to other or additional external networks such as an external telephone network or other types of data networks.

[0076] Figure 2 An example base station according to an embodiment of the present disclosure is shown. Figure 2 The embodiment of the gNB 102 shown in FIG. is for illustration only, and Figure 1 gNBs 101 and 103 may have the same or similar configurations. However, gNBs appear in a variety of configurations, and Figure 2 The scope of the present disclosure is not limited to any particular implementation of the gNB.

[0077] like Figure 2 As shown in FIG. 1 , gNB 102 includes multiple antennas 200a-200n, multiple radio frequency (RF) transceivers 201a-201n, a transmit (TX) processing circuit 203, and a receive (RX) processing circuit 204. gNB 102 also includes a controller / processor 205, a memory 206, and a backhaul or network interface (IF) 207.

[0078] The RF transceivers 201a-201n receive incoming RF signals from the antennas 200a-200n, such as signals sent by UEs in the network 100. The RF transceivers 201a-201n down-convert the incoming RF signals to generate intermediate frequency (IF) or baseband signals. The IF or baseband signals are sent to the RX processing circuit 204, which generates processed baseband signals by filtering, decoding and / or digitizing the baseband or IF signals. The RX processing circuit 204 sends the processed baseband signals to the controller / processor 205 for further processing.

[0079] TX processing circuit 203 receives analog or digital data (such as voice data, web data, email, or interactive video game data) from controller / processor 205. TX processing circuit 203 encodes, multiplexes, and / or digitizes the outgoing baseband data to generate processed baseband or IF signals. RF transceivers 201a-201n receive the outgoing processed baseband or IF signals from TX processing circuit 203 and up-convert the baseband or IF signals to RF signals that are transmitted via antennas 201a-201n.

[0080] Controller / processor 205 may include one or more processors or other processing devices that control the overall operation of gNB 102. For example, controller / processor 205 may control the reception of forward channel signals and the transmission of reverse channel signals by RF transceivers 201a-201n, RX processing circuitry 204, and TX processing circuitry 203 in accordance with well-known principles. Controller / processor 205 may also support additional functionality, such as more advanced wireless communication functionality.

[0081] For example, the controller / processor 205 may support beamforming or directional routing operations in which outgoing signals from multiple antennas 200a-200n are weighted differently to effectively steer the outgoing signals in a desired direction. Any of a variety of other functions may be supported in the gNB 102 by the controller / processor 205.

[0082] The controller / processor 205 is also capable of executing programs and other processes, such as an operating system (OS), located in the memory 206. The controller / processor 205 can move data into or out of the memory 206 as required by the executing process.

[0083] The controller / processor 205 is also connected to a backhaul or network interface 207. The backhaul or network interface 207 allows the gNB 102 to communicate with other devices or systems via a backhaul connection or via a network. The interface 207 may support communication over any appropriate (one or more) wired or wireless connections. For example, when the gNB 102 is implemented as part of a cellular communication system (such as a cellular communication system supporting 5G, LTE, or LTE-A), the interface 207 may allow the gNB 102 to communicate with other gNBs via a wired or wireless backhaul connection. When the gNB 102 is implemented as an access point, the interface 207 may allow the gNB 102 to communicate over a wired or wireless local area network or over a wired or wireless connection to a larger network (such as the Internet). The interface 207 includes any appropriate structure that supports communication over a wired or wireless connection such as an Ethernet or RF transceiver.

[0084] Memory 206 is connected to controller / processor 205. A portion of memory 206 may include random access memory (RAM), and another portion of memory 206 may include flash memory or other read-only memory (ROM).

[0085] although Figure 2 An example of gNB 102 is shown, but the Figure 2 For example, gNB 102 may include any number of Figure 2 As a specific example, the access point may include multiple interfaces 207, and the controller / processor 205 may support routing functions to route data between different network addresses. As another specific example, although shown as including a single instance of TX processing circuitry 203 and a single instance of RX processing circuitry 204, the gNB 102 may include multiple instances of each (such as one for each RF transceiver). For example, Figure 2 The various components in may be combined, further subdivided, or omitted, and additional components may be added according to specific needs.

[0086] Figure 3 An example user device according to an embodiment of the present disclosure is shown. Figure 3 The embodiment of UE 116 shown in FIG. 1 is for illustration only, and Figure 1 UEs 111-115 and 117-119 may have the same or similar configurations. However, UEs may appear in a variety of configurations, and Figure 3 The scope of the present disclosure is not limited to any particular implementation of the UE.

[0087] like Figure 3 As shown in FIG. 1 , UE 116 includes an antenna 301, a radio frequency (RF) transceiver 302, a TX processing circuit 303, a microphone 304, and a receive (RX) processing circuit 305. UE 116 also includes a speaker 306, a controller or processor 307, an input / output (I / O) interface (IF) 308, an input device 309, a touch screen display 310, and a memory 311. The memory 311 includes an OS 312 and one or more applications 313.

[0088] RF transceiver 302 receives incoming RF signals transmitted by a gNB of network 100 from antenna 301. RF transceiver 302 downconverts the incoming RF signals to generate an IF or baseband signal. The IF or baseband signal is sent to RX processing circuitry 305, which generates a processed baseband signal by filtering, decoding, and / or digitizing the baseband or IF signal. RX processing circuitry 305 sends the processed baseband signal to speaker 306 (such as for voice data) or processor 307 for further processing (such as for web browsing data).

[0089] The TX processing circuit 303 receives analog or digital voice data from the microphone 304 or other outgoing baseband data (such as web data, email, or interactive video game data) from the processor 307. The TX processing circuit 303 encodes, multiplexes, and / or digitizes the outgoing baseband data to generate a processed baseband or IF signal. The RF transceiver 302 receives the outgoing processed baseband or IF signal from the TX processing circuit 303 and up-converts the baseband or IF signal to an RF signal that is transmitted via the antenna 301.

[0090] The processor 307 may include one or more processors or other processing devices and executes the OS 312 stored in the memory 311 to control the overall operation of the UE 116. For example, the processor 307 may control the reception of forward channel signals and the transmission of reverse channel signals by the RF transceiver 302, the RX processing circuit 305, and the TX processing circuit 303 according to well-known principles. In some embodiments, the processor 307 includes at least one microprocessor or microcontroller.

[0091] The processor 307 is also capable of executing other processes and programs located in the memory 311, such as a process for CSI (Channel State Information) reporting on the uplink channel. The processor 307 can move data into or out of the memory 311 as needed to execute the process. In some embodiments, the processor 307 is configured to execute the application 313 based on the OS 312 or in response to a signal received from the gNB or operator. The processor 307 is also coupled to the I / O interface 308, which provides the UE 116 with the ability to connect to other devices such as laptops and portable computers. The I / O interface 308 is the communication path between these accessories and the processor 307.

[0092] Processor 307 is also connected to touch screen display 310. A user of UE 116 may use touch screen display 310 to enter data into UE 116. Touch screen display 310 may be a liquid crystal display, light emitting diode display, or other display capable of rendering text and / or at least limited graphics, such as from a website.

[0093] The memory 311 is connected to the processor 307. A portion of the memory 311 may include RAM, and another portion of the memory 311 may include flash memory or other ROM.

[0094] although Figure 3 An example of UE 116 is shown, but the Figure 3 Make various changes. For example, Figure 3 The various components in the embodiment may be combined, further subdivided, or omitted, and additional components may be added as required. As a specific example, processor 307 may be divided into multiple processors, such as one or more central processing units (CPUs) and one or more graphics processing units (GPUs). Furthermore, although Figure 3 The UE 116 is shown configured as a mobile phone or smart phone, but the UE may be configured to operate as other types of mobile or stationary devices.

[0095] In the communication mechanism, for the GBR (Guaranteed Bit Rat) QoS (Quality of Service) Flow that has been established by the user equipment (UE), the network side reserves corresponding resources. However, sometimes the resources reserved for the QoS Flow are temporarily not used during one or certain periods of time, resulting in a waste of resources. Releasing and then establishing the QoS Flow is a possible solution, but this method may lead to frequent release and establishment of the QoS Flow, which will generate a large signaling overhead.

[0096] Various embodiments of the present disclosure provide a method performed by a first network node, by receiving a first message sent by a second node, the first message including information for indicating deactivation of a quality of service QoS flow, and sending the first message to a third node, so that when the resources reserved for the QoS Flow are temporarily not used during a certain period of time, the quality of service QoS flow can be deactivated by the information for indicating deactivation of the quality of service QoS flow included in the first message (for example, the network side retains the configuration of the GBR QoS Flow but no longer reserves the corresponding resources, and these resources can be used for other transmissions, thus avoiding waste of resources). Since there is no need to frequently release and then establish the QoS Flow, the large signaling overhead generated by the frequent release and establishment of the QoS Flow is improved / solved.

[0097] In addition, as the digitalization process accelerates, a large amount of data will be generated. Machine learning technology can automatically mine the treasures hidden in the data. Machine learning models trained with a large amount of data have been applied in various scenarios and are profoundly changing our world, such as precision medicine, clinical auxiliary diagnosis, new drug development, portrait recognition, voiceprint recognition, recommendation algorithms for thousands of people, pictures, voice, natural language and other multimodal learning. In the application, the accuracy and generalization ability of the model are crucial, and all of these rely on the machine's learning of large amounts of data.

[0098] Due to data privacy and security constraints such as laws and regulations, policy supervision, commercial secrets, and personal privacy, multiple data sources cannot directly exchange data, forming a "data island" phenomenon, which restricts the further improvement of artificial intelligence model capabilities. Federated Learning (FL) was born to solve this problem.

[0099] The combination of the federated learning mechanism and the mobile communication system is an application trend in the future. In the following, when the federated learning mechanism is combined with the 5G mobile communication system, the situation in which the resources reserved for QoSFlow are temporarily not utilized at a certain stage of the federated learning will be described as an implementation method of the present disclosure. However, it can be understood that the present disclosure is not limited to the federated learning mechanism, and any mechanism that may cause the resources reserved for QoSFlow to be temporarily not utilized at a certain time period and mechanisms that meet similar characteristics are within the scope of protection of the present disclosure. In addition, the present disclosure is not limited to the 5G communication system, and the present disclosure can also be extended to the 6G communication system or other communication systems that can implement the inventive concept of the present disclosure.

[0100] In the FL mode, the cloud server performs model averaging based on iterations, aggregates the partially trained local models of each terminal device, and thus trains a global model. In each training iteration, the device uses local training data to train according to the model downloaded from the FL server. The device then reports the mid-term training results to the cloud server via the UL channel of 5G or other wireless communication systems, and then the FL server aggregates the training results from each device and updates the global model. Next, the updated global model is distributed to each device via the DL channel of 5G or other wireless communication systems, and then the device can perform the next iteration of training. The distribution of the global model or the uploading of the device training results is closely related to the services of 5G or related wireless communication systems.

[0101] The following will be combined Figure 4 Describe in detail the integration of the federated learning mechanism with the mobile communication system. Please refer to Figure 4 , Figure 4 The basic process of FL is shown, which is basically divided into the following stages in one iteration:

[0102] 1. Federated learning training UE selection stage

[0103] a) For multiple candidate UEs for federated learning training, each UE reports the training resource status to the FL server before each round of training tasks, and then the FL server decides whether the UE participates in or leaves this round of training tasks. As an implementation method, when the UE's status does not change much, the UE does not need to notify the FL server of the training resource status in each iteration. For example, after the UE reports the training resource status in a certain round, the FL server decides whether the UE joins or leaves the federation in this round of iteration. In subsequent iterations, if the UE's status does not change, it can choose not to report, and the FL server continues to execute the last selection result. When the UE's training resource status changes, the updated status will be reported to the FL server.

[0104] 2. Model distribution stage

[0105] a) For UEs participating in this round of federated learning training, 5GS (5G system) needs to establish communication resources in the core network, radio access network and UE side, such as establishing corresponding PDU sessions, QoS flows and data radio bearers (DRBs). In the synchronous federated learning training task, in order to ensure consistent communication performance in each UE, to ensure that the model distribution time is basically consistent, and to control the reporting of training results within a certain time, the FL server may require 5GS to establish a GBR (Guaranteed Bit Rat) QoS Flow for the UEs participating in the federated learning training.

[0106] b) The following describes the possible disadvantages of establishing a non-GBR QoS Flow for UEs participating in federated learning training. If the FL server establishes a non-GBR QoS Flow for UEs participating in federated learning training, the cell and base station where each UE is located may be different, or even in the same cell or base station, the configuration of the wireless resources for the UE cannot be guaranteed to be the same. In this way, for the distribution of the federated model, it is very likely that the time for each UE to complete the model download is different, and some UEs complete the model download quickly, while some UEs complete the model download slowly. In this way, when each UE completes the training and reports the training results, similar problems will also occur. The time when the training results of each UE arrive at the FL server is different, and the time span between the first arrival and the last arrival may be large, which will affect the FL server's model aggregation process. Of course, in order to avoid affecting the FL server's model aggregation process due to the late reporting of the results of some UEs, the FL server may implement some strategies, such as deciding to stop waiting for UEs that have not reported results after a certain time threshold is exceeded from the start of model distribution to the start of model aggregation. However, this will lead to other problems. Since the training results of the UE in this iteration did not successfully participate in the aggregation of the model, this resulted in a waste of UE resources. Therefore, establishing a GBRQoS Flow for the UE participating in the federated learning training is conducive to the advancement of the entire process of the synchronous federation and will facilitate the rapid convergence of the model synchronization. The solutions involved in the present invention are all implemented under the premise that the FL server requires the 5GS to establish a GBR QoS Flow for the federated learning training UE.

[0107] 3. UE training phase of federated learning training

[0108] a) After completing the model download and configuration of federated learning training, the UE participating in this iteration of federated learning training performs the corresponding training according to the configuration. For the UE in this process, if there is no other business, from the perspective of UE energy saving, the UE may notify the network by carrying certain indications (for example, only the federated learning training task or the indication of expecting to enter the inactive state), and expect the network to decide that the UE enters the RRC_INACTIVE (RRC inactive) state as soon as possible.

[0109] 4.UE training result reporting stage

[0110] a) For UEs participating in this round of training, after completing the training task, they need to report the training results to the FL server according to the configuration of the federated learning training. When the UE completes the training in the RRC_INACTIVE state, it enters the RRC_CONNECTED state through the RRC resume process. The UE reports the training results.

[0111] 5. Model aggregation stage

[0112] a) When the FL server receives the training results of each UE participating in the federated learning training, or adopts some strategies to receive the training results of some UEs at the same time, it decides to aggregate the model.

[0113] After completing an iterative training task, the FL server obtains a new aggregated global model, and the FL server will decide to start the next iterative task. This process is repeated until the convergence of the model meets the corresponding accuracy requirements, and the federated learning training is stopped, thus completing the federated learning training task.

[0114] pass Figure 4 According to the analysis, regarding the basic process of federated learning training, there are several stages involved in one iteration. The requirements and use of GBR QoS only involve the model distribution stage and the UE training result reporting stage. When the UE is in the training stage, the model download has been completed, and it is waiting for the output of the training results before reporting. At this stage, the resources reserved for GBR QoS required by the federated learning training task will not be used, and depending on different models and training tasks, the duration of UE training is uncertain, which may be seconds or minutes. Therefore, if in one iteration, the resources of GBR QoS required by the federated learning training task are always reserved but not used during the UE training process, a large waste of resources will occur. Based on this problem, this paper proposes an optimization processing mechanism for GBR QoS in federated learning training tasks.

[0115] Please refer to Figure 5 , Figure 5An example process of an activation and deactivation mechanism based on GBRQoS Flow of a UE based on federated learning training according to an embodiment of the present disclosure is shown.

[0116] like Figure 5 As shown in the figure, the core idea of ​​the optimization processing mechanism of GBR QoS in the federated learning training task is to introduce the UE-based GBR QoS Flow-based activation and deactivation mechanism, or the corresponding PDU Session activation and deactivation mechanism.

[0117] -When the GBR QoS Flow or the corresponding PDU Session is in the deactivated state, the 5G Core Network (5G Core Network), RAN (Radio Access Network) and UE side save the configuration of the existing GBR QoS Flow, but the RAN side and the 5GC side no longer reserve corresponding resources. The previously reserved resources can be allocated to other UEs to avoid resource waste.

[0118] -When the GBR QoS Flow or the corresponding PDU Session is in the activated state, the 5GC, RAN and UE sides activate the configuration of the existing GBR QoS Flow, and the RAN side and the 5GC side need to reserve corresponding resources to meet the GBR QoS requirements.

[0119] The current specification on the establishment of GBR QoS Flow can be achieved by establishing a new PDU Session, or by establishing a new QoS Flow on an existing PDU Session. Regardless of the method, the introduced UE-based GBR QoS Flow-based activation and deactivation mechanism, or the UE-based activation and deactivation mechanism of the PDU Session corresponding to the GBR QoS Flow are the ideas embodied in the present invention and are within the protection scope of the present invention.

[0120] When the FL server decides that a certain UE joins this iterative training task, it notifies 5GS to establish the corresponding GBRQoS Flow and may indicate that the initial state of the GBR QoS Flow is configured as "activated". Optionally, the initial state of the GBR QoS Flow may also be configured as "deactivated", and then when the FL server decides to start the model distribution task, it notifies 5GS to activate the GBR QoS Flow. Optionally, the existing mechanism of the specification is used. When the GBR QoS Flow is established, the default state is activated and no indication is required. When the RAN receives the deactivation indication of the GBR QoS Flow, the activation indication is required to restore the GBR QoS Flow to the activated state. When the GBR QoS Flow is in the activated state, the FL server distributes the model. For a certain UE, when the model distribution is completed, the FL server knows this event and notifies 5GC to configure the corresponding GBR QoS Flow deactivation state so that the 5GC, RAN and UE sides can save the configuration of the existing GBR QoS Flow, but the RAN side and the 5GC side no longer reserve the corresponding resources.

[0121] Regarding the indication of GBR QoS Flow deactivation, it is optional that it is triggered by the AF / FL server and notified to the 5GC, and then the 5GC notifies the RAN through the NGAP message, and then notifies the UE through the RRC message. The indication message may contain one or more of the following information:

[0122] -PDU Session ID or QoS Flow ID (QFI)

[0123] -PDU Session or QoS Flow Status="deactivated"

[0124] Regarding the GBR QoS Flow activation indication, it is optional and is triggered by the AF / FL server to notify the 5GC, which then notifies the RAN through an NGAP message, and then notifies the UE through an RRC message. The indication message may contain one or more of the following information:

[0125] -PDU Session ID or QoS Flow ID (QFI)

[0126] -PDU Session or QoS Flow Status="activated"

[0127] Regarding the GBR QoS Flow status indication information, the NG interface can be used to indicate from the 5GC to the RAN through a new Class1 or Class2 process, or the existing NG interface message can be used to indicate when the initial QoS Flow is established (such as NGAP (NG Application Protocol): INITIAL CONTEXT SETUP REQUEST (initial context establishment request) or PDU SESSION RESOURCE SETUP REQUEST (PDU session resource establishment request)), or it can be indicated through QoS Flow modification signaling (such as NGAP: UE CONTEXT MODIFICATION REQUEST (UE context modification request) or PDU SESSION RESOURCE MODIFY REQUEST (PDU session resource modification request)). If the Class1 process is used, for the indication of GBR QoS Flow deactivation, the corresponding response message may include one or more of the following information:

[0128] -PDU Session ID or QoS Flow ID (QFI)

[0129] - Successful response or confirmation message of "PDU Session deactivated" or "QoS Flow deactivated"

[0130] - "PDU Session deactivated" or "QoS Flow deactivated" rejection information and may indicate Cause information

[0131] For the indication of GBR QoS Flow activation, the corresponding response message may include one or more of the following information:

[0132] -PDU Session ID or QoS Flow ID (QFI)

[0133] - Successful response or confirmation message of "PDU Session activated" or "QoS Flow activated"

[0134] - "PDU Session activated" or "QoS Flow activated" rejection information and may indicate Cause information

[0135] Optionally, when the UE completes the model download, the UE knows this event and notifies the 5GC through NAS layer signaling, and then the 5GC notifies the RAN side to deactivate the corresponding GBR QoS Flow through the above solution. The enhancement of NAS signaling can be optionally implemented in the following ways, such as by adding an IE (information element) to an existing message (such as NAS PDU session modification request), which can include one or more of the following information:

[0136] -PDU Session ID or QoS Flow ID (QFI)

[0137] -PDU Session or QoS Flow Status="deactivated"

[0138] Optionally, the UE notifies the RAN via an RRC message, and then the RAN notifies the 5GC. The corresponding RRC message or NGAP message may include one or more of the following information:

[0139] -PDU Session ID or QoS Flow ID (QFI)

[0140] -PDU Session or QoS Flow Status="deactivated"

[0141] The UE can notify the RAN through a new RRC message or by using an existing RRC message, such as UE Assistance Information. The RAN can also notify the 5GC through a new NGAP message or by using an existing NGAP message (such as PDU Session Resource Modify Indication or PDU Session Resource Notify).

[0142] For the UE training result upload phase, when the UE completes the training, the UE can notify the 5GC through NAS layer signaling, and then the 5GC notifies the RAN side to activate the corresponding GBR QoS Flow through the above solution to upload the training results to the FL server. When the UE completes the training result upload, the UE can notify the 5GC through NAS layer signaling, and then the 5GC notifies the RAN side to deactivate the corresponding GBR QoS Flow through the above solution.

[0143] Optionally, when the UE completes the training and is about to upload the training results, the UE may notify the RAN via an RRC message, and then the RAN notifies the 5GC to activate the corresponding GBR QoS Flow. The corresponding RRC message or NGAP message may contain one or more of the following information:

[0144] -PDU Session ID or QoS Flow ID (QFI)

[0145] -PDU Session or QoS Flow Status="activated"

[0146] Optionally, when the UE completes uploading the training results, the UE can notify the RAN through an RRC message, and then the RAN notifies the 5GC to deactivate the corresponding GBR QoS Flow. The corresponding RRC message or NGAP message may contain one or more of the following information:

[0147] -PDU Session ID or QoS Flow ID (QFI)

[0148] -PDU Session or QoS Flow Status="deactivated"

[0149] Optionally, 5GC may notify the AF / FL server whether the status of a GBR QoS Flow or a corresponding PDU Session of a UE trained by federated learning is activated or deactivated.

[0150] Through the above-mentioned method, the problem that the GBR QoS resources of the UE undergoing federated learning training are always in a reserved but unused state during the training process, resulting in a large waste of resources is solved / improved, and through different message transmission methods, the embodiments of the present disclosure can adapt to more flexible scenarios and various different network architectures.

[0151] In addition, through Figure 4 The analysis of the basic process of federated learning training involves several different stages in one iteration. However, to achieve model convergence and meet the accuracy requirements and complete model training, N different iterations are required, and the processes involved in different iterations are basically the same. In this way, each iteration can be set to be periodic, and the activation and deactivation time of GBR QoS does not need to be configured in each iteration, but can be configured in advance. This can reduce signaling overhead.

[0152] Please refer to Figure 6, Figure 6 Another example process of the activation and deactivation mechanism based on GBRQoS Flow of UE based on federated learning training according to an embodiment of the present disclosure is shown. Figure 6 As shown, the configuration parameters for activation and deactivation can be set in advance when the PDUSession or GBR QoS Flow is initially established, and the settings can be changed when adjustments are required later, or the configuration parameters can be set during one of multiple training iterations, and the settings can be changed when adjustments are required later. Figure 6 As shown in the figure, assuming that in one iteration, the model distribution starts at time T0, the distribution is completed at time T1, the training results start reporting at time T2, and the reporting is completed at time T3. Assuming that the iteration period is T, and N iterations are experienced. Then in the subsequent iteration process, the activation and deactivation operations of GBR QoSFlow can be performed at the following times.

[0153] For the initial distribution of the model, the activation of the GBR QoS Flow can be set at time T0, and at times that increase by an integer multiple of the period T, that is, at time T0, T0+T, T0+2T, ..., T0+(N-1)*T.

[0154] When the model distribution ends, the deactivation of the GBR QoS Flow can be set at time T1, and at times that increase by an integer multiple of the period T, that is, at time T1, T1+T, T1+2T, ..., T1+(N-1)*T.

[0155] For the start of reporting of training results, the activation of GBR QoS Flow can be set at time T2, and the time when the integer multiple period T is increased, that is, at time T2, T2+T, T2+2T, ..., T2+(N-1)*T.

[0156] When the reporting of the training results is finished, the deactivation of the GBR QoS Flow can be set at time T3, and at times that increase the integer multiple of the period T, that is, at time T3, T3+T, T3+2T, ..., T3+(N-1)*T.

[0157] Therefore, the pre-settings for activation and deactivation of GBR QoS Flow can be configured in the NG interface message or RRC message through the following information:

[0158] -PDU Session ID or QoS Flow ID (QFI)

[0159] - In one training iteration, the QoS flow is activated for the first time at the first time T0, deactivated for the first time at the second time T1, activated for the second time at the third time T2, and deactivated for the second time at the fourth time T3 (T0 and T1 can set the start and end time of the model distribution phase respectively, and T2 and T3 can set the start and end time of the training result reporting phase respectively)

[0160] -T (time length, representing the iteration cycle)

[0161] -N (positive integer, representing the number of iterations)

[0162] The above-mentioned pre-configured information may be received when a PDU session (Session) or a QoS Flow (eg, GBR QoS Flow) is initially established or may be received during one training iteration among multiple training iterations.

[0163] Regarding the pre-setting information of GBR QoS Flow activation and deactivation, it can be indicated from 5GC to RAN by means of the new Class1 or Class2 process of the NG interface, or it can be indicated by means of the existing NG interface message at the time of initial QoS Flow establishment (such as NGAP: INITIAL CONTEXT SETUP REQUEST or PDU SESSION RESOURCE SETUP REQUEST), or it can be indicated by means of QoS Flow modification signaling (such as NGAP: UECONTEXT MODIFICATION REQUEST or PDU SESSION RESOURCE MODIFY REQUEST). If the Class1 process is used, for the pre-setting of GBR QoS Flow activation and deactivation, the corresponding response message (indicated by RAN to 5GC) may include one or more of the following information:

[0164] - Successful response or confirmation message

[0165] -Rejection information and possibly indication of Cause information

[0166] By pre-uniformly configuring the activation and deactivation mechanism of the GBR QoS Flow, it is not necessary to configure the activation and deactivation time of the GBR QoS in each iteration, thereby further reducing the signaling overhead. In addition, it can be understood that the pre-configuration method can be pre-configured at the beginning of the iteration, or uniformly pre-configured after multiple iterations (for example, after determining appropriate parameters), and the present disclosure does not limit this.

[0167] As explained above, when the FL server decides to establish a GBR QoS Flow for the UE participating in the federated learning training, it helps to ensure that the time process of model distribution and training result upload in the federated learning training process is roughly consistent in each iteration of training, thereby achieving better model aggregation effect and shortening the convergence speed of the entire federated model. In order to obtain independent and identically distributed samples from all devices, that is, to give all devices a fair opportunity to contribute to the aggregated model. A large number of UEs that meet the training requirements and conditions may need to participate in the federated learning training. The environment in which the UE is located under wireless communication is time-varying and complex. The UE participating in the federated learning training may experience the following situations:

[0168] 1. After the UE has successfully joined the federated learning training task, in the next iteration, certain situations may occur (such as the UE's location changes, or the quality of the wireless channel deteriorates) and the UE may be determined by the FL server to not meet the training requirements during the selection phase of this iteration, causing the UE to leave this iteration. Later, the situation may recover, or the conditions may meet the requirements of the federated learning training, and the UE may re-join the federated learning training.

[0169] 2. When the UE is in the model distribution phase or the training result reporting phase, network congestion or deterioration of wireless channel quality may occur. Or if the model data is not received within a certain time window during the start and end time of the model distribution phase (based on UE implementation), or if the uplink training result data is not received within a certain time window during the start and end time of the training result reporting phase (based on RAN implementation), model distribution failure or training result reporting failure may occur.

[0170] 3. When the UE is in the training phase, training failure may occur due to some conditions of the UE, such as increased computing load or other abnormal computing conditions.

[0171] The following will describe the three abnormal situations mentioned above and the corresponding processing methods.

[0172] In the first case, the final result is that the UE leaves the training task for this iteration and fails to successfully participate in the federated learning training task for this iteration. However, for the GBR QoS Flow established by the UE, the network side is still in the state of reserving resources, but the resources will not be used in this iteration. How to deal with it to avoid wasting wireless resources and core network resources?

[0173] Releasing the GBR QoS Flow through signaling is a possible solution, and then re-establishing the corresponding GBR QoS Flow when the UE re-joins the federation. The corresponding signaling for releasing and establishing the GBR QoS Flow involves NG interface signaling, NAS signaling, and RRC signaling. Figure 7 As shown, Figure 7 An example process of a network-triggered QoS flow release or establishment process according to an embodiment of the present disclosure is shown. Figure 7 In this case, the QoS Flow can be released and established through the PDU session resource modify process without releasing the PDU session. Figure 8 As shown, Figure 8 An example process of a network-triggered PDU session release or establishment process according to an embodiment of the present disclosure is shown. Figure 8 In the process, the QoS Flow can be released and established by releasing the PDU Session resources of the entire FL task and then re-establishing it. Regardless of the signaling mechanism used, when one or more UEs in federated learning training tasks frequently leave and re-join the federation in different iterations, a large signaling overhead will be generated. In order to avoid the problems caused by the above situation, the present application proposes a processing mechanism for GBR QoS of UEs that optimize federated learning training.

[0174] Please refer to Fig. 9 , Fig. 9 An example process of GBR QoS processing of a UE trained by federated learning in an abnormal scenario according to an embodiment of the present disclosure is shown. A flowchart of a processing mechanism 1 for optimizing GBR QoS of a UE trained by federated learning. Fig. 9 As shown, the deactivation operation for FL GBR QoS is:

[0175] Alt.1 (optional solution 1): When the AF / FL server decides that the UE leaves the federated learning training task of this iteration, it notifies the 5GC, and then the network element in the 5GC notifies the RAN side of the deactivation of the GBR QoS Flow of the corresponding FL training task of this UE, or the deactivation of the corresponding PDU Session. The meaning of deactivation is that the RAN side no longer reserves the corresponding resources to avoid wasting resources while retaining the configuration of the existing GBR QoS Flow. There are many ways to notify the RAN, which can be done by defining a new class1 or class2 message, or by adding a new IE to an existing NGAP message (such as PDUSession Resource Modify Request). The message may contain one or more of the following information:

[0176] -PDU Session ID or QoS Flow ID (QFI)

[0177] -PDU Session or QoS Flow Status="deactivated"

[0178] For class1 process, the corresponding response information may include one or more of the following information:

[0179] -PDU Session ID or QoS Flow ID (QFI)

[0180] - Successful response confirmation message of "PDU Session deactivated" or "QoS Flow deactivated"

[0181] - A rejection response message of "PDU Session deactivated" or "QoS Flow deactivated" and may indicate the Cause information

[0182] In order to avoid the impact on the existing messages or existing IEs (such as GBR QoS information IE) on the RAN side in the specification, a new message can be defined by the CN (core network) to notify the RAN, such as Downlink FL information Transfer, or to notify the RAN of the event of UE leaving the federation for a certain federation task, and then the RAN side will configure the corresponding GBR QoS Flow of this UE to retain but release the reserved resources. The newly defined message can contain one or more of the following information:

[0183] -PDU Session ID or QoS Flow ID (QFI)

[0184] -FL UE Status="leave the federation"

[0185] -FL Application ID (used to identify FL training tasks, because a UE may join multiple FL Application training tasks at the same time, and each application task establishes a different PDUSession or a different GBR QoS Flow)

[0186] In short, as long as the purpose of notifying the RAN side is to allow the RAN side to no longer reserve corresponding resources while retaining the configuration of the existing specified GBR QoS Flow, it falls within the protection scope of the present invention.

[0187] Optionally, the RAN side may then notify the UE through an RRC message as needed, and the corresponding PDU session or QoSFlow or DRB enters the deactivated state.

[0188] Alt.2 (optional solution 2): When the AF / FL server decides that the UE leaves the federated learning training task for this iteration, it notifies the UE through application layer information, and then the UE notifies the 5GC through NAS layer signaling, and then the 5GC notifies the RAN side through the solution of Alt.1. The enhancement of NAS signaling can be optionally implemented in the following ways, such as by adding IE to the existing message (such as: NAS PDU session modification request), which can include one or more of the following information:

[0189] -PDU Session ID or QoS Flow ID (QFI)

[0190] -PDU Session or QoS Flow Status="deactivated"

[0191] -FL UE Status="leave the federation"

[0192] -FL Application ID (used to identify FL training tasks, because a UE may join multiple FL Application training tasks at the same time, and each application task establishes a different PDUSession or a different GBR QoS Flow)

[0193] Alt.3 (optional solution 3): When the AF / FL server decides that the UE leaves the federated learning training task for this iteration, it notifies the UE through application layer information, and then the UE notifies the RAN through an RRC message, and then the RAN notifies the CN. The corresponding RRC message or NGAP message may contain one or more of the following information:

[0194] -PDU Session ID or QoS Flow ID (QFI)

[0195] -PDU Session or QoS Flow Status="deactivated"

[0196] -FL UE Status="leave the federation"

[0197] -FL Application ID (used to identify FL training tasks, because a UE may join multiple FL Application training tasks at the same time, and each application task establishes a different PDUSession or a different GBR QoS Flow)

[0198] The UE may notify the RAN through a new RRC message or by using an existing RRC message, such as UE AssistanceInformation. The RAN may also notify the CN through a new NGAP message or by using an existing NGAP message (such as PDUSession Resource Modify Indication or PDU Session Resource Notify process).

[0199] When the UE in federated learning training is determined by the FL server to not meet the training requirements and the UE changes from the previous join to leave this iteration, after the GBR QoS Flow or the corresponding PDU Session has been in a deactivated state through the above-mentioned Option 1, Option 2 or Option 3, in a subsequent iteration, the UE's participation in the federated learning training may be restored, and its conditions meet the requirements of the federated learning training. The UE re-joins the federated learning training and needs to restore the corresponding PDU Session or GBR QoS Flow to an activated state. Figure 6 As shown, the operation for activating FL GBRQoS is:

[0200] Alt.11 (optional solution 11): When the AF / FL server decides that the UE re-joins the federated learning training task of this iteration, it notifies 5GC, and then the network element in 5GC notifies the RAN side of the GBR QoS Flow activation of the corresponding FL training task of this UE, or the corresponding PDU Session activation. The meaning of activation is that when the RAN side restores the configuration of the existing GBR QoS Flow, the RAN side needs to reserve corresponding resources to meet the corresponding GBR QoS requirements. There are many ways to notify the RAN, which can be done by defining a new class1 or class2 message, or by adding a new IE to an existing NGAP message (such as PDU Session Resource Modify Request). It can include one or more of the following information:

[0201] -PDU Session ID or QoS Flow ID (QFI)

[0202] -PDU Session or QoS Flow Status="activated"

[0203] For class1 process, the corresponding response information may include one or more of the following information:

[0204] -PDU Session ID or QoS Flow ID (QFI)

[0205] - Successful response confirmation message of "PDU Session activated" or "QoS Flow activated"

[0206] - A rejection response message of "PDU Session activated" or "QoS Flow activated" and may indicate the Cause information

[0207] In order to avoid the impact on the existing messages or IEs (such as GBR QoS information IE) on the RAN side in the specification, the CN can notify the RAN by defining a new message, such as Downlink FL informationTransfer, or notify the RAN of the event of UE re-joining the federation for a certain federation task, and then the RAN side restores the corresponding GBR QoS Flow configuration of the UE and reserves the corresponding resources. It can contain one or more of the following information:

[0208] -PDU Session ID or QoS Flow ID (QFI)

[0209] -FL UE Status="join the federation"

[0210] -FL Application ID (used to identify the FL training task, because a UE may join multiple FL Application training tasks at the same time, and each application task establishes a different PDU Session or a different GBR QoS Flow)

[0211] In short, as long as the function of notifying the RAN side is to allow the RAN side to reserve corresponding resources when restoring the configuration of the existing specified GBR QoS Flow, it falls within the protection scope of the present invention.

[0212] Optionally, the RAN side may then notify the UE through an RRC message as needed, and the corresponding PDU session or QoS Flow or DRB is restored to the activated state.

[0213] Alt.12 (optional solution 12): When the AF / FL server decides that the UE re-joins the federated learning training task of this iteration, it notifies the UE through application layer information, and then the UE notifies the 5GC through NAS layer signaling, and then the 5GC notifies the RAN side through the solution of Alt.11. The enhancement of NAS signaling can be optionally implemented in the following ways, such as by adding IE to the existing message (such as: NAS PDU session modification request), which can include one or more of the following information:

[0214] -PDU Session ID or QoS Flow ID (QFI)

[0215] -PDU Session or QoS Flow Status="activated"

[0216] -FL UE Status="join the federation"

[0217] -FL Application ID (used to identify the FL training task, because a UE may join multiple FL Application training tasks at the same time, and each application task establishes a different PDU Session or a different GBR QoS Flow)

[0218] Alt.13 (optional solution 13): When the AF / FL server decides that the UE re-joins the federated learning training task of this iteration, it notifies the UE through application layer information, and then the UE notifies the RAN through an RRC message, and then the RAN notifies the CN. The corresponding RRC message or NGAP message may contain one or more of the following information:

[0219] -PDU Session ID or QoS Flow ID (QFI)

[0220] -PDU Session or QoS Flow Status="activated"

[0221] -FL UE Status="join the federation"

[0222] -FL Application ID (used to identify the FL training task, because a UE may join multiple FL Application training tasks at the same time, and each application task establishes a different PDU Session or a different GBR QoS Flow)

[0223] The UE may notify the RAN through a new RRC message or by using an existing RRC message, such as UE AssistanceInformation. The RAN may also notify the CN through a new NGAP message or by using an existing NGAP message (such as PDUSession Resource Modify Indication or PDU Session Resource Notify process).

[0224] Through the above-mentioned approach, the problem of resource waste and / or large signaling overhead caused by the frequent leaving and re-joining of the federation by UEs with one or more federated learning training tasks in different iterations is solved / improved, and through different message transmission methods, the embodiments of the present disclosure can adapt to more flexible scenarios and various network architectures.

[0225] For the second case, when the UE is in the model distribution stage or the training result reporting stage, network congestion or deterioration of wireless channel quality may occur. Or during the start and end time of the model distribution stage, if the model data is not received within a certain time window (based on UE implementation), or during the start and end time of the training result reporting stage, if the uplink training result data is not received within a certain time window (based on RAN implementation), it may cause model distribution failure or training result reporting failure. The final result is that the UE failed to successfully participate in the federated learning training task of this iteration. However, for the GBR QoS Flow established by the UE, the network side is still in the state of reserving resources, but this iteration will not utilize resources. How to deal with it to avoid wasting wireless resources and core network resources.

[0226] Releasing the GBR QoS Flow through signaling is a possible solution, and then re-establishing the corresponding GBR QoS Flow when the UE re-joins the federation. The corresponding signaling for releasing and establishing the GBR QoS Flow involves NG interface signaling, NAS signaling, and RRC signaling. Figure 7As shown in FIG. 1 , the QoS Flow can be released and established through the PDU session resource modify process without releasing the PDU session. Figure 8 As shown, the QoS Flow can also be released and established by releasing the PDU Session resources of the entire FL task and then re-establishing it. However, no matter which signaling mechanism is used, a large signaling overhead will be generated. In order to avoid the problems caused by the above situation, the present application proposes a processing mechanism for GBR QoS of UEs that optimize federated learning training.

[0227] Please refer to Fig.10 , Fig.10 Another example process of GBR QoS processing of a UE trained by federated learning in an abnormal scenario according to an embodiment of the present disclosure is shown. Fig.10 As shown, the deactivation operation for FL GBR QoS is:

[0228] Alt.4 (optional solution 4): When the RAN side monitors that the service requirements of the FL GBR QoS Flow cannot be met, the existing mechanism can be used to notify the CN. The CN notifies the AF / FL server, and then the AF / FL server decides that the UE leaves the federated learning training task of this iteration, notifies the 5GC, and then the network element in the 5GC notifies the RAN side of the deactivation of the GBR QoS Flow of the corresponding FL training task of this UE, or the deactivation of the corresponding PDU Session. The meaning of deactivation is that the RAN side will no longer reserve the corresponding resources to avoid wasting resources while retaining the existing GBR QoS Flow configuration. The notification method can be used Fig. 9 Option 1 in the illustrated embodiment implements corresponding actions to notify the RAN and the UE.

[0229] Of course, Alt.5 (optional solution 5) can be used. In the event of the above abnormal situation, RAN can notify CN and UE to deactivate the GBR QoS Flow corresponding to the FL training task, which may include one or more of the following information:

[0230] -PDU Session ID or QoS Flow ID (QFI)

[0231] -PDU Session or QoS Flow Status="deactivated"

[0232] RAN can notify CN through a new NGAP message or by using an existing NGAP message (such as PDU Session Resource Modify Indication or PDU Session Resource Notify process). Then CN notifies the AF / FL server, and the FL server decides that UE leaves the federated learning training of this iteration.

[0233] Among them, for the start and end time of the model distribution phase, based on the implementation of the UE, if the model data is not received within a certain time window, the UE can notify the FL server of the model distribution failure event through an application layer message, and then the FL server decides that the UE leaves the federated learning training task of this iteration, and then notifies the 5GC. Then the network element in the 5GC notifies the RAN side of the deactivation of the GBR QoS Flow corresponding to the FL training task of this UE, or the corresponding PDUSession deactivation. The meaning of deactivation is that the RAN side no longer reserves the corresponding resources while retaining the configuration of the existing GBR QoS Flow to avoid wasting resources. There can be multiple ways of notification, and the optional scheme 1 in Example 3 can be used to implement the corresponding action to notify the RAN.

[0234] When the UE in federated learning training is decided by the FL server to change from the previous join to leave this iteration, after the GBR QoS Flow or the corresponding PDU Session has been in a deactivated state through the above-mentioned optional solution 4 or optional solution 5, in a subsequent iteration, the UE's participation in the federation may be restored, and its conditions meet the requirements of the federated learning training. The UE re-joins the federated learning training and needs to restore the corresponding PDU Session or GBR QoS Flow to an activated state. The operation of activating the FL GBR QoS can be implemented using the mechanism of optional solution 11, optional solution 12, or optional solution 13 in Example 3.

[0235] Through the above-mentioned method, the problem of resource waste and / or large signaling overhead caused by network congestion or deterioration of wireless channel quality in the user equipment during the federated learning model distribution stage or the training result reporting stage is solved / improved, and through different message transmission methods, the embodiments of the present disclosure can adapt to more flexible scenarios and various different network architectures.

[0236] For the third case, when the UE is in the training phase, there may be some problems of its own, such as increased computing load or other abnormal situations in the calculation, which may cause training failure. The final result is that the UE fails to successfully participate in the federated learning training task of this iteration. However, for the GBR QoS Flow established by the UE, the network side is still in the state of reserving resources, but this iteration will not use resources. How to deal with it to avoid wasting wireless resources and core network resources.

[0237] Releasing the GBR QoS Flow through signaling is a possible solution, and then re-establishing the corresponding GBR QoS Flow when the UE re-joins the federation. The corresponding signaling for releasing and establishing the GBR QoS Flow involves NG interface signaling, NAS signaling, and RRC signaling. Figure 4 As shown in FIG. 1 , the QoS Flow can be released and established through the PDU session resource modify process without releasing the PDU session. Figure 5 As shown, the QoS Flow can also be released and established by releasing the PDU Session resources of the entire FL task and then re-establishing it. Regardless of which signaling mechanism is used, a large signaling overhead will be generated. In order to avoid the problems caused by the above situation, the present application proposes a processing mechanism for GBR QoS of UEs that optimize federated learning training.

[0238] Please refer to Fig.11 , Fig.11 Another example process of GBR QoS processing of a UE trained by federated learning in an abnormal scenario according to an embodiment of the present disclosure is shown. For the deactivation operation of FL GBR QoS, the corresponding QoS processing mechanism can utilize Option 1, Option 2, or Option 3 in the third embodiment.

[0239] The UE notifies the FL server of the training failure event through an application layer message, and then the FL server decides that the UE leaves the federated learning training task for this iteration, and then notifies the 5GC. The network element in the 5GC then notifies the RAN side of the deactivation of the GBR QoS Flow corresponding to the FL training task of this UE, or the deactivation of the corresponding PDU Session. The meaning of deactivation is that the RAN side no longer reserves the corresponding resources to avoid wasting resources while retaining the configuration of the existing GBR QoS Flow. There are many ways to notify, and the optional solution 1 in Example 3 can be used to implement the corresponding actions to notify the RAN and UE.

[0240] Optionally, when a UE fails in training, NAS signaling can be used to inform 5GC that the GBR QoS Flow corresponding to the FL training task of this UE is deactivated, or the corresponding PDU Session is deactivated, and then 5GC notifies the RAN side using Option 1. The CN notifies the FL server, and then the FL server decides whether the UE leaves the federated task of this iteration.

[0241] Optionally, when a UE fails in training, the RAN can be notified via an RRC message, which then notifies the CN, and the CN notifies the FL server of the deactivation of the GBR QoS Flow corresponding to the FL training task of this UE, or the deactivation of the corresponding PDUSession. The FL server then decides whether the UE leaves the federated task of this iteration.

[0242] When the UE in the federated learning training fails and the FL server decides to change from the previous join to leave this iteration, after the GBR QoS Flow or the corresponding PDU Session has been in a deactivated state through the above-mentioned Option 1, Option 2, or Option 3, in a subsequent iteration, the UE's participation in the federation may be restored, and its conditions meet the requirements of the federated learning training. The UE re-joins the federated learning training and needs to restore the corresponding PDU Session or GBR QoS Flow to an activated state. The operation of activating the FL GBR QoS can be implemented using the mechanism of Option 11, Option 12, or Option 13 in Example 3.

[0243] Through the above-mentioned method, the problem of resource waste and / or large signaling overhead caused by training failure of user equipment during the federated learning training phase is solved / improved, and through different message transmission methods, the embodiments of the present disclosure can adapt to more flexible scenarios and various different network architectures.

[0244] Please continue to refer to the following Fig.12 , Fig.12 Another example process of federated learning training of UE for GBR QoS processing when switching based on Xn interface according to an embodiment of the present disclosure is shown.

[0245] The current standard mechanism is that for Xn-based switching, the GBR QoS Flow information and Alternative QoS Parameters Set List information to be established are carried in the switching request message. If the target base station cannot meet the required QoS and Alternative QoS due to the current status (such as network congestion), the establishment of this QoS Flow will be rejected.

[0246] For UEs undergoing federated learning training, when the UE is in the model distribution phase or the training result reporting phase, switching may occur due to mobility. If during the switching process, the target base station refuses to establish the GBRQoS Flow due to temporary network congestion, and then rejoins the federation in a certain iteration after the status is restored, then the GBR QoS Flow for the federated learning training service will be reestablished at the target base station. In order to save signaling overhead, a new mechanism can be considered to be introduced. In the above situation, the target base station still accepts the establishment of the corresponding GBR QoS Flow for the federated learning training service, but the QoS status is configured to be deactivated, and no resources need to be reserved on the RAN side. The corresponding switching request confirmation message may include one or more of the following information:

[0247] -PDU Session ID or QoS Flow ID (QFI)

[0248] -PDU Session or QoS Flow Status="deactivated"

[0249] Accordingly, the source base station notifies the UE by sending an RRC reconfiguration message, which may include one or more information:

[0250] -PDU Session ID or QoS Flow ID (QFI)

[0251] -PDU Session or QoS Flow Status="deactivated"

[0252] Accordingly, the target base station notifies the CN through the NG message that the GBR QoS Flow corresponding to the FL training task is deactivated. The path switching request message may include one or more of the following information:

[0253] -PDU Session ID or QoS Flow ID (QFI)

[0254] -PDU Session or QoS Flow Status="deactivated"

[0255] Then CN notifies the AF / FL server, and the FL server decides that the UE leaves the federated learning training for this iteration. In a subsequent iteration, the network congestion status may be restored, and the CN is notified through the existing mechanism, and then the AF / FL server is notified. In this way, the UE re-joins the federated learning training, and the corresponding PDU Session or GBR QoS Flow needs to be restored to the activated state. The operation of activating FL GBR QoS can be implemented using the mechanism of Option 11, Option 12, or Option 13 in Embodiment 3.

[0256] Through the above method, the problem of resource waste and / or large signaling overhead caused by the target base station refusing to establish the GBR QoS Flow due to temporary network congestion when the UE is in the model distribution stage or the training result reporting stage and switching occurs due to mobility is solved / improved, and then after the state is restored, the target base station rejoins the federation in a certain iteration, and causes large signaling overhead. In addition, through different message transmission methods, the embodiments of the present disclosure can adapt to more flexible scenarios and various network architectures.

[0257] Please continue to refer to the following Fig.13 , Fig.13 Another example process of federated learning training for GBR QoS processing when UE switches based on Ng interface according to an embodiment of the present disclosure is shown.

[0258] The current standard mechanism is that for Ng-based switching, the GBR QoS Flow information and Alternative QoS Parameters Set List information to be established are carried in the switching request message. If the target base station cannot meet the required QoS and Alternative QoS due to the current status (such as network congestion), the establishment of this QoS Flow will be rejected.

[0259] For federated learning training UE, when the UE is in the model distribution stage or the training result reporting stage, switching may occur due to mobility. If during the switching process, the target base station refuses to establish the GBRQoS Flow due to temporary network congestion, and then rejoins the federation in a certain iteration after the state is restored, then the GBR QoS Flow for the federated learning training service will be reestablished in the target base station. In order to save signaling overhead, it is possible to consider introducing a new mechanism. In the above case, the target base station still accepts the establishment of the corresponding GBR QoS Flow for the federated learning training service, but the QoS status is configured to be deactivated, and no resources need to be reserved on the RAN side. The corresponding switching request confirmation message may include one or more of the following information:

[0260] -PDU Session ID or QoS Flow ID (QFI)

[0261] -PDU Session or QoS Flow Status="deactivated"

[0262] Accordingly, the CN forwards the establishment status of the QoS Flow of the target base station to the source base station, and the corresponding handover command may include one or more of the following information:

[0263] -PDU Session ID or QoS Flow ID (QFI)

[0264] -PDU Session or QoS Flow Status="deactivated"

[0265] Accordingly, the source base station notifies the UE by sending an RRC reconfiguration message, which may include one or more information:

[0266] -PDU Session ID or QoS Flow ID (QFI)

[0267] -PDU Session or QoS Flow Status="deactivated"

[0268] Then CN notifies the AF / FL server, and the FL server decides that the UE leaves the federated learning training for this iteration. In a subsequent iteration, the network congestion status may be restored, and the CN is notified through the existing mechanism, and then the AF / FL server is notified. In this way, the UE re-joins the federated learning training, and the corresponding PDU Session or GBR QoS Flow needs to be restored to the activated state. The operation of activating FL GBR QoS can be implemented using the mechanism of Option 11, Option 12, or Option 13 in Embodiment 3.

[0269] Through the above method, the problem of resource waste and / or large signaling overhead caused by the target base station refusing to establish the GBR QoS Flow due to temporary network congestion when the UE is in the model distribution stage or the training result reporting stage and switching occurs due to mobility is solved / improved, and then after the state is restored, the target base station rejoins the federation in a certain iteration, and causes a waste of resources and / or a large signaling overhead. In addition, by setting a different message transmission method under another network architecture, the embodiments of the present disclosure can adapt to more flexible scenarios and various different network architectures.

[0270] In the above embodiments, the core idea of ​​the optimization processing mechanism of GBR QoS in the federated learning training task introduces the activation and deactivation mechanism based on UE based GBR QoS Flow, or the activation and deactivation mechanism of the corresponding PDU Session. In the case of split-gNB, when CU-CP receives the status indication information about GBR QoS Flow from 5GC through the NG interface, it can further indicate to CU-UP through the E1 interface, because CU-UP is responsible for the user plane resources of the UE, and may add a new IE to an existing message through a new message or in an existing message, such as BEARERCONTEXT SETUP REQUEST (bearer context establishment request) or BEARERCONTEXTMODIFICATIONREQUEST (bearer context modification request) and other messages, including one or more of the following information in the new message or in the existing message:

[0271] -PDU Session ID or QoS Flow ID (QFI)

[0272] -PDU Session or QoS Flow Status = "activated or deactivated"

[0273] When the CU-CP receives the status indication information about the GBR QoS Flow from the 5GC through the NG interface, it can further indicate to the DU through the F1 interface, because the DU is also responsible for the user plane resources of the UE. It may be through a new message or by adding a new IE to an existing message, such as UE CONTEXT SETUP REQUEST or UE CONTEXT MODIFICATION REQUEST. One or more of the following information may be added to the new message or to the existing message:

[0274] -PDU Session ID or QoS Flow ID (QFI)

[0275] -PDU Session or QoS Flow Status = "activated or deactivated"

[0276] For UEs in dual connectivity, when the MN master node receives the status indication information about the GBR QoS Flow from the 5GC through the NG interface, it can further indicate to the SN slave node through the Xn interface, possibly through a new message or by adding a new IE to an existing message, such as S-NODE ADDITION REQUEST (slave node addition request) or S-NODE MODIFICATION REQUEST (slave node modification request initiated by the master node) or NOTIFICATION CONTROL INDICATION (notification control indication) and other messages, including one or more of the following information in the new message or adding one or more of the following information to the existing message:

[0277] -PDU Session ID or QoS Flow ID (QFI)

[0278] -PDU Session or QoS Flow Status = "activated or deactivated"

[0279] If the secondary node is also a split-gNB, the CU-CP of the secondary node can also indicate the CU-UP and DU in the corresponding secondary node through the E1 and F1 interfaces.

[0280] For the activation and deactivation instructions of GBR QoS Flow initiated by the RAN side, in the DC dual connection scenario, the SN can notify the MN through the Xn interface, and then the MN notifies the 5GC. New messages can be added to existing messages by adding new IEs, such as S-NODE MODIFICATION REQUIRED (secondary node modification request initiated by the secondary node) or NOTIFICATION CONTROL INDICATION, etc., and one or more of the following information can be included in the new message or added to the existing message:

[0281] -PDU Session ID or QoS Flow ID (QFI)

[0282] -PDU Session or QoS Flow Status = "activated or deactivated"

[0283] Through the above-mentioned approach, the embodiments of the present disclosure can adapt to more flexible scenarios and various network architectures.

[0284] Of course, the activation and deactivation mechanism of the UE-based GBR QoS Flow, or the corresponding activation and deactivation mechanism of the PDUSession, is also applicable to the existing GBR QoS Flow. For the indication of activation and deactivation of the GBR QoS Flow, in addition to defining a new message, a new IE can also be added to the existing message. One possible method is to add status indication information based on the GBR QoS Flow Information IE, which specifically involves the IE in the following specifications.

[0285] ◆NGAP 9.3.1.10GBRQoS Flow Information in TS 38413-h40

[0286] ◆XnAP 9.2.3.6GBR QoS Flow Information in TS 38423-h40

[0287] ◆F1AP 9.3.1.46GBR QoS Flow Information in TS 38473-h41

[0288] ◆E1AP 9.3.1.30GBR QoS Flow Information in TS 37483-h40

[0289] Two implementation methods of implementing the activation and deactivation mechanism based on GBR QoS Flow by adding a new IE (the IE indicates the QoS parameters of the GBR QoS flow for the downlink and uplink) in the current specification are given in the form of a table. The first implementation method is shown in Table 1 below.

[0290]

[0291]

[0292]

[0293] Table 1

[0294] Optionally, indication information (QoS Flow Status) may be added as the status indication information of GBR QoS Flow at the same level as the QoS Flow Identifier IE, as shown in Table 2 below.

[0295]

[0296]

[0297] Table 2

[0298] Through the above-mentioned manner, the embodiments of the present disclosure can be improved under the existing specification definition, so that the embodiments of the present disclosure can be compatible with the current specification.

[0299] For the status indication information of the GBR QoS Flow of the UE, during the mobility of the UE, such as a handover based on the Xn interface or the Ng interface, or an RRC recovery process occurs on the new base station after the UE in the RRC_INACTIVE state is mobilized, the status of the GBR QoS Flow on the source base station will be transferred to the new base station and maintained. For example, if the status of the GBR QoS Flow on the source base station is deactivated, the status of the GBR QoS Flow will be transferred and maintained as the deactivated state after the UE switches to the new base station; if the status of the GBR QoS Flow on the source base station is activated, the status of the GBR QoS Flow will be transferred and maintained as the activated state after the UE switches to the new base station, but in some cases, it may also become a deactivated state.

[0300] Based on the process of federated learning training, it is found that for the distribution of the model, each UE participating in the training receives the same data. In order to further save the user plane resources on the RAN side and the 5GC side, the federated learning FL can introduce the MBS (Multicast / Broadcast service) mechanism for model distribution. Since FL has not only downlink data (model distribution) but also uplink data (training result report), and different UEs may be selected for different iterations, not all UEs in each cell participate in the federation, so it is more appropriate to use multicast service (Multicast).

[0301] Please refer to Fig.14 , Fig.14 An example process of UE-triggered federated learning training for UE to join a multicast MBS for model distribution according to an embodiment of the present disclosure is shown. Fig.14 As shown, the traditional MBS multicast mechanism is that when a user receives a notification of a multicast service, if the user is interested in the multicast service, the UE sends a NAS signaling to initiate a request to participate in the multicast service. However, for federated learning, the selection of UEs participating in federated learning training in each iteration is determined by the FL server. In order to unify the mechanism of using MBS in FL, for the purpose of enabling MBS to better serve FL, and to avoid the situation where multiple UEs selected by FL could have used the MBS mechanism to distribute downlink model data but the UE did not initiate a request to join MBS, the FL server can further determine which UEs participating in the federation are suitable for using MBS for model distribution, and then notify CN, and then CN initiates the establishment of an MBS session to the corresponding base station, and then the base station configures multicast MBS wireless resources for the corresponding UE.

[0302] Please refer to Fig.15 , Fig.15 FIG. 4 shows an example process of a federated learning training UE joining a multicast MBS for model distribution triggered by an FL server according to an embodiment of the present disclosure. Fig.15As shown, in the UE selection phase, in addition to reporting computing resources and wireless channel status, the UE can also report more detailed location information, such as reporting the cell ID information, tracking area ID (TAI), base station ID information, etc. to which the UE is currently attached, so that the FL server can determine whether the UE participates in the federation and decide whether to apply multicast MBS for model distribution, thereby saving resources on the user plane of the RAN side and the 5GC side. For example, when the FL server finds that different UEs participating in federated learning training are in the same base station or the same cell, applying multicast MBS for model distribution can save network bearer resources. Using the existing mechanism, the CN can know whether the RAN node supports the multicast MBS function. Then, the CN side sends a multicast MBS session resource establishment request to the RAN. This signaling is non-UE-associated and MBS session-based signaling. It prepares RAN-side resources for UEs under the RAN node that use MBS for model distribution. The multicast MBS session resource establishment request may include one or more of the following information:

[0303] -MBS Session ID

[0304] -Shared NG-U TNL Information at UPF

[0305] -MBS QoS Flows To Be Setup List (including GBR QoS information required by FL)

[0306] -MBS start and end times, and / or the sequence of scheduled activation times and deactivation times of MBS sessions

[0307] (e.g., the first start and end time, and period) of the MBS session) (indicates the start and end time of different iteration model distributions)

[0308] -MBS Session Status(activated,deactivated,…)

[0309] -MBS UE ID list (corresponding to the UE ID under the base station to which MBS will be applied, which can be represented by RAN UE NGAP ID, and other representation methods are not excluded. The purpose is to indicate which UE under the base station applies MBS)

[0310] After the RAN side successfully configures the corresponding UE through the RRC message, it sends a multicast MBS session resource establishment response message to the CN. The multicast MBS session resource establishment response message may include one or more of the following information:

[0311] -MBS Session ID

[0312] -Shared NG-U TNL Information at NG-RAN

[0313] -MBS QoS Flows Setup List (including successfully established GBR QoS Flow information)

[0314] Then the FL server starts distributing the model data using the UEs that have joined the multicast.

[0315] In order to better adapt to the application of MBS on FL and save resources, the above strategy in which the FL server decides which UEs participating in the federation need to use MBS for model distribution simplifies the signaling process compared to the traditional mechanism of joining and exiting the multicast MBS initiated by the UE, and better reflects the idea of ​​Network For AI.

[0316] For the establishment process of multicast MBS session resources initiated by the CN side, if the RAN side may reject it due to some abnormal conditions, the RAN side sends a multicast MBS session resource establishment failure message. The message may include one or more of the following information:

[0317] -MBS Session ID

[0318] -Cause

[0319] The CN side may also make necessary modifications to certain parameters (such as changes in the selected UE list, GBR QoS required by the FL, MBS session status, MBS start and end time, etc.) in different iterations or at appropriate times. The CN side may trigger a multicast MBS session resource update process. Under normal circumstances, the RAN side sends a response to a successful update, such as a multicast MBS session resource update response message; or due to some abnormal circumstances, the RAN side sends a response to an update failure, such as a multicast MBS session resource update failure message. The multicast MBS session resource update failure message may include one or more of the following information:

[0320] -MBS Session ID

[0321] -Cause

[0322] Correspondingly, if MBS is no longer used for model distribution or federated learning training is completed, CN receives corresponding instructions from the AF / FL server, and the CN side can send a multicast MBS session resource release request message to the RAN. After the RAN side releases the MBS session resources, it sends a multicast MBS session resource release response to the CN.

[0323] -MBS Session ID

[0324] -Cause

[0325] The above method is different from the method triggered by the UE. Instead, it is triggered by the FL server to train the UE to join the multicast MBS for model distribution through federated learning, which solves / improves the problem that multiple UEs selected by the FL can use the MBS mechanism to distribute downlink model data but the UE does not initiate a request to join the MBS.

[0326] It can be understood that the above-mentioned embodiments can be combined with each other to form new embodiments, and various combinations can adapt to more flexible scenarios and various network architectures.

[0327] In addition, it can be understood that the message names in the present invention are only examples, and other message names can also be used. For example, the first message can also be called a deactivation indication message, the second message can also be called an activation indication message, the third message can also be called a deactivation response message, the fourth message can also be called an activation response message, the fifth message can also be called a deactivation preconfiguration message, the sixth message can also be called a deactivation preconfiguration response message, and so on, and the present disclosure does not limit this. Moreover, the "first", "second", etc. contained in the message names of the present invention are only examples of messages and do not represent the execution order.

[0328] In addition, it can be understood that the contents contained in each message in the above-mentioned embodiments are only examples of the present disclosure. The contents contained in various similar or identical messages can be replaced with each other according to the situation, or named with different names, or some items can be added or deleted. For the sake of brevity, the meanings of similar content items can be explained to each other according to the context without further elaboration. The modified and interpreted implementation methods still fall within the scope of protection of the present invention.

[0329] At least one embodiment of the present disclosure also provides a non-transitory computer-readable recording medium having stored thereon a program for executing the above method when executed by a computer.

[0330] Various embodiments of the present disclosure may be implemented as computer-readable codes embodied on a computer-readable recording medium from a specific perspective. A computer-readable recording medium is any data storage device that can store data readable by a computer system. Examples of computer-readable recording media may include read-only memory (ROM), random access memory (RAM), compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, optical data storage device, carrier wave (e.g., data transmission via the Internet), and the like. Computer-readable recording media may be distributed via a computer system connected via a network, and therefore computer-readable codes may be stored and executed in a distributed manner. Moreover, the functional programs, codes, and code segments for implementing the various embodiments of the present disclosure may be easily interpreted by a technician in the field of the embodiments of the present disclosure.

[0331] It will be understood that the embodiments of the present disclosure may be implemented in the form of hardware, software, or a combination of hardware and software. The software may be stored as program instructions or computer-readable codes executable on a controller on a non-transient computer-readable medium. Examples of non-transient computer-readable recording media include magnetic storage media (e.g., ROM, floppy disk, hard disk, etc.) and optical recording media (e.g., CD-ROM, digital video disk (DVD), etc.). The non-transient computer-readable recording medium may also be distributed on a network-coupled computer system so that the computer-readable code is stored and executed in a distributed manner. The medium may be read by a computer, stored in a memory, and executed by a controller. Various embodiments may be implemented by a computer or a portable terminal including a controller and a memory, and the memory may be an example of a non-transient computer-readable recording medium suitable for storing (multiple) programs with instructions for implementing the embodiments of the present disclosure. The present disclosure may be implemented by a program having a code for specifically implementing the apparatus and method described in the claims, the program being stored in a machine (or computer) readable storage medium. The program may be electronically carried on any medium, such as a communication signal transmitted via a wired or wireless connection, and the present disclosure suitably includes its equivalents.

[0332] The above description is only a specific implementation of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any technician familiar with the technical field can make various changes or substitutions within the technical scope disclosed in the present disclosure, and these changes or substitutions should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A method performed by a first network node in a communication system, the method comprising: receiving a first message sent by a first node, wherein the first message includes information for indicating deactivation of a quality of service QoS flow of a user equipment UE, The information for instructing to deactivate the QoS flow of the UE is sent to the second node, and the QoS flow of the UE is deactivated.

2. The method according to claim 1, wherein: The method further comprises: receiving a second message sent by the first node, wherein the second message includes information for indicating activation of a QoS flow of the UE, and activating the QoS flow of the UE, and sending the information indicating activation of the QoS flow of the UE to the second node.

3. The method according to claim 1, wherein: The method further comprises: receiving a fifth message sent by the second network node, The fifth message includes configuration information related to deactivating or activating the QoS flow.

4. The method according to claim 3, wherein: The configuration information is received when a PDU session or a QoS flow is initially established or during one of a plurality of training iterations.

5. The method according to claim 4, wherein: The configuration information includes at least one of the following: Protocol data unit PDU session identification ID or QoS flow identification ID; In one training iteration, the time when the QoS flow is activated for the first time, the time when the QoS flow is deactivated for the first time, the time when the QoS flow is activated for the second time, and the time when the QoS flow is deactivated for the second time; Model training iteration cycle; or The number of model training iterations.

6. The method according to claim 1, wherein: The method further comprises: receiving the seventh message sent by the first node, The seventh message includes information indicating deactivation of the UE's QoS flow when the user equipment leaves federated learning.

7. The method according to claim 6, wherein: The seventh message includes at least one of the following: Protocol data unit PDU session identification ID or QoS flow identification ID; The PDU session or QoS flow state is set to deactivated; The state of the user device of the federated learning is set to indicate that the user device has left the federated learning; Federated learning application identifier, used to identify federated learning training tasks.

8. The method according to claim 1, wherein: The method further comprises: receiving the ninth message sent by the first node, The ninth message includes information indicating activation of the UE's QoS flow when the user equipment rejoins the federated learning after leaving the federated learning.

9. The method according to claim 8, wherein: The ninth message includes at least one of the following: Protocol data unit PDU session identification ID or QoS flow identification ID; The PDU session or QoS flow state is set to active; The user device state of the federated learning is set to indicate that the user device has joined the federated learning; Federated learning application identifier, used to identify federated learning training tasks.

10. The method according to claim 6, wherein: The user equipment leaving the federated learning includes at least one of the following: After joining the federated learning, the user device leaves the federated learning in one iteration because it does not meet the federated learning training requirements; The user equipment leaves the federated learning due to network congestion or deterioration of wireless channel quality during the federated learning model distribution phase or the training result reporting phase, or when the learning model related data is not received within a predetermined time window during the model distribution phase, or when the first network node does not receive the training result data within a predetermined time window during the training result reporting phase, resulting in the failure of model distribution or training result reporting; During the federated learning training phase, the user device leaves the federated learning due to training failure; or When the first network node detects that the service requirement of the QoS flow cannot be met, the user equipment leaves the federated learning.

11. The method according to claim 1, wherein: The method further comprises: sending an eleventh message to the first node and the second node, The eleventh message includes information indicating deactivation of the QoS flow when the first network node detects that a service requirement of the QoS flow cannot be met.

12. The method according to claim 1 or 11, wherein: The first message and the eleventh message include at least one of the following: Protocol data unit PDU session identification ID or QoS flow identification ID; or The PDU session or QoS flow state is set to deactivated.

13. The method according to claim 2, wherein: The second message includes at least one of the following: Protocol data unit PDU session identification ID or QoS flow identification ID; or The PDU session or QoS flow state is set to active.

14. The method according to any one of claims 1 to 11, wherein: The first network node is a base station, the first node includes one of a user equipment or a second network node, and the second node includes the other of the user equipment or the second network node, and the second network node is a core network device; or The first network node is a centralized unit control plane CU-CP of a base station, the first node is a core network device, and the second node is a centralized unit user plane CU-UP or a distributed unit DU of the base station; or The first network node is a main node MN, the first node is a core network device, and the second node is a secondary node SN.

15. The method according to claim 1, wherein: The method further comprises: receiving a handover request confirmation message sent by the third network node, and Sending a radio resource control (RRC) reconfiguration message to a user equipment, The handover request confirmation message and the RRC reconfiguration message include at least one of the following: Protocol data unit PDU session identification ID or QoS flow identification ID; or The PDU session or QoS flow state is set to deactivated.

16. The method according to any one of claims 1 to 15, wherein: The method further comprises: receiving a switching command sent by the second network node, and Sending a radio resource control (RRC) reconfiguration message to a user equipment, The handover command and the RRC reconfiguration message include at least one of the following: Protocol data unit PDU session identification ID or QoS flow identification ID; or The PDU session or QoS flow state is set to deactivated.

17. A method performed by a second network node, the method comprising: Sending a first message to a first network node, and deactivating the QoS flow of the UE, wherein the first message includes information for indicating deactivation of the quality of service QoS flow; and / or A second message is sent to the first network node, and the QoS flow of the UE is activated, wherein the second message includes information for indicating activation of the QoS flow.

18. The method according to claim 17, further comprising: receiving a path switching request message and / or a switching request confirmation sent by a third network node, The path switching request message includes at least one of the following: Protocol data unit PDU session identification ID or QoS flow identification ID; or The PDU session or QoS flow state is set to deactivated, The handover request confirmation includes at least one of the following: Protocol data unit PDU session identification ID or QoS flow identification ID; or The PDU session or QoS flow state is set to deactivated.

19. A method performed by a user equipment UE in a communication system, the method comprising: determining that a condition for deactivating and / or activating a quality of service QoS flow of the UE is met, Sending a first message to a first network node or a second network node, wherein the first message includes information for indicating deactivation of a QoS flow, and / or A second message is sent to the first network node or the second network node, wherein the second message includes information indicating activation of the QoS flow.

20. A node device, comprising: Transceiver; as well as A controller coupled to the transceiver and configured to execute the method described in any one of claims 1-16 or 17-18 or 19.

Images