Method and apparatus for vertical federated learning for network analytics service

Abstract

A network exposure function (NEF) can be configured to receive a service request for vertical federated learning (VFL) initialization. The NEF can determine a list of wireless transmit / receive units (WTRUs) to be used as samples for VFL training. The NEF can also send a service response for VFL initialization. The service response for VFL initialization can indicate the list of WTRUs to be used as samples for VFL training. Further, the NEF can receive an acknowledgement of the VFL initialization and perform VFL operations.

Description

[0001] Cross-reference to related applications This application claims the benefit of U.S. Provisional Application No. 63 / 574,669, filed April 4, 2024, the entire contents of which are incorporated herein by reference. Background Technology

[0002] Network data analytics features in cellular networks can provide statistics and / or predictions based on specific requests from entities consuming that information. Some examples of the types of information these features may provide include statistics and predictions regarding gNB status information, gNB resource usage, and communication / mobility performance in areas of interest. The targets of such analyses may include a single radio transmit / receive unit (WTRU), a group of WTRUs, or any WTRU in an area of ​​interest. Furthermore, network data analytics can be provided to characterize network function load and / or network slice load, and can provide predictions and statistics regarding WTRU mobility, expected WTRU behavior, and even service experience observed at multiple levels, including network slices, service experience for specific applications, or service experience for specific applications on specific access types (e.g., radio access technology (RAT) type or frequency). Summary of the Invention

[0003] The Network Exposure Function (NEF) can be configured (e.g., from the Application Function (AF)) to receive service requests for Vertical Federated Learning (VFL) initialization and / or alignment. Requests may include initialization and / or alignment information (e.g., including sample selection criteria), and / or requests for machine learning models. The NEF can execute alignment procedures with the network function (e.g., based on sample selection criteria). The NEF can determine a list of Wireless Transmit / Receive Units (WTRUs) as samples. The NEF can also send service responses for VFL initialization or alignment. Responses may include parameters for VFL alignment. Furthermore, the NEF can receive confirmation of VFL initialization and perform VFL operations (e.g., VFL training, sharing of VFL results).

[0004] Service requests used for VFL initialization can be based on the analytics ID, the requested machine learning (ML) model, a list of requested (e.g., available) features, the time period of the request, and selection criteria for sample WTRUs. The NEF can evaluate one or more WTRUs and determine whether they meet the sample selection criteria. If the NEF determines that one or more WTRUs do not meet the sample selection criteria, the NEF can exclude said one or more WTRUs. Parameters used for VFL alignment can indicate the analytics service and / or the associated ML model.

[0005] NEF can also discover and select network functions, such as the Federated Learning (FL) Server Network Data Analysis Function (NWDAF). In one example, the service response for VFL initialization is based on the analysis ID, available ML models, a list of available features, and a recommended WTRU list as samples. In one embodiment, confirmation of VFL initialization is based on the list of available features and the WTRU list as samples. In another example, the WTRU list meets selection criteria.

[0006] The NEF can receive initiation messages for VFL training. For example, an initiation message may include a list of available features and / or a list of WTRUs selected for VFL training. The NEF can perform VFL operations based on the initiation message (e.g., using the feature list and / or WTRU list included in the initiation message). VFL operations may include VFL training and / or sharing of VFL results trained locally. Attached Figure Description

[0007] Figure 1A This is a system diagram illustrating an example communication system that can implement one or more of the disclosed embodiments.

[0008] Figure 1B The illustration is based on an embodiment and can be used Figure 1A The diagram shows a system diagram of an example wireless transmit / receive unit (WTRU) used in a communication system.

[0009] Figure 1C The illustration is based on an embodiment and can be used Figure 1A The diagram shows a system diagram of an example radio access network (RAN) and an example core network (CN) used in a communication system.

[0010] Figure 1D The illustration is based on an embodiment and can be used Figure 1A The diagram shows another example RAN and another example CN used in the communication system.

[0011] Figure 2 A flowchart illustrating an exemplary procedure for a longitudinal federated learning (VFL) operation initiated by an application function (AF) for network (NW) analysis is provided. Detailed Implementation

[0012] Figure 1AThis is a diagram illustrating an exemplary communication system 100 that can implement one or more of the disclosed embodiments. The communication system 100 may be a multiple access system that provides content, such as voice, data, video, messaging, broadcasting, etc., to multiple wireless users. The communication system 100 enables multiple wireless users to access this content through the sharing of system resources, including wireless bandwidth. For example, the communication system 100 may employ one or more channel access methods, such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal FDMA (OFDMA), Single Carrier FDMA (SC-FDMA), Zero-Tail Unique Word DFT Spread Spectrum OFDM (ZT UW DTS-s OFDM), Unique Word OFDM (UW-OFDM), Resource Block Filtered OFDM, Filter Library Set Multicarrier (FBMC), etc.

[0013] like Figure 1A As shown, the communication system 100 may include wireless transmit / receive units (WTRUs) 102a, 102b, 102c, 102d, RAN 104 / 113, CN 106 / 115, Public Switched Telephone Network (PSTN) 108, Internet 110, and other networks 112. However, it will be understood that the disclosed embodiments contemplate any number of WTRUs, base stations, networks, and / or network elements. Each of the WTRUs 102a, 102b, 102c, and 102d may be any type of device configured to operate and / or communicate in a wireless environment. As examples, WTRUs 102a, 102b, 102c, and 102d (any of which may be referred to as a “station” and / or “STA”) may be configured to transmit and / or receive wireless signals and may include user equipment (UE), mobile stations, fixed or mobile subscriber units, subscription-based units, pagers, cellular phones, personal digital assistants (PDAs), smartphones, laptops, netbooks, personal computers, wireless sensors, hotspots or Mi-Fi devices, Internet of Things (IoT) devices, watches or other wearable devices, head-mounted displays (HMDs), vehicles, drones, medical devices and applications (e.g., remote surgery), industrial devices and applications (e.g., robots and / or other wireless devices operating in industrial and / or automated processing chain scenarios), consumer electronics devices, devices operating on commercial and / or industrial wireless networks, etc. Any of WTRUs 102a, 102b, 102c, and 102d may be interchangeably referred to as WTRUs. Furthermore, any descriptions herein made with reference to UEs may equally apply to WTRUs (or vice versa). For example, the WTRU can be configured to execute any procedure or program described herein as being executed by the UE (or vice versa).

[0014] The communication system 100 may also include base station 114a and / or base station 114b. Each of the base stations 114a and 114b may be any type of device configured to wirelessly interface with at least one of the WTRUs 102a, 102b, 102c, and 102d to facilitate access to one or more communication networks (such as CN 106 / 115, Internet 110, and / or other networks 112). As an example, base stations 114a and 114b may be base transceivers (BTS), Node-B, eNode B, home Node B, home eNode B, gNB, NR Node B, field controllers, access points (APs), wireless routers, etc. Although base stations 114a and 114b are depicted as single elements, it will be understood that base stations 114a and 114b may include any number of interconnected base stations and / or network elements.

[0015] Base station 114a may be part of RAN 104 / 113, which may also include other base stations and / or network elements (not shown), such as base station controllers (BSCs), radio network controllers (RNCs), relay nodes, etc. Base station 114a and / or base station 114b may be configured to transmit and / or receive radio signals on one or more carrier frequencies, which may be referred to as cells (not shown). These frequencies may be in licensed spectrum, unlicensed spectrum, or a combination of licensed and unlicensed spectrum. A cell may provide coverage of a specific geographic area, which may be relatively fixed or may change over time. A cell may be further divided into cell sectors. For example, the cell associated with base station 114a may be divided into three sectors. Therefore, in one embodiment, base station 114a may include three transceivers, i.e., one transceiver for each sector of the cell. In embodiments, base station 114a may employ multiple-input multiple-output (MIMO) technology and may use multiple transceivers for each sector of the cell. For example, beamforming can be used to transmit and / or receive signals in a desired spatial direction.

[0016] Through air interface 116, base stations 114a and 114b can communicate with one or more WTRUs among WTRUs 102a, 102b, 102c, and 102d. Air interface 116 can be any suitable wireless communication link (e.g., radio frequency (RF), microwave, centimeter wave, micrometer wave, infrared (IR), ultraviolet (UV), visible light, etc.). Air interface 116 can be established using any suitable radio access technology (RAT).

[0017] More specifically, as described above, the communication system 100 can be a multiple access system and can employ one or more channel access schemes, such as CDMA, TDMA, FDMA, OFDMA, SC-FDMA, etc. For example, base stations 114a and WTRUs 102a, 102b, and 102c in RAN 104 / 113 can implement radio technologies such as Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access (UTRA), which can use Wideband CDMA (WCDMA) to establish air interfaces 115 / 116 / 117. WCDMA may include communication protocols such as High-Speed ​​Packet Access (HSPA) and / or Evolved HSPA (HSPA+). HSPA may include High-Speed ​​Downlink (DL) Packet Access (HSDPA) and / or High-Speed ​​UL Packet Access (HSUPA).

[0018] In the embodiment, base station 114a and WTRUs 102a, 102b, 102c can implement radio technologies such as evolved UMTS terrestrial radio access (E-UTRA), which can use Long Term Evolution (LTE) and / or LTE-Advanced (LTE-A) and / or LTE-Advanced Pro (LTE-A Pro) to establish air interface 116.

[0019] In the embodiment, base station 114a and WTRUs 102a, 102b, 102c can implement radio technologies such as NR radio access, which can use New Radio (NR) to establish air interface 116.

[0020] In the embodiments, base station 114a and WTRUs 102a, 102b, and 102c can implement multiple radio access technologies. For example, using, for instance, the dual connectivity (DC) principle, base station 114a and WTRUs 102a, 102b, and 102c can implement both LTE radio access and NR radio access. Therefore, the air interface used by WTRUs 102a, 102b, and 102c can be characterized by multiple types of radio access technologies and / or transmissions sent to / from multiple types of base stations (e.g., eNBs and gNBs).

[0021] In other embodiments, base station 114a and WTRUs 102a, 102b, 102c may implement radio technologies such as IEEE 802.11 (i.e., WiFi), IEEE 802.16 (i.e., WiMAX), CDMA2000, CDMA2000 1X, CDMA2000 EV-DO, transitional standard 2000 (IS-2000), transitional standard 95 (IS-95), transitional standard 856 (IS-856), Global System for Mobile Communications (GSM), Enhanced Data Rate GSM Evolution (EDGE), GSMEDGE (GERAN), etc.

[0022] Figure 1A Base station 114b can be, for example, a wireless router, a home NodeB, a home eNodeB, or an access point, and can use any suitable RAT to facilitate wireless connectivity in a local area, such as commercial locations, homes, vehicles, campuses, industrial facilities, air corridors (e.g., used by drones), roads, etc. In one embodiment, base station 114b and WTRUs 102c, 102d can implement radio technologies (such as IEEE 802.11) to establish a wireless local area network (WLAN). In another embodiment, base station 114b and WTRUs 102c, 102d can implement radio technologies (such as IEEE 802.15) to establish a wireless personal area network (WPAN). In yet another embodiment, base station 114b and WTRUs 102c, 102d can use cellular-based RATs (e.g., WCDMA, CDMA2000, GSM, LTE, LTE-A, LTE-APro, NR, etc.) to establish picocells or femtocells. Figure 1A As shown, base station 114b can have a direct connection to Internet 110. Therefore, base station 114b may not need to access Internet 110 via CN 106 / 115.

[0023] RAN 104 / 113 can communicate with CN 106 / 115, which can be any type of network configured to provide voice, data, application, and / or Voice over Internet Protocol (VoIP) services to one or more WTRUs of WTRUs 102a, 102b, 102c, and 102d. Data may have varying Quality of Service (QoS) requirements, such as different throughput requirements, latency requirements, fault tolerance requirements, reliability requirements, data throughput requirements, mobility requirements, etc. CN106 / 115 can provide call control, billing services, mobile location-based services, prepaid telephony, internet connectivity, video distribution, etc., and / or perform advanced security functions such as user authentication. Although not explicitly stated... Figure 1A As shown, but will be understood, RAN 104 / 113 and / or CN 106 / 115 may communicate directly or indirectly with other RANs that use the same RAT as or a different RAT than RAN 104 / 113. For example, in addition to being connected to RAN 104 / 113, which may use NR radio technology, CN 106 / 115 may also communicate with another RAN (not shown) that uses GSM, UMTS, CDMA2000, WiMAX, E-UTRA, or WiFi radio technology.

[0024] CN 106 / 115 can also serve as a gateway for WTRU 102a, 102b, 102c, 102d to access PSTN 108, the Internet 110, and / or other networks 112. PSTN 108 may include a circuit-switched telephone network providing Common Old-Style Telephone Service (POTS). Internet 110 may include a globally interconnected computer network and apparatus system using common communication protocols such as Transmission Control Protocol (TCP), User Datagram Protocol (UDP), and / or Internet Protocol (IP) from the TCP / IP Internet Protocol suite. Network 112 may include wired and / or wireless communication networks owned and / or operated by other service providers. For example, network 112 may include another CN connected to one or more RANs, which may use the same RAT as RAN 104 / 113 or a different RAT.

[0025] Some or all of the WTRUs 102a, 102b, 102c, and 102d in communication system 100 may include multi-mode capability (e.g., WTRUs 102a, 102b, 102c, and 102d may include multiple transceivers for communicating with different wireless networks on different wireless links). For example, Figure 1A The WTRU 102c shown can be configured to communicate with base station 114a, which can employ cellular-based radio technology, and with base station 114b, which can employ IEEE 802 radio technology.

[0026] Figure 1B This is a system diagram illustrating an exemplary WTRU 102. (See diagram below.) Figure 1B As shown, WTRU 102 may include a processor 118, a transceiver 120, a transmit / receive element 122, a speaker / microphone 124, a keypad 126, a display / touchpad 128, non-removable memory 130, removable memory 132, a power supply 134, a Global Positioning System (GPS) chipset 136, and / or other peripheral devices 138, etc. It will be understood that WTRU 102 may include any sub-combination of the foregoing elements while remaining consistent with the embodiments.

[0027] Processor 118 may be a general-purpose processor, a special-purpose processor, a conventional processor, a digital signal processor (DSP), multiple microprocessors, one or more microprocessors associated with a DSP core, a controller, a microcontroller, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) circuit, any other type of integrated circuit (IC), a state machine, etc. Processor 118 may perform signal decoding, data processing, power control, input / output processing, and / or any other functions that enable WTRU 102 to operate in a wireless environment. Processor 118 may be coupled to transceiver 120, and transceiver 120 may be coupled to transmitting / receiving element 122. Although... Figure 1B While the processor 118 and transceiver 120 are depicted as separate components, it will be understood that the processor 118 and transceiver 120 may be integrated together in an electronic package or chip.

[0028] Transmitting / receiving element 122 may be configured to transmit signals to or receive signals from a base station (e.g., base station 114a) via air interface 116. For example, in one embodiment, transmitting / receiving element 122 may be an antenna configured to transmit and / or receive RF signals. In embodiments, for example, transmitting / receiving element 122 may be a transmitter / detector configured to transmit and / or receive IR, UV, or visible light signals. In another embodiment, transmitting / receiving element 122 may be configured to transmit and / or receive both RF and optical signals. It will be understood that transmitting / receiving element 122 may be configured to transmit and / or receive any combination of wireless signals.

[0029] Although the transmitting / receiving element 122 is in Figure 1B While depicted as a single element, WTRU 102 may include any number of transmit / receive elements 122. More specifically, WTRU 102 may employ MIMO technology. Thus, in one embodiment, WTRU 102 may include two or more transmit / receive elements 122 (e.g., multiple antennas) for transmitting and receiving wireless signals via air interface 116.

[0030] Transceiver 120 can be configured to modulate signals to be transmitted by transmitting / receiving element 122 and demodulate signals received by transmitting / receiving element 122. As described above, WTRU 102 may have multi-mode capability. Therefore, transceiver 120 may include multiple transceivers to enable WTRU 102 to communicate over various RATs (such as, for example, NR and IEEE 802.11).

[0031] The processor 118 of WTRU 102 can be coupled to a speaker / microphone 124, a keypad 126, and / or a display / touchpad 128 (e.g., a liquid crystal display (LCD) unit or an organic light-emitting diode (OLED) display unit), and can receive user input data from the speaker / microphone 124, keypad 126, and / or display / touchpad 128. The processor 118 can also output user data to the speaker / microphone 124, keypad 126, and / or display / touchpad 128. Furthermore, the processor 118 can access information from any type of suitable memory (such as non-removable memory 130 and / or removable memory 132), and store data in any type of suitable memory. Non-removable memory 130 may include random access memory (RAM), read-only memory (ROM), a hard disk, or any other type of memory storage device. Removable memory 132 may include a subscriber identity module (SIM) card, a memory stick, a secure digital storage (SD) card, etc. In other embodiments, the processor 118 may access information from memory that is not physically located on the WTRU 102 (such as on a server or home computer (not shown)) and store the data in the memory.

[0032] The processor 118 may receive power from the power supply 134 and may be configured to distribute and / or control power to other components in the WTRU 102. The power supply 134 may be any suitable device for powering the WTRU 102. For example, the power supply 134 may include one or more dry cell battery packs (e.g., nickel-cadmium (NiCd), nickel-zinc (NiZn), nickel metal hydride (NiMH), lithium-ion (Li-ion), etc.), solar cells, fuel cells, etc.

[0033] Processor 118 may also be coupled to GPS chipset 136, which may be configured to provide location information (e.g., longitude and latitude) about the current location of WTRU 102. In addition to or alternative to information from GPS chipset 136, WTRU 102 may receive location information via air interface 116 from base stations (e.g., base stations 114a, 114b) and / or determine its location based on the timing of signals received from two or more nearby base stations. It will be understood that WTRU 102 may acquire location information using any suitable location determination method while remaining consistent with the embodiments.

[0034] The processor 118 may be further coupled to other peripheral devices 138, which may include one or more software and / or hardware modules providing additional features, functions, and / or wired or wireless connectivity. For example, peripheral devices 138 may include accelerometers, electronic compasses, satellite transceivers, digital cameras (for photos and / or videos), Universal Serial Bus (USB) ports, vibration devices, television transceivers, hands-free headsets, Bluetooth® modules, frequency modulation (FM) radio units, digital music players, media players, video game player modules, internet browsers, virtual reality and / or augmented reality (VR / AR) devices, activity trackers, etc. Peripheral devices 138 may include one or more sensors, which may be one or more of the following: gyroscopes, accelerometers, Hall effect sensors, magnetometers, orientation sensors, proximity sensors, temperature sensors, time sensors, geolocation sensors, altimeters, light sensors, touch sensors, magnetometers, barometers, gesture sensors, biometric sensors, and / or humidity sensors.

[0035] WTRU 102 may include a full-duplex radio device for which the transmission and reception of some or all of the signals (e.g., associated with a specific subframe of both UL (e.g., for transmission) and downlink (e.g., for reception)) may be concurrent and / or simultaneous. The full-duplex radio device may include an interference management unit 139 to reduce and / or substantially eliminate self-interference through hardware (e.g., a choke) or through signal processing (e.g., a separate processor (not shown) or through processor 118). In an embodiment, WTRU 102 may include a half-duplex radio device for which the transmission and reception of some or all of the signals (e.g., associated with a specific subframe of either UL (e.g., for transmission) or downlink (e.g., for reception)) may be concurrent and / or simultaneous.

[0036] Figure 1C This is a system diagram illustrating RAN 104 and CN 106 according to an embodiment. As described above, RAN 104 can communicate with WTRUs 102a, 102b, and 102c via air interface 116 using E-UTRA radio technology. RAN 104 can also communicate with CN 106.

[0037] RAN 104 may include eNode-Bs 160a, 160b, and 160c, but it will be understood that RAN 104 may include any number of eNode-Bs while remaining consistent with the embodiments. Each eNode-B 160a, 160b, and 160c may include one or more transceivers for communicating with WTRUs 102a, 102b, and 102c via air interface 116. In one embodiment, eNode-Bs 160a, 160b, and 160c may implement MIMO technology. Therefore, eNode-B 160a may, for example, use multiple antennas to transmit and / or receive radio signals from WTRU 102a.

[0038] Each of the eNode-B 160a, 160b, and 160c can be associated with a specific cell (not shown) and can be configured to handle radio resource management decisions, handover decisions, and user scheduling in the UL and / or DL, etc. Figure 1C As shown, eNode-B 160a, 160b, and 160c can communicate with each other via the X2 interface.

[0039] Figure 1C CN 106 shown may include a Mobility Management Entity (MME) 162, a Serving Gateway (SGW) 164, and a Packet Data Network (PDN) Gateway (or PGW) 166. Although each of the foregoing elements is described as part of CN 106, it will be understood that any of these elements may be owned and / or operated by an entity other than a CN operator.

[0040] The MME 162 can be connected to each of the eNode-Bs 162a, 162b, and 162c in RAN 104 via the S1 interface and can be used as a control node. For example, the MME 162 can be responsible for authenticating users of WTRUs 102a, 102b, and 102c, activating / deactivating bearers, selecting a specific serving gateway during the initial attachment of WTRUs 102a, 102b, and 102c, etc. The MME 162 can provide control plane functions for handover between RAN 104 and other RANs (not shown) employing other radio technologies such as GSM and / or WCDMA.

[0041] The SGW 164 can be connected to each of the eNodeBs 160a, 160b, and 160c in RAN 104 via the S1 interface. The SGW 164 can typically route and forward user packets to / from WTRUs 102a, 102b, and 102c. The SGW 164 can also perform other functions, such as anchoring the user plane during inter-eNodeB handover, triggering paging when DL data is available for WTRUs 102a, 102b, and 102c, and managing and storing the context of WTRUs 102a, 102b, and 102c.

[0042] SGW 164 can be connected to PGW 166, which provides WTRU 102a, 102b, 102c with access to packet-switched networks (such as Internet 110) to facilitate communication between WTRU 102a, 102b, 102c and IP-enabled devices.

[0043] CN 106 can facilitate communication with other networks. For example, CN 106 can provide WTRUs 102a, 102b, and 102c with access to a circuit-switched network (such as PSTN 108) to facilitate communication between WTRUs 102a, 102b, and 102c and conventional landline communication devices. For example, CN 106 may include an IP gateway (e.g., an IP Multimedia Subsystem (IMS) server), or can communicate with said IP gateway, which serves as an interface between CN 106 and PSTN 108. Furthermore, CN 106 can provide WTRUs 102a, 102b, and 102c with access to other networks 112, which may include other wired and / or wireless networks owned and / or operated by other service providers.

[0044] Although WTRU is Figure 1A-1D While described as a wireless terminal, it is conceivable that in some representative embodiments, such a terminal may (e.g., temporarily or permanently) use a wired communication interface with a communication network.

[0045] In a representative embodiment, the other network 112 may be a WLAN.

[0046] A WLAN in an Infrastructure Basic Services Set (BSS) mode may have an Access Point (AP) for the BSS and one or more Stations (STAs) associated with the AP. The AP may access a Distribution System (DS) or another type of wired / wireless network, or have an interface with a Distribution System (DS) or another type of wired / wireless network that carries traffic into and / or out of the BSS. Traffic originating outside the BSS destined for a STA can be delivered to the AP via the AP. Traffic originating from a STA destined for a destination outside the BSS can be sent to the AP for delivery to the appropriate destination. Traffic between STAs within the BSS can be sent via the AP, for example, where a source STA can send traffic to the AP and the AP can deliver traffic to the destination STA. Traffic between STAs within the BSS may be considered and / or referred to as peer-to-peer traffic. Peer-to-peer traffic can be sent (e.g., directly between the source and destination STAs) using Direct Link Establishment (DLS). In some representative embodiments, the DLS may use 802.11e DLS or 802.11z Tunneled DLS (TDLS). WLANs using Standalone BSS (IBSS) mode may not have access points (APs), and STAs within the IBSS or using the IBSS (e.g., all STAs within a STA) can communicate directly with each other. IBSS communication mode may sometimes be referred to herein as "ad-hoc" ("self-organizing") communication mode.

[0047] When operating in 802.11ac infrastructure mode or a similar mode, the AP can transmit beacons on a fixed channel, such as a primary channel. The primary channel can be of fixed width (e.g., a 20 MHz bandwidth) or dynamically set via signaling. The primary channel can be the operating channel of the BSS and can be used by the STA to establish a connection with the AP. In some representative embodiments, such as in an 802.11 system, carrier-sense multiple access (CSMA / CA) with collision avoidance can be implemented. For CSMA / CA, each STA (including the AP) can sense the primary channel. If the primary channel is sensed / detected and / or determined to be busy by a particular STA, that particular STA can back off. In a given BSS, one STA (e.g., only one station) can transmit at any given time.

[0048] High-throughput (HT) STAs can communicate using a 40 MHz wide channel, for example, by combining a primary 20 MHz channel with adjacent or non-adjacent 20 MHz channels to form a 40 MHz wide channel.

[0049] Very High Throughput (VHT) STAs can support channels with widths of 20 MHz, 40 MHz, 80 MHz, and / or 160 MHz. 40 MHz and / or 80 MHz channels can be formed by combining adjacent 20 MHz channels. A 160 MHz channel can be formed by combining eight adjacent 20 MHz channels, or by combining two non-adjacent 80 MHz channels (this can be referred to as an 80+80 configuration). For the 80+80 configuration, after channel coding, data is delivered via a segment resolver, which divides the data into two streams. Inverse Fast Fourier Transform (IFFT) processing and time-domain processing can be performed separately on each stream. The streams can be mapped onto two 80 MHz channels, and the data can be transmitted by the transmitting STA. At the receiver of the receiving STA, the above operations for the 80+80 configuration can be reversed, and the combined data can be sent to Media Access Control (MAC).

[0050] Operating modes below 1 GHz are supported by 802.11af and 802.11ah. In 802.11af and 802.11ah, the channel operating bandwidth and carrier are reduced compared to those used in 802.11n and 802.11ac. 802.11af supports 5 MHz, 10 MHz, and 20 MHz bandwidths in the TV white space (TVWS) spectrum, and 802.11ah supports 1 MHz, 2 MHz, 4 MHz, 8 MHz, and 16 MHz bandwidths using non-TVWS spectrum. According to a representative embodiment, 802.11ah may support instrument-type control / machine-type communication, such as MTC devices in macro coverage areas. MTC devices may have certain capabilities, such as limited capabilities, including support (e.g., only support) certain and / or limited bandwidths. MTC devices may include batteries with a battery life exceeding a threshold (e.g., to maintain a very long battery life).

[0051] WLAN systems that support multiple channels and channel bandwidths (such as 802.11n, 802.11ac, 802.11af, and 802.11ah) include a channel that can be designated as the primary channel. The primary channel may have a bandwidth equal to the maximum common operating bandwidth supported by all STAs in the BSS. The bandwidth of the primary channel may be set and / or limited by the STA that supports the minimum bandwidth operating mode among all STAs operating in the BSS. In the example of 802.11ah, for STAs that support (e.g., only support) the 1MHz mode (e.g., MTC type devices), the primary channel may be 1 MHz wide even if the AP and other STAs in the BSS support 2 MHz, 4 MHz, 8 MHz, 16 MHz, and / or other channel bandwidth operating modes. Carrier sensing and / or Network Allocation Vector (NAV) settings may depend on the status of the primary channel. If, for example, the primary channel is busy due to a STA (that only supports the 1 MHz operating mode) transmitting to the AP, the entire available band may be considered busy even if most of the band remains idle and may be available.

[0052] In the United States, the available frequency bands for 802.11ah are from 902 MHz to 928 MHz. In South Korea, the available frequency bands are from 917.5 MHz to 923.5 MHz. In Japan, the available frequency bands are from 916.5 MHz to 927.5 MHz. Depending on the country code, the total bandwidth available for 802.11ah is 6 MHz to 26 MHz.

[0053] Figure 1D This is a system diagram illustrating RAN 113 and CN 115 according to an embodiment. As described above, RAN 113 can communicate with WTRUs 102a, 102b, and 102c via air interface 116 using NR radio technology. RAN 113 can also communicate with CN 115.

[0054] RAN 113 may include gNBs 180a, 180b, and 180c, but it will be understood that RAN 113 may include any number of gNBs while remaining consistent with the embodiments. gNBs 180a, 180b, and 180c may each include one or more transceivers for communicating with WTRUs 102a, 102b, and 102c via air interface 116. In one embodiment, gNBs 180a, 180b, and 180c may implement MIMO technology. For example, gNBs 180a and 180b may use beamforming to transmit signals to and / or receive signals from WTRU 102a. Therefore, gNB 180a may, for example, use multiple antennas to transmit radio signals to and / or receive radio signals from WTRU 102a. In an embodiment, gNBs 180a, 180b, and 180c may implement carrier aggregation technology. For example, gNB 180a can transmit multiple component carriers to WTRU 102a (not shown). A subset of these component carriers may be on unlicensed spectrum, while the remaining component carriers may be on licensed spectrum. In embodiments, gNBs 180a, 180b, and 180c can implement Coordinated Multipoint (CoMP) technology. For example, WTRU 102a can receive coordinated transmissions from gNBs 180a and 180b (and / or gNB 180c).

[0055] Using transmissions associated with scalable digital theory, WTRU 102a, 102b, and 102c can communicate with gNB180a, 180b, and 180c. For example, OFDM symbol spacing and / or OFDM subcarrier spacing can vary for different transmissions, different cells, and / or different portions of the radio transmission spectrum. Using various or scalable length subframes or transmit time intervals (TTIs) (e.g., containing a varying number of OFDM symbols and / or an absolute time of continuously varying length), WTRU 102a, 102b, and 102c can communicate with gNB180a, 180b, and 180c.

[0056] gNBs 180a, 180b, and 180c can be configured to communicate with WTRUs 102a, 102b, and 102c in standalone and / or non-standalone configurations. In standalone configuration, WTRUs 102a, 102b, and 102c can communicate with gNBs 180a, 180b, and 180c without also accessing other RANs (e.g., eNode-Bs 160a, 160b, and 160c). In standalone configuration, WTRUs 102a, 102b, and 102c can use one or more gNBs from gNBs 180a, 180b, and 180c as mobility anchors. In standalone configuration, WTRUs 102a, 102b, and 102c can communicate with gNBs 180a, 180b, and 180c using signals in unlicensed frequency bands. In a non-standalone configuration, WTRUs 102a, 102b, and 102c can communicate with / connect to gNBs 180a, 180b, and 180c, and also communicate with / connect to another RAN, such as eNode-Bs 160a, 160b, and 160c. For example, WTRUs 102a, 102b, and 102c can implement DC principles to communicate substantially simultaneously with one or more gNBs 180a, 180b, and 180c and one or more eNode-Bs 160a, 160b, and 160c. In a non-standalone configuration, eNode-B160a, 160b, and 160c can be used as mobility anchors for WTRU 102a, 102b, and 102c, and gNB180a, 180b, and 180c can provide additional coverage and / or throughput for serving WTRU 102a, 102b, and 102c.

[0057] Each gNB in ​​gNBs 180a, 180b, and 180c can be associated with a specific cell (not shown) and can be configured to handle radio resource management decisions, handover decisions, user scheduling in UL and / or DL, network slicing support, dual connectivity, interoperability between NR and E-UTRA, routing of user plane data to User Plane Functions (UPFs) 184a and 184b, routing of control plane information to Access and Mobility Management Functions (AMFs) 182a and 182b, etc. Figure 1D As shown, gNB180a, 180b, and 180c can communicate with each other via the Xn interface.

[0058] Figure 1DThe CN 115 shown may include at least one AMF 182a, 182b, at least one UPF 184a, 184b, at least one Session Management Function (SMF) 183a, 183b, and possibly a Data Network (DN) 185a, 185b. Although each of the foregoing elements is described as part of CN 115, it will be understood that any of these elements may be owned and / or operated by an entity other than a CN operator.

[0059] AMF 182a and 182b can be connected to one or more gNBs (gNBs) 180a, 180b, and 180c in RAN 113 via the N2 interface and can be used as control nodes. For example, AMF 182a and 182b can be responsible for authenticating users of WTRU 102a, 102b, and 102c, supporting network slices (e.g., handling different PDU sessions with different requirements), selecting specific SMF 183a and 183b, managing registration areas, terminating NAS signaling, mobility management, etc. Network slices can be used by AMF 182a and 182b to customize CN support for WTRU 102a, 102b, and 102c based on the service types used by WTRU 102a, 102b, and 102c. For example, different network slices can be established for different use cases, such as services relying on Ultra Reliable Low Latency (URLLC) access, services relying on Enhanced Massive Mobile Broadband (eMBB) access, and services for Machine Type Communication (MTC) access. AMF 162 can provide control plane functions for switching between RAN 113 and other RANs (not shown) that employ other radio technologies, such as LTE, LTE-A, LTE-A Pro, and / or non-3GPP access technologies such as WiFi.

[0060] SMF 183a and 183b can be connected to AMF 182a and 182b in CN 115 via the N11 interface. SMF 183a and 183b can also be connected to UPF 184a and 184b in CN 115 via the N4 interface. SMF 183a and 183b can select and control UPF 184a and 184b, and configure the routing of services through UPF 184a and 184b. SMF 183a and 183b can perform other functions, such as managing and allocating WTRU IP addresses, managing PDU sessions, controlling policy enforcement and QoS, and providing downlink data notifications. PDU session types can be IP-based, non-IP-based, or Ethernet-based.

[0061] UPF 184a and 184b can be connected via the N3 interface to one or more gNBs (gNBs) 180a, 180b, and 180c in RAN 113. This provides WTRU 102a, 102b, and 102c with access to packet-switched networks (such as the Internet 110) to facilitate communication between WTRU 102a, 102b, 102c and IP-enabled devices. UPF 184 and 184b can perform other functions such as routing and forwarding packets, enforcing user plane policies, supporting multi-homed PDU sessions, handling user plane QoS, buffering downlink packets, and providing mobility anchoring.

[0062] CN 115 can facilitate communication with other networks. For example, CN 115 may include an IP gateway (e.g., an IP Multimedia Subsystem (IMS) server), or may communicate with said IP gateway, which serves as an interface between CN 115 and PSTN 108. Furthermore, CN 115 can provide WTRUs 102a, 102b, and 102c with access to other networks 112, which may include other wired and / or wireless networks owned and / or operated by other service providers. In one embodiment, WTRUs 102a, 102b, and 102c can be connected to DNs 185a and 185b via UPFs 184a and 184b, through their N3 interfaces and the N6 interface between UPFs 184a and 184b and local data networks (DNs) 185a and 185b.

[0063] Considering Figure 1A-1D and Figure 1A-1D As described herein, one or all of the functions described for one or more of the following devices—WTRU 102a-d, base station 114a-b, eNode-B 160a-c, MME 162, SGW 164, PGW 166, gNB180a-c, AMF 182a-ab, UPF184a-b, SMF183a-b, DN185a-b, and / or any other device described herein—may be performed by one or more emulation devices (not shown). An emulation device can be one or more devices configured to emulate one or more of the functions described herein. For example, an emulation device may be used to test other devices and / or simulate network and / or WTRU functions.

[0064] The simulation device may be designed to perform one or more tests on other devices in a laboratory environment and / or in a carrier network environment. For example, while being fully or partially implemented and / or deployed as part of a wired and / or wireless communication network, the one or more simulation devices may perform one or more of the functions to test other devices within the communication network. While being temporarily implemented / deployed as part of a wired and / or wireless communication network, the one or more simulation devices may perform one or more of the functions. The simulation device may be directly coupled to another device for testing purposes and / or may use over-the-air wireless communication to perform the tests.

[0065] When not implemented / deployed as part of a wired and / or wireless communication network, the one or more simulation devices may perform the one or more functions (including all functions). For example, the simulation devices may be used in test scenarios in test laboratories and / or non-deployed (e.g., test) wired and / or wireless communication networks to perform testing of one or more components. The one or more simulation devices may be test equipment. Wireless communication via direct RF coupling and / or through an RF circuit system (e.g., the RF circuit system may include one or more antennas) may be used by the simulation devices to transmit and / or receive data.

[0066] This document describes an implementation of Vertical Federated Learning (VFL) initiated by an Application Function (AF). The AF may request the 5G Core (5GC) to provide a recommended sample set for model training, with a requested sample size and / or requested selection criteria. Based on this request, the Network Exposure Function or Network Data Analysis Function (NEF / NWDAF) may collect data from the Network Function (NF) related to the requested selection criteria. Based on the collected data, the NEF / NWDAF may select samples of WTRUs to satisfy the selection criteria. The NEF / NWDAF may respond to the AF with a list of WTRUs as samples (e.g., via a Common Public Subscription Identifier (GPSI), etc.). The AF and 5GC may perform VFL based on the negotiated / selected samples and features.

[0067] Federated learning among multiple Network Data Analysis Functions (NWDAFs) can be specified by 3GPP. 3GPP can specify how NWDAF functions (including model training functions) can utilize federated learning (FL) techniques to train machine learning (ML) models.

[0068] For horizontal federated learning, the NWDAF function enabled for federated learning can register with the Network Repository Function (NRF) to perform the registration. For example, the NWDAF can provide the NRF with its Network Function (NF) profile, information about the analytics services it supports (e.g., one or more analytics IDs), the NWDAF's address information, service area, and / or its federated learning capabilities. This registered information can be used to find appropriate NWDAF functions to join the federated learning for analytics services, leveraging candidate ML models and requested service areas.

[0069] Model filter information can be defined to indicate the conditions when an ML model is requested for the analysis service, as well as one or more objectives of the ML model, such as one or more specific WTRUs, a set of WTRUs, or any WTRU.

[0070] For horizontal federated learning, a Federated Learning (FL) server Network Data Analysis Function (NWDAF) and FL client NWDAFs can be defined. When an analysis service is requested, federated learning can be requested from the FL server NWDAF using the accuracy of the ML model. The FL server NWDAF can discover and select appropriate FL client NWDAFs for a specific analysis service with the requested ML model(s) based on selection criteria (e.g., in a service area, the network function (NF) type of the data, the source from which the NWDAF collects data for local model training, the time period of interest, etc.). The FL server NWDAF can also provide local ML models to the FL client NWDAFs and request the FL client NWDAFs to perform local model training. Each FL client NWDAF can collect its local data, perform local model training using its own data, and report intermediate local ML model information to the FL server NWDAF. The FL server NWDAF can update the global ML model based on the aggregated local ML models and provide an appropriate global ML model for the requested analysis service.

[0071] This paper describes an example of a use case for longitudinal federated learning in 5GC. Longitudinal federated learning can be a federated learning approach where multiple participants perform training on datasets that share the same sample space but differ in their feature spaces. Due to this characteristic, alignment of samples and feature spaces among participating entities may often be required before applying VFL (e.g., as part of VFL initialization). VFL allows joint training to be performed without exposing the original data, where each entity has its own model. Therefore, VFL can be a suitable approach for federated learning when privacy is required among the participants performing training.

[0072] The Virtual Functions (VFLs) between the Application Function (AF) and the Network Function (NF) in 5GC can be used to provide network analysis features in 5GC. To support the use of VFLs for analysis derivation, sample and feature alignment among entities participating in the VFLs may be required (e.g., as part of VFL initialization).

[0073] In one example, 5GC can provide observed service experience analytics as VFL.

[0074] For observed service experience analysis services performed by NWDAF, the AF can provide quality of service and service experience data as input data for each WTRU of interest. The 5GC NF can provide network data at the Quality of Service (QoS) flow level as input data. The Operations, Administration and Maintenance (OAM) system can provide WTRU-level network (NW) data related to the QoS profile.

[0075] However, when the NWDAF or AF initiates a VFL training process for observed service experience analysis, it is desirable not to directly exchange raw data between the NWDAF and the external AF. This could be because the NWDAF is in a Public Land Mobile Network (PLMN) and the AF is outside the PLMN, and user data has high privacy protection requirements.

[0076] Furthermore, NWDAF and AF can have different features with the same sample identifiers for local training, although VFL application between two entities requires alignment of samples and features.

[0077] Furthermore, because VFL's inference is also distributed, the original data does not need to be shared during inference or training. Each entity can use local data to determine the inference, and the outputs can be aggregated to obtain the final result.

[0078] AF or 5GC can trigger VFL for network analytics services (e.g., VFL-based observed service experience analysis). Each entity involved in the VFL can perform local training based on locally collected data. However, AF and 5GC may require alignment of samples and features before performing VFL operations (e.g., local training).

[0079] To improve model training reliability, a representative dataset can be selected for model training. A representative sample set can be selected for network analysis services based on the VFL (Virtual Dataset). Therefore, the question arises regarding how to select an appropriate sample set for the VFL, where the sample set represents the dataset used for model training reliability. The embodiments described herein present a solution to this problem concerning VFL.

[0080] When requesting network analysis services for longitudinal federated learning (e.g., in a sample alignment request), sample alignment can be requested by the consumer NF. When longitudinal federated learning is triggered (e.g., initialized) by 5GC NW (e.g., NWDAF / NEF) or AF, 5GC NF (e.g., NWDAF / NEF) or AF can also request sample alignment.

[0081] When requesting sample alignment for VFL, a request can be made to 5GC for sampling of WTRUs (or any other relevant samples, such as network slices or data networks), which have selection criteria for sampling WTRUs. When requesting WTRU sampling with selection criteria, 5GC can collect data from the relevant NF based on the selection criteria and the requested data collection period, and determine a list of WTRUs that meet the selection criteria as samples.

[0082] Selection criteria for sampling WTRUs may include one or more of the following: requested sample size, requested conditions, requested aspects, and / or requested dataset statistical properties. Selection criteria may instruct the selection of each WTRU in the sample that satisfies the requested conditions. If the data of the WTRUs in the sample satisfy a requested aspect (e.g., a requested statistical distribution), the selection criteria may instruct the selection of one or more (e.g., all) WTRUs in the sample. For example, an AF or NWDAF might want to select WTRUs as samples based on requested aspects such as WTRU interval distance. WTRU interval distance may be the minimum distance between WTRUs in a group. Examples of requested dataset statistical properties may include uniform distribution, having outliers, or not having outliers.

[0083] If a list of WTRUs (e.g., a target WTRU) is provided along with selection criteria, the WTRU list (e.g., the target WTRU) is validated. Validating the WTRU list may include determining whether the data of the target WTRUs for the requested aspects meet the requested statistical properties. The WTRU list may be updated (e.g., a down-selection list of WTRUs, adding other WTRUs not already in the list) to meet the requested statistical properties.

[0084] The request conditions may include NW slice information, application information, data network name (DNN), data type (e.g., IPv4, IPv6, etc.), and location information of interest.

[0085] User consent (e.g., data collection for analysis of requests targeting the VFL model) can also be considered a condition of the request.

[0086] The requested aspects may include one or more of the following: the RAT used (e.g., whether 5G, LTE, or N3GPP is used); the requested QoS range (e.g., the range of average or maximum bit rate, the range of average or maximum latency, and the range of average or maximum packet error rate); the Protocol Data Unit (PDU) session type (e.g., Session and Service Continuity (SSC) mode 1, 2, or 3); the use of a remote data server or edge server; a list of Data Network Access Identifiers (DNAIs); and / or the service area of ​​interest.

[0087] This document describes an example of a procedure for an AF-initiated VFL operation used in NW analysis. Figure 2 A flowchart illustrating an exemplary procedure for a longitudinal federated learning (VFL) operation initiated by an application function (AF) for network (NW) analysis at point 200 is provided. At point 202, AF 230 can be triggered to perform a network analysis service using longitudinal federated learning. Based on the application, the characteristics of the target user for the machine learning request can be determined. For example, applications utilizing adaptive data rates (such as machine learning-based voice calls and / or video streaming services) will require the user to be in a range of channel conditions covering both low and high data rates during the machine learning process. Based on the characteristics of the target user for the machine learning request, selection criteria for the sampling WTRU can be determined.

[0088] Once AF 230 decides to use VFL for network analysis services with 5GC, AF 230 can send a service request for VFL initialization or alignment (e.g., at 204). The service request may include information about the requested analysis service (e.g., alignment information), candidate ML models for the requested analysis service, and a list of features of the input data to be used in the requested ML models for federated learning. AF 230 may include information about the requested service area and / or the time period for data collection. AF 230 may (e.g., implicitly) request 5GC to provide a recommended sample set of WTRUs by including selection criteria for sampling WTRUs (e.g., in the alignment information request). Selection criteria for sampling WTRUs may include the requested sample size, requested conditions, requested aspects, and / or requested dataset statistics. Requested conditions may include an indication of which requested conditions each WTRU in the sample should satisfy (e.g., requested network slice information, region of interest, etc.). Requested aspects may include an indication of which data from the sampled WTRUs satisfy the requested statistical properties in that aspect (e.g., average data rate range, location within the region of interest, etc.). Examples of statistical properties of the requested dataset may include uniform distribution, having outliers, or not having outliers.

[0089] AF 230 can provide a target WTRU list with selection criteria (e.g., in a request for alignment information). If the target WTRU list and selection criteria are received together, NWDAF 234 / NEF 232 can verify whether the sample WTRUs meet the selection criteria, reduce the WTRU list from the sample WTRUs to meet the selection criteria, and / or provide a list of additional WTRUs that are not part of the initial list to meet the selection criteria. For example, an additional WTRU list can be selected when a determined sample size needs to be increased to meet a requirement, or when additional WTRUs need to be provided so that the dataset distribution meets the requested statistical distribution. The additional WTRUs may not initially be in the target WTRU list. The dataset distribution may have WTRUs as samples.

[0090] At 206, after receiving a service request for VFL initialization, the Network Exposure Function (NEF) 232 may (e.g., as part of an alignment procedure) query the Network Repository Function (NRF) to discover the Network Data Analysis Function (NWDAF) 234 to perform federated learning for the requested analysis service (e.g., if the requested service area information is included in the request from AF 230, then federated learning is performed at the requested service area). Alternatively, NEF 232 may send a request for VFL initialization to a selected NWDAF 234. NEF 232 may use a feature set provided by the AF to deduce possible samples to be aligned. The request may include previously received information.

[0091] At point 208, upon receiving a request for VFL initialization (e.g., as part of an alignment procedure), NWDAF 234 may collect data for the requested analytics service from relevant 5GC NFs (e.g., Unified Data Management (UDM) 238, AMF 236, SMF 240, UPF 242, etc.) and check the availability of the requested features. When NWDAF 234 receives selection criteria for sampling WTRUs, NWDAF 234 may construct a candidate WTRU list and collect data for the candidate WTRUs. The candidate WTRU list may include WTRUs that meet the conditions requested in the selection criteria. The collected data for the candidate WTRUs may be collected from relevant 5GC NFs (e.g., UDM 238, AMF 236, SMF 240, UPF 242, etc.), which are related to the requested features and aspects in the selection criteria.

[0092] For example, (as part of an alignment procedure), NWDAF 234 may contact Access and Mobility Management Function (AMF) 236 (e.g., at 210) serving the area of ​​interest to construct a list of WTRUs in that area. NWDAF 234 may (as part of an alignment procedure) query the Network Slice Selection Function (NSSF) or UDM 238 (e.g., at 212) to query network slice information for WTRUs in the list, collect data on features related to the NW slice information, and / or verify whether each WTRU in the list meets the conditions for network slicing (e.g., if the NW slice information is included in the requested conditions). NWDAF 234 may (as part of an alignment procedure) contact SMF 240 (e.g., at 214) to query PDU session-related information, collect data on features related to the PDU session, and / or verify whether UEs in the list meet any PDU session-related conditions (e.g., DNN, SSC mode, etc.). NWDAF234 may (e.g., as part of an alignment procedure) contact the UPF and Radio Access Network (RAN) 242 (e.g., at 216) for user plane-related information of WTRUs in the list, collect data on user plane-related characteristics, and verify whether the WTRUs in the list meet any user plane-related conditions (e.g., error rate, data rate, etc.).

[0093] The NWDAF 234 can contact the Gateway Mobility Location Center (GMLC) to query the location information for each WTRU in the list.

[0094] If a requested time period was previously included, data collection can be performed within that requested time period.

[0095] At 218, (e.g., as part of an alignment procedure), based on the collected data, NWDAF 234 may select a sample of WTRUs that meet the selection criteria. In the example, each WTRU meets the requested conditions, and the data for the WTRUs regarding the requested aspects meet the requested statistical distribution. NWDAF 234 may select at least as many WTRUs as the requested sample size. NWDAF 234 may send a response to NEF for VFL initialization. The response may include a list of available ML models and / or available features for the requested analysis ID. If all requested features are supported, the list of available features may be omitted. The response may also include a list of recommended WTRUs as samples.

[0096] If the size of the WTRU sample list is smaller than the requested sample size, a service request from AF 230 may be rejected with a rejection code indicating that the requested sample size has not been met. In other embodiments, when the size of the WTRU sample list is smaller than the requested sample size, the list is reported to AF 230, and AF 230 may determine whether to continue VFL operations using these samples.

[0097] In another embodiment, NEF 232 may execute a prior procedure. That is, (e.g., as part of an alignment procedure), at 218, NEF 232 may select WTRU samples that meet the selection criteria. NEF 232 may select at least as many samples as requested.

[0098] At 220, NEF 232 may send a service response for VFL initialization to AF 230 in response to a previously received request. The service response may include an indication of whether the requested ML model for the requested analytics service is available, a list of available ML models for the requested analytics service, a list of available features (e.g., parameters, parameters for VFL alignment) for the available ML models as input data (this list may be omitted if all requested features are supported), and / or a recommended list of WTRUs as samples (e.g., as part of the parameters for VFL alignment). When including a list of WTRUs as samples, identifiers of WTRUs that can be understood by AF 230 (e.g., through the use of GPSI conversion services, etc.) may be included. For each available ML model, recommended hyperparameters for ML may be added.

[0099] Optionally, at 222, AF 230 may send an acknowledgment for VFL initialization. The acknowledgment may include the list of available features and the list of WTRUs as samples, if they are updated relative to previously received lists. The response may include the selected ML model and the hyperparameters for selecting the ML model. For example, AF 230 may narrow down the list of recommended WTRUs as samples based on the availability of data for WTRU features used in AF 230 for model training.

[0100] At point 224, AF 230 can perform longitudinal federated learning against the analytics service, based on available or selected ML models, hyperparameters for selecting the ML model, and / or previously received and updated lists of available features and WTRU samples, together with 5GC. In the example, AF 230 can send an initiation message for VFL to NEF 232, including the list of available features and / or the list of WTRUs for VFL training. After performing federated learning, AF 230, NEF 232, and 5GC can share the results of their local training. If no further requests for data collection are received, previously collected data can be used for local training within 5GC.

[0101] Based on the shared results of local training, AF 230 can determine an ML model using training results from both AF 230 and 5GC. If the performance of the ML model based on the training results (e.g., the compliance level of the ML model's output) is insufficient, AF 230 can request an updated VFL from 5GC, which has an updated ML model (e.g., using the determined ML model), updated features, and / or updated samples. Based on the request for the updated VFL from AF 230, 5GC can perform local training of the requested updated ML model and share the results of the local training. This procedure can be repeated until an ML model with performance that meets the requested requirements is developed.

[0102] In another embodiment, AF and 5GC can negotiate the time period and / or sampling rate for data collection targeting negotiated features and WTRU samples. Alternatively, AF and 5GC can use the newly collected data for local training from AF and 5GC.

[0103] This document describes an example of a procedure for VFL operations initiated by a Consumer Network Function (NF) for NW analysis. The consumer NF may request VFL operations for NW analysis. In this case, the consumer NF can provide selection criteria for sampling the WTRU for the requested VFL for NW analysis.

[0104] When a service request for VFL operation for NW analysis is received from the consumer NF, the NWDAF / NEF can execute any of the procedures described previously to determine the ML model, features, and sample WTRU based on the received selection criteria.

[0105] Based on the selected sample WTRUs, features, and ML model, NWDAF / NEF can initiate a VFL operation with AF for NW analysis. NWDAF / NEF may include a list of candidate ML models, features, and / or a list of sample WTRUs. AF can respond to indicate whether AF can join the VFL and whether AF can provide an updated list of sample WTRUs and features based on the availability of data for the requested features and samples. For example, when some data for a WTRU of a requested feature is unavailable to AF, AF can discard that WTRU and narrow down the sample WTRUs based on the availability of data for the requested feature.

Claims

1. A network node, comprising: Processor and memory, wherein the processor and memory are configured as follows: Receive a request for alignment information for longitudinal federated learning (VFL), wherein the request for alignment information includes sample selection criteria; Based on the sample selection criteria included in the alignment information, an alignment procedure with network functions is executed; as well as Send a response to the request for alignment information, the response indicating parameters for VFL alignment, wherein the parameters for VFL alignment include a recommended list of wireless transmit / receive units (WTRUs) for VFL training.

2. The network node of claim 1, wherein the processor and memory are configured to receive the request for alignment information from the application function (AF).

3. The network node of claim 1, wherein the parameters for VFL alignment indicate the analysis service and the associated machine learning (ML) model.

4. The network node of claim 1, wherein the request for alignment information indicates a request for a recommended WTRU set for VFL training of a machine learning (ML) model.

5. The network node according to claim 1, wherein the sample selection criteria include network slice information, location or region information, date range or statistical conditions.

6. The network node of claim 1, wherein the alignment procedure includes determining whether the WTRUs in the recommended WTRU list for VFL training satisfy the sample selection criteria included in the alignment information.

7. The network node of claim 1, wherein the alignment procedure includes excluding the WTRU from the list of recommended WTRUs for VFL training based on determining that the WTRU does not meet the sample selection criteria included in the alignment information.

8. The network node of claim 1, wherein the processor and memory are further configured to: receive an initiation message for VFL training.

9. The network node of claim 8, wherein the initiating message includes a list of available features or a list of WTRUs selected for VFL training.

10. The network node of claim 9, wherein the processor and memory are further configured to: perform VFL operations using the list of available features or the list of WTRUs selected for VFL training, wherein the VFL operations include sharing VFL results trained locally.

11. A method performed by a cellular network node, the method comprising: Receive a request for alignment information for longitudinal federated learning (VFL), wherein the request for alignment information includes sample selection criteria; Based on the sample selection criteria included in the alignment information, an alignment procedure with network functions is executed; as well as Send a response to the request for alignment information, the response indicating parameters for VFL alignment, wherein the parameters for VFL alignment include a recommended list of wireless transmit / receive units (WTRUs) for VFL training.

12. The method of claim 11, further comprising: The request for alignment information is received from the application function (AF).

13. The method of claim 11, wherein the parameters for VFL alignment indicate a machine learning (ML) model.

14. The method of claim 11, wherein the request for alignment information indicates a request for a machine learning (ML) model or a request for a recommended WTRU set for VFL training.

15. The method of claim 11, wherein the sample selection criteria include network slice information, location or region information, date range or statistical conditions.

16. The method of claim 11, wherein performing the alignment procedure includes determining whether the WTRUs in the list of recommended WTRUs for VFL training satisfy the sample selection criteria included in the alignment information.

17. The method of claim 11, wherein performing the alignment procedure includes excluding the WTRU from the list of recommended WTRUs for VFL training based on determining that the WTRU does not meet the sample selection criteria included in the alignment information.

18. The method of claim 11, further comprising: Receive the initiation message for VFL training.

19. The method of claim 18, wherein the initiating message includes a list of available features or a list of WTRUs selected for VFL training.

20. The method of claim 19, further comprising: Perform VFL operations, wherein the VFL operations include sharing the VFL results trained locally.

Images