PC5-based AI / ML discovery and enhancement of PC5 connectivity

Abstract

Discovery of PC5-based AI / ML and enhancement of PC5 connectivity are described herein. As would be more detailed below, the WTRU performs ProSe discovery of other WTRUs that support PC5-based AI / ML splitting while taking into account the supported AI / ML operation, i.e., the capability of the WTRU. ProSe discovery codes may be enhanced to include an AI / ML model for support for applications using PC5-based AI / ML. Capability information of the WTRU or level information of the capability of the WTRU may be included for discovery. Alternatively, some of those information may be negotiated during or after PC5 connection establishment.

Description

[0001] Cross - reference to related applications

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 422,298, filed on November 3, 2022, and U.S. Provisional Application No. 63 / 537,332, filed on September 8, 2023. The entire contents of the U.S. Provisional Applications are incorporated herein by reference. Summary of the Invention

[0003] Described herein are enhancements to PC5 - based AI / ML discovery and PC5 connections. As will be described in more detail below, the WTRU performs ProSe discovery of other WTRUs that support PC5 - based AI / ML split, taking into account the supported AI / ML operations, i.e., the capabilities of the WTRU. The ProSe discovery code can be enhanced to include supported AI / ML models for applications using PC5 - based AI / ML. Information about the capabilities of the WTRU or the level of the capabilities of the WTRU can be included for discovery. As an alternative, some of the capability information can be negotiated during or after the establishment of the PC5 connection.

[0004] Described herein is dynamic WTRU configuration for local AI / ML services. As will be described in more detail below, based on the location of the WTRU, when there are any local AI / ML services that support or do not support PC5 - based AI / ML operations, the 5GS configuration can notify the configuration parameters of the local AI / ML services, such as the application server information and configuration parameters for PC5 - based AI / ML, and discover, for example, the code for PC5 - based AI / ML services.

[0005] Described herein is dynamic AI / ML split negotiation between the WTRU and the AF considering the availability of PC5 - based AI / ML and server - based AI / ML. As will be described in more detail below, the WTRU and the AI / ML application server use the expected performance and communication performance analysis information of PC5 - based AI / ML to negotiate the AI / ML split point. The AI / ML application server or the 5GS can trigger the discovery process for PC5 - based AI / ML services and request a report on the expected performance of PC5 - based AI / ML. Brief Description of the Drawings

[0006] A more detailed understanding can be obtained from the following description given by way of example in conjunction with the drawings, in which like reference numerals in the figures indicate like elements, and in which:

[0007] Figure 1A is a system diagram illustrating an example communication system in which one or more of the disclosed embodiments can be implemented;

[0008] Figure 1B is a system diagram illustrating an example wireless transmit / receive unit (WTRU) that can be used within the communication system illustrated in Figure 1A ;

[0009] Figure 1C is a system diagram illustrating an example radio access network (RAN) and an example core network (CN) that can be used within the communication system illustrated in Figure 1A ;

[0010] Figure 1D is a system diagram illustrating additional example RANs and additional example CNs that can be used within the communication system illustrated in Figure 1A ;

[0011] Figure 2 illustrates a reference model of a potential architecture for 5G or a next generation network;

[0012] Figure 3 illustrates a PC5-based AI / ML operation split according to certain examples;

[0013] Figure 4 illustrates model A-based discovery and PC5 connection establishment for PC5-based AI / ML;

[0014] Figure 5 illustrates model B-based discovery and PC5 connection establishment for PC5-based AI / ML;

[0015] Figure 6 illustrates an enhanced method for discovery and PC5 connection for PC5-based AI / ML;

[0016] Figure 7 illustrates a WTRU configuration update procedure for PC5-based AI / ML;

[0017] Figure 8 illustrates the configuration of local AI / ML and PC5-based AI / ML during PDU session establishment;

[0018] Figure 9 illustrates a method for dynamic WTRU configuration for local AI / ML services;

[0019] Figure 10 illustrates AI / ML split negotiation with network analysis;

[0020] Figure 11 illustrates AI / ML split negotiation with availability reporting for PC5-based AI / ML;

[0021] Figure 12Illustrated is a method for dynamic AI / ML split negotiation between a WTRU and an AF considering the availability of PC5-based AI / ML and server-based AI / M;

[0022] Figure 13 Illustrated is an example of a two-stage AI / ML split process;

[0023] Figure 14 Illustrated is an AI / ML split complementary split process; and

[0024] Figure 15 Illustrated is an AI / ML split complementary split process enhanced with aggregated information. Detailed implementation mode

[0025] The discovery of PC5-based AI / ML and the enhancement of PC5 connection are described herein. As will be described in more detail below, the WTRU performs ProSe discovery of other WTRUs that support PC5-based AI / ML split, while considering the supported AI / ML operations, i.e., the capabilities of the WTRU. The ProSe discovery code can be enhanced to include supported AI / ML models for applications using PC5-based AI / ML. The capability information of the WTRU or the level information of the capabilities of the WTRU can be included for discovery. As an alternative, some of the capability information can be negotiated during or after the establishment of the PC5 connection.

[0026] The dynamic WTRU configuration for local AI / ML services is described herein. As will be described in more detail below, based on the location of the WTRU, when there are any local AI / ML services that support or do not support PC5-based AI / ML operations, the 5GS configuration can notify the configuration parameters of the local AI / ML services, such as the application server information and configuration parameters of PC5-based AI / ML, and discover, for example, the code of the PC5-based AI / ML services.

[0027] The dynamic AI / ML split negotiation between the WTRU and the AF considering the availability of PC5-based AI / ML and server-based AI / ML is described herein. As will be described in more detail below, the WTRU and the AI / ML application server negotiate the AI / ML split point using the expected performance and communication performance analysis information of PC5-based AI / ML. The AI / ML application server or the 5GS can trigger the discovery process of the PC5-based AI / ML service and request a report on the expected performance of PC5-based AI / ML.

[0028] A PC5-based AI / ML system, a wireless transmission and reception unit (WTRU), and a method are described. The method includes: transmitting a discovery request message that includes a discovery code for PC5-based AI / ML, a requested AI / ML model, and a performance level of the AI / ML; receiving a discovery response message that includes an identification of the AI / ML model, a performance level, and target user information; configuring a PC5 connection based on the discovery response message; and negotiating a PC5-based AI / ML split. The discovery request message may allow for the selection of a device for the AI / ML. The device may be a split WTRU. The device may be selected based on the discovery response. The discovery response may be from a device. The method may further include performing a registration process that can provide an indication of the ability for the split. The method may further include determining the availability of local services in the area of the WTRU for use in the split. The negotiated split may be based on the expected performance of the WTRU. The negotiated split may be based on the performance data of the WTRU. The configured connection may be configured based on the performance level of the AI / ML.

[0029] A WTRU may include a processor and a transceiver communicatively coupled to the processor to: transmit a discovery request message that includes a discovery code for PC5-based AI / ML, a requested AI / ML model, and a performance level of the AI / ML; receive a discovery response message that includes an identification of the AI / ML model, a performance level, and target user information; configure a PC5 connection based on the discovery response message; and negotiate a PC5-based AI / ML split. The discovery request message may allow for the selection of a device for the AI / ML. The device may be a split WTRU. The device may be selected based on the discovery response. The discovery response may be from a device. The processor and transceiver further operate to perform a registration process that can provide an indication of the ability for the split. The processor and transceiver further operate to determine the availability of local services in the area of the WTRU for use in the split. The negotiated split may be based on the expected performance of the WTRU. The negotiated split may be based on the performance data of the WTRU. The configured connection may be configured based on the performance level of the AI / ML.

[0030] Figure 1AFIG. is a diagram of an example communication system 100 in which one or more of the disclosed embodiments may be implemented. The communication system 100 may be a multi-access system that provides content such as voice, data, video, messaging, broadcast, etc. to a plurality of wireless users. The communication system 100 may enable a plurality of wireless users to access such 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 discrete Fourier transform spread OFDM (ZT UW DFT-S OFDM), unique word OFDM (UW-OFDM), resource block filtered OFDM, filter bank multicarrier (FBMC), and the like.

[0031] As Figure 1A shown, the communication system 100 may include wireless transmit / receive units (WTRUs) 102a, 102b, 102c, 102d, a radio access network (RAN) 104, a core network (CN) 106, a public switched telephone network (PSTN) 108, the Internet 110, and other networks 112, although 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, 102d may be any type of device configured to operate and / or communicate in a wireless environment. By way of example, the WTRUs 102a, 102b, 102c, 102d (any one of which may be referred to as a station (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 telephones, personal digital assistants (PDA), smart phones, laptop computers, netbooks, personal computers, wireless sensors, hotspots or Mi-Fi devices, Internet of Things (IoT) devices, watches or other wearable devices, head-mounted displays (HMD), vehicles, drones, medical devices and applications (e.g., remote surgery), industrial devices and applications (e.g., robots and / or other wireless devices operating in an industrial and / or automation processing chain environment), consumer electronic devices, devices operating on commercial and / or industrial wireless networks, and the like. Any one of the WTRUs 102a, 102b, 102c, and 102d may be interchangeably referred to as a UE.

[0032] The communication system 100 may further include base station 114a and / or base station 114b. Each of base stations 114a, 114b may be any type of device configured to wirelessly interface with at least one of WTRUs 102a, 102b, 102c, 102d to facilitate access to one or more communication networks such as CN 106, the Internet 110, and / or other networks 112. By way of example, base stations 114a, 114b may be base transceiver stations (BTSs), NodeBs, eNode Bs (eNBs), home NodeBs, home eNode Bs, next-generation NodeBs (such as gNode Bs (gNBs), New Radio (NR) NodeBs), site controllers, access points (APs), wireless routers, and the like. Although base stations 114a, 114b are each depicted as a single element, it will be understood that base stations 114a, 114b may include any number of interconnected base stations and / or network elements.

[0033] Base station 114a may be part of RAN 104, which may further include other base stations and / or network elements (not shown), such as base station controllers (BSCs), radio network controllers (RNCs), relay nodes, and the like. Base station 114a and / or base station 114b may be configured to transmit and / or receive wireless 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 wireless services to a particular 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. Thus, in one embodiment, base station 114a may include three transceivers, i.e., one transceiver for each sector of the cell. In an embodiment, base station 114a may employ multiple-input multiple-output (MIMO) technology and may utilize multiple transceivers for each sector of the cell. For example, beamforming may be used to transmit and / or receive signals in a desired spatial direction.

[0034] Base stations 114a, 114b may communicate with one or more of WTRUs 102a, 102b, 102c, 102d via air interface 116, which may be any suitable wireless communication link (e.g., radio frequency (RF), microwave, centimeter wave, millimeter wave, infrared (IR), ultraviolet (UV), visible light, etc.). Any suitable radio access technology (RAT) may be used to establish air interface 116.

[0035] More specifically, as noted above, the communication system 100 can be a multi-access system and can employ one or more channel access schemes such as CDMA, TDMA, FDMA, OFDMA, SC-FDMA, and the like. For example, the base stations 114a in the RAN 104 and the WTRUs 102a, 102b, 102c can implement radio technologies such as Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access (UTRA), which can use Wideband CDMA (WCDMA) to establish the air interface 116. WCDMA can include communication protocols such as High-Speed Packet Access (HSPA) and / or Evolved HSPA (HSPA+). HSPA can include High-Speed Downlink (DL) Packet Access (HSDPA) and / or High-Speed Uplink (UL) Packet Access (HSUPA).

[0036] In an embodiment, the base stations 114a and the 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 Advanced LTE (LTE-A) and / or Advanced LTE Pro (LTE-APro) to establish the air interface 116.

[0037] In an embodiment, the base stations 114a and the WTRUs 102a, 102b, 102c can implement a radio technology (such as NR radio access), which can use NR to establish the air interface 116.

[0038] In an embodiment, the base stations 114a and the WTRUs 102a, 102b, 102c can implement multiple radio access technologies. For example, the base stations 114a and the WTRUs 102a, 102b, 102c can implement LTE radio access and NR radio access together using, for example, the Dual Connectivity (DC) principle. Thus, the air interface utilized by the WTRUs 102a, 102b, 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).

[0039] In other embodiments, the base station 114a and the WTRUs 102a, 102b, 102c may implement radio technologies such as IEEE 802.11 (i.e., Wireless Fidelity (WiFi)), IEEE 802.16 (i.e., Worldwide Interoperability for Microwave Access (WiMAX)), CDMA2000, CDMA2000 1X, CDMA2000 EV-DO, Interim Standard 2000 (IS-2000), Interim Standard 95 (IS-95), Interim Standard 856 (IS-856), Global System for Mobile Communications (GSM), GSM Enhanced Data Rate Evolution (EDGE), GSM EDGE (GERAN), and the like.

[0040] Figure 1A The base station 114b in may be, for example, a wireless router, a Home Node B, a Home eNode B, or an access point, and may utilize any suitable RAT to facilitate wireless connectivity in a local area such as a business location, a home, a vehicle, a campus, an industrial facility, an air corridor (e.g., for drones), a road, etc. In one embodiment, the base station 114b and the WTRUs 102c, 102d may implement radio technologies such as IEEE 802.11 to establish a Wireless Local Area Network (WLAN). In an embodiment, the base station 114b and the WTRUs 102c, 102d may implement radio technologies such as IEEE 802.15 to establish a Wireless Personal Area Network (WPAN). In yet another embodiment, the base station 114b and the WTRUs 102c, 102d may utilize a cellular-based RAT (e.g., WCDMA, CDMA2000, GSM, LTE, LTE-A, LTE-A Pro, NR, etc.) to establish a pico cell or a femto cell. As Figure 1A shown in Figure 1A , the base station 114b may have a direct connection to the Internet 110. Thus, the base station 114b may not need to access the Internet 110 via the CN 106.

[0041] The RAN 104 may communicate with the CN 106, which may be any type of network configured to provide voice, data, applications, and / or Voice over Internet Protocol (VoIP) services to one or more of the WTRUs 102a, 102b, 102c, 102d. The data may have different Quality of Service (QoS) requirements such as different throughput requirements, latency requirements, error tolerance requirements, reliability requirements, data throughput requirements, mobility requirements, and the like. The CN 106 may provide call control, billing services, location-based services for mobile devices, prepaid calling, Internet connectivity, video distribution, etc., and / or perform advanced security functions such as user authentication. Although not shown in Figure 1Ais shown, it will be understood that RAN 104 and / or CN 106 may communicate directly or indirectly with other RANs employing the same RAT as RAN 104 or a different RAT. For example, in addition to being connected to RAN 104 which may utilize NR radio technology, CN 106 may also communicate with another RAN (not shown) employing GSM, UMTS, CDMA 2000, WiMAX, E-UTRA or WiFi radio technology.

[0042] CN 106 may also act as a gateway for WTRUs 102a, 102b, 102c, 102d to access the PSTN 108, the Internet 110 and / or other networks 112. The PSTN 108 may include a circuit-switched telephone network providing plain old telephone service (POTS). The Internet 110 may include a global system of interconnected computer networks and devices using common communication protocols such as Transmission Control Protocol (TCP), User Datagram Protocol (UDP) and / or Internet Protocol (IP) in the TCP / IP Internet protocol suite. The network 112 may include wired and / or wireless communication networks owned and / or operated by other service providers. For example, the network 112 may include another CN connected to one or more RANs, which may employ the same RAT as RAN 104 or a different RAT.

[0043] Some or all of the WTRUs 102a, 102b, 102c, 102d in the communication system 100 may include multi-mode capabilities (e.g., the WTRUs 102a, 102b, 102c, 102d may include multiple transceivers for communicating with different wireless networks via different wireless links). For example, Figure 1A the WTRU 102c shown in may be configured to communicate with a base station 114a that may employ a cellular-based radio technology and with a base station 114b that may employ IEEE 802 radio technology.

[0044] Figure 1B is a system diagram illustrating an example WTRU 102. As Figure 1B shown, among other things, the WTRU 102 may further include a processor 118, a transceiver 120, a transmit / receive element 122, a speaker / microphone 124, a keypad 126, a display / touchpad 128, a non-removable memory 130, a removable memory 132, a power supply 134, a Global Positioning System (GPS) chipset 136 and / or other peripheral devices 138. It will be understood that the WTRU 102 may include any sub-combination of the foregoing elements while remaining consistent with the embodiments.

[0045] The processor 118 can be a general-purpose processor, a dedicated 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), any other type of integrated circuit (IC), a state machine, and the like. The processor 118 can perform signal encoding, data processing, power control, input / output processing, and / or any other functions that enable the WTRU 102 to operate in a wireless environment. The processor 118 can be coupled to a transceiver 120, which can be coupled to a transmit / receive element 122. Although Figure 1B the processor 118 and the transceiver 120 are depicted as separate components, it will be understood that the processor 118 and the transceiver 120 can be integrated together in an electronic package or chip.

[0046] The transmit / receive element 122 can be configured to transmit signals to a base station (e.g., base station 114a) or receive signals from a base station (e.g., base station 114a) via an air interface 116. For example, in one embodiment, the transmit / receive element 122 can be an antenna configured to transmit and / or receive RF signals. In an embodiment, the transmit / receive element 122 can be a transmitter / detector configured to transmit and / or receive, for example, IR, UV, or visible light signals. In yet another embodiment, the transmit / receive element 122 can be configured to transmit and / or receive both RF signals and optical signals. It will be understood that the transmit / receive element 122 can be configured to transmit and / or receive any combination of wireless signals.

[0047] Although the transmit / receive element 122 is depicted as a single element in Figure 1B the WTRU 102 can include any number of transmit / receive elements 122. More specifically, the WTRU 102 can employ MIMO technology. Thus, in one embodiment, the WTRU 102 can include two or more transmit / receive elements 122 (e.g., multiple antennas) for transmitting and receiving wireless signals via the air interface 116.

[0048] The transceiver 120 can be configured to modulate the signals to be transmitted by the transmit / receive element 122 and demodulate the signals received by the transmit / receive element 122. As noted above, the WTRU 102 can have multi-mode capabilities. Thus, the transceiver 120 can include multiple transceivers to enable the WTRU 102 to communicate via multiple RATs (such as NR and IEEE802.11), for example.

[0049] The processor 118 of the WTRU 102 may be coupled to a speaker / microphone 124, a keypad 126, and / or a display / touchpad 128 (e.g., a liquid crystal display (LCD) display unit or an organic light emitting diode (OLED) display unit) and may receive user input data from the speaker / microphone 124, the keypad 126, and / or the display / touchpad 128 (e.g., a liquid crystal display (LCD) display unit or an organic light emitting diode (OLED) display unit). The processor 118 may also output user data to the speaker / microphone 124, the keypad 126, and / or the display / touchpad 128. In addition, the processor 118 may 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. The 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. The removable memory 132 may include a subscriber identity module (SIM) card, a memory stick, a secure digital (SD) memory card, etc. In other embodiments, the processor 118 may access information from a memory that is not physically located on the WTRU 102 (such as, a server or a home computer (not shown)) and store data in that memory.

[0050] The processor 118 may receive power from a 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 battery packs (e.g., nickel cadmium (NiCd), nickel zinc (NiZn), nickel metal hydride (NiMH), lithium ion (Li-ion), etc.), a solar cell, a fuel cell, and the like.

[0051] The processor 118 may also be coupled to a GPS chipset 136, which may be configured to provide location information (e.g., longitude and latitude) regarding the current location of the WTRU 102. In addition to or instead of the information from the GPS chipset 136, the WTRU 102 may receive location information from a base station (e.g., base stations 114a, 114b) via an air interface 116 and / or determine its location based on the timing of signals received from two or more nearby base stations. It will be understood that the WTRU 102 may obtain location information by any suitable location determination method while remaining consistent with the embodiments.

[0052] The processor 118 may be further coupled to other peripheral devices 138, which may include one or more software modules and / or hardware modules that provide additional features, functionality, and / or wired or wireless connections. For example, the peripheral device 138 may include an accelerometer, an electronic compass, a satellite transceiver, a digital camera (for photos and / or videos), a universal serial bus (USB) port, a vibration device, a television transceiver, a hands-free headset, modules, a frequency modulation (FM) radio unit, a digital music player, a media player, a video game player module, an Internet browser, a virtual reality and / or augmented reality (VR / AR) device, an activity tracker, and the like. The peripheral device 138 may include one or more sensors. The sensor may be one or more of the following: a gyroscope, an accelerometer, a Hall effect sensor, a magnetometer, an orientation sensor, a proximity sensor, a temperature sensor, a time sensor; a geographic location sensor; an altimeter, a light sensor, a touch sensor, a magnetometer, a barometer, a gesture sensor, a biometric sensor, a humidity sensor, and the like.

[0053] The WTRU 102 may include a full-duplex radio, for which the transmission and reception of some or all signals (e.g., associated with a particular subframe for UL (e.g., for transmission) and DL (e.g., for reception) both) may be concurrent and / or simultaneous. The full-duplex radio may include an interference management unit for reducing and / or substantially eliminating self-interference via signal processing performed by hardware (e.g., a choke) or via a processor (e.g., a separate processor (not shown) or via the processor 118). In an embodiment, the WTRU 102 may include a half-duplex radio, for which the transmission and reception of some or all signals (e.g., associated with a particular subframe for UL (e.g., for transmission) or DL (e.g., for reception)).

[0054] Figure 1C is a system diagram illustrating a RAN 104 and a CN 106 according to an embodiment. As noted above, the RAN 104 may employ E-UTRA radio technology to communicate with the WTRU 102a, 102b, 102c via the air interface 116. The RAN 104 may also communicate with the CN 106.

[0055] The RAN 104 may include eNode-Bs 160a, 160b, 160c, but it will be understood that, while remaining consistent with the embodiments, the RAN 104 may include any number of eNode-Bs. Each of the eNode-Bs 160a, 160b, 160c may include one or more transceivers for communicating with the WTRUs 102a, 102b, 102c via the air interface 116. In one embodiment, the eNode-Bs 160a, 160b, 160c may implement MIMO technology. Thus, the eNode-B 160a, for example, may use multiple antennas to transmit wireless signals to and / or receive wireless signals from the WTRU 102a.

[0056] Each of the eNode-Bs 160a, 160b, 160c may be associated with a particular cell (not shown) and may be configured to handle radio resource management decisions, handover decisions, scheduling of users in the UL and / or DL, and the like. As Figure 1C shown, the eNode-Bs 160a, 160b, 160c may communicate with each other via the X2 interface.

[0057] Figure 1C The CN 106 shown may include a Mobility Management Entity (MME) 162, a Serving Gateway (SGW) 164, and a Packet Data Network (PDN) Gateway (PGW) 166. Although the foregoing elements are depicted as part of the CN 106, it will be understood that any of these elements may be owned and / or operated by entities other than the CN operator.

[0058] The MME 162 may be connected to each of the eNode-Bs 162a, 162b, 162c in the RAN 104 via the S1 interface and may act as a control node. For example, the MME 162 may be responsible for authenticating users of the WTRUs 102a, 102b, 102c, bearer activation / deactivation, selecting a particular serving gateway during the initial attachment of the WTRUs 102a, 102b, 102c, and the like. The MME 162 may provide control plane functions for handover between the RAN 104 and other RANs (not shown) employing other radio technologies such as GSM and / or WCDMA.

[0059] The SGW 164 can be connected to each of the eNode Bs 160a, 160b, 160c in the RAN 104 via the S1 interface. The SGW 164 can generally route and forward user data packets to / from the WTRUs 102a, 102b, 102c. The SGW 164 can perform other functions, such as anchoring the user plane during handovers between eNode Bs, triggering paging when DL data is available for the WTRUs 102a, 102b, 102c, managing and storing the contexts of the WTRUs 102a, 102b, 102c, and the like.

[0060] The SGW 164 can be connected to the PGW 166, which can provide the WTRUs 102a, 102b, 102c with access to a packet switched network (such as the Internet 110) to facilitate communication between the WTRUs 102a, 102b, 102c and IP-enabled devices.

[0061] The CN 106 can facilitate communication with other networks. For example, the CN 106 can provide the WTRUs 102a, 102b, 102c with access to a circuit switched network (such as the PSTN 108) to facilitate communication between the WTRUs 102a, 102b, 102c and traditional landline communication devices. For example, the CN 106 can include an IP gateway (such as an IP Multimedia Subsystem (IMS) server) that serves as an interface between the CN 106 and the PSTN 108 or can communicate with the IP gateway. Additionally, the CN 106 can provide the WTRUs 102a, 102b, 102c with access to other networks 112, which can include other wired and / or wireless networks owned and / or operated by other service providers.

[0062] Although the WTRU is described as a wireless terminal in Figures 1A - 1D it is contemplated that in some representative embodiments, such a terminal can (e.g., temporarily or permanently) use a wired communication interface with the communication network.

[0063] In a representative embodiment, the other network 112 can be a WLAN.

[0064] A WLAN in infrastructure basic service set (BSS) mode can have an access point (AP) for the BSS and one or more stations (STAs) associated with the AP. The AP can have access or an interface to a distribution system (DS) or another type of wired / wireless network that carries traffic to and / or from the BSS. Traffic originating from outside the BSS and destined for an STA can reach the STA through the AP and can be delivered to the STA. Traffic originating from an STA and destined for a destination outside the BSS can be sent to the AP for delivery to the corresponding destination. Traffic between STAs within the BSS can be sent through the AP. For example, the source STA can send traffic to the AP, and the AP can deliver the traffic to the destination STA. Traffic between STAs within the BSS can be considered and / or referred to as peer-to-peer traffic. Peer-to-peer traffic can be sent between a source and a destination STA (e.g., directly between them) using direct link setup (DLS). In some representative embodiments, DLS can use 802.11e DLS or 802.11z tunnel DLS (TDLS). A WLAN using independent BSS (IBSS) mode may not have an AP, and STAs within the IBSS or using the IBSS (e.g., all STAs) can communicate directly with each other. The IBSS communication mode can sometimes be referred to in this document as an "ad-hoc" communication mode.

[0065] When using 802.11ac infrastructure operation mode or a similar operation mode, the AP can transmit beacons on a fixed channel, such as the primary channel. The primary channel can be of a fixed width (e.g., 20 MHz bandwidth) or dynamically set width. The primary channel can be the operating channel of the BSS and can be used by STAs to establish a connection with the AP. In some representative embodiments, for example, carrier sense multiple access / collision avoidance (CSMA / CA) can be implemented in an 802.11 system. For CSMA / CA, STAs (e.g., 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, the particular STA can back off. Only one STA (e.g., only one station) can transmit at any given time in a given BSS.

[0066] High throughput (HT) STAs can communicate using a 40 MHz wide channel, e.g., by combining the primary 20 MHz channel with an adjacent or non-adjacent 20 MHz channel to form a 40 MHz wide channel.

[0067] A very high throughput (VHT) STA can support channels that are 20 MHz, 40 MHz, 80 MHz, and / or 160 MHz wide. 40 MHz and / or 80 MHz channels can be formed by combining contiguous 20 MHz channels. A 160 MHz channel can be formed by combining eight contiguous 20 MHz channels, or by combining two non - contiguous 80 MHz channels (which can be referred to as an 80+80 configuration). For the 80+80 configuration, after channel coding, data can pass through a segment parser that can split the data into two streams. The inverse fast Fourier transform (IFFT) processing and time - domain processing can be performed separately on each stream. These streams can be mapped to two 80 MHz channels, and the data can be transmitted by a 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 the media access control (MAC).

[0068] 802.11af and 802.11ah support operation modes below 1 GHz. The channel operation bandwidth and carriers are reduced in 802.11af and 802.11ah 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 can support meter - type control / machine - type communication (MTC), such as MTC devices in a macro - coverage area. MTC devices can have certain capabilities, such as limited capabilities, including supporting (e.g., only supporting) certain bandwidths and / or limited bandwidths. MTC devices can include a battery with a battery life above a threshold (e.g., for maintaining a very long battery life).

[0069] A WLAN system that can support multiple channels and channel bandwidths such as 802.11n, 802.11ac, 802.11af, and 802.11ah include channels that can be designated as the primary channel. The primary channel can have a bandwidth equal to the maximum common operating bandwidth supported by all STAs in the BSS. The bandwidth of the primary channel can be set and / or restricted by the STA (which supports the minimum bandwidth operation mode) from all STAs operating in the BSS. In the example of 802.11ah, for an STA that supports (e.g., only supports) the 1MHz mode (e.g., an MTC-type device), the primary channel can be 1MHz wide, even if the AP and other STAs in the BSS support 2MHz, 4MHz, 8MHz, 16MHz, and / or other channel bandwidth operation modes. Carrier sensing and / or Network Allocation Vector (NAV) settings can depend on the state of the primary channel. If the primary channel is busy, for example, because an STA (which only supports the 1MHz operation mode) is transmitting to the AP, then all available frequency bands can be considered busy even if most of the available frequency bands remain idle.

[0070] In the United States, the available frequency band for 802.11ah is from 902MHz to 928MHz. In Korea, the available frequency band is from 917.5MHz to 923.5MHz. In Japan, the available frequency band is from 916.5MHz to 927.5MHz. The total bandwidth available for 802.11ah is 6MHz to 26MHz, depending on the country code.

[0071] Figure 1D FIG. is a system diagram illustrating RAN 104 and CN 106 according to an embodiment. As noted above, RAN104 can employ NR radio technology to communicate with WTRUs 102a, 102b, 102c via air interface 116. RAN 104 can also communicate with CN 106.

[0072] The RAN 104 may include gNBs 180a, 180b, 180c, but it will be understood that while remaining consistent with the embodiments, the RAN 104 may include any number of gNBs. Each of the gNBs 180a, 180b, 180c may include one or more transceivers for communicating with the WTRUs 102a, 102b, 102c via the air interface 116. In one embodiment, the gNBs 180a, 180b, 180c may implement MIMO technology. For example, the gNBs 180a, 180b may utilize beamforming to transmit signals to and / or receive signals from the gNBs 180a, 180b, 180c. Thus, the gNB 180a, for example, may use multiple antennas to transmit wireless signals to and / or receive wireless signals from the WTRU 102a. In an embodiment, the gNBs 180a, 180b, 180c may implement carrier aggregation technology. For example, the gNB 180a may transmit multiple component carriers to the 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 an embodiment, the gNBs 180a, 180b, 180c may implement coordinated multi-point (CoMP) technology. For example, the WTRU 102a may receive a coordinated transmission from the gNB 180a and the gNB 180b (and / or gNB 180c).

[0073] The WTRUs 102a, 102b, 102c may communicate with the gNBs 180a, 180b, 180c using transmissions associated with scalable numerology. For example, the OFDM symbol interval and / or the OFDM subcarrier interval may vary for different transmissions, different cells, and / or different portions of the radio transmission spectrum. The WTRUs 102a, 102b, 102c may communicate with the gNBs 180a, 180b, 180c using various or scalable length subframes or transmission time intervals (TTIs) (e.g., containing different numbers of OFDM symbols and / or continuously varying absolute time lengths).

[0074] gNBs 180a, 180b, 180c can be configured to communicate with WTRUs 102a, 102b, 102c in a stand-alone configuration and / or a non-stand-alone configuration. In the stand-alone configuration, WTRUs 102a, 102b, 102c can communicate with gNBs 180a, 180b, 180c without accessing other RANs (e.g., such as eNode-Bs 160a, 160b, 160c). In the stand-alone configuration, WTRUs 102a, 102b, 102c can use one or more of gNBs 180a, 180b, 180c as a mobility anchor point. In the stand-alone configuration, WTRUs 102a, 102b, 102c can communicate with gNBs 180a, 180b, 180c using signals in an unlicensed band. In the non-stand-alone configuration, WTRUs 102a, 102b, 102c can communicate with or connect to gNBs 180a, 180b, 180c while also communicating with or connecting to other RANs (such as eNode-Bs 160a, 160b, 160c). For example, WTRUs 102a, 102b, 102c can implement the DC principle to communicate with one or more gNBs 180a, 180b, 180c and one or more eNode-Bs 160a, 160b, 160c substantially simultaneously. In the non-stand-alone configuration, eNode-Bs 160a, 160b, 160c can be used as a mobility anchor for WTRUs 102a, 102b, 102c, and gNBs 180a, 180b, 180c can provide additional coverage and / or throughput for serving WTRUs 102a, 102b, 102c.

[0075] Each of gNBs 180a, 180b, 180c can be associated with a specific cell (not shown) and can be configured to handle radio resource management decisions, handover decisions, scheduling of users in UL and / or DL, support for network slicing, DC, interworking between NR and E-UTRA, routing of user plane data towards user plane functions (UPFs) 184a, 184b, routing of control plane information towards access and mobility management functions (AMFs) 182a, 182b, and the like. As Figure 1D shown, gNBs 180a, 180b, 180c can communicate with each other via the Xn interface.

[0076] Figure 1DCN 106 shown in the figure may include at least one AMF 182a, 182b, at least one UPF 184a, 184b, at least one session management function (SMF) 183a, 183b, and may include data networks (DN) 185a, 185b. Although the foregoing elements are depicted as part of CN 106, it will be understood that any of these elements may be owned and / or operated by entities other than the CN operator.

[0077] AMF 182a, 182b may be connected to one or more of gNBs 180a, 180b, 180c in RAN 104 via the N2 interface and may serve as control nodes. For example, AMF 182a, 182b may be responsible for authenticating users of WTRUs 102a, 102b, 102c, support for network slicing (e.g., handling of different protocol data unit (PDU) sessions with different requirements), selection of a particular SMF 183a, 183b, management of the registration area, termination of non-access stratum (NAS) signaling, mobility management, and the like. AMF 182a, 182b may use network slicing to customize CN support for WTRUs 102a, 102b, 102c based on the type of service used by WTRUs 102a, 102b, 102c. For example, different network slices may 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, services for machine type communication (MTC) access, and the like). AMF 182a, 182b may provide control plane functions for handover between RAN 104 and other RANs (not shown) employing other radio technologies such as LTE, LTE-A, LTE-A Pro, and / or non-3GPP access technologies such as WiFi.

[0078] SMF 183a, 183b may be connected to AMF 182a, 182b in CN 106 via the N11 interface. SMF 183a, 183b may also be connected to UPF 184a, 184b in CN 106 via the N4 interface. SMF 183a, 183b may select and control UPF 184a, 184b and configure the routing of traffic passing through UPF 184a, 184b. SMF 183a, 183b may perform other functions such as managing and allocating UE IP addresses, managing PDU sessions, controlling policy enforcement and QoS, providing DL data notifications, and the like. The PDU session type may be IP-based, non-IP-based, Ethernet-based, and the like.

[0079] UPF 184a and 184b can be connected to one or more of gNBs 180a, 180b, 180c in the RAN 104 via the N3 interface, and the N3 interface can provide access to a packet switched network (such as the Internet 110) to WTRUs 102a, 102b, 102c to facilitate communication between WTRUs 102a, 102b, 102c and IP-enabled devices. UPF 184a, 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 DL packets, providing mobility anchoring, and the like.

[0080] The CN 106 can facilitate communication with other networks. For example, the CN 106 can include an IP gateway (e.g., an IP Multimedia Subsystem (IMS) server) that serves as an interface between the CN 106 and the PSTN 108 or can communicate with the IP gateway. Additionally, the CN 106 can provide access to other networks 112 to WTRUs 102a, 102b, 102c, and the other networks can include other wired and / or wireless networks owned and / or operated by other service providers. In one embodiment, WTRUs 102a, 102b, 102c can be connected to local DNs 185a, 185b via UPF 184a, 184b through the N3 interface to UPF 184a, 184b and the N6 interface between UPF 184a, 184b and DNs 185a, 185b.

[0081] In view of Figures 1A - 1D and Figures 1A - 1D the corresponding descriptions, one or more or all of the functions described herein with reference to one or more of the following: one or more functions or all functions of WTRUs 102a - 102d, base stations 114a - 114b, eNode-Bs 160a - 160c, MME 162, SGW 164, PGW 166, gNBs 180a - 180c, AMFs 182a - 182b, UPFs 184a - 184b, SMFs 183a - 183b, DNs 185a - 185b, and / or any other device(s) described herein (one or more) can be performed by one or more emulation devices (not shown). The emulation device(s) can be one or more devices configured to emulate one or more or all of the functions described herein. For example, the emulation device(s) can be used to test other devices and / or simulate network and / or WTRU functions.

[0082] Emulation devices can be designed to implement one or more tests of other devices in a laboratory environment and / or an operator network environment. For example, the one or more emulation devices can perform one or more or all functions while being fully or partially implemented and / or deployed as part of a wired and / or wireless communication network to test other devices within the communication network. The one or more emulation devices can perform one or more functions or all functions while being temporarily implemented / deployed as part of a wired and / or wireless communication network. Emulation devices can be directly coupled to another device for testing purposes and / or perform tests using over-the-air wireless communication.

[0083] The one or more emulation devices can perform one or more (including all) functions while not being implemented / deployed as part of a wired and / or wireless communication network. For example, emulation devices can be used in a test laboratory and / or test scenarios in a non-deployed (e.g., test) wired and / or wireless communication network to implement tests of one or more components. The one or more emulation devices can be test equipment. Direct RF coupling and / or wireless communication via an RF circuit system (e.g., which can include one or more antennas) can be used by the emulation device to transmit and / or receive data.

[0084] Figure 2 Reference model 200 illustrates a potential architecture of a 5G or next-generation network. The architecture of model 200 specifies discrete interfaces between control plane elements. RAN 210 refers to a radio access network based on 5G RAT or evolved E-UTRA that is connected to a next-generation core network. The access and mobility management function (AMF) 220 includes at least the following functions: registration management, connection management, reachability management, mobility management, etc. The session management function (SMF) 230 includes at least the following functions: session management (including session establishment, modification, and release), WTRU IP address allocation, selection and control of the UPF function, etc. The user plane function (UPF) 240 includes at least the following functions: packet routing and forwarding, packet inspection, traffic usage reporting, etc.

[0085] 5G Location Service (LCS) can provide the function of providing location information of the WTRU 250. The location of the WTRU 250 can be supported by RAT-related location methods. RAT-related location methods can rely on, for example, 3GPP RAT measurements obtained by the target WTRU and / or measurements obtained by the access network on 3GPP RAT signals transmitted by the target WTRU. The location of the WTRU can be supported by RAT-independent location methods. RAT-independent location methods can rely on non-RAT measurements obtained by the WTRU and / or other information. The location information of one or more target WTRUs can be requested by an LCS client or application function (AF) 260 inside or outside the 3GPP operator network or a control plane NF within the 3GPP system and reported to an LCS client or application function (AF) 260 inside or outside the 3GPP operator network or a control plane NF within the 3GPP system. For a location request from an LCS client or AF 260, privacy verification of the target WTRU can be enabled to check whether it is allowed to obtain WTRU location information.

[0086] Several different types of location requests can be supported. A Mobile Terminated Location Request (MT-LR) that may occur together with a Mobile Terminated Location Request (MT-LR), where the LCS client or AF sends a location request to the 5G network to obtain the location of the target WTRU. A Mobile Originated Location Request (MO-LR) that may occur together with a Mobile Originated Location Request (MO-LR), where the WTRU sends a request to the 5G network to obtain location-related information of the WTRU. An Immediate Location Request that occurs together with an Immediate Location Request, where the LCS client or AF 260 sends or initiates a location request for (one or more) target WTRUs and expects to receive a response containing the location information of (one or more) target WTRUs within a short period of time. The Immediate Location Request can be used for MT-LR or MO-LR. A Delayed Location Request that occurs together with a Delayed Location Request, where the LCS client or AF 260 sends a location request for (one or more) target WTRUs to the 5G network and expects to receive a response when an indication event occurs for the target WTRU at a future time. It can be used for MT-LR.

[0087] The Authentication Server Function (AUSF) 270 verifies the identity of the user and provides access to network resources based on its security level.

[0088] The Unified Data Management (UDM) 280 stores and manages user data, including its IMSI and authentication data. When requested, the UDM 280 provides user data, such as authentication data, to other network functions (i.e., e.g., AMF 220, SMF 230).

[0089] The Policy Control Function (PCF) 290 is responsible for implementing policies that manage a user's access to network resources. When requested, the PCF 290 provides a user's policy data to other network functions (i.e., e.g., the AMF 220, the SMF 230). As illustrated and described herein, a Data Network (DN) 295 is located within the system.

[0090] Supporting AI / ML operations in the 5GC may be beneficial. Because the performance of AI / ML applications, including split computing and model transfer scenarios, can be significantly improved when an estimation of network conditions before / during an operation can be given to the AI / ML application. The solution can monitor the performance data and / or the analysis data regarding performance between the WTRU and the 5GC and provide the results to the WTRU or the AF so that the WTRU or the AF can initiate an AI / ML split operation.

[0091] For example, according to a KPI table for AI / ML split, when the uplink E2E delay is below 2 ms and the data rate is higher than 1.08 Gbps, the AI / ML image recognition job can be split, and that information can be provided to the WTRU or the AF to initiate the operations as elaborated in Table 1 below.

[0092] Table 1

[0093]

[0094] As an extension of the current AI / ML support in 3GPP, D2D can be leveraged to enhance AI / ML operations to provide benefits. For example, even if it is decided that the WTRU or the application server decides to split the AI / ML operation such that for the AI / ML split, a WTRU (such as WTRU-A) can be responsible for the computations of layers 1 - 15, while the application server can be responsible for the computations of layers 16 - 24, the WTRU (such as WTRU-A) can offload some of the AI / ML operations to another nearby WTRU (such as WTRU-B) when a PC5 connection is acceptable, and that other WTRU (such as WTRU-B) can be responsible for the computations of layers 5 - 15, while the WTRU (such as WTRU-A) can only be responsible for the computations of layers 1 - 4.

[0095] Figure 3Illustrates an example 300 of PC5-based AI / ML operation splitting. As illustrated in example 300, when WTRU-A 350 is responsible for the calculations of layers 1-15, the AI / ML operations can be split between WTRU-A 350 and WTRU-B 340 such that WTRU-A 350 is responsible for AI / ML operations 1-4, while WTRU-B 340 is responsible for AI / ML operations 5-15. WTRU-A 350 can benefit as it may require less power consumption in WTRU-A 350 and can provide better AI / ML services, such as reduced latency, as WTRU-B 340 has a closer connection to the network 320, such as via NG-RAN 330 to the application server 310. The AI / ML operations using multiple WTRUs in combination with PC5 connections are depicted as PC5-based AI / ML operations, and the AI / ML services using the application server in the network 320 are depicted as server 310-based AI / ML operations.

[0096] For PC5-based AI / ML supported discovery and PC5 connection establishment, this ProSe mechanism configures the WTRU to discover other WTRUs supporting the same application using the ProSe application code. When the ProSe application codes match, the WTRU can select another WTRU for further communication. The WTRU can also consider the target user information for discovery.

[0097] For PC5-based AI / ML operations, more information can be considered when the WTRU selects another WTRU. For example, the WTRU can check whether the same AI / ML model and AI / ML logic are available at the other WTRU. The WTRU may need to know the ability of the peer WTRU to perform PC5 AI / ML operations.

[0098] Dynamic negotiation of AI / ML splitting between the WTRU and the AI / ML application server can be used. When both PC5-based AI / ML operations and server-based AI / ML operations are available and the WTRU triggers the AI / ML service, the WTRU can decide whether to use PC5-based AI / ML operations, server-based AI / ML operations, or any hybrid type of these two operations. To compare between PC5-based AI / ML operations and server-based AI / ML operations, the WTRU can use the PC5-based AI / ML operations to measure the expected performance, thus allowing the WTRU to know the ability of the peer WTRU.

[0099] When the WTRU has low computing power, the application can change the split point so that the WTRU calculates fewer layers while increasing the data rate in the UVU to transmit higher-load intermediate data to the network. In other words, based on the NW performance and the state of the WTRU, the AI / ML split can be determined and negotiated. Therefore, when the WTRU knows about the PC5-based AI / ML operation, the availability of the PC5-based AI / ML operation can be considered to obtain a better AI / ML split between the WTRU and the application server.

[0100] When the PC5-based AI / ML operation and the server-based AI / ML operation are available, the AI / ML split operation can be enhanced.

[0101] Support for local AI / ML services can be used. There are several use cases for leveraging the local services of the 5GC. For example, in an amusement park, stadium, or exhibition, there may be a Local Area Data Network (LADN) to provide a connection to a local application server, as well as customized services based on event and location characteristics. The AI / ML service can be one of the customized services for the local environment. For example, if the target is restricted to a specific group of objects in a local area, image recognition can be enhanced to provide better object recognition. For offloading, the PC5-based AI / ML service can be considered. In such a local AI / ML service scenario, the availability of the local AI / ML service can be dynamically changed and configured (e.g., the application server address, DNS server address, possible AI / ML operations, AI / ML models to be supported, etc.) can be dynamically provided to the users in the area.

[0102] For local AI / ML services, configuration data can be provided for the availability of AI / ML operations and PC5-based AI / ML operations. AI / ML operations mean AI / ML model distribution, AI / ML model split for offloading, and federated learning support. PC5-based AI / ML operations mean AI / ML operations between multiple WTRUs, and the WTRU can communicate using the PC5 connection. For PC5-based AI / ML operations, the WTRU can use the ProSe mechanism to discover other WTRUs, and for the selection of WTRUs for PC5-based AI / ML operations, the ProSe discovery and PC5 connection establishment mechanism are enhanced.

[0103] To provide configuration and policy updates, the 5GS can support the WTRU configuration update process. For localized services, the Local Area Data Network (LADN) may be supported. When the WTRU supports the LADN, the availability of the PDU session using the LADN can be notified to the WTRU based on the location of the WTRU, and the WTRU can initiate a PDU session using the LADN.

[0104] Based on the dynamics of local AI / ML services and their limited availability at each location and time, the WTRU can be configured to use local AI / ML services for the required information, and the UCU mechanism and PDU session establishment mechanism can be enhanced to support dynamic configuration updates.

[0105] Enhancements to PC5-based AI / ML discovery and PC5 connection establishment are described. When the WTRU discovers and selects another WTRU for PC5-based AI / ML operations, this can be characterized by the service name, AI / ML operation name, and / or QoS requirements of the service and operation (e.g., processing latency, communication latency, data rate for some AI / ML split operations). In addition to ProSe application information, other parameters can be considered to select an appropriate WTRU for PC5-based AI / ML operations. The supported AI / ML framework, AI / ML model, and / or AI / ML algorithm can be considered. The capabilities of the WTRU can be considered, such as the available time in terms of power, battery level, battery charge level, available memory, and / or computing power. The possible role of the WTRU for AI / ML operations can be considered, such as the AI / ML model distribution function, AI / ML model split and / or combination function, AI / ML result determination function, etc. The link quality of the PC5 channel can be considered, such as signal strength, bit rate, and error rate, for example.

[0106] When performing ProSe discovery between a WTRU and another WTRU, such as Model A discovery or Model B discovery, the ProSe discovery code is used to identify the application using the ProSe service. For PC5-based AI / ML operations, the information can be included in the discovery code or added as additional information in the discovery message. The information can include application information related to the AI / ML operation, such as application ID, PLMNID, advertised user information ID, target user information ID to be discovered. The information can include identification information of the supported AI / ML algorithm or model (e.g., AlexNet, linear regression, deep neural network, logistic regression, decision tree, etc.), and the supported AI / ML framework, such as Tensorflow, TFLearn, etc. The information can include the capabilities of the WTRU, such as the available time in terms of power, battery level, battery charge level, available memory, and / or computing power.

[0107] Alternatively or additionally, the capabilities of the WTRU may be included. WTRU capability levels may be defined and pre-configured in the WTRU, and level values may be used. For example, levels based on terminal capabilities may be defined and represent a combination of capabilities including battery power, available memory, and / or computing power. For example, Level 1 may be provided when the battery power > 80%, the available memory > 16 GB, and the computing power > 8 cores * 1 GHz; Level 2 may be provided when the battery power > 80%, the available memory > 12 GB, and the computing power > 4 cores * 1 GHz; and Level 3 may be provided when the battery power > 80%, the available memory > 8 GB, and the computing power > 4 cores * 1 GHz.

[0108] The possible roles of the WTRU for AI / ML operations may also be included, e.g., together with an AI / ML model distribution function, an AI / ML model splitting and / or combining function, an AI / ML result determination function, etc. The AI / ML model distribution function is the function of distributing the AI / ML model to other WTRUs involved in the AI / ML operation. AI / ML model splitting is the function of splitting the AI / ML model for execution at each entity. For example, in an AI / ML model involving a server and a user terminal, the AI / ML model splitting function is used to split the used AI / ML model into two sub-models, with one part executed at the server and the other part executed at the user terminal. In this case, the result of the AI / ML model operation at one side may be used to execute the AI / ML model operation at the other side. The AI / ML model combination is the function of making one AI / ML model by combining two AI / ML models. The AI / ML result determination is the function of deriving the final result of the AI / ML operation based on the results collected from the entities involved in the AI / ML operation.

[0109] Alternatively or additionally, the identification information of the supported AI / ML operations or models may be included in the discovery code or included as additional information under the application information related to the AI / ML operation. Alternatively or additionally, the above-mentioned information may be negotiated with the WTRU during the establishment of the PC5 connection for the PC5-based AI / ML operation.

[0110] Additionally or alternatively, the WTRU may indicate its intention to support PC5-based AI / ML operations, e.g., the intention for other parties to join the PC5-based AI / ML splitting. This intention may have conditions for allowing or rejecting requests to join the operation, e.g., in certain link characteristics (such as UL / DL latency, data rate, etc.). When the WTRU decides whether to establish a PC5 connection with the discovered WTRU for the PC5-based AI / ML operation, the WTRU may consider the indication from the discovered WTRU.

[0111] Figure 4 Figure 400 shows a signaling diagram for discovery and PC5 connection establishment for PC5-based AI / ML based on Model A. In Signaling Diagram 400, a first WTRU (WTRU-1 405) is communicatively coupled to a second WTRU (WTRU-2 415).

[0112] At 410, Figure 400 includes WTRU-1 405 sending an advertisement message to WTRU-2 415. The advertisement message may include the type of discovery message, the ProSe application code, and application information related to AI / ML operations. The ProSe application code may include information about the application that supports AI / ML, and it may include additional information such as supported PC5-based AI / ML operations. For example, such additional information may include PC5-based AI / ML splitting and / or supported AI / ML algorithms.

[0113] The application information related to AI / ML operations may include the following information: user information of the application (such as user ID), identification information of the supported AI / ML algorithms or models (e.g., AlexNet, linear regression, deep neural network, logistic regression, decision tree, etc.), and the supported AI / ML frameworks (e.g., Tensorflow, TFLearn, etc.). It may include information representing the device capabilities of the WTRU, such as level information about the capabilities of the WTRU, the value of the battery power of the WTRU, available memory, and / or computing power. It may include the possible roles of the WTRU for AI / ML operations, for example, the AI / ML model distribution function, the AI / ML model splitting and / or combination function, the AI / ML result determination function, etc. The AI / ML model distribution function is the function of distributing the AI / ML model to other WTRUs involved in the AI / ML operation. The AI / ML model splitting is the function of splitting the AI / ML model for execution at each entity. For example, in an AI / ML model involving a server and a user terminal, the AI / ML model splitting function is used to split the used AI / ML model into two sub-models, with one part executed at the server and the other part executed at the user terminal. In this case, the result of the AI / ML model operation on one side can be used to execute the AI / ML model operation on the other side. The AI / ML model combination is the function of making one AI / ML model by combining two AI / ML models. The AI / ML result determination is the function of deriving the final result of the AI / ML operation based on the results collected from the entities involved in the AI / ML operation.

[0114] The WTRU-1405 may indicate its intention to support PC5-based AI / ML operations, e.g., for the intention of other parties to join a PC5-based AI / ML split. When there are any conditions related to the intention, they may be discovered together, exchanged during PC5 connection establishment, or checked by the 5GS to notify whether the discovered WTRU is suitable for PC5-based AI / ML operations.

[0115] At 420, the WTRU-2415 may monitor the advertisement message of the WTRU-1405 at 410. If the WTRU-2415 is interested in the received ProSe application code, the WTRU-2415 may perform a direct PC5 link establishment procedure with the WTRU-1405. If the advertisement message includes some application information related to AI / ML operations, such as supported PC5-based AI / ML operations, AI / ML algorithms, the capabilities of the WTRU, then for example, the WTRU-2415 may consider the provided application information when deciding whether the WTRU-2415 performs a direct PC5 link establishment with the WTRU-1405. For example, if the WTRU-2415 is interested in a PC5-based AI / ML split, and the supported AI / ML algorithms and / or the capabilities of the WTRU from the WTRU-1405 may not meet the acceptable threshold for the PC5-based AI / ML split. The WTRU-2415 may not select the WTRU-1405 for PC5 link establishment, and the WTRU-2415 may attempt to discover other WTRUs. When the WTRU-2415 receives multiple advertisement messages from multiple WTRUs, the WTRU-2415 may, based on whether the content of the advertisement message meets the requirements of the WTRU-2415 to participate in the expected AI / ML operations of the application represented by the ProSe application code, e.g., whether the device capabilities of the WTRU are higher than a certain level value, whether the supported AI / ML algorithms of the WTRU include the AI / ML algorithms of the interested WTRU-2415, whether the WTRU supports some roles of the WTRU for AI / ML operations (e.g., supports AI / ML model split), or whether all the mentioned requirements are met or only some of the requirements are met, and other parameters (such as signal quality), select one or more WTRUs that meet the requirements of the ProSe service and application. The WTRU-2415 may perform a direct PC5 link establishment procedure with those selected WTRUs.

[0116] Additionally or alternatively, the PC5 link quality between the WTRU-1405 and the WTRU-2415 may be one of the criteria for determining whether the WTRU-2415 performs a direct PC5 link establishment with the WTRU-1405. For example, if the PC5 link quality is below a certain threshold, the WTRU-2415 may not select the WTRU-1405.

[0117] When the WTRU-1405 selects the WTRU-2415 as the entity for establishing the PC5 link for performing PC5-based AI / ML operations, the WTRU-1405 may consider the intention of the WTRU-2415 to participate in the PC5-based AI / ML operations (if available regarding the conditions). The WTRU-1405 may check its availability based on this condition by referring to the 5GS or signaling exchange between the WTRU-1405 and the WTRU-2415 during or after the PC5 connection establishment.

[0118] At 430, the WTRU-2415 and the WTRU-1405 may negotiate the capabilities of the WTRUs for PC5-based AI / ML during or after the PC5 link establishment process. The negotiated capabilities of the WTRUs may include some application information related to the AI / ML operations, such as the supported PC5-based AI / ML operations, the supported AI / ML algorithms, and the device capabilities of the WTRUs, such as power, battery level, available memory, and / or computing power. When there are multiple selected WTRUs at 420 and the PC5 connection is established, the WTRU-2415 may perform the capability negotiation with the multiple WTRUs. And based on the capability negotiation, the WTRU-2415 may select some of the WTRUs downwards. For example, when some WTRUs may report their respective device capabilities as being connected to the power line. Such WTRUs may be selected, and other WTRUs reporting their device capabilities as being battery-powered may be discarded. For example, when some WTRUs support multiple roles of the WTRUs (e.g., both the AI / ML split function and the AI / ML model distribution function), and the capabilities of the WTRUs are high enough to support multiple roles of the WTRUs. Considering the operation efficiency, the WTRUs supporting multiple capabilities may be selected, and other WTRUs supporting only one role (e.g., the AI / ML split function or the AI / ML model distribution function) may be discarded.

[0119] The WTRU-2415 may perform further operations for the applications regarding the selected WTRU.

[0120] For example, if the target information is the reference WTRU to be discovered, the reference WTRU is interested in the ranging service and supports the ranging / sidelink positioning capabilities of the target WTRU (if any), and the reference WTRU performs the direct PC5 link establishment process with the target WTRU.

[0121] Figure 5Figure 500 is a signaling diagram illustrating PC5 connection establishment for Model B-based discovery and PC5-based AI / ML. In signaling diagram 500, a first WTRU (WTRU-1505) may be communicatively coupled to a second WTRU (WTRU-2515). In signaling diagram 500, at 510, ProSe discovery is enhanced to include discovery codes dedicated to PC5-based AI / ML via a discovery request message. The requested AI / ML model, the requested WTRU performance level, and PC5-based AI / ML operations may be added in the discovery message via 510 and 520.

[0122] As Figure 5As illustrated, at 510, the WTRU 1505 sends a discovery request message to the WTRU 2515. The discovery request message may include the type of discovery message and a ProSe query code. The ProSe query code may include information such as an interested ProSe application code and interested application information related to AI / ML operations. The ProSe application code may include information about the interested application that supports AI / ML, such as an application ID and a PLMN ID, and may include additional information, such as, for example, supported PC5-based AI / ML operations, PC5-based AI / ML splitting, and / or supported AI / ML algorithms. The interested application information related to AI / ML operations may include advertised user information, target user information to be discovered, information about the requested PC5-based AI / ML operations for the application ID (such as the requested AI / ML operation name or ID or the QoS requirements of the requested AI / ML operation), identification information of the interested AI / ML algorithms or models (such as AlexNet, linear regression, deep neural network, logistic regression, decision tree, etc.), information requesting the device capabilities of the WTRU (such as level information about the capabilities of the WTRU, the value of the battery power of the WTRU, the available time in terms of battery power, indication of the source, available memory, and / or computing power), the role of the requested WTRU for AI / ML operations (such as AI / ML model distribution function, AI / ML model splitting and / or combination function, AI / ML result determination function, etc.), and the role of the supported UE1 for AI / ML operations (such as AI / ML model distribution function, AI / ML model splitting and / or combination function, AI / ML result determination function, etc.). The AI / ML model distribution function is the function of distributing the AI / ML model to other WTRUs involved in the AI / ML operation. AI / ML model splitting is the function of splitting the AI / ML model for execution at each entity. For example, in an AI / ML model involving a server and a user terminal, the AI / ML model splitting function is used to split the used AI / ML model into two sub-models, with one part executed at the server and the other part executed at the user terminal. In this case, the result of the AI / ML model operation at one side of the communication can be used to execute the AI / ML model operation at the other side of the communication. AI / ML model combination is the function of making one AI / ML model by combining two AI / ML models. AI / ML result determination is the function of deriving the final result of the AI / ML operation based on the results collected from the entities involved in the AI / ML operation.

[0123] Alternatively or additionally, the identification information of the supported AI / ML algorithms or models may be included in the application information related to AI / ML operations.

[0124] The discovery request message sent by WTRU 1505 can be received by multiple WTRUs including WTRU2515. At 520, if WTRU2515 receives the discovery request message from WTRU1505 and determines a match, WTRU2515 responds with a discovery response message. For example, when WTRU1505 includes ProSe application code for PC5-based AI / ML splitting, the requested AI / ML algorithm (including logistic regression), the requested device capabilities (with a certain requested computing power level), and the role of the WTRU requested in the request message as an AI / ML model splitting function or an AI / ML model distribution function, if WTRU2515 supports the ProSe application for PC5-based AI / ML splitting, and WTRU2515 supports the AI / ML algorithm including logistic regression, the WTRU2515 capabilities (such as computing power, available memory, and power) are higher than the requested device capabilities, and WTRU2515 is authorized by the application provider or the mobile network service provider as an AI / ML model splitting function or an AI / ML model distribution function, then WTRU2515 determines a match with the discovery request message. The discovery response message can include the type of the discovery message, the ProSe application code, and application information related to AI / ML operations. The ProSe application code can include information about the application that supports AI / ML, and it can include more information such as supported PC5-based AI / ML operations, such as PC5-based AI / ML splitting, and / or supported AI / ML algorithms. The application information related to AI / ML operations can include user information of the application (such as user ID), identification information of the supported AI / ML algorithms or models (such as AlexNet, linear regression, deep neural network, logistic regression, decision tree, etc.), information representing the device capabilities of the WTRU (for example, level information about the capabilities of the WTRU, the value of the battery power of the WTRU, available memory, and / or computing power), and the possible role of the WTRU for AI / ML operations (such as AI / ML model distribution function, AI / ML model splitting and / or combination function, AI / ML result determination function, etc.).

[0125] Under some conditions, when WTRU2515 is available for the requested PC5-based AI / ML operation, WTRU2515 can check whether the conditions are met before responding. WTRU2515 can rely on network analysis or prediction services to check whether the conditions are met. When there are other WTRUs that receive the discovery request message from WTRU1505, the other WTRUs also perform the operations described at 520.

[0126] At 530, when the WTRU 1505 receives a discovery response message from the WTRU 2515, the WTRU 1505 may perform a direct PC5 link establishment procedure with the WTRU 2515. For example, based on the content of the discovery response message from the WTRU 2515, the WTRU 1505 determines whether the WTRU 2515 can be selected for the desired ProSe service and application represented by the ProSe query code at 510, and when the WTRU 2515 is selected for the desired ProSe service, the WTRU 1505 may perform a direct PC5 link establishment procedure with the WTRU 2515.

[0127] When sending discovery response messages to multiple WTRUs in response to a discovery request message sent at 510, WTRU 1505 may select one or more WTRUs that are considered suitable for the desired ProSe services and applications based on the content of the discovery response messages and other parameters such as signal quality. If the discovery response message includes some application information related to AI / ML operations (such as supported PC5-based AI / ML operations, AI / ML algorithms, capabilities of the WTRU), then WTRU 1505 may consider such information to determine whether WTRU 1505 performs a direct PC5 link establishment with WTRU 2515. For example, if WTRU 1505 can operate in a PC5-based AI / ML split and the supported AI / ML algorithms and / or the capabilities of the WTRU of WTRU 2515 do not seem good enough for the interested PC5-based AI / ML split, then WTRU 1505 may not select WTRU 2515 for PC5 link establishment and WTRU 1505 may attempt to discover other WTRUs. For example, when receiving multiple discovery response messages and the responding WTRUs support the requested AI / ML model / algorithm and the role of the requested WTRU, WTRU 1505 may select capable WTRUs based on their supported capabilities. For example, WTRU 1505 may select one or more WTRUs with higher capabilities (such as a higher capability level, or higher computing power) compared to other WTRUs with larger available memory and / or more sustainable power. For example, when WTRU 1505 requests multiple roles (such as an AI / ML model split function and an AI / ML model distribution function) in the discovery request message at 510 and receives multiple discovery response messages from WTRU 2515, ……, WTRU 10 (not shown), WTRU 1505 may select capable WTRUs for each different requested role based on their supported roles and capabilities. For example, when WTRU 2515 and other WTRUs support the AI / ML model split function and WTRU 2515 has higher capabilities than other WTRUs that support the AI / ML model split, WTRU 2515 may be selected for the AI / ML model split function. When WTRU 3 and other WTRUs support the AI / ML model distribution function and WTRU 3 has higher capabilities than other WTRUs that support the AI / ML model distribution function, WTRU 3 (not shown) may be selected for the AI / ML model distribution function.

[0128] Additionally or alternatively, the PC5 link quality between WTRU 1505 and WTRU 2515 can be one of the criteria for determining whether WTRU 1505 performs direct PC5 link establishment with WTRU 2515. For example, if the PC5 link quality is below a certain threshold, WTRU 1505 may not select WTRU 2515.

[0129] At 540, WTRU 1505 and WTRU 2515 can negotiate the PC5-based AI / ML capabilities of the WTRUs during or after the PC5 link establishment process. The negotiated capabilities of the WTRUs can include some application information related to AI / ML operations, such as supported PC5-based AI / ML operations, supported AI / ML algorithms, and the device capabilities of the WTRUs, such as power supply, battery level, available memory, and / or computing power. When there are multiple selected WTRUs at 530 and a PC5 connection is established, WTRU 1505 performs capability negotiation with the multiple WTRUs. Based on the capability negotiation, WTRU 1505 can select some of the WTRUs downwards. For example, when some WTRUs report their device capabilities to be connected to the power line, such WTRUs can be selected, and other WTRUs that report their device capabilities to be battery-powered can be discarded.

[0130] WTRU 1505 can perform further operations for the application regarding the selected WTRU.

[0131] Figure 6 An enhanced method 600 for PC5-based AI / ML discovery and PC5 connection is illustrated. Method 600 includes the establishment of PC5 communication. The establishment can include one WTRU sending a discovery request message to another WTRU at 610. Then, at 620, the other WTRU can send a discovery response message back to the WTRU. Once the discovery messages are sent / received, the PC5 connection can be configured at 630. Method 600 can include negotiating the split capabilities between the WTRUs at 640.

[0132] Dynamic WTRU configuration using local AI / ML services is possible. When a WTRU enters a location where local AI / ML services are available, the 5G network can notify the WTRU of the availability of the local AI / ML services and the configuration parameters of the local AI / ML services. For local AI / ML services, if PC5-based AI / ML services are available, discovery parameters for ProSe services can be provided to the WTRU. When the WTRU sends a registration request, the WTRU can notify its support for AI / ML services (including PC5-based AI / ML services). After receiving the WTRU's ability to support AI / ML services and / or PC5-based AI / ML services, the 5GC can notify any configuration parameters of the local AI / ML services. When the 5GC knows that the WTRU is located in the service area of the local AI / ML services, signaling overhead can be reduced by updating the configuration parameters of the local AI / ML services to the WTRU.

[0133] Additionally or alternatively, during registration, the WTRU can indicate its intention to support AI / ML operations over PC5, e.g., the intention to support PC5-based AI / ML splitting for other parties. The intention can include conditions for allowing or denying requests for join operations, e.g., in certain link characteristics (e.g., UL / DL latency, data rate, etc.). The 5GS can refer to the indication to decide whether to share configuration information for certain specific AI / ML operations with the WTRU, or select a list of potentially available WTRUs for certain PC5-based AI / ML operations, which can be shared with other WTRUs to assist in discovering other WTRUs for PC5-based AI / ML operations.

[0134] As another solution, when a specific DN is defined for a local AI / ML service and the WTRU supports the service and the connection to the DN, if the DN is available in the registration area, the WTRU can establish a PDU session for the DN. After receiving a PDU session establishment / modification request for the DN, the 5GC can notify any configuration information of the AI / ML services supported in the DN in the PDU session establishment / modification response, e.g., the configuration information, the address of the application server or the AI / ML server can be included in the PCO of the PDU session establishment / modification response.

[0135] Using the configuration information, the WTRU can access an application server to download applications and supported AI / ML models. By downloading the applications and AI / ML models, the WTRU supporting PC5-based AI / ML can engage in PC5-based AI / ML services for applications with the supported AI / ML models. And when PC5-based AI / ML services are possible for an application, any discovery information such as application information, discovery codes for ProSe services can be downloaded from the application server.

[0136] Figure 7 A signaling diagram 700 illustrating a WTRU configuration update process for PC5-based AI / ML is shown. Signaling diagram 700 includes WTRU-A 715, WTRU-B 705, AMF 735, PCF 745, and (R)AN 725, which are communicatively coupled for the signaling described below.

[0137] At 710, signaling diagram 700 includes the registration of a WTRU (e.g., WTRU-A 715) with the ability to support AI / ML operations. When a WTRU (such as WTRU-A 715) supports PC5-based AI / ML, WTRU-A 715 can notify its capabilities to the 5GC during registration or any other process. Such registration can be from WTRU-A 715 to (R)AN 725 and AMF 735. For example, during initial registration and / or mobility registration, WTRU-A 715 can include information about its support for PC5-based AI / ML, supported AI / ML operations and models (e.g., support for image recognition, support for PC5-based AI / ML splitting, etc.). If there are any policies or configurations regarding PC5-based AI / ML managed by the 5GC, WTRU-A 715 can report its status regarding the received policies or configurations for PC5-based AI / ML. This status can include whether WTRU-A 715 has any saved policies or configurations regarding PC5-based AI / ML, the version of its policies or configurations.

[0138] When WTRU-A 715 provides its ability to support PC5-based AI / ML operations, WTRU-A 715 can provide capabilities in terms of details such as its processing power, maximum data rate. For example, the capability can be an indication that WTRU-A 715 "is capable of performing PC5 AI / ML splitting operations" and WTRU-A 715 can process some "AI / ML models" with parameters such as "processing power of X GHz" and "processing latency of Y milliseconds". For example, the capability can indicate that WTRU-A 715 is capable of performing "PC5 AI / ML model sharing" within a "delay of [1s - 1min]" for an AI / ML model with a "maximum size of [100 - 500] MB".

[0139] At 720, the AMF 735 may provide the WTRU-A 715 capabilities to the PCF 745. When receiving information regarding the PC5-based AI / ML WTRU-A 715 capabilities and / or status, the AMF 735 may notify a network function (e.g., the PCF 745) that processes the WTRU-A 715 policies and / or configurations regarding PC5-based AI / ML of this information. The AMF 735 may provide the location of the WTRU-A 715.

[0140] Additionally or alternatively, when the AMF 735 knows that the WTRU-A 715 may enter a certain area (e.g., an area of interest requested by the PCF 745, or an area where AI / ML services are available), the AMF 735 may notify a network function (e.g., the PCF 745) that processes the WTRU-A 715 policies and / or configurations regarding PC5-based AI / ML of the location of the WTRU-A 715.

[0141] The AMF 735 may consider the capabilities of the PC5-based AI / ML WTRU-A 715 in view of specific conditions such as data rate, link quality, etc. The AMF 735 may refer to the gNB or the NWDAF to determine whether the conditions of the WTRU-A 715 can be met or are expected to be met for PC5-based AI / ML operations. When the WTRU-A 715 is capable of supporting local AI / ML services and / or PC5-based AI / ML, the AMF 735 may notify a network function to process the WTRU-A 715 policies and / or configurations regarding AI / ML services.

[0142] At 730, the PCF 745 may decide to update the configuration of the WTRU-A 715 regarding PC5-based AI / ML operations. The PCF 745 may utilize PC5-based AI / ML in the WTRU to determine the availability of local services. At 740, a configuration update request may be sent from the PCF 745 to the AMF 735. At 730 - 740, the PCF 745 or other responsible NF may decide to update the configuration or policy of the WTRU-A 715 regarding PC5-based AI / ML operations, e.g., because of the new availability of an AI / ML service in the location of the WTRU-A 715 or an updated configuration regarding the AI / ML service. At 740, the PCF 745 may send a configuration update request to the AMF 735. The configuration update request may include policy or configuration information regarding the PC5-based AI / ML service, such as available applications related to the PC5 AI / ML service, available area information associated with the application, discovery codes for discovering other WTRUs supporting the PC5-based AI / ML service, supported AI / ML operations (e.g., AI / ML splitting) and / or supported AI / ML models.

[0143] At 750, when the AMF 735 receives the configuration update request at 740, the AMF 735 may send a WTRU configuration update request to the WTRU-A 715, including the configuration and / or policy update information received at 740.

[0144] At 760, after receiving the WTRU configuration update request message including the configuration and / or policy update information, the WTRU-A 715 discovers the availability of some applications using AI / ML and the availability of PC5-based AI / ML. If the WTRU-A 715 decides to use the ProSe service for PC5-based AI / ML, the WTRU-A 715 may use the discovery code and any application-related information in the received configuration message to discover other WTRUs (such as the WTRU-B 705) supporting the PC5-based AI / ML.

[0145] Figure 8 A signaling diagram 800 of local AI / ML and PC5-based AI / ML during PDU session establishment is illustrated. A WTRU configuration procedure using PDU session establishment may be used. The diagram 800 includes the WTRU-A 815, WTRU-B 805, AMF 835, PCF 845, SMF 865, UPF 855, AF 885, and (R)AN 825, which are communicatively coupled for the following signaling.

[0146] At 810, WTRU-A 815 provides a PDU session establishment request to SMF 865. The PDU session establishment request may include a requested DNN for AI / ML. WTRU-A 815 may send a PDU session establishment request with the requested DNN, e.g., based on a URSP rule, based on a dedicated DNN for an AI / ML application, or based on configuration information from the 5GS for each location, where the requested DNN is allocated for traffic exchange for some AI / ML applications.

[0147] At 820, SMF 865 may signal an SM policy association request to PCF 845. When SMF 865 receives a PDU session establishment request with the requested DNN at 810, SMF 865 may send an SM policy association request for WTRU-A 815 to the PCF and send the requested DNN to PCF 845.

[0148] At 830, when PCF 845 receives the SM policy association request for WTRU-A 815 and the requested DNN at 820, PCF 845 may reply with an SM policy association response that includes service flow information of the applications and data that WTRU-A 815 is authorized to use via the DNN. PCF 845 may include application-related information of the associated AI / ML service for the DNN in the SM policy association response. The application-related information may include, for example, an application identifier of the AI / ML service and an application server address through which WTRU-A 815 may retrieve any configuration information of the application. Additionally or alternatively, other NFs may provide configuration information for the application associated with the DNN upon request from PCF 845 or SMF 865.

[0149] At 840, SMF 865 may reply to WTRU-A 815 with a PDU session establishment response that has QoS flows of the authorized applications and / or services to be used at the DNN and configuration information of the application that may be received from PCF 845 at 830 or from another network function.

[0150] At 850, after receiving the PDU session establishment response in 840, WTRU-A 815 may access application server 885 to retrieve configuration information of the application to be used on the DNN. The configuration information may include AI / ML model information, supported AI / ML service operation information, such as AI / ML splitting and / or PC5-based AI / ML services, supported PC5-based AI / ML services, ProSe information of the PC5-based AI / ML service including discovery codes.

[0151] At 860, after retrieving configuration information for the PC5-based AI / ML service from the application server 885 at 850, when the WTRU-A 815 uses the PC5-based AI / ML service (such as AI / ML split), the WTRU-A 815 may attempt to discover a WTRU (such as the WTRU-B 805) that supports the PC5-based AI / ML service in the configuration information from the AF 885.

[0152] Figure 9 Method 900 for dynamic WTRU configuration for local AI / ML services is illustrated. Method 900 includes registering the ability to perform AI / ML operations at 910. At 920, method 900 includes updating the configuration regarding PC5-based operations. At 930, method 900 includes performing discovery for PC5-based AI / ML split.

[0153] Dynamic AI / ML split negotiation between a WTRU and an AF is described herein, where the availability of PC5-based AI / ML and server-based AI / ML is considered. As will be described in more detail below, the WTRU and the AI / ML application server utilize the expected performance and communication performance analysis information of PC5-based AI / ML to negotiate the AI / ML split point. The AI / ML application server or the 5GS may trigger the discovery process for the PC5-based AI / ML service and request a report on the expected performance of PC5-based AI / ML. When the WTRU is triggered for AI / ML split, the WTRU may attempt to discover any WTRU that supports the PC5-based AI / ML split. After checking the availability of the PC5-based AI / ML split, the WTRU may use the PC5-based AI / ML to measure the expected performance of the AI / ML operation based on the performance data of the discovered WTRU.

[0154] After measuring the expected performance, the WTRU may negotiate the AI / ML split with the AI / ML server. When determining the split point, the WTRU and the application server may consider the expected AI / ML performance including PC5-based AI / ML as the expected performance of the WTRU.

[0155] Figure 10 A signaling diagram 1000 for AI / ML split negotiation with network analysis is illustrated. Diagram 1000 includes WTRU-1 1015, WTRU-2 1005, AMF 1035, PCF 1045, SMF 1065, UPF 1055, UDR 1095, NWDAF 1085, NEF 1075, and AI / ML AF 1025, which are communicatively coupled for the signaling described below.

[0156] At 1010, signaling diagram 1000 includes that WTRU-11015 can determine that it is appropriate to perform AI / ML splitting on at least one in-use AI / ML service.

[0157] At 1020, when WTRU-11015 knows that a PC5-based AI / ML service is available, WTRU-11015 attempts to discover a WTRU (e.g., WTRU-21005) that supports the PC5-based AI / ML service. Based on the WTRU-21005 capability information and the supported AI / ML model information, WTRU-11015 can derive the expected performance value of the PC5-based AI / ML.

[0158] At 1030, WTRU-11015 can send a request for AI / ML splitting to AI / ML AF 1025. WTRU-11015 can include the expected performance for the PC5-based AI / ML operations available to WTRU-11015, as well as information on any discovered WTRUs for the PC5-based AI / ML operations.

[0159] At 1040, after receiving the AI / ML splitting request via request 1030, AI / ML AF1025 can request 5GC (NWDAF 1085 or NEF 1075) to provide any analysis information on the communication performance of WTRU-11015.

[0160] At 1050, when 5GC (e.g., NWDAF 1085 or NEF 1075) receives the analysis information request from AI / ML AF 1025, 5GC can request data related to the WTRU from relevant network functions (e.g., SMF 1065, UPF 1055).

[0161] At 1060, after collecting the data, 5GC (e.g., NWDAF 1085 or NEF 1075) can, according to the request of AF 1025, derive the analysis information related to the communication performance of WTRU-11015 and notify the analysis result to AI / ML AF1025.

[0162] At 1070, the AI / ML AF 1025 can determine the AI / ML split point based on the analysis information from the 5GC and the expected performance of PC5-based AI / ML from the WTRU-11015. Based on the expected performance of the WTRU and the analysis data of the NW's communication performance of the WTRU, the AI / ML AF can determine an appropriate split point for the AI / ML split. For example, based on the communication performance of the WTRU-11015, the latency of sending split data is greater than the processing latency of the PC5-based AI / ML operation. Based on the expected performance of the PC5-based AI / ML, the AI / ML AF 1025 can determine to let the WTRU-11015 be responsible for more computational layers for the AI / ML operation.

[0163] At 1080, the AI / ML AF 1025 can provide an AI / ML split response to the WTRU-11015 with the recommended AI / ML split operation.

[0164] Alternatively or additionally, the 5GC can use a network function (here the AI / ML assistance function) to determine the AI / ML split between the WTRU and the AI / ML AF 1025. In this case, at 1040, the AI / ML AF 1025 can request a recommendation on the AI / ML split operation from the AI / ML assistance function. The AI / ML AF 1025 can notify the AI / ML assistance function of the expected performance of the PC5-based AI / ML of the WTRU-11015. At 1040, the AI / ML assistance function can request analysis information from another NF in the 5GC. After receiving the analysis information, at 1060, the AI / ML assistance function can determine the recommended AI / ML split point and notify the result to the AI / ML AF 1025 at 1060.

[0165] Figure 11 A signaling diagram 1100 for AI / ML split negotiation with availability reporting for PC5-based AI / ML is illustrated. An AI / ML split negotiation process with network-triggered PC5-based AI / ML discovery can be used. The diagram 1100 includes the WTRU-11115, WTRU-21105, AMF 1135, PCF 1145, SMF 1165, UPF 1155, UDR 1195, NWDAF1185, NEF 1175, and AI / ML AF 1125, which are communicatively coupled for the following signaling.

[0166] At 1110, the signaling diagram 1100 includes that the WTRU-11115 can determine that it is appropriate to perform an AI / ML split on at least one of the in-use AI / ML services.

[0167] At 1120, the WTRU-11115 may request an AI / ML split from the AI / ML AF 1125. The request may include the expected performance of PC5-based AI / ML.

[0168] At 1130, after receiving the AI / ML split request at 1120, the AI / ML AF 1125 requests the 5G Core Network (referred to as 5GC) (e.g., NWDAF 1185 or NEF 1175) to provide any analytical information on the communication performance of the WTRU-11115.

[0169] At 1140, when the 5GC (e.g., NWDAF 1185 or NEF 1175) receives the analytical information request from the AF 1125 at 1130, the 5GC (e.g., NWDAF 1185 or NEF 1175) may request data related to the WTRU from relevant network functions (e.g., SMF 1165, UPF1155, PCF 1145).

[0170] At 1150, when the WTRU-11115 is available for PC5-based AI / ML split, the 5GC (e.g., PCF 1145) may send a request for a status report on the availability of PC5-based AI / ML split to the WTRU-11115. The status report can be requested through a control network function (e.g., PCF1145 or NWDAF1185).

[0171] At 1160, after receiving the request for a status report on the availability of PC5-based AI / ML split at 1150, the WTRU-11115 may attempt to discover any WTRU that supports PC5-based AI / ML split.

[0172] At 1170, based on the capability information of the discovered WTRU (such as WTRU-21105) and the supported AI / ML model information, the WTRU-11115 may derive the expected performance value of PC5-based AI / ML. In response to 1150, the WTRU-11115 may send a discovery result report of the PC5-based AI / ML split with the expected performance and / or information of the discovered WTRU to the PC51145 or NWDAF 1185.

[0173] At 1180, the NWDAF 1185 may derive analytical information related to the communication performance of the WTRU-11115 according to the request of the AF and send the analysis result to the AI / ML AF 1125. The 5GC may include a status report of PC5-based AI / ML, including the expected performance value.

[0174] At 1190, the AI / ML AF 1125 can determine the AI / ML split point based on the analysis information from the 5GC and the expected performance of PC5-based AI / ML. That is, the AI / ML AF (which is an external entity) receives information from the 5GC (such as, for example, the NEF 1175 or the NWDAF 1185). For example, based on the communication performance of the WTRU-11115, the latency of sending split data is greater than the processing latency of PC5-based AI / ML operations. Based on the expected performance of PC5-based AI / ML, the AI / ML AF 1125 can decide to let the WTRU-11115 be responsible for more computational layers for AI / ML operations.

[0175] At 1195, the AI / ML AF 1125 can provide an AI / ML split response to the WTRU-11115 with the recommended AI / ML split operation.

[0176] Alternatively or additionally, the 5GC can use a network function (such as an AI / ML assistance function) to determine the AI / ML split between the WTRU-11115 and the AI / ML AF 1125. In this case, at 1130, the AI / ML AF 1125 can request a recommendation on the AI / ML split operation from the AI / ML assistance function. The AI / ML assistance function can request analysis information from another NEF 1175 in the 5GC at 1130. After receiving the analysis information including the communication performance analysis information and the expected performance of PC5-based AI / ML, the AI / ML assistance function can determine the recommended AI / ML split point at 1180 and notify the result to the AI / ML AF 1125 at 1180.

[0177] Figure 12 Method 1200 for dynamic AI / ML split negotiation between a WTRU and an AF considering the availability of PC5-based AI / ML and server-based AI / ML is illustrated. At 1210, method 1200 includes determining the AI / ML split. At 1220, method 1200 includes performing discovery associated with the AI / ML split. At 1230, method 1200 includes deciding on an appropriate split point for the AI / ML split.

[0178] Complementary split decisions between the AS and the WTRU can be used. In some scenarios, especially if the PC5 assistance is to be handled by the network / WTRU and may be transparent to the application server, it may be beneficial and useful to make the WTRU part of the split point decision together with the server. Understand the split process and focus on the 5GS management of this complementary split process. For example, in a simple scenario, the application server can decide the AI / ML split point between the AS and the WTRU-A, such that the AS processes layers 16 - 24 and the WTRU processes layers 1 - 15. The AS may not necessarily know that PC5 is used and does not process in this case.

[0179] To assist with its tasks (layers 1 - 15), the WTRU can initiate PC5 discovery and PC5-assisted AI / ML, such that layers 1 - 15 are split. The WTRU can perform local splitting of its own part (layers 1 - 15) between the WTRU and other assisting WTRUs. The network can assist the WTRU by providing PC5 information such as discovery codes without involving the AS in the PC5 aspect. This may be a sub-optimal usage scenario but is beneficial because sometimes some applications do not need to know the PC5 aspect when performing model split AI / ML operations. This can be referred to as the AI / ML split decision performed by the AS at the first stage (as a broad or AS-initiated split decision), and the UE-side AI / ML split decision (as a local or WTRU-initiated split decision).

[0180] Figure 13 Figure 1300 illustrates an example of a two-stage AI / ML split process. Figure 13 Figure 1300 illustrates an example of the process of AI / ML split point decision. For example, at step 11310, first, the AI / ML application server can decide to perform AI / ML operations from layer 16 to layer 24 at the application server and the UE performs AI / ML operations for layers 1 to 15. Based on this, the AI / ML model including layers 1 to 15 can be distributed to the WTRU. After this decision, the WTRU can decide to perform further splitting of the AI / ML operations between the WTRUs by using PC5-based AI / ML cooperation. For example, at step 21320, WTRU-A can perform 1 - 5, while WTRU-B can perform layers 6 - 15. Based on this split decision, WTRU-A and WTRU-B can share the AI / ML model including layers 1 - 5 and layers 6 - 15.

[0181] In this example, FIG. 1300 illustrates a wide / AS-initiated split decision at 1310 and a local / WTRU-initiated split decision at 1320. Based on the AS-initiated split decision, it can be decided that the AS performs AI / ML operations including layers 16-24, while the WTRU performs AI / ML operations including layers 1-15. Based on the AS-initiated split decision, an appropriate AI / ML model including layers 1-15 can be shared with the WTRU. After receiving the AS-initiated split decision, at 1320, the local / WTRU-initiated split point can be utilized. For example, based on the WTRU-initiated split decision, it can be decided that the AS with layers 16-24 remains unchanged from 1310, and WTRU-A performs AI / ML operations including layers 1-5, and WTRU-B performs AI / ML operations including layers 6-15.

[0182] Figure 14 Signaling diagram 1400 illustrating the AI / ML split complementary split process. FIG. 1400 includes WTRU-A 1415, WTRU-B 1405, AMF 1435, UPF 1455, SMF 1465, PCF 1445, and AS 1475, which are communicatively coupled for the following signaling.

[0183] At 1410, FIG. 1400 includes WTRU-A 1415 determining that the AI / ML split is beneficial for some of the AI / ML services in use. For example, WTRU-A 1515 may want to use a certain service to initiate the AI / ML split.

[0184] At 1420, WTRU-A 1515 may send a request for the AI / ML split to AS 1475. For example, request 1420 may include AI / ML based on the expected WTRU performance.

[0185] At 1450, AS 1475 may perform a refined wide split decision. For example, AS 1475 may determine the AI / ML split point based on the expected performance from WTRU-A 1415. For example, AS 1475 may process layers 16-24, while the WTRU side processes layers 1-15. AS 1475 may use analysis information from the 5GC network (not shown in this figure for simplicity).

[0186] At 1460, AS 1475 may signal a response for the AI / ML split to WTRU-A 1415. The response may include the wide AI / ML split operation.

[0187] At 1480, the WTRU-A 1415 may retrieve PC5 information from the PCF 1445. This PC5 information may be information related to AI / ML operations, such as, for example, discovery codes.

[0188] At 1430, when the WTRU-A 1415 knows that a PC5-based AI / ML service is available, the WTRU-A 1415 may attempt to discover a WTRU (such as the WTRU-B 1405) that supports the PC5-based AI / ML service.

[0189] At 1470, based on the capability information of the WTRU-B 1405 and the supported AI / ML model information, the WTRU-A 1415 may further include the WTRU-B 1405 in the AI / ML operation by further splitting the WTRU side layer.

[0190] In a second scenario, for example, the WTRU-A 1415 may make its own local splitting decision independent of the AS 1475. The WTRU-A 1415 may pre-provide information including aggregation information and performance expectations to the AS 1475 (such as not providing PC5-related information of other WTRUs to the AS 1475). For example, if the WTRU-A 1415 notifies the AS 1475 that it has more capabilities (after checking nearby available devices), the WTRU-A 1415 may send new expected performance, and the AS 1475 may provide more layers for the WTRU-A 1415 to process (such as providing layers 1 - 20 to process). The WTRU-A 1415 may use its own local splitting decision mechanism to further split this decision to the discovered WTRUs.

[0191] Figure 15 A signaling diagram 1500 for an AI / ML splitting complementary splitting process enhanced with aggregation information is illustrated. The diagram 1500 includes a WTRU-A 1515, a WTRU-B 1505, an AMF 1535, a UPF 1555, an SMF 1565, a PCF 1545, and an AS 1575, which are communicatively coupled for the following signaling.

[0192] At 1510, the diagram 1500 includes the WTRU-A 1515 determining that AI / ML splitting is beneficial for some AI / ML services in use. For example, when the WTRU-A 1515 is experiencing poor communication conditions and is attempting to send a large amount of data to an AI / ML application server for AI / ML operations, the WTRU-A 1515 may decide to perform AI / ML splitting to reduce the radio resources required to send the results to the AI / ML AS 1575 for sending to the AI / ML application server.

[0193] For example, without AI / ML splitting, WTRU-A 1515 may need to send the results to AI / ML AS 1575 even when there is poor channel quality. With AI / ML splitting, WTRU-A 1515 can send partial results to WTRU-B 1505, and WTRU-B 1505 can perform the remaining AI / ML task(s). WTRU-B 1505 can send the results to AI / ML AS 1575 with better channel quality than WTRU-A 1515.

[0194] At 1520, PCF 1545 communicates with WTRU-A 1515 to retrieve PC5 AI / ML information from 5GC / PCF. At 1530, WTRU-A 1515 performs discovery of additional WTRUs such as, for example, WTRU-B 1505. At 1520 and 1530, when WTRU-A 1515 knows that PC5-based AI / ML services are available, WTRU-A 1515 can attempt to discover WTRUs (such as WTRU-B 1505) that support PC5-based AI / ML services. Based on the capability information of WTRU-B 1505 and the supported AI / ML model information, WTRU-A 1515 can derive the expected performance value of PC5-based AI / ML.

[0195] At 1540, WTRU-a 1515 can send or aggregate the expected performance to AS1575. For example, WTRU-A 1515 can request AI / ML splitting of the AI / ML application function by means of an aggregated performance indication. For example, by comparing the transmission latency of the raw data used to perform some layers of AI / ML operations in the server, and the aggregated computing power and available memory of the discovered WTRU are sufficient to perform some layers of AI / ML operations, WTRU-A 1515 can request the AI / ML splitting operation of the AI / ML application function by indicating the aggregated computing power and available memory.

[0196] At 1550, AS1575 can perform a fine-grained extensive splitting decision. The AI / ML AF can determine the AI / ML splitting point based on the analysis information from 5GC and the expected performance from WTRU-A 1515. For example, AS1575 can process layers 21-24 while the WTRU side processes layers 1-20.

[0197] At 1560, AS1575 can respond to WTRU-A 1515 with an AI / ML splitting response that has the recommended AI / ML splitting operation.

[0198] At 1570, the WTRU-A 1515 can perform a local AI / ML splitting mechanism based on information received from the AS 1575 to split layers among WTRUs (such as, for example, the WTRU (WTRU-B 1505) that has been discovered at 1530).

[0199] Although the features and elements have been described above in a particular combination, one of ordinary skill in the art will understand that each feature or element can be used alone or in combination with other features and elements. In addition, the methods described herein can be implemented in a computer program, software, or firmware that is incorporated into a computer-readable medium for execution by a computer or processor. Examples of computer-readable media include electronic signals (transmitted via wired or wireless connections) and computer-readable storage media. Examples of computer-readable storage media include, but are not limited to, read-only memory (ROM), random access memory (RAM), registers, cache memory, semiconductor memory devices, magnetic media (such as internal hard disks and removable disks), magneto-optical media, and optical media (such as CD-ROM disks and digital versatile disks (DVDs)). A processor associated with the software can be used to implement a radio frequency transceiver for use in a WTRU, UE, terminal, base station, RNC, or any host computer.

Claims

1. A method for PC5-based AI / ML performed by a wireless transmission and reception unit (WTRU), the method comprising: Transmit a discovery request message, the discovery request message including a discovery code for PC5-based AI / ML and a requested AI / ML model and performance level of the AI / ML; Receive a discovery response message, the discovery response message including an identification of the AI / ML model, a performance level, and target user information; Configure a PC5 connection based on the discovery response message; and Negotiate a PC5-based AI / ML split.

2. The method according to claim 1, wherein, The discovery request message allows selection of a device for AI / ML.

3. The method according to any one of the above claims, wherein, The device is a split WTRU.

4. The method according to any one of the above claims, wherein, Select a device based on the discovery response.

5. The method according to any one of the above claims, wherein, The discovery response is from the device.

6. The method according to any one of the above claims, further comprising performing a registration process capable of providing an ability indication for splitting.

7. The method according to any one of the above claims, further comprising determining the availability of local services in the area of the WTRU to be used in the splitting.

8. The method according to any one of the above claims, wherein, The negotiated split is based on the expected performance of the WTRU.

9. The method according to any one of the above claims, wherein, The negotiated split is based on the performance data of the WTRU.

10. The method according to any one of the above claims, wherein, The configured connection is configured based on the performance level of the AI / ML.

11. A wireless transmission and reception unit (WTRU) comprising: A processor; and A transceiver communicatively coupled to the processor for: Transmit a discovery request message, the discovery request message including a discovery code for PC5-based AI / ML and a requested AI / ML model and performance level of the AI / ML; Receive a discovery response message, the discovery response message including an identification of the AI / ML model, a performance level, and target user information; Configure a PC5 connection based on the discovery response message; and Negotiate a PC5-based AI / ML split.

12. The WTRU according to claim 11, wherein, The discovery request message allows selection of a device for AI / ML.

13. The WTRU according to any one of claims 11-12, wherein, The WTRU is a split WTRU.

14. The WTRU according to any one of claims 11-13, wherein, Select the WTRU based on the discovery response message.

15. The WTRU according to any one of claims 11 - 14, wherein, The discovery response message is from the WTRU.

16. The WTRU according to any one of claims 11 - 15, wherein, The processor and transceiver further operate to perform a registration process capable of providing an indication of the ability for the split.

17. The WTRU according to any one of claims 11 - 16, wherein, The processor and transceiver further operate to determine the availability of local services in the area of the WTRU to be used in the split.

18. The WTRU according to any one of claims 11 - 17, wherein, The negotiated split is based on the expected performance of the WTRU.

19. The WTRU according to any one of claims 11 - 18, wherein, The negotiated split is based on the performance data of the WTRU.

20. The WTRU according to any one of claims 11 - 19, wherein, The configured connection is configured based on the performance level of the AI / ML.

Images