Artificial intelligence machine learning functionality in a wireless communication system

Abstract

Various aspects of the present disclosure relate to a first network entity for wireless communication. The first network entity may be configured to, capable of, or operable to receive, from an application entity, a first request to support an AIML functionality for a UE served by a first wireless communication network; determine a second network entity for supporting the AIML functionality whilst the UE is to be served by a second wireless communication network; and transmit, to the second network entity, a second request to support the AIML functionality.

Description

ARTIFICIAL INTELLIGENCE MACHINE LEARNING FUNCTIONALITY IN A WIRELESS COMMUNICATION SYSTEMTECHNICAL FIELD

[0001] The present disclosure relates generally to wireless communication, including the Artificial Intelligence / Machine Learning (AI / ML) functionality in the wireless communication system.BACKGROUND

[0002] A wireless communications system may include one or multiple network communication devices, which may be otherwise knowns as network equipment (NE) supporting wireless communications for one or multiple user communication devices, which may be otherwise known as user equipment (UE), or other suitable terminology. The wireless communications system may support wireless communications with one or multiple user communication devices by utilizing resources of the wireless communication system (e.g., time resources (e.g., symbols, slots, subframes, frames, or the like) or frequency resources (e.g., subcarriers, carriers, or the like). Additionally, the wireless communications system may support wireless communications across various radio access technologies including third generation (3G) radio access technology, fourth generation (4G) radio access technology, fifth generation (5G) radio access technology, among other suitable radio access technologies beyond 5G (e.g., sixth generation (6G)).SUMMARY

[0003] An article “a” before an element is unrestricted and understood to refer to “at least one” of those elements or “one or more” of those elements. The terms “a,” “at least one,” “one or more,” and “at least one of one or more” may be interchangeable. As used herein, including in the claims, “or” as used in a list of items (e.g., a list of items prefaced by a phrase such as “at least one of’ or “one or more of’ or “one or both of’) indicates an inclusive list such that, for example, a list of at least one of A, B, or C means A or B or C or AB or AC or BC or ABC (i.e., A and B and C). Also, as used herein, the phrase “based on” shall not be construed as a reference to a closed set of conditions. For example, an exampleDocket No. SMM920250065-GR-NPstep that is described as “based on condition A” may be based on both a condition A and a condition B without departing from the scope of the present disclosure. In other words, as used herein, the phrase “based on” shall be construed in the same manner as the phrase “based at least in part on. Further, as used herein, including in the claims, a “set” may include one or more elements.

[0004] The following abbreviations are herewith defined, at least some of which are referred to within the following description: 3 GPP - Third Generation Partnership Project; 5GC - 5G Core Network; 5GS - 5G System; AD AES - Application Data Analytics Enablement Server; AF - Application Function; Al - Artificial Intelligence; AIMLE - AI / ML Enablement; API - Application Programming Interface; ASP - Application Service Provider; CSP - Cloud Service Provider; ECSP - Edge Computing Service Provider; ECS- Edge Configuration Server; ECS-ER - Edge Configuration Server Edge Relay; EDN - Edge Data Network (DN); EEC - Edge Enabling Client; EEL - Edge Enabler Layer; eV2X- Evolved Vehicle-to-everything; FL - Federated Learning; H-ECS - Hardware-based Edge Computing Server; HFL - Horizontal Federated Learning; HPLMN - Home PLMN; HR - Home Routed; LBO - Local Breakout; LMC - Local Management Client; LMF - Location Management Function; LMS - Location Management Service; ML - Machine Learning; MNO - Mobile Network Operator; NE - Network Entity; NF - Network Function; NEF - Network Exposure Function; NPN - Non-Public Network; NSCE - Network Slice Capability Enablement; NWDAF - Network Data Analytics Function; 0AM - Operations, Administration and Maintenance; PaaS - Platform as a Service; PLMN - Public Land Mobile Network; SEAL - Service Enabler Architecture Layer; SNPN - Standalone NPN; TL - Transfer Learning; V2X - Vehicle-to-everything; VAL - Vertical Application Layer; VAS - Vertical Application layer Server; V-ECS - Virtualized Edge Computing Server; V- EES - Virtualized Edge Enabling Server; VFL - Vertical Federated Learning; VPLMN - Visiting PLMN.

[0005] A first network entity for wireless communication is described. The first network entity may be configured to, capable of, or operable to perform one or more operations as described herein. For example, the first network entity may include at least one memory, and at least one processor coupled with the at least one memory andDocket No. SMM920250065-GR-NPconfigured to cause the first network entity to: receive, from an application entity, a first request to support an AIML functionality for a UE served by a first wireless communication network; determine a second network entity for supporting the AIML functionality whilst the UE is to be served by a second wireless communication network; and transmit, to the second network entity, a second request to support the AIML functionality.

[0006] A method performed or performable by the first network entity is described herein. The method may comprise receiving, from an application entity, a first request to support an AIML functionality for a UE served by a first wireless communication network; determining a second network entity for supporting the AIML functionality whilst the UE is to be served by a second wireless communication network; and transmitting, to the second network entity, a second request to support the AIML functionality.

[0007] A processor for wireless communication is described. The processor may be configured to, capable of, or operable to perform one or more operations as described herein. For example, the processor may comprise at least one controller coupled with at least one memory and configured to cause the processor to: receive, from an application entity, a first request to support an AIML functionality for a UE served by a first wireless communication network; determine a second network entity for supporting the AIML functionality whilst the UE is to be served by a second wireless communication network; and transmit, to the second network entity, a second request to support the AIML functionality.

[0008] A UE for wireless communication is described. The UE may be configured to, capable of, or operable to perform one or more operations as described herein. For example, the UE may include at least one memory, and at least one processor coupled with the at least one memory and configured to cause the UE to: determine to support a first network entity with an AIML functionality, wherein the UE is served by a first wireless communication network; determine that the UE is to be served by a second wireless communication network; and determine a second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network.Docket No. SMM920250065-GR-NP

[0009] A method performed or performable by the UE is described herein. The method may comprise determining to support a first network entity with an AIML functionality, wherein the UE is served by a first wireless communication network; determining that the UE is to be served by a second wireless communication network; and determining a second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network.

[0010] A processor for wireless communication is described. The processor may be configured to, capable of, or operable to perform one or more operations as described herein. For example, the processor may comprise at least one controller coupled with at least one memory and configured to cause the processor to: determine to support a first network entity with an AIML functionality, wherein the UE is served by a first wireless communication network; determine that the UE is to be served by a second wireless communication network; and determine a second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network.BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 illustrates an example of a wireless communications system in accordance with aspects of the present disclosure.

[0012] Figure 2 illustrates an example of an on-network AIMLE functional model in accordance with aspects of the present disclosure.

[0013] Figure 3 illustrates an example of ML model lifecycle enablement in accordance with aspects of the present disclosure.

[0014] Figure 4 illustrates an example of relationships involved in an edge computing service in accordance with aspects of the present disclosure.

[0015] Figure 5 illustrates an example of a UE roaming between wireless communication networks with different AIMLE servers in accordance with aspects of the present disclosure.

[0016] Figure 6 illustrates an example architecture for supporting AIML-enabled ADAE analytics in accordance with aspects of the present disclosure.Docket No. SMM920250065-GR-NP

[0017] Figure 7 illustrates an example of a process flow for AIMLE in accordance with aspects of the present disclosure.

[0018] Figure 8 illustrates an example of a process flow for AIMLE in accordance with aspects of the present disclosure.

[0019] Figure 9 illustrates an example of a UE 900 in accordance with aspects of the present disclosure.

[0020] Figure 10 illustrates an example of a processor 1000 in accordance with aspects of the present disclosure.

[0021] Figure 11 illustrates an example of a NE 1100 in accordance with aspects of the present disclosure.

[0022] Figure 12 illustrates a flowchart of a method 1200 performed by a NE in accordance with aspects of the present disclosure.DETAILED DESCRIPTION

[0023] A wireless communication system (which may be referred to as a wireless communication network herein), may comprise one or more UE and NE. In a roaming scenario, a UE may move from coverage of a first PLMN to coverage of a second PLMN. Whilst roaming, the UE, the first PLMN and the second PLMN may be configured to maintain connectivity between the UE and the NE.

[0024] The wireless communication system may be configured to support AIMLE. Such scenarios may involve one or more AIMLE servers which support the wireless communication system (and the UE) with an AI / ML functionality (e.g., an AIMLE service, an Al service or an AI / ML task). However, roaming from the first PLMN to the second PLMN tends to affect the performance (e.g., continuity) of the AI / ML functionality.

[0025] Examples described herein generally relate to a first AIMLE server arranged to support the AI / ML functionality whilst the UE is roaming. In some examples described herein, the first AIMLE server is arranged to determine a second AIMLE server for supporting the AI / ML functionality whilst the UE us being served by the second PLMN.Docket No. SMM920250065-GR-NPThis tends to reduce the impact of UE roaming on the performance of the AI / ML functionality.

[0026] Aspects of the present disclosure are described in the context of a wireless communications system.

[0027] Figure 1 illustrates an example of a wireless communications system 100 in accordance with aspects of the present disclosure. The wireless communications system 100 may include one or more NE 102, one or more UE 104, and a core network (CN) 106. The wireless communications system 100 may support various radio access technologies. In some implementations, the wireless communications system 100 may be a 4G network, such as an LTE network or an LTE- Advanced (LTE-A) network. In some other implementations, the wireless communications system 100 may be a NR network, such as a 5G network, a 5G- Advanced (5G-A) network, or a 5G ultrawideband (5G-UWB) network. In other implementations, the wireless communications system 100 may be a combination of a 4G network and a 5G network, or other suitable radio access technology including Institute of Electrical and Electronics Engineers (IEEE) 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20. The wireless communications system 100 may support radio access technologies beyond 5G, for example, 6G. Additionally, the wireless communications system 100 may support technologies, such as time division multiple access (TDMA), frequency division multiple access (FDMA), or code division multiple access (CDMA), etc.

[0028] The one or more NE 102 may be dispersed throughout a geographic region to form the wireless communications system 100. One or more of the NE 102 described herein may be or include or may be referred to as a network node, a base station, a network element, a network function, a network entity, a radio access network (RAN), a NodeB, an eNodeB (eNB), a next-generation NodeB (gNB), or other suitable terminology. An NE 102 and a UE 104 may communicate via a communication link, which may be a wireless or wired connection. For example, an NE 102 and a UE 104 may perform wireless communication (e.g., receive signalling, transmit signalling) over a Uu interface.

[0029] An NE 102 may provide a geographic coverage area for which the NE 102 may support services for one or more UEs 104 within the geographic coverage area. ForDocket No. SMM920250065-GR-NPexample, an NE 102 and a UE 104 may support wireless communication of signals related to services (e.g., voice, video, packet data, messaging, broadcast, etc.) according to one or multiple radio access technologies. In some implementations, an NE 102 may be moveable, for example, a satellite associated with a non-terrestrial network (NTN). In some implementations, different geographic coverage areas associated with the same or different radio access technologies may overlap, but the different geographic coverage areas may be associated with different NE 102.

[0030] The one or more UE 104 may be dispersed throughout a geographic region of the wireless communications system 100. A UE 104 may include or may be referred to as a remote unit, a mobile device, a wireless device, a remote device, a subscriber device, a transmitter device, a receiver device, or some other suitable terminology. In some implementations, the UE 104 may be referred to as a unit, a station, a terminal, or a client, among other examples. Additionally, or alternatively, the UE 104 may be referred to as an Internet-of-Things (loT) device, an Internet-of-Everything (loE) device, or machine-type communication (MTC) device, among other examples.

[0031] A UE 104 may be able to support wireless communication directly with other UEs 104 over a communication link. For example, a UE 104 may support wireless communication directly with another UE 104 over a device-to-device (D2D) communication link. In some implementations, such as vehicle-to-vehicle (V2V) deployments, V2X deployments, or cellular-V2X deployments, the communication link may be referred to as a sidelink. For example, a UE 104 may support wireless communication directly with another UE 104 over a PC5 interface.

[0032] An NE 102 may support communications with the CN 106, or with another NE 102, or both. For example, an NE 102 may interface with other NE 102 or the CN 106 through one or more backhaul links (e.g., SI, N2, N2, or network interface). In some implementations, the NE 102 may communicate with each other directly. In some other implementations, the NE 102 may communicate with each other or indirectly (e.g., via the CN 106. In some implementations, one or more NE 102 may include subcomponents, such as an access network entity, which may be an example of an access node controller (ANC). An ANC may communicate with the one or more UEs 104 through one or more otherDocket No. SMM920250065-GR-NPaccess network transmission entities, which may be referred to as a radio heads, smart radio heads, or transmission-reception points (TRPs).

[0033] The CN 106 may support user authentication, access authorization, tracking, connectivity, and other access, routing, or mobility functions. The CN 106 may be an evolved packet core (EPC), or a 5GC, which may include a control plane entity that manages access and mobility (e.g., a mobility management entity (MME), an access and mobility management functions (AMF)) and a user plane entity that routes packets or interconnects to external networks (e.g., a serving gateway (S-GW), a Packet Data Network (PDN) gateway (P-GW), or a user plane function (UPF)). In some implementations, the control plane entity may manage non-access stratum (NAS) functions, such as mobility, authentication, and bearer management (e.g., data bearers, signal bearers, etc.) for the one or more UEs 104 served by the one or more NE 102 associated with the CN 106.

[0034] The CN 106 may communicate with a packet data network over one or more backhaul links (e.g., via an SI, N2, N2, or another network interface). The packet data network may include an application server. In some implementations, one or more UEs 104 may communicate with the application server. A UE 104 may establish a session (e.g., a protocol data unit (PDU) session, or the like) with the CN 106 via an NE 102. The CN 106 may route traffic (e.g., control information, data, and the like) between the UE 104 and the application server using the established session (e.g., the established PDU session). The PDU session may be an example of a logical connection between the UE 104 and the CN 106 (e.g., one or more network functions of the CN 106).

[0035] In the wireless communications system 100, the NEs 102 and the UEs 104 may use resources of the wireless communications system 100 (e.g., time resources (e.g., symbols, slots, subframes, frames, or the like) or frequency resources (e.g., subcarriers, carriers)) to perform various operations (e.g., wireless communications). In some implementations, the NEs 102 and the UEs 104 may support different resource structures. For example, the NEs 102 and the UEs 104 may support different frame structures. In some implementations, such as in 4G, the NEs 102 and the UEs 104 may support a single frame structure. In some other implementations, such as in 5 G and among other suitable radio access technologies, the NEs 102 and the UEs 104 may support various frame structuresDocket No. SMM920250065-GR-NP(i.e., multiple frame structures). The NEs 102 and the UEs 104 may support various frame structures based on one or more numerologies.

[0036] One or more numerologies may be supported in the wireless communications system 100, and a numerology may include a subcarrier spacing and a cyclic prefix. A first numerology (e.g., fi=O) may be associated with a first subcarrier spacing (e.g., 15 kHz) and a normal cyclic prefix. In some implementations, the first numerology (e.g., .=Q) associated with the first subcarrier spacing (e.g., 15 kHz) may utilize one slot per subframe. A second numerology (e.g., jU=l ) may be associated with a second subcarrier spacing (e.g., 30 kHz) and a normal cyclic prefix. A third numerology (e.g., .=2) may be associated with a third subcarrier spacing (e.g., 60 kHz) and a normal cyclic prefix or an extended cyclic prefix. A fourth numerology (e.g., jU=3) may be associated with a fourth subcarrier spacing (e.g., 120 kHz) and a normal cyclic prefix. A fifth numerology (e.g., / r=4) may be associated with a fifth subcarrier spacing (e.g., 240 kHz) and a normal cyclic prefix.

[0037] A time interval of a resource (e.g., a communication resource) may be organized according to frames (also referred to as radio frames). Each frame may have a duration, for example, a 10 millisecond (ms) duration. In some implementations, each frame may include multiple subframes. For example, each frame may include 10 subframes, and each subframe may have a duration, for example, a 1 ms duration. In some implementations, each frame may have the same duration. In some implementations, each subframe of a frame may have the same duration.

[0038] Additionally or alternatively, a time interval of a resource (e.g., a communication resource) may be organized according to slots. For example, a subframe may include a number (e.g., quantity) of slots. The number of slots in each subframe may also depend on the one or more numerologies supported in the wireless communications system 100. For instance, the first, second, third, fourth, and fifth numerologies (i.e., / r=0, jU=l , / r=2, JU=3, / I=4) associated with respective subcarrier spacings of 15 kHz, 30 kHz, 60 kHz, 120 kHz, and 240 kHz may utilize a single slot per subframe, two slots per subframe, four slots per subframe, eight slots per subframe, and 16 slots per subframe, respectively. Each slot may include a number (e.g., quantity) of symbols (e.g., OFDM symbols). In some implementations, the number (e.g., quantity) of slots for a subframe may depend on aDocket No. SMM920250065-GR-NPnumerology. For a normal cyclic prefix, a slot may include 14 symbols. For an extended cyclic prefix (e.g., applicable for 60 kHz subcarrier spacing), a slot may include 12 symbols. The relationship between the number of symbols per slot, the number of slots per subframe, and the number of slots per frame for a normal cyclic prefix and an extended cyclic prefix may depend on a numerology. It should be understood that reference to a first numerology (e.g., / r=0) associated with a first subcarrier spacing (e.g., 15 kHz) may be used interchangeably between subframes and slots.

[0039] In the wireless communications system 100, an electromagnetic (EM) spectrum may be split, based on frequency or wavelength, into various classes, frequency bands, frequency channels, etc. By way of example, the wireless communications system 100 may support one or multiple operating frequency bands, such as frequency range designations FR1 (410 MHz - 7.125 GHz), FR2 (24.25 GHz - 52.6 GHz), FR3 (7.125 GHz - 24.25 GHz), FR4 (52.6 GHz - 114.25 GHz), FR4a or FR4-1 (52.6 GHz - 71 GHz), and FR5 (114.25 GHz - 300 GHz). In some implementations, the NEs 102 and the UEs 104 may perform wireless communications over one or more of the operating frequency bands. In some implementations, FR1 may be used by the NEs 102 and the UEs 104, among other equipment or devices for cellular communications traffic (e.g., control information, data). In some implementations, FR2 may be used by the NEs 102 and the UEs 104, among other equipment or devices for short-range, high data rate capabilities.

[0040] FR1 may be associated with one or multiple numerologies (e.g., at least three numerologies). For example, FR1 may be associated with a first numerology (e.g., / r=0), which includes 15 kHz subcarrier spacing; a second numerology (e.g., / r=l), which includes 30 kHz subcarrier spacing; and a third numerology (e.g., / r=2), which includes 60 kHz subcarrier spacing. FR2 may be associated with one or multiple numerologies (e.g., at least 2 numerologies). For example, FR2 may be associated with a third numerology (e.g., / r=2), which includes 60 kHz subcarrier spacing; and a fourth numerology (e.g., / r=3), which includes 120 kHz subcarrier spacing.

[0041] The wireless communication system 100 may comprise multiple PLMN operators. Each of the multiple PLMN operators may operate different NEs 102. In roaming, the UEs 104 may move out of coverage of a first NE 102 operated by a firstDocket No. SMM920250065-GR-NPPLMN operator and into coverage of a second NE 102 operated by a second PLMN operator. If the wireless communication system 100 is configured to support an AI / ML functionality, roaming of the UEs 104 tends to impact on the performance of the AI / ML functionality. Examples described herein generally relate to a first NE 102 in the first PLMN arranged to determine the second NE 102 in the second PLMN that is suitable for supporting the AI / ML functionality whilst the UEs 104 roaming.

[0042] Figure 2 is a diagram 200 illustrating an example of an on-network AIMLE functional model in accordance with aspects of the present disclosure. The diagram 200 comprises a UE 205, a 3GPP Network System 240, VAL server(s) 216, ML Repository 241 and an AIMLE server 236. The UE 205 comprises a VAL 210 and a SEAL 220. The VAL 210 comprises a VAL Client(s) 212. The SEAL 220 comprises an AIML-C 230 and an AIMLE client 232. The VAL server(s) 216 connects to the VAL Client(s) 212 via a VAL- UU 214 reference point. The AIMLE client 232 connects to the AIMLE server 236 via an AIML-UU 234 reference point. The 3GPP Network System 240 connects to the AIMLE server 236 via Network interfaces 235. The AIMLE server 236 connects to the VAL server(s) 216 via an AIML-S 231 reference point. The AIMLE server 236 connects to the ML Repository 241 via AIML-R 233 reference point. The AIMLE server 236 comprises an AIML-E 238 reference point.

[0043] Vertical-specific applications and edge applications tend to be the major consumers of 3 GPP-provided data analytics and AI / ML support services. The AIMLE service plays a role on the exposure of AI / ML services from different 3 GPP domains to the vertical / ASP in a unified manner on top of 3 GPP core network and 0AM. The AIMLE service also plays a role on defining, at a SEAL layer, value-add support services for assisting AI / ML services provided by the VAL 210, while being complementary to AI / ML support solutions provided in other 3 GPP domains.

[0044] Figure 2 illustrates the on-network functional model of AIML enablement. In the VAL 210, the VAL client 212 communicates with the VAL server 216 over VAL-UU reference point 214. The VAL-UU reference point 214supports both unicast and multicast delivery modes. The AIML enablement functional entities on the UE 205 and the server are grouped into AIMLE client(s) 232 and AIMLE server(s) 236 respectively.Docket No. SMM920250065-GR-NP

[0045] The AIMLE server 236 may be a SEAL server which includes of a common set of services for comprehensive enablement of AIML functionality. The AIMLE server 236 may comprise at least one of the following group of capabilities: support for applicationlayer ML model related aspects (e.g., model retrieval, model training, model monitoring, model selection, model update and / or model storage / discovery); assistance in AI / ML task transfer and split AI / ML operations; support HFL / VTL operations (e.g., FL member registration, FL grouping and FL-related events notification, VFL feature alignment, and / or HFL training); and support for AIMLE client registration, discovery, participation and / or selection.

[0046] The AIMLE client 232 functional entity acts as the application client supporting AIMLE services. The ML repository 241 is an entity that serves as: a registry for ML / FL members (e.g., application layer entities participating in an AI / ML operation) and as a repository for application layer ML model related information.

[0047] AIMLE is applicable to ML model lifecycle enablement which provides assistance for use cases where an ASP / VAL layer 210 aims to find and use other application entities to perform some ML operations (e.g., ML model inference) and the AIMLE server 236 acts as a mediator to accomplish that.

[0048] Figure 3 is a diagram 300 illustrating an example of ML model lifecycle enablement in accordance with aspects of the present disclosure. The ML model lifecycle enablement diagram 300 comprises a VAL server(s) 316 for ML model operational workflow and an AIMLE 330 for ML model lifecycle enablement. The VAL Server(s) 316 comprises data management, model training, model evaluation, model deployment and model inference. The AIMLE 330 comprises ML model related support which comprises model retrieval, model discovery and model storage. The AIMLE 330 further comprises ML operation related support which comprises VFL / HFL enablement, TL enablement, split AI / ML operation, Data management support and FL member support e.g., grouping, register and events. The AIMLE 330 further comprises AIMLE client support which comprises AIMLE register, discovery, participate and monitoring.Docket No. SMM920250065-GR-NP

[0049] The ML model lifecycle enablement diagram 300 is an exemplary grouping of AIMLE capabilities with respect to enablement. The AIMLE capabilities are described in Annex C.4 of TS 23.482 V19.1.0.

[0050] AIMLE may undertake at least one of: ML model related support capabilities (e.g., model retrieval, discovery and / or storage; for example, as described in procedures in clauses 8.2 and 8.11 of TS 23.482 V19.1.0); ML operation related support capabilities (e.g., VFL / HFL and TL enablement, Split AI / ML Operation support, Data management assistance, AI / ML task transfer, FL assistance in member grouping, registration and / or event notification; for example, as described in procedures in clauses 8.4, 8.6, 8.12, 8.14, 8.15-8.18 of TS 23.482 V19.1.0); and AIMLE client related support capabilities (e.g., AIMLE client registration, discovery, participation, monitoring, selection; for example, as described in procedures in clauses 8.7-8.10, 8.13 of TS 23.482 V19.1.0).

[0051] Figure 4 is a diagram 400 illustrating an example of relationships involved in an edge computing service in accordance with aspects of the present disclosure.

[0052] The diagram 400 comprises an ASP 401 which has an ASP service agreement(s) 411 with an End User 402. The ASP 401 also has an ECSP service agreement(s) 412 with ECSP A 403 and ECSP B 404. ECSP A 403 and ECSP B 404 have a Federation agreement 415. ECSP A 403 and ECSP B 404 also have an Edge service authorization(s) 413 with the End User 402. ECSP A 403 and ECSP B 404 also have a PLMN operator service agreement(s) 416 with the PLMN operator X (e.g., HPLMN) 406 and PLMN operator Y (e.g., VPLMN) 407. The PLMN operator X 406 and PLMN operator Y 407 have a service agreement 417. The PLMN operator X 406 and PLMN operator Y 407 have a PLMN subscription arrangement 414 with the End User 402.

[0053] There may be different business relationships that exist with respect to federation. In TS 23.558 VI 9.5.0 Annex B.2, the relationship of edge computing service providers 403 / 404, PLMN operators 406 / 407, ASPs 401 and end users 402 is illustrated in Figure 4, taking federation and roaming into account.

[0054] The end user 402 is the consumer of the applications provided by the ASP 401. The end user 402: may have the ASP service agreement 411 with a single or multipleDocket No. SMM920250065-GR-NPapplication service providers. The end user 402 may have a PLMN subscription arrangement 414 with a PLMN operator (e.g., HPLMN 406), and the UE used by the end user may register on the HPLMN 406 and network of its roaming partners; or an SNPN subscription arrangement with a SNPN operator (subscribed SNPN), and the UE used by the end user 402 may register on the subscribed SNPN and a serving SNPN. The end user 402 may have authorization to access edge services of a single or multiple ECSPs (e.g., ECSP A 403 and / or ECSP B 404).

[0055] The ASP 401 consumes the edge services (e.g., infrastructure, platform) provided by the ECSP (e.g., ECSP A 403 and / or ECSP B 404). The ASP 401 may have the ECSP service agreement 412 with a single or multiple ECSPs (e.g., ECSP A 403 and / or ECSP B 404).

[0056] The PLMN operator (e.g., PLMN operator X 406 and / or PLMN operator Y 407) provides connectivity between the end user 402 and the edge services provided by the ECSP (e.g., ECSP A 403 and / or ECSP B 404). The PLMN operator may have the PLMN operator service agreement 416 with a single or multiple ECSPs 403 / 404. The PLMN operator may have a service agreement for roaming including agreements for Edge Computing services, and / or federation with a single or multiple PLMN operators 406 / 407.

[0057] The ECSP 403 / 404 provides the edge services. The ECSP 403 / 404 may have the PLMN operator service agreement 416 with a single or multiple PLMN operators 406 / 407 which provide edge computing support. The ECSP 403 / 404 have federation partnership to share edge services with a single or multiple ECSPs 403 / 404. The ECSP 403 / 404 and the PLMN operator 406 / 407 may be part of the same organization.

[0058] Figure 5 is a diagram 500 illustrating an example of a UE roaming between wireless communication networks with different AIMLE servers in accordance with aspects of the present disclosure. The diagram 500 includes UE 505 being served by a first 3GPP Network (e.g., PLMN1, HPLMN) 540. The UE 505 comprises an AIMLE client 532 that connects to a first AIMLE server 536 via the first 3GPP Network 540. The UE 505 roams from the first 3 GPP Network 540 to a second 3 GPP Network (e.g., PLMN2, VPLMN) 542. The AIMLE client 532 then connects to a second AIMLE server 538. The first AIMLE server 536 and second AIMLE server 538 are connected to VAL servers 516. The firstDocket No. SMM920250065-GR-NPAIMLE server 536 connects to a first ML repository 541. The second AIMLE server 538 connects to a second ML repository 543.

[0059] In diagram 500, the UE 505 is roaming or expected to roam from PLMN1 540 to PLMN2 542 during an ongoing application session (e.g., AIML service, AIMLE service, ML process, FL process, ML / FL process)

[0060] AIMLE server 1 536 is mapped to PLMN1 540. Roaming may lead to an update of the serving AIMLE server for the UE. AIMLE client 532 may change AIMLE server while roaming to the target area covered by a different PLMN (e.g., PLMN2 542).

[0061] A change of the AIMLE server may comprise a change of the ML / FL server role for an ML / FL process. The second AIMLE Server 538 may undertake the ML / FL server role / functionality for the ongoing ML / FL process. A change of AIMLE server may not comprise a change to the ML / FL server role; e.g., the ML / FL server role remains the same in the first AIMLE server 536. This may not require a context transfer for the role / capability change, and only the serving AIMLE server changes.

[0062] The ML repository 541 / 543 may store ML models and ML / FL member information. If the AIMLE server changes (e.g., due to roaming), the ML repository 541 / 543 may also change based on the deployment (e.g., from the first ML repository 541 to the second ML repository 543). There are two scenarios which may require interactions among ML repositories:

[0063] Each AIMLE server 536 / 538 is deployed at the edge / cloud by the edge / cloud provider, and the ML repository is a single entity which is expected to store roaming info (e.g., multi-MNO info), and there is no update required based on the switching of the AIMLE server for the ongoing AI / ML service.

[0064] Each AIMLE server 536 / 538 is deployed by each MNO, and the ML repository is different per PLMN. The AIMLE server 536 / 538 may discover and subscribe to the target ML repository (e.g., the second ML repository 543) to receive information on the ML model and FL / ML members for the ongoing ML / FL process. This may involve migration and an app context transfer between ML repositories 541 / 543 or via the AIMLE servers 536 / 538.Docket No. SMM920250065-GR-NP

[0065] Figure 6 is a diagram 600 illustrating an example architecture for supporting AIML-enabled ADAE analytics in accordance with aspects of the present disclosure. The diagram comprises a VAL 610 and SEAL 620. The VAL 610 comprises VAL server(s) 616 connected to an AD AES 622 in SEAL 620 via ADAE-S 623. The AD AES 622 connects to AIMLE server 636 via AIML-X 637. The AIMLE server 636 connects to an ML Repository 641 via a AIML-R 633 reference point. ADAE-S 623 may be an architecture and AIML-X 637 may be a conceptual framework.

[0066] 3GPP SA6 in Rel-18 specified an application data analytics enablement functionality (e.g., AD AES 622, TS 23.436 V19.3.0). AD AES 622 is a SEAL functionality for providing end-to-end performance analytics (e.g., VAL server 616 performance). There are multiple analytics services defined in Rel-18 including: application performance analytics; slice-specific application performance analytics; UE-to-UE application performance analytics; location accuracy analytics Service; API analytics; slice usage pattern analytics; edge load analytics; edge computing preparation analytics; support for server-to-server performance analytics; support for collision detection analytics; support for location-related; UE group analytics; support for VAL performance analytics for tethered UEs; support for Application Layer AI / ML Member Capability Analytics; support DN Energy Efficiency analytics; and support ML Model Performance Degradation Detection.

[0067] With respect to ML-enabled analytics, diagram 600 illustrates the architecture representation including AIMLE (as specified in 3 GPP TS 23.482 VI 9.1.0) for supporting ML-enabled analytics in AD AES 622. In this representation, the AIML support capabilities serve AD AES 622 to enhance its analytics services. Based on the VAL request to provide ML-enabled analytics, AD AES 622 may consume AIMLE services (e.g., for ML model training for a given analytics ID) to derive application layer data analytics.

[0068] For the interaction between AIMLE server 636 and AD AES 622, AIML-X is introduced to support consuming AIMLE services for deriving ADAE analytics (e.g., VAL server 616 performance analytics).

[0069] The ML repository 641 is specified in 3GPP TS 23.482 V19.1.0 as a repository for the ML model and registry for the ML-related information (such as ML / FL members). This repository may be utilized by AD AES 622 via AIMLE server 636 (via AIML-R 633)Docket No. SMM920250065-GR-NPfor fetching ML-related information (e.g., trained ML model, ML / FL members) which is used for a given ADAE analytics event.

[0070] In 3GPP SA6, as part of EDGEAPP (see TS 23.558 V19.5.0) there is support for roaming (e.g., cross-ECS impacts). To support UEs that are roaming, the EEL uses ECSs provided in HPLMN and VPLMN. The EEC in the UE obtains EEL services from V-ECS and V-EES. EDGE-10 reference point is used between the H-ECS and the V-ECS. Two roaming models are supported for edge enabling applications: Local breakout (LBO) roaming architecture; and Home routed (HR) roaming architecture.

[0071] In both architecture options, the EDN is located in the V-PLMN and is accessed via an LBO or HR-SBO session. More details on the two options are provided in TS 23.558 VI 9.5.0 clause 6.2a. In both cases, there is no change of EES while roaming to the VPLMN. In addition, in SA6 as part of NSCE, there is support for service continuity while UE roaming. This functionality is provided to the VAL 610 to make predictive slice modification where NSCE service provider provides its services when connected to two PLMNs and has SLA with them. The NSCE server checks with 5GS (e.g., 0AM, 5GC) whether the serving slice is available and can offer the same performance at the target PLMN and make slice modification decision if needed.

[0072] Predictive slice modification in Inter-PLMN based slice service continuity procedures are specified in clause 9.13 of TS 23.435 V19.3.0. In this feature, the NSCE server initially receives an expected / predicted UE location / mobility change request outside a PLMN1 slice service area for one or more UEs within the VAL application session (e.g., such session can be a V2X session). Then, the NSCE server checks with 5GS (e.g., 0AM, 5GC) whether the serving slice is available and can offer the same performance at the target PLMN. The NSCE server evaluates the need for a slice modification (e.g., a slice lifecycle related trigger change). Based on this decision / recommendation, it provides the action to the 0AM of PLMN2 proactively, before UE mobility happens.

[0073] In Rel-20, in TR 23.700-83 V0.1.0, there is a key issue (#3) to study how AIMLE services can support mobility, including roaming scenarios where UEs access AI / ML services in visited networks, and cross-PLMN environments where AI / MLDocket No. SMM920250065-GR-NPoperations involve entities from different operators. This relates to both the continuity of AI / ML services and the federation across MNOs and CSPs / ECSPs.

[0074] In particular, as ML model inference increasingly moves toward the network edge to meet ultra-low latency and bandwidth efficiency demands (e.g., in real-time video analytic, industrial automation), maintaining inference continuity across mobility events becomes critical. When VAL UEs move across network boundaries or service areas, edge AIMLE instances may need to be reselected, dynamically discovered, or seamlessly replaced to ensure uninterrupted AI / ML inference execution and delivery.

[0075] The current AIMLE specifications and ongoing studies focus on static deployments and assume the VAL UE remains within its home network. However, real- world deployments may require: 1. VAL UEs to maintain access to AIMLE services when roaming into visited PLMNs or NPNs; 2. Cross-border AI / ML service collaboration (e.g., in eV2X or multiplayer gaming scenarios); and / or 3. Edge inference or Al model access to continue seamlessly even as the UE moves between coverage areas.

[0076] There is an ongoing requirement for developments in the following aspects: 1. identify the use cases for supporting multiple PLMN / NPN, and across MNOs and CSPs / ECSPs scenarios; 2. whether and how AIMLE services can support VAL UEs that roam into visited PLMNs or NPNs, e.g., registration, discovery, and access of AIMLE in the visited domain, as well as go back to home domain; 3. whether and how to ensure service continuity and switching for inference tasks or Al service sessions during UE mobility across MNOs and CSPs / ECSPs; 4. whether and how UEs or VAL functions may discover the most suitable AIMLE instance in edge-specific scenarios while UEs move across network boundaries or service areas, considering latency, capability, and service policy.

[0077] In the context of AIML services, the UE may be expected or predicted to move to a different PLMN while the AIML service is running. Examples described herein generally relate to supporting the AIMLE service continuity while roaming to different PLMN, and in particular how to change the service AIMLE server without impacting the AI / ML service performance while the UE is moving to a target PLMN. Examples described herein may also relate to discovering and selecting a target AIMLE server in the VPLMNDocket No. SMM920250065-GR-NPfor undertaking at least part of the AIMLE service which involves the UE of interest (e.g., a roaming UE).

[0078] Figure 7 illustrates an example of a process flow 700 for AIMLE in accordance with aspects of the present disclosure. The process flow 700 may implement or be implemented by aspects of the wireless communication system 100. For example, the process flow 700 may include an AIMLE client 732, a Source AIMLE Server (e.g., H- PLMN) 736, a VAL server / AIMLE consumer 716, a Target AIMLE server (e.g., V- PLMN) 738 and ML repository (or multiple ML repositories) 741, which may be one or more examples of devices described herein with reference to Figure 1.

[0079] The process flow 700 may be referred to as a procedure, including one or more operations performed by one or more of the AIMLE client 732, Source AIMLE Server 736, VAL server / AIMLE consumer 716, Target AIMLE server 738 and ML repository 741. In the example of Figure 7, the process flow 700 may include server enabled continuity support for ML model workflow operation.

[0080] In the following description of the process flow 700, the operations or signalling performed between one or more of the AIMLE client 732, Source AIMLE Server 736, VAL server / AIMLE consumer 716, Target AIMLE server 738 and ML repository 741 may be performed or signalled (e.g., transmitted, received) in a different order than the example order shown, or the operations or signalling performed by one or more of the AIMLE client 732, Source AIMLE Server 736, VAL server / AIMLE consumer 716, Target AIMLE server 738 and ML repository 741 may be performed or signalled (e.g., transmitted, received) in different orders or at different times. Some operations or signalling may also be omitted from the process flow 700. Additionally, although some operations or signalling may be shown to occur at different times, these operations or signalling may occur at the same time or in overlapping time periods.

[0081] Process flow 700 and Process flow 800 (described below in relation to Figure 8) are both examples of mechanisms for supporting the continuity of the ML training / inference capability in scenarios where the ML model training / inference happens at the UE (e.g., AIMLE client 732) while the UE client is expected to move to an area served by another PLMN. Process flow 700 generally relates to server enabled continuity support forDocket No. SMM920250065-GR-NPML model workflow operation. Process flow 800 generally relates to client enabled continuity support for ML model workflow operation.

[0082] Process flow 700 begins at step 771. The VAL server 716 sends (e.g., transmits, outputs) an ML model inference request to the Source AIMLE server 736, requesting to assist in its ML model inference. This request comprises ML model information or ML model requirement information, one or more VAL UE IDs to support the ML model inference, and the allowed MNO info per UE for supporting serving continuing in case of UE roaming etc.

[0083] Step 772. After authorizing the request, the Source AIMLE server 736 discovers and selects (e.g., determines) the AIMLE client(s) 732 to perform the ML model inference. This happens at the Source AIMLE server 736 (e.g., S-AIMLE server or AIMLE server @H-PLMN) via interacting with the ML repository 741 (e.g., ML model repository) to fetch the ML model info and optionally AIMLE client 732 info (shown as step 772a) as well as with the AIMLE client 732 to request / subscribe for performing the ML model inference task (shown as step 772b).

[0084] Step 773. The Source AIMLE server 736 monitors (e.g., determines, detects) UE mobility or analytics (e.g., from core network, NWDAF analytics on UE mobility, or location information from SEAL LMS, LMF via NEF, AMF, or UDM via NEF). The Source AIMLE server 736 detects (e.g., determines, predicts, estimates) a mobility of the UE to an area covered by different PLMN (e.g., VPLMN). The Source AIMLE server 736 may not know which PLMN the UE is roaming to; for example, if there are multiple PLMNs serving the same area. The AIMLE client 732 may notify the Source AIMLE server 736 that it has moved to a different PLMN (e.g., via app layer signaling or via the enablement layer API).

[0085] Step 774. The Source AIMLE server 736 identifies (e.g., determines, checks with the Core, the ECS-ER or the VAL UE) the V-PLMN 738 info and the time when the VAL UE is expected to reach V-PLMN 738. In case of multiple VPLMNs in the target area, the Source AIMLE server 736 may have detected the one or more VPLMNs via the UE or via VAL server 716 or this information can be pre-configured at the Source AIMLE server 736.Docket No. SMM920250065-GR-NP

[0086] Step 775. The Source AIMLE server 736 fetches (e.g., receives, retrieves, requests), from ML repository 741, context information relating to available AIMLE servers in the target V-PLMN.

[0087] If the ML repository 741 is deployed by PLMN / MNO, then step 775 may require interaction among ML repositories of the same provider based on the service agreements or synchronization via the AIMLE servers.

[0088] Step 776. The Source AIMLE server 736 selects the Target AIMLE server 738 (e.g., T-AIMLE server) from a list of AIMLE servers, based on a selection criteria. The selection criteria may comprise the context information (fetched from the ML repository 741 in step 775).

[0089] The context information may comprise at least one of: capabilities (e.g., computational, processing) and coverage (e.g., geographical or topological area covered by the Target AIMLE server 738); load and energy status of the Target AIMLE server 738 based on monitoring or querying possible overload or high energy consumption (this may be provided by the target AIMLE service provider, e.g., edge platform provider); expected delay / latency for accessing the Target AIMLE server 738 (e.g., in case of relaying ML inference task); vendor compatibility (e.g., when the selection of a Target AIMLE server 738 is based on which vendor provides the server ( this may be either the same vendor only, or a list of selected vendors based on the service agreements among MNOs / SEAL providers); cost for task migration (e.g., the computational effort and / or processing required or time required or energy required for offloading a task to the Target AIMLE server 738) (such cost may be based on the distance among servers (e.g., edge to regional cloud) or based on the task to be offloaded) (this may be also a parameter to identify whether to use Option 1 or Option 2 as discussed below in step 777); and a rate factor (e.g., ranking factor, weighting factor) of the Target AIMLE server 738 based on fulfilment of previous tasks and statistics. The rate factor may be stored and updated at the ML repository 741 (or shared among repositories) after fulfilling each task based on consumer feedback or based on internal evaluation at the AIMLE (e.g., based on success or failure of tasks under similar conditions).Docket No. SMM920250065-GR-NP

[0090] The expected delay / latency for accessing the Target AIMLE server 738 may be estimated based on probes. The expected delay / latency for accessing the Target AIMLE server 738 may be estimated based on relative location of platform hosting the Target AIMLE server 738. The expected delay / latency for accessing the Target AIMLE server 738 may be estimated based on the interface / API statistics on delay or failures based on high delay.

[0091] Step 777. The Source AIMLE server 736 requests (e.g., negotiates) with the Target AIMLE server 738 to undertake the task for the one or more UEs of interest, which may be at least one of: migrating the capability (e.g., app, service, or PaaS) for continuing the ML inference aggregation (referred to as Option 1); and relaying the ML inference output to the AIMLE consumer 716 via VPLMN 738 (referred to as Option 2).

[0092] Step 778. The Source AIMLE server 736 requests (e.g., informs) the AIMLE client 732 to change from serving the Source AIMLE server 736 to serving the Target AIMLE server 738. The Source AIMLE server 736 (e.g., in the request) sends (e.g., transmits, passes) the Target AIMLE server 738 info to the AIMLE client 732. The Source AIMLE server 736 (e.g., in the request) may also send (e.g., transmits, passes) possible limitations in the target V-PLMN (e.g., lower priority, or capacity limitations for ML operation). The AIMLE client 732 confirms (e.g., acknowledges) the change and prepares to connect to the Target AIMLE server 738 via the Source AIMLE server 736. The AIMLE client 732 then connects to Target AIMLE server 738.

[0093] Step 779. The Target AIMLE server 738 sends (e.g., transmits, outputs) a notification (e.g., indication) to the Source AIMLE server 736 indicating roaming of the UE(s) of interest (see step 779a). The Source AIMLE Server 736 may forward (e.g., send, transmit, output) the notification to the VAL server 716 (as expected consumer of the ML inference output) (see step 779b).

[0094] Step 780. The Target AIMLE server 738 may provide the ML inference output to the VAL server 716 via the Source AIMLE server 736; if the task involves relaying the ML inference output over the VPLMN. Alternatively, the Target AIMLE server 738 sends (e.g., transmits, outputs) the ML inference output to VAL server 716; if the Target AIMLE server 738 is the new ML inference server.Docket No. SMM920250065-GR-NP

[0095] The ML inference output may be reported to a local VAL server (not shown). The local VAL server may be located close to the UE location. The local VAL server may be in a cloud. The local VAL server may be an AIMLE consumer.

[0096] Figure 8 illustrates an example of a process flow 800 for AIMLE in accordance with aspects of the present disclosure.

[0097] The process flow 800 may implement or be implemented by aspects of the wireless communication system 100. For example, the process flow 800 may include an AIMLE client 832, a Source AIMLE Server (e.g., H-PLMN) 836, a VAL server / AIMLE consumer 816, a Target AIMLE server (e.g., V-PLMN) 838 and ML repository (or multiple ML repositories) 841, which may be one or more examples of devices described herein with reference to Figure 1.

[0098] The process flow 800 may be referred to as a procedure, including one or more operations performed by one or more of the AIMLE client 832, Source AIMLE Server 836, VAL server / AIMLE consumer 816, Target AIMLE server 838 and ML repository 841. In the example of Figure 8, the process flow 800 may include client enabled continuity support for ML model workflow operation.

[0099] In the following description of the process flow 800, the operations or signalling performed between one or more of the AIMLE client 832, Source AIMLE Server 836, VAL server / AIMLE consumer 816, Target AIMLE server 838 and ML repository 841 may be performed or signalled (e.g., transmitted, received) in a different order than the example order shown, or the operations or signalling performed by one or more of the AIMLE client 832, Source AIMLE Server 836, VAL server / AIMLE consumer 816, Target AIMLE server 838 and ML repository 841 may be performed or signalled (e.g., transmitted, received) in different orders or at different times. Some operations or signalling may also be omitted from the process flow 800. Additionally, although some operations or signalling may be shown to occur at different times, these operations or signalling may occur at the same time or in overlapping time periods.

[0100] Process flow 800 starts at step 871. The VAL server 816 sends (e.g., transmits, outputs) an ML model inference or training request to the Source AIMLE server 836,Docket No. SMM920250065-GR-NPrequesting to assist in its ML model inference or training. For training, this procedure may be as described in TS 23.482 V19.1.0 clause 8.2. The Source AIMLE server 836, with the support of the AIMLE client 832, starts the ML model inference or training (e.g., using conventional AIMLE client discovery and selection processes based on TS 23.482 VI 9.1.0). The ML model inference or training request may comprise at least one of: one or more VAL UE IDs to support the ML model training or inference; and the allowed MNO info per UE for supporting serving continuing in case of UE roaming.

[0101] The Source AIMLE server 836 (e.g., as part of the ML model training / inference process at the AIMLE client 832) may subscribe to or request notifications relating to cross-PLMN or roaming mobility events for the UE(s) of interest.

[0102] Step 872. The AIMLE client 832 (e.g., via the VAL client or other SEAL clients (e.g., LMC, ADAEC)) monitors UE mobility or analytics related to the expected or predicted mobility of the UE. The AIMLE client 832 detects (e.g., determines, predicts, estimates) a mobility of the UE to an area covered by a different PLMN (e.g., VPLMN).

[0103] Step 873. The AIMLE client 832 sends (e.g., transmits, outputs) a notification (e.g., indication) indicating that the AIMLE client 832 has detected mobility of the UE to an area covered by a different PLMN (e.g., cross-PLMN mobility event notify).

[0104] Step 874. The Source AIMLE server 836 identifies (e.g., determines, retrieves, checks with Core or ECS-ER or the VAL UE) the VPLMN info and the time when the VAL UE is expected to reach VPLMN.

[0105] Step 875. The Source AIMLE server 836 fetches (e.g., retrieves, receives, inputs), from the ML repository 841, context information relating to available AIMLE servers in the target VPLMN. If the ML repository 841 is deployed by PLMN / MNO, then step 875 may require interaction among ML repositories of the same provider based on the service agreements or synchronization via the AIMLE servers.

[0106] Step 876. The Source AIMLE server 836 sends (e.g., transmits, outputs), to the AIMLE client 832, a notification (e.g., an alert, a command, a message or a request) for selecting a new AIMLE server. The notification may comprise a list of available AIMLEDocket No. SMM920250065-GR-NPservers in the VPLMN and / or context information (e.g., rating, ranking, weighting, vendor info, capability information, energy status, and / or load status).

[0107] Step 877. The AIMLE client 832 based on the received list, selects (e.g., determines) the Target AIMLE server 838 from a list of AIMLE servers, based on a selection criteria. The selection criteria may comprise the context information (fetched from the ML repository 841 in step 875).

[0108] The context information may comprise at least one of: capabilities (e.g., computational, processing) and coverage (e.g., geographical or topological area covered by the Target AIMLE server 838) load and energy status of the Target AIMLE server 838, based on monitoring or querying possible overload or high energy consumption (this may be provided by the target AIMLE service provider, e.g., edge platform provider); expected delay / latency for accessing the Target AIMLE server 838 (e.g., in case of relaying ML inference task); vendor compatibility (e.g., when the selection of a Target AIMLE server 838 is based on which vendor provides the server. It can be either the same vendor only, or a list of selected vendors based on the service agreements among MNOs / SEAL providers); cost for task migration (e.g., the computational effort and / or processing required or time required or energy required for offloading a task to the Target AIMLE server 838. Such cost may be based on the distance among servers (e.g., edge to regional cloud) or based on the task to be offloaded. This may be also a parameter to identify whether to use Option 1 or Option 2; discussed below in step 877. A rate factor (e.g., ranking factor, weighting factor) of the Target AIMLE server 838 based on fulfilment of previous tasks and statistics. The rate factor may be stored and updated at the ML repository 841 (or shared among repositories) after fulfilling each task based on consumer feedback or based on internal evaluation at the AIMLE (e.g., based on success or failure of tasks under similar conditions).

[0109] Pre-defined policies from the Source AIMLE server 836 or VAL server 816 may be provided in steps 871 to 877 to dictate the AIMLE server selection criteria (e.g., policies on minimum rating, allowed vendors, preferred AIMLE servers).Docket No. SMM920250065-GR-NP

[0110] Step 878. The AIMLE client 832 sends (e.g., transmits, outputs) to the Source AIMLE server 836 the selected Target AIMLE server 838 and the expected time and area of roaming.

[0111] Steps 879 to 882 of Process flow 800 correspond to (e.g., are identical to) steps 777 to 780 of Process flow 700.

[0112] Figure 9 illustrates an example of a UE 900 in accordance with aspects of the present disclosure. The UE 900 may include a processor 902, a memory 904, a controller 906, and a transceiver 908. The processor 902, the memory 904, the controller 906, or the transceiver 908, or various combinations thereof or various components thereof may be examples of means for performing various aspects of the present disclosure as described herein. These components may be coupled (e.g., operatively, communicatively, functionally, electronically, electrically) via one or more interfaces.

[0113] The processor 902, the memory 904, the controller 906, or the transceiver 908, or various combinations or components thereof may be implemented in hardware (e.g., circuitry). The hardware may include a processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), or other programmable logic device, or any combination thereof configured as or otherwise supporting a means for performing the functions described in the present disclosure.

[0114] The processor 902 may include an intelligent hardware device (e.g., a general- purpose processor, a DSP, a CPU, an ASIC, an FPGA, or any combination thereof). In some implementations, the processor 902 may be configured to operate the memory 904. In some other implementations, the memory 904 may be integrated into the processor 902.The processor 902 may be configured to execute computer-readable instructions stored in the memory 904 to cause the UE 900 to perform various functions of the present disclosure.

[0115] The memory 904 may include volatile or non-volatile memory. The memory 904 may store computer-readable, computer-executable code including instructions when executed by the processor 902 cause the UE 900 to perform various functions described herein. The code may be stored in a non-transitory computer-readable medium such the memory 904 or another type of memory. Computer-readable media includes both non-Docket No. SMM920250065-GR-NPtransitory computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A non-transitory storage medium may be any available medium that may be accessed by a general-purpose or special-purpose computer.

[0116] In some implementations, the processor 902 and the memory 904 coupled with the processor 902 may be configured to cause the UE 900 to perform one or more of the functions described herein (e.g., executing, by the processor 902, instructions stored in the memory 904). For example, the processor 902 may support wireless communication at the UE 900 in accordance with examples as disclosed herein. The UE 900 may be configured to support means for determining to support a first network entity with an AIML functionality, wherein the UE is served by a first wireless communication network; determining that the UE is to be served by a second wireless communication network; and determining a second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network.

[0117] The controller 906 may manage input and output signals for the UE 900. The controller 906 may also manage peripherals not integrated into the UE 900. In some implementations, the controller 906 may utilize an operating system such as iOS®, ANDROID®, WINDOWS®, or other operating systems. In some implementations, the controller 906 may be implemented as part of the processor 902.

[0118] In some implementations, the UE 900 may include at least one transceiver 908. In some other implementations, the UE 900 may have more than one transceiver 908. The transceiver 908 may represent a wireless transceiver. The transceiver 908 may include one or more receiver chains 910, one or more transmitter chains 912, or a combination thereof.

[0119] A receiver chain 910 may be configured to receive signals (e.g., control information, data, packets) over a wireless medium. For example, the receiver chain 910 may include one or more antennas for receive the signal over the air or wireless medium. The receiver chain 910 may include at least one amplifier (e.g., a low-noise amplifier (LNA)) configured to amplify the received signal. The receiver chain 910 may include at least one demodulator configured to demodulate the receive signal and obtain the transmitted data by reversing the modulation technique applied during transmission of theDocket No. SMM920250065-GR-NPsignal. The receiver chain 910 may include at least one decoder for decoding the processing the demodulated signal to receive the transmitted data.

[0120] A transmitter chain 912 may be configured to generate and transmit signals (e.g., control information, data, packets). The transmitter chain 912 may include at least one modulator for modulating data onto a carrier signal, preparing the signal for transmission over a wireless medium. The at least one modulator may be configured to support one or more techniques such as amplitude modulation (AM), frequency modulation (FM), or digital modulation schemes like phase-shift keying (PSK) or quadrature amplitude modulation (QAM). The transmitter chain 912 may also include at least one power amplifier configured to amplify the modulated signal to an appropriate power level suitable for transmission over the wireless medium. The transmitter chain 912 may also include one or more antennas for transmitting the amplified signal into the air or wireless medium.

[0121] Figure 10 illustrates an example of a processor 1000 in accordance with aspects of the present disclosure. The processor 1000 may be an example of a processor configured to perform various operations in accordance with examples as described herein. The processor 1000 may include a controller 1002 configured to perform various operations in accordance with examples as described herein. The processor 1000 may optionally include at least one memory 1004, which may be, for example, an L1 / L2 / L3 cache. Additionally, or alternatively, the processor 1000 may optionally include one or more arithmetic-logic units (ALUs) 1006. One or more of these components may be in electronic communication or otherwise coupled (e.g., operatively, communicatively, functionally, electronically, electrically) via one or more interfaces (e.g., buses).

[0122] The processor 1000 may be a processor chipset and include a protocol stack (e.g., a software stack) executed by the processor chipset to perform various operations (e.g., receiving, obtaining, retrieving, transmitting, outputting, forwarding, storing, determining, identifying, accessing, writing, reading) in accordance with examples as described herein. The processor chipset may include one or more cores, one or more caches (e.g., memory local to or included in the processor chipset (e.g., the processor 1000) or other memory (e.g., random access memory (RAM), read-only memory (ROM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), static RAM (SRAM), ferroelectricDocket No. SMM920250065-GR-NPRAM (FeRAM), magnetic RAM (MRAM), resistive RAM (RRAM), flash memory, phase change memory (PCM), and others).

[0123] The controller 1002 may be configured to manage and coordinate various operations (e.g., signalling, receiving, obtaining, retrieving, transmitting, outputting, forwarding, storing, determining, identifying, accessing, writing, reading) of the processor 1000 to cause the processor 1000 to support various operations in accordance with examples as described herein. For example, the controller 1002 may operate as a control unit of the processor 1000, generating control signals that manage the operation of various components of the processor 1000. These control signals include enabling or disabling functional units, selecting data paths, initiating memory access, and coordinating timing of operations.

[0124] The controller 1002 may be configured to fetch (e.g., obtain, retrieve, receive) instructions from the memory 1004 and determine subsequent instruction(s) to be executed to cause the processor 1000 to support various operations in accordance with examples as described herein. The controller 1002 may be configured to track memory address of instructions associated with the memory 1004. The controller 1002 may be configured to decode instructions to determine the operation to be performed and the operands involved. For example, the controller 1002 may be configured to interpret the instruction and determine control signals to be output to other components of the processor 1000 to cause the processor 1000 to support various operations in accordance with examples as described herein. Additionally, or alternatively, the controller 1002 may be configured to manage flow of data within the processor 1000. The controller 1002 may be configured to control transfer of data between registers, arithmetic logic units (ALUs), and other functional units of the processor 1000.

[0125] The memory 1004 may include one or more caches (e.g., memory local to or included in the processor 1000 or other memory, such RAM, ROM, DRAM, SDRAM, SRAM, MRAM, flash memory, etc. In some implementations, the memory 1004 may reside within or on a processor chipset (e.g., local to the processor 1000). In some other implementations, the memory 1004 may reside external to the processor chipset (e.g., remote to the processor 1000).Docket No. SMM920250065-GR-NP

[0126] The memory 1004 may store computer-readable, computer-executable code including instructions that, when executed by the processor 1000, cause the processor 1000 to perform various functions described herein. The code may be stored in a non-transitory computer-readable medium such as system memory or another type of memory. The controller 1002 and / or the processor 1000 may be configured to execute computer-readable instructions stored in the memory 1004 to cause the processor 1000 to perform various functions. For example, the processor 1000 and / or the controller 1002 may be coupled with or to the memory 1004, the processor 1000, the controller 1002, and the memory 1004 may be configured to perform various functions described herein. In some examples, the processor 1000 may include multiple processors and the memory 1004 may include multiple memories. One or more of the multiple processors may be coupled with one or more of the multiple memories, which may, individually or collectively, be configured to perform various functions herein.

[0127] The one or more ALUs 1006 may be configured to support various operations in accordance with examples as described herein. In some implementations, the one or more ALUs 1006 may reside within or on a processor chipset (e.g., the processor 1000). In some other implementations, the one or more ALUs 1006 may reside external to the processor chipset (e.g., the processor 1000). One or more ALUs 1006 may perform one or more computations such as addition, subtraction, multiplication, and division on data. For example, one or more ALUs 1006 may receive input operands and an operation code, which determines an operation to be executed. One or more ALUs 1006 be configured with a variety of logical and arithmetic circuits, including adders, subtractors, shifters, and logic gates, to process and manipulate the data according to the operation. Additionally, or alternatively, the one or more ALUs 1006 may support logical operations such as AND, OR, exclusive-OR (XOR), not-OR (NOR), and not-AND (NAND), enabling the one or more ALUs 1006 to handle conditional operations, comparisons, and bitwise operations.

[0128] The processor 1000 may support wireless communication in accordance with examples as disclosed herein. The processor 1000 may be configured to support a means for determining to support a first network entity with an AIML functionality, wherein the UE is served by a first wireless communication network; determining that the UE is to beDocket No. SMM920250065-GR-NPserved by a second wireless communication network; and determining a second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network. The processor 1000 may be configured to or operable to support a means for receiving, from an application entity, a first request to support an AIML functionality for a UE served by a first wireless communication network; determining a second network entity for supporting the AIML functionality whilst the UE is to be served by a second wireless communication network; and transmitting, to the second network entity, a second request to support the AIML functionality.

[0129] Figure 11 illustrates an example of a NE 1100 in accordance with aspects of the present disclosure. The NE 1100 may include a processor 1102, a memory 1104, a controller 1106, and a transceiver 1108. The processor 1102, the memory 1104, the controller 1106, or the transceiver 1108, or various combinations thereof or various components thereof may be examples of means for performing various aspects of the present disclosure as described herein. These components may be coupled (e.g., operatively, communicatively, functionally, electronically, electrically) via one or more interfaces.

[0130] The processor 1102, the memory 1104, the controller 1106, or the transceiver 1108, or various combinations or components thereof may be implemented in hardware (e.g., circuitry). The hardware may include a processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), or other programmable logic device, or any combination thereof configured as or otherwise supporting a means for performing the functions described in the present disclosure.

[0131] The processor 1102 may include an intelligent hardware device (e.g., a general- purpose processor, a DSP, a CPU, an ASIC, an FPGA, or any combination thereof). In some implementations, the processor 1102 may be configured to operate the memory 1104. In some other implementations, the memory 1104 may be integrated into the processor 1102. The processor 1102 may be configured to execute computer-readable instructions stored in the memory 1104 to cause the NE 1100 to perform various functions of the present disclosure.Docket No. SMM920250065-GR-NP

[0132] The memory 1104 may include volatile or non-volatile memory. The memory 1104 may store computer-readable, computer-executable code including instructions when executed by the processor 1102 cause the NE 1100 to perform various functions described herein. The code may be stored in a non-transitory computer-readable medium such the memory 1104 or another type of memory. Computer-readable media includes both non- transitory computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A non-transitory storage medium may be any available medium that may be accessed by a general-purpose or special-purpose computer.

[0133] In some implementations, the processor 1102 and the memory 1104 coupled with the processor 1102 may be configured to cause the NE 1100 to perform one or more of the functions described herein (e.g., executing, by the processor 1102, instructions stored in the memory 1104). For example, the processor 1102 may support wireless communication at the NE 1100 in accordance with examples as disclosed herein. The NE 1100 may be configured to support a means for receiving, from an application entity, a first request to support an AIML functionality for a UE served by a first wireless communication network; determining a second network entity for supporting the AIML functionality whilst the UE is to be served by a second wireless communication network; and transmitting, to the second network entity, a second request to support the AIML functionality.

[0134] The controller 1106 may manage input and output signals for the NE 1100. The controller 1106 may also manage peripherals not integrated into the NE 1100. In some implementations, the controller 1106 may utilize an operating system such as iOS®, ANDROID®, WINDOWS®, or other operating systems. In some implementations, the controller 1106 may be implemented as part of the processor 1102.

[0135] In some implementations, the NE 1100 may include at least one transceiver 1108. In some other implementations, the NE 1100 may have more than one transceiver 1108. The transceiver 1108 may represent a wireless transceiver. The transceiver 1108 may include one or more receiver chains 1110, one or more transmitter chains 1112, or a combination thereof.Docket No. SMM920250065-GR-NP

[0136] A receiver chain 1110 may be configured to receive signals (e.g., control information, data, packets) over a wireless medium. For example, the receiver chain 1110 may include one or more antennas for receive the signal over the air or wireless medium. The receiver chain 1110 may include at least one amplifier (e.g., a low-noise amplifier (LNA)) configured to amplify the received signal. The receiver chain 1110 may include at least one demodulator configured to demodulate the receive signal and obtain the transmitted data by reversing the modulation technique applied during transmission of the signal. The receiver chain 1110 may include at least one decoder for decoding the processing the demodulated signal to receive the transmitted data.

[0137] A transmitter chain 1112 may be configured to generate and transmit signals (e.g., control information, data, packets). The transmitter chain 1112 may include at least one modulator for modulating data onto a carrier signal, preparing the signal for transmission over a wireless medium. The at least one modulator may be configured to support one or more techniques such as amplitude modulation (AM), frequency modulation (FM), or digital modulation schemes like phase-shift keying (PSK) or quadrature amplitude modulation (QAM). The transmitter chain 1112 may also include at least one power amplifier configured to amplify the modulated signal to an appropriate power level suitable for transmission over the wireless medium. The transmitter chain 1112 may also include one or more antennas for transmitting the amplified signal into the air or wireless medium.

[0138] Figure 12 illustrates a flowchart of a method 1200 in accordance with aspects of the present disclosure. The operations of the method 1200 may be implemented by a NE as described herein. In some implementations, the NE may execute a set of instructions to control the function elements of the NE to perform the described functions.

[0139] At 1202, the method 1200 may include receiving, from an application entity, a first request to support an AIML functionality for a UE served by a first wireless communication network. The operations of 1202 may be performed in accordance with examples as described herein. In some implementations, aspects of the operations of 1202 may be performed by a NE as described with reference to Figure 11.

[0140] At 1204, the method 1200 may include determining a second network entity for supporting the AIML functionality whilst the UE is to be served by a second wirelessDocket No. SMM920250065-GR-NPcommunication network. The operations of 1204 may be performed in accordance with examples as described herein. In some implementations, aspects of the operations of 1204 may be performed by a NE as described with reference to Figure 11.

[0141] At 1206, the method 1200 may include transmitting, to the second network entity, a second request to support the AIML functionality. The operations of 1206 may be performed in accordance with examples as described herein. In some implementations, aspects of the operations of 1206 may be performed a NE as described with reference to Figure 11.

[0142] It should be noted that the method 1200 described herein describes a possible implementation, and that the operations and the steps may be rearranged or otherwise modified and that other implementations are possible.

[0143] There is provided a first network entity for wireless communication, comprising: at least one memory; and at least one processor coupled with the at least one memory and configured to cause the first network entity to: receive, from an application entity, a first request to support an AIML functionality for a UE served by a first wireless communication network; determine a second network entity for supporting the AIML functionality whilst the UE is to be served by a second wireless communication network; and transmit, to the second network entity, a second request to support the AIML functionality. Such a first network entity tends to improve continuity of the AIML functionality whilst the UE roams from the first wireless communication network to the second wireless communication network.

[0144] The first network entity may comprise at least one of: a first application enablement entity, a first AIMLE server, and a first application function. The first network entity may comprise AI / ML functionality. The second network entity may comprise at least one of: a second application enablement entity, a second AIMLE server, and a second application function. The second network entity may comprise AI / ML functionality. AI / ML functionality may comprise undertaking or assisting the operation and / or preparation of the AI / ML functionality. The AI / ML functionality may comprise at least one of: an AIMLE service, an Al service, and an AI / ML task. The AIML functionality may comprise at least one of: a model operation; a model retrieval, a model training, a model monitoring, a modelDocket No. SMM920250065-GR-NPselection, a model update; a model storage; a model discovery; a ML operation, an ML training operation, a ML model operation; a ML model inference, and an ML model inference operation.

[0145] The UE may comprise an AIMLE client. The UE may comprise a VAL client. The VAL client may comprise a VAL UE. The first request to support the AIML functionality for the UE may comprise a first request to support the AIML functionality for the VAL client of the UE. The method may further comprise discovering the AIMLE client of the UE in response to receiving the first request to support the AIML functionality for the UE. Discovering the AIMLE client of the UE may comprise transmitting, to an ML repository, a request for information relating to the AIMLE client. The method may further comprise selecting the AIMLE client of the UE. The method may further comprise transmitting, to the ML repository, a request for information relating to the AIML functionality. The method may further comprise transmitting to the AIMLE client a request for performing the AIML functionality. The method may further comprise subscribing to the AIMLE client for performing the AIML functionality.

[0146] The first wireless communication network may comprise a PLMN. The second wireless communication network may comprise a PLMN. The first wireless communication network may comprise a HPLMN. The second wireless communication network may comprise a VPLMN. The first wireless communication network may comprise a NPN. The second wireless communication network may comprise a NPN. The first network entity may comprise a source AIMLE server. The second network entity may comprise a target AIMLE server. The application entity may comprise an external application entity. The application entity may comprise a VAL server. The application entity may comprise an AIMLE consumer. The VAL server may comprise the AIMLE consumer. The application entity may comprise an app server. The application entity may comprise a non-vertical app server. The first network entity may be connected to the first wireless communication network. The second network entity may be connected to the second wireless communication network.

[0147] The first request to support an AIML functionality for the UE may comprise a VAL UE ID. The VAL UE ID may comprise an identifier for the UE. The VAL UE IDDocket No. SMM920250065-GR-NPmay comprise an identifier for the VAL client of the UE. The VAL UE may be suitable to support the AIML functionality. The first request to support an AIML functionality for the UE may comprise information relating to one or more MNO for supporting the AIML functionality for the UE. The first request may comprise allowed MNO info per UE for supporting serving continuing in case of UE roaming.

[0148] The second request to support the AIML functionality may comprise a request for the second network entity to migrate capability for continuing the AIML functionality. The second request to support the AIML functionality may comprise a request for the second network entity to relay output from the AIML functionality to the application entity via the second wireless communication network.

[0149] The at least one processor may be further configured to cause the first network entity to: determine that the UE is to be served by the second wireless communication network. The at least one processor may be further configured to cause the first network entity to: receive information relating to the mobility of the UE.

[0150] Receiving information relating to the mobility of the UE may comprise receiving, from a location management entity, the information relating to the mobility of the UE. Receiving information relating to the mobility of the UE may comprise subscribing to the location management entity for the information relating to the mobility of the UE. The location management entity may be at least one of: a NF, AF, server entity, client entity and LMF.

[0151] Receiving information relating to the mobility of the UE may comprise receiving the information relating to the mobility of the UE from at least one of: a core network, an NWDAF, SEAL LMS, SEAL LMF and NEF. Determining that the UE is to be served by the second wireless communication network may further comprise detecting, based on the information relating to the mobility of the UE, the mobility of the UE to an area covered by the second wireless communication network. Determining that the UE is to be served by the second wireless communication network may further comprise may comprise predicting mobility of the UE to a geographical or topological area covered by the second wireless communication network.Docket No. SMM920250065-GR-NP

[0152] The at least one processor may be further configured to cause the first network entity to determine information relating to the second wireless communication network. Determining the information relating to the second wireless communication network may comprise transmitting a request for the information relating to the second wireless communication network from at least one of: the core network, ECS-ER, and VAL UE. The at least one processor may be further configured to cause the first network entity to determine a time for the UE to enter the area covered by the second wireless communication network.

[0153] To determine the second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network, the at least one processor may be further configured to cause the first network entity to: transmit, to an ML repository, a request for context information for one or more available AIMLE server(s) in the second wireless communication network. The context information may comprise at least one of: a capability; a coverage; a load status; an energy status; an expected delay for access; an expected latency for access; vendor compatibility; cost for task migration; and a weighting factor.

[0154] The at least one processor may be further configured to cause the first network entity to: select, based on the context information, the second network entity from the one or more available AIMLE server(s) in the second wireless communication network. The at least one processor may be further configured to cause the first network entity to: transmit, to the UE, information relating to the second network entity.

[0155] The information relating to the second network entity may comprise information for identifying the second network entity. The information relating to the second network entity may comprise information for connecting to the second network entity. The information relating to the second network entity may comprise target AIMLE server information. The information relating to the second network entity may comprise limitations in the second wireless communication network. The limitations in the second wireless communication network may comprise a priority for the AIML functionality and / or a capacity limit for the AIML functionality.Docket No. SMM920250065-GR-NP

[0156] The at least one processor may be further configured to cause the first network entity to: relay an output of the AIML functionality from the second network entity to the application entity. To determine that the UE is to be served by the second wireless communication network, the at least one processor may be further configured to cause the first network entity to: receive, from the UE, an indication that the UE is to be served by the second wireless communication network.

[0157] The indication that the UE is to be served by the second wireless communication network may comprise a cross-PLMN mobility event notify. The indication that the UE is to be served by the second wireless communication network may comprise a cross-PLMN mobility event notification. The indication that the UE is to be served by the second wireless communication network may comprise an indication that the UE has detected a mobility of the UE to an area covered by the second wireless communication network. The indication that the UE is to be served by the second wireless communication network may comprise an indication that the UE has detected a mobility of the UE to an area covered by a different wireless communication network. Determining that the UE is to be served by the second wireless communication network may comprise: receiving for the UE a notification relating to the UE expected and / or predicted mobility to the second wireless communication network.

[0158] To determine the second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network, the at least one processor may be further configured to cause the first network entity to: transmit, to the UE, a request to select an AIMLE server from one or more available AIMLE server(s) in the second wireless communication network.

[0159] The request to select the AIMLE server from the one or more available AIMLE server(s) in the second wireless communication network may comprise a list of the one or more available AIMLE server(s) in the second wireless communication network. The request to select the AIMLE server from the one or more available AIMLE server(s) in the second wireless communication network may comprise context information for the one or more available AIMLE server(s) in the second wireless communication network.Docket No. SMM920250065-GR-NP

[0160] To determine the second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network, the at least one processor may be further configured to cause the first network entity to: receive, from the UE, an indication that the second network entity has been selected.

[0161] There is further provided a method performed or performable by a first network entity, the method comprising: receiving, from an application entity, a first request to support an AIML functionality for a user equipment, UE served by a first wireless communication network; determining a second network entity for supporting the AIML functionality whilst the UE is to be served by a second wireless communication network; and transmitting, to the second network entity, a second request to support the AIML functionality. Such a method performed or performable by the first network entity tends to improve continuity of the AIML functionality whilst the UE roams from the first wireless communication network to the second wireless communication network.

[0162] The method may further comprise determining that the UE is to be served by the second wireless communication network. Determining that the UE is to be served by the second wireless communication network may comprise receiving information relating to the mobility of the UE. Determining the second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network may comprise transmitting, to an ML repository, a request for context information for one or more available AIMLE server(s) in the second wireless communication network. The context information may comprise at least one of: a capability; a coverage; a load status; an energy status; an expected delay for access; an expected latency for access; vendor compatibility; cost for task migration; and a weighting factor.

[0163] The method may further comprise selecting, based on the context information, the second network entity from the one or more available AIMLE server(s) in the second wireless communication network. The method may further comprise transmitting, to the UE, information relating to the second network entity. The method may further comprise relaying an output of the AIML functionality from the second network entity to the application entity. Determining that the UE is to be served by the second wirelessDocket No. SMM920250065-GR-NPcommunication network may comprise receiving, from the UE, an indication that the UE is to be served by the second wireless communication network.

[0164] Determining the second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network may comprise transmitting, to the UE, a request to select an AIMLE server from one or more available AIMLE server(s) in the second wireless communication network. Determining the second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network may comprise receiving, from the UE, an indication that the second network entity has been selected.

[0165] There is further provided a method performed or performable by a user equipment, UE, the method comprising: determining to support a first network entity with an AIML functionality, wherein the UE is served by a first wireless communication network; determining that the UE is to be served by a second wireless communication network; and determining a second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network. Such a method performed or performable by the UE tends to improve continuity of the AIML functionality whilst the UE roams from the first wireless communication network to the second wireless communication network.

[0166] The method may further comprise transmitting, to the first network entity, an indication that the UE is to be served by the second wireless communication network. The method may further comprise receiving, from the first network entity, a request to select an AIMLE server from one or more available AIMLE server(s) in the second wireless communication network. Determining the second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network may comprise selecting the second network entity from the one or more available AIMLE server(s) in the second wireless communication network. The method may further comprise transmitting, to the first network entity, an indication that the second network entity has been selected. The method may further comprise connecting to the second network entity.

[0167] A UE may be expected or predicted to move to a different PLMN while an AIML service is running. Examples described herein tend to support AIMLE serviceDocket No. SMM920250065-GR-NPcontinuity while roaming to a different PLMN. Examples described herein may relate to changing the service AIMLE server whilst the UE is moving to a target PLMN. Such examples tend to reduce the impact to AI / ML service performance.

[0168] Examples described herein generally relate to a mechanism for supporting the continuity of the ML training / inference capability in scenarios where the ML model training / inference happens at the UE [which comprises the AIMLE client] while the UE client is expected to move to an area served by another PLMN. Previous solutions did not consider the roaming aspect for the AIMLE service continuity.

[0169] Some examples described herein relate to server enabled continuity support for ML model workflow operation. Some examples described herein relate to client enabled continuity support for ML model workflow operation.

[0170] There is further provided a method for supporting the continuity of an ML operation involving one or more UEs, wherein the one or more UEs are served by a first network, the method comprising: detecting an event related to the mobility of the one or more UEs, wherein the event is for an expected or predictive mobility to a remote area covered by a second network; identifying a condition for expected or predicted roaming of the at least one UE to a second network; determining the adaption of at least one parameter of the ML operation based on the identified condition; and sending the adapted parameter to the involved entities in the ML operation.

[0171] The ML operation may be an ML model training operation, a ML model inference operation or combination thereof. The one or more UE may be served by an ML server entity at a first network. The ML server may be an AIMLE server connected to the first network. The first and / or the second network may be a PLMN or NPN

[0172] Detecting the event may comprise identifying an expected or predicted mobility to a geographical or topological area. The area may be covered by the second network.Identifying the mobility may comprise subscribing to a location management entity [at UE or server side] for obtaining mobility information for the UE. The method may further comprise obtaining mobility information based on the subscription. The location management entity may be one or more of a NF, AL, Server entity, and Client entity.Docket No. SMM920250065-GR-NP

[0173] The method may further comprise identifying at least one second network. Identifying the at least one second network may comprise fetching information on the second network and / or its capability related to supporting the ML operation. The adaption of at least one parameter may comprise at least one of: adaption of ML server in ML operation; adaption of the communication to the ML server via the second network. This may involve adaption of network endpoint serving the at least one UE; and adaption of one or more ML operation configuration parameters, the configuration parameters relate to the performance / KPI of the ML operation in the second network.

[0174] The adaption of the at least one parameter may be based on one or more of the context information described herein. Sending the adapted parameter may comprise notifying an entity acting as ML server [+providing the configuration] [+negotiating with the T-AIMLE server]; notifying an entity acting as the new network endpoint for the ML operation; notifying VAL server; and notifying the UE.

[0175] It should be noted that the method described herein describes a possible implementation, and that the operations and the steps may be rearranged or otherwise modified and that other implementations are possible.

[0176] The description herein is provided to enable a person having ordinary skill in the art to make or use the disclosure. Various modifications to the disclosure will be apparent to a person having ordinary skill in the art, and the generic principles defined herein may be applied to other variations without departing from the scope of the disclosure. Thus, the disclosure is not limited to the examples and designs described herein but is to be accorded the broadest scope consistent with the principles and novel features disclosed herein.Docket No. SMM920250065-GR-NP

Claims

CLAIMSWhat is claimed is:

1. A first network entity for wireless communication, comprising: at least one memory; and at least one processor coupled with the at least one memory and configured to cause the first network entity to: receive, from an application entity, a first request to support an artificial intelligence / machine learning, AIML, functionality for a user equipment, UE served by a first wireless communication network; determine a second network entity for supporting the AIML functionality whilst the UE is to be served by a second wireless communication network; and transmit, to the second network entity, a second request to support the AIML functionality.

2. The first network entity of claim 1, wherein the at least one processor is further configured to cause the first network entity to: determine that the UE is to be served by the second wireless communication network.

3. The first network entity of claim 2, wherein the at least one processor is further configured to cause the first network entity to: receive information relating to mobility of the UE.

4. The first network entity of any one of claims 1 to 3, wherein to determine the second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network, the at least one processor is further configured to cause the first network entity to: transmit, to an ML repository, a request for context information for one or more available AIMLE server(s) in the second wireless communication network.Docket No. SMM920250065-GR-NP5. The first network entity of claim 4, wherein the context information comprises at least one of: a capability; a coverage; a load status; an energy status; an expected delay for access; an expected latency for access; vendor compatibility; cost for task migration; and a weighting factor.

6. The first network entity of claim 4 or claim 5, wherein the at least one processor is further configured to cause the first network entity to: select, based on the context information, the second network entity from the one or more available AIMLE server(s) in the second wireless communication network.

7. The first network entity of any one of claims 1 to 6, wherein the at least one processor is further configured to cause the first network entity to: transmit, to the UE, information relating to the second network entity.

8. The first network entity of any one of claims 1 to 7, wherein the at least one processor is further configured to cause the first network entity to: relay an output of the AIML functionality from the second network entity to the application entity.

9. The first network entity of any one of claims 2 to 8, when dependent on claim 2, wherein to determine that the UE is to be served by the second wireless communication network, the at least one processor is further configured to cause the first network entity to: receive, from the UE, an indication that the UE is to be served by the second wireless communication network.Docket No. SMM920250065-GR-NP10. The first network entity of any one of claims 1 to 9, wherein to determine the second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network, the at least one processor is further configured to cause the first network entity to: transmit, to the UE, a request to select an AIMLE server from one or more available AIMLE server(s) in the second wireless communication network.

11. The first network entity of any one of claims 1 to 10, wherein to determine the second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network, the at least one processor is further configured to cause the first network entity to: receive, from the UE, an indication that the second network entity has been selected.

12. A method performed or performable by a first network entity, the method comprising: receiving, from an application entity, a first request to support an artificial intelligence / machine learning, AIML, functionality for a user equipment, UE served by a first wireless communication network; determining a second network entity for supporting the AIML functionality whilst the UE is to be served by a second wireless communication network; and transmitting, to the second network entity, a second request to support the AIML functionality.

13. The method of claim 12, further comprising: determining that the UE is to be served by the second wireless communication network.

14. The method of claim 13, wherein determining that the UE is to be served by the second wireless communication network comprises: receiving information relating to mobility of the UE.Docket No. SMM920250065-GR-NP15. The method of any one of claims 12 to 14, wherein determining the second network entity for supporting the AIML functionality whilst the UE is being served by the second wireless communication network comprises: transmitting, to an ML repository, a request for context information for one or more available AIMLE server(s) in the second wireless communication network.Docket No. SMM920250065-GR-NP

Images