Device and method for efficient international ai service in wireless communication system

Abstract

A method for providing an artificial intelligence (AI) roaming service, according to one embodiment of a first aspect of the present invention, comprises the steps of: acquiring, from a terminal located in a residence nation, a request for an AI service provided by a prescribed AI service provider within the provider's own nation; selecting, according to the request, a subject performing AI inference required for the AI service from among a plurality of local data centers provided in the residence nation; controlling a resource required for the AI inference to be transferred from the AI service provider to the selected local data center; and controlling a result of the AI inference performed by using the resource by the selected local data center to be provided to the terminal located in the residence nation.

Description

Device and method for efficient AI services between countries in wireless communication systems

[0001] The present disclosure relates generally to wireless communication systems, and more particularly to devices and methods for efficient AI services between countries in wireless communication systems.

[0002] For reference, this application claims priority to Korean Patent Application No. 10-2024-0119368, filed September 3, 2024. The entire contents of that application, which serves as the basis for this priority claim, are incorporated herein by reference.

[0003] Currently, accessing AI services located in domestic data centers while roaming abroad requires latency equal to, or greater than, that of roaming data services. This is because, if AI model operation is required, the time it takes for AI calculations to be executed is included in the latency. For example, real-time AI services such as language / image recognition, real-time translation, video / image / X-ray analysis, and optical character recognition (OCR) are often sensitive to latency, and roaming makes this even more so.

[0004] The current roaming system architecture has limitations in improving these latencies, and this is particularly important given the increasing use of AI services alongside rising roaming service usage. To address this issue, a new roaming service architecture capable of rapidly providing AI services in roaming environments is needed.

[0005] Based on the discussion described above, the present disclosure provides a device and method for efficient AI service between countries in a wireless communication system.

[0006] In addition, the present disclosure provides a device and method for providing necessary AI services without going through a local communication network by transferring an AI model required for AI (artificial intelligence) in a wireless communication system to a data center close to a roaming terminal.

[0007] In addition, the present disclosure provides a device and method for providing necessary AI services by transferring an AI model to a data center close to a terminal within a wireless communication system.

[0008] Additionally, the present disclosure provides a device and method for deleting an AI model transferred for providing AI services in a wireless communication system.

[0009] A method for providing an AI roaming service according to one embodiment of the first aspect of the present invention includes the steps of: obtaining a request for an AI (artificial intelligence) service provided by a predetermined AI service provider within a country from the terminal located in the country of residence; selecting, in response to the request, an entity performing AI inference required for the AI ​​service from among a plurality of local data centers provided in the country of residence; controlling resources required for the AI ​​inference to be transferred from the AI ​​service provider to the selected local data center; and controlling a result of AI inference performed by the selected local data center using the resources to be provided to the terminal located in the country of residence.

[0010] The above method may further include a step of verifying whether the terminal is subscribed to the AI ​​roaming service. In this case, the entity performing the AI ​​inference may be the selected local data center for terminals subscribed to the AI ​​roaming service, and the AI ​​service provider for terminals not subscribed to the AI ​​roaming service.

[0011] In the above selection process, at least one of whether each of the plurality of local data centers can perform the AI ​​inference, the latency until the AI ​​roaming service is provided, the distance between the terminal located in the country of residence and each of the plurality of local data centers, and the amount of resources required for each of the plurality of local data centers to perform the AI ​​inference may be considered.

[0012] The resources transmitted from the AI ​​service provider to the selected local data center may include at least one of information about an AI model trained for the AI ​​service, environmental information required for executing the AI ​​model, and data formats of input or output of the AI ​​model.

[0013] The type of resource to be transmitted may be determined dependently on the specifications provided in the selected local data center.

[0014] Resources transferred to the selected local data center may be prevented from being used in the selected local data center under certain conditions.

[0015] A method for receiving an AI roaming service according to another embodiment of the first aspect of the present invention includes the steps of requesting an AI (artificial intelligence) service provided by a given AI service provider within a country while located in the country of residence, and obtaining the result of AI inference corresponding to the request from a local data center selected from among a plurality of local data centers provided in the country of residence. In this case, the AI ​​inference corresponding to the request is performed by the selected local data center utilizing the resources required for the AI ​​inference, which have been transferred from the AI ​​service provider.

[0016] A core network of a residence country for receiving an AI roaming service according to a second aspect of the present invention comprises a transceiver and a control unit operably connected to the transceiver, wherein the control unit obtains a request for an AI (artificial intelligence) service provided by a predetermined AI service provider within the country from the terminal located in the residence country, selects an AI inference performer required for the AI ​​service from among a plurality of local data centers provided in the residence country according to the request, controls resources required for the AI ​​inference to be transferred from the AI ​​service provider to the selected local data center, and controls the result of the AI ​​inference performed by the selected local data center using the resources to be provided to the terminal located in the residence country.

[0017] The control unit can verify whether the terminal is subscribed to the AI ​​roaming service. In this case, the entity performing the AI ​​inference may be the selected local data center for terminals subscribed to the AI ​​roaming service, and the AI ​​service provider for terminals not subscribed to the AI ​​roaming service.

[0018] In the above selection process, at least one of whether each of the plurality of local data centers can perform the AI ​​inference, the latency until the AI ​​roaming service is provided, the distance between the terminal located in the country of residence and each of the plurality of local data centers, and the amount of resources required for each of the plurality of local data centers to perform the AI ​​inference may be considered.

[0019] The resources transmitted from the AI ​​service provider to the selected local data center may include at least one of information about an AI model trained for the AI ​​service, environmental information required for executing the AI ​​model, and data formats of input or output of the AI ​​model.

[0020] The type of resource to be transmitted may be determined dependently on the specifications provided in the selected local data center.

[0021] Resources transferred to the selected local data center may be prevented from being used in the selected local data center under certain conditions.

[0022] Devices and methods according to various embodiments of the present disclosure enable the provision of necessary AI services by transferring AI models required for AI (artificial intelligence) to a data center close to the terminal.

[0023] The effects that can be obtained from the present disclosure are not limited to the effects mentioned above, and other effects that are not mentioned can be clearly understood by a person having ordinary skill in the art to which the present disclosure belongs from the description below.

[0024] FIG. 1 illustrates a data movement path for AI service use during existing roaming, according to various embodiments of the present disclosure.

[0025] FIG. 2 illustrates a data movement path for using AI services while roaming, according to one embodiment of the present disclosure.

[0026] FIG. 3 illustrates a signal flow diagram between a roaming terminal, a resident core network or base station, and a domestic core network or base station, according to one embodiment of the present disclosure.

[0027] FIG. 4 illustrates a signal flow diagram for transferring an AI model to a data center close to a roaming terminal, according to one embodiment of the present disclosure.

[0028] FIG. 5 illustrates a signal flow diagram for transferring an AI model to a local data center according to one embodiment of the present disclosure.

[0029] FIG. 6 illustrates a signal flow diagram for deleting an AI model received by a terminal adjacent data center according to an embodiment of the present disclosure.

[0030] FIG. 7 illustrates an operation method of a core network of a country of residence in which a roaming terminal is located, according to one embodiment of the present disclosure.

[0031] FIG. 8 illustrates an operation method of a roaming terminal according to one embodiment of the present disclosure.

[0032] FIG. 9 illustrates a configuration diagram of a terminal in a wireless communication system according to various embodiments of the present disclosure.

[0033] FIG. 10 illustrates a configuration of a core network device in a wireless communication system according to various embodiments of the present disclosure.

[0034] The terms used in this disclosure are used only to describe specific embodiments and may not be intended to limit the scope of other embodiments. The singular expression may include plural expressions unless the context clearly indicates otherwise. Terms used herein, including technical or scientific terms, may have the same meaning as commonly understood by those of ordinary skill in the art described in this disclosure. Terms defined in general dictionaries among the terms used in this disclosure may be interpreted as having the same or similar meaning in the context of the relevant technology, and shall not be interpreted in an idealized or overly formal sense unless explicitly defined in this disclosure. In some cases, even if a term is defined in this disclosure, it cannot be interpreted to exclude embodiments of the present disclosure.

[0035] The various embodiments of the present disclosure described below illustrate a hardware-based approach as an example. However, since the various embodiments of the present disclosure include techniques utilizing both hardware and software, the various embodiments of the present disclosure do not exclude a software-based approach.

[0036] Additionally, in the detailed description and claims of the present disclosure, “at least one of A, B, and C” can mean “only A,” “only B,” “only C,” or “any combination of A, B, and C.” Additionally, “at least one of A, B, or C” or “at least one of A, B, and / or C” can mean “at least one of A, B, and C.”

[0037] The present disclosure relates to a device and method for providing efficient AI services across countries in a wireless communication system. Specifically, the present disclosure describes a technology for providing necessary AI services by transferring AI models required for AI (artificial intelligence) to a data center close to the terminal.

[0038] The terms used in the following description, including terms referring to signals, channels, control information, network entities, and device components, are provided for convenience of explanation. Therefore, the present disclosure is not limited to the terms described below, and other terms with equivalent technical meanings may be used.

[0039] Additionally, while this disclosure describes various embodiments using terminology used in certain communication standards (e.g., 3rd Generation Partnership Project (3GPP)), these are merely illustrative examples. The various embodiments of this disclosure can be easily modified and applied to other communication systems.

[0040] A base station may be a network infrastructure that provides wireless access to terminals. A base station may have coverage defined as a certain geographic area based on the distance at which a signal can be transmitted. In addition to a base station, a base station may be referred to as an "access point (AP)", an "eNodeB (eNB)", a "5th generation node", a "wireless point", a "transmission / reception point (TRP)", a "next generation Node B (gNodeB, gNB)", or other terms having equivalent technical meanings.

[0041] Each terminal is a device used by a user and can communicate with a base station via a wireless channel. In some cases, at least one of the terminals can be operated without the user's intervention. That is, at least one of the terminals is a device that performs machine type communication (MTC) and may not be carried by the user. Each of the terminals may be referred to as a 'user equipment (UE)', a 'mobile station', a 'subscriber station', a 'remote terminal', a 'wireless terminal', a 'user device', or other terms having equivalent technical meanings.

[0042] A network refers to the entire system for providing wireless communication services, and may include both a radio access network (RAN) and a core network (CN). The support for edge AI infrastructure in this disclosure indicates the capabilities of the entire network system and can be provided through collaboration between the RAN and CN.

[0043] FIG. 1 illustrates a data movement path for AI service use during existing roaming, according to various embodiments of the present disclosure.

[0044] Referring to FIG. 1, in order to use a data service desired to be used in a roaming terminal (hereinafter, roaming terminal), data is sent to a data center through a path in the following order: a roaming terminal (101), a resident country communication base station (e.g., a country A communication base station) (103), a resident country communication core (e.g., a country A communication core) (105), a domestic communication core (107), and a domestic or overseas data center (109, 111) (e.g., an AI service provider).

[0045] A roaming terminal (101) may be a terminal that a user wishes to use data in the country where the user is currently located. The resident country communication base station (103) may be a communication base station in the country where the roaming terminal is located. The resident country communication base station (103) may receive data transmitted from the roaming terminal and transmit the data to the resident country communication core.

[0046] The resident country communication core (105) may be a communication network of the country where the roaming terminal is located. The resident country communication core (105) may process data and transmit the data to the local country communication core (107).

[0047] The domestic data center (109) and overseas data center (111) may be data centers that provide AI services. The domestic data center may be located in the country where the roaming terminal originally belongs, and the overseas data center may be located in a country other than the country where the roaming terminal is located. Data may be transmitted from the domestic communication core (109) to the data center for processing.

[0048] Accordingly, regardless of whether the data center (109, 111) is located within the country or overseas, the roaming terminal (101) must go through the domestic communication server to receive services, and therefore, regardless of the service received, the roaming terminal inevitably has to experience delay time.

[0049] In the existing roaming system shown in Figure 1, each service required a user to pass through their home country's telecommunications network. Therefore, if a terminal could receive services through the same telecommunications network as the local terminal, even while roaming, this delay could be significantly reduced. Figure 2 illustrates the data transfer path for using an AI service that applies the aforementioned idea.

[0050] FIG. 2 illustrates a data movement path for using AI services while roaming, according to one embodiment of the present disclosure.

[0051] Referring to FIG. 2, when a roaming terminal (201) wants to use an AI service held in a data center in the home country or overseas (a country other than the country of residence), the AI ​​model required for the AI ​​service can be transferred to a data center close to the roaming terminal (201) (e.g., a terminal-adjacent data center (209) (Edge data center)) to perform AI inference, and the related AI service can be provided from the terminal-adjacent data center (209) to the roaming terminal (201).

[0052] That is, the AI ​​service held in the domestic data center (211) or overseas data center (213) is transferred to the terminal-adjacent data center (209), and the roaming terminal (201) can perform the AI ​​service from the terminal-adjacent data center via the resident country communication base station (203) (e.g., country A communication base station) and the resident country communication core (205) (e.g., country A communication core).

[0053] FIG. 3 illustrates a signal flow diagram between a roaming terminal, a resident core network or base station, and a domestic core network or base station, according to one embodiment of the present disclosure.

[0054] Referring to FIG. 3, when a roaming terminal (301) wishes to use an AI service while roaming, it may need to subscribe to a separate roaming service (306) to receive AI service-related support from the core network or base station (303) of the country of residence. This separate roaming service is a method for improving the quality of AI services even while roaming, in which the core network or base station (303) of the country of residence provides support that would otherwise be provided by the domestic core network or base station (305). This service is hereinafter referred to as an "AI roaming service."

[0055] A roaming terminal (301) can request AI roaming from a resident country core network or base station (303) (307).

[0056] The resident country core network, or base station (303), may request roaming terminal subscription information (e.g., AI Roaming service) from the local core network, or base station (305) to check whether the roaming terminal (301) is subscribed to the AI ​​roaming service (309).

[0057] The home country core network or base station (305) can transmit roaming terminal subscription information to the resident country core network or base station (303) (311). For example, if the roaming terminal is subscribed to the AI ​​roaming service, 'AI Roaming service=true' can be transmitted as roaming terminal subscription information, and if not, 'AI Roaming service=false' can be transmitted.

[0058] In one embodiment, roaming terminal subscription information (e.g., AI Roaming service) may be mapped to the international mobile subscriber identity (IMSI) of the roaming terminal.

[0059] Referring to FIG. 3, when a roaming terminal (301) cancels AI roaming, the roaming terminal (301) can request cancellation of AI roaming to the local core network or base station (305) (313).

[0060] Accordingly, the home country core network or base station (305) can transmit whether or not to cancel the AI ​​roaming service to the home country core network (315).

[0061] FIG. 4 illustrates a signal flow diagram for transferring an AI model to a data center close to a roaming terminal, according to one embodiment of the present disclosure.

[0062] When a roaming terminal wishes to use an AI service, the resident communication core can search for a data center physically close to the roaming terminal (hereinafter referred to as a "terminal-adjacent data center"). The resident core can request the data center that holds the AI ​​service to transmit the AI ​​model that performs computations for the AI ​​service to the terminal-adjacent data center.

[0063] Terminal-adjacent data centers can receive the AI ​​model, process the computations required for AI services, and provide AI services to roaming terminals based on this data. This process of providing AI services to roaming terminals can be performed independently within the telecommunications core of the country of residence, rather than through the local telecommunications core.

[0064] Specifically, referring to FIG. 4, a roaming terminal can transmit a message to a resident communication core to notify that it wishes to use an AI service (401).

[0065] The resident communication core can recognize that a roaming terminal wants to use an AI service based on a service request of operation (401) (403).

[0066] According to one embodiment, to indicate that a roaming terminal wishes to utilize an AI service (request an AI service), the roaming terminal may transmit a message defining a separate message format to the communication core of the resident country. The message defining the separate message format described above may include at least one of the AI ​​service provider to be utilized, the IP address of the AI ​​service provider to be utilized, or the port of the AI ​​service provider to be utilized.

[0067] In another embodiment, to determine that a roaming terminal intends to use an AI service (request an AI service), the resident communication core may check data packets exchanged with the roaming terminal and, based on information about the AI ​​service provider to which the terminal is connected, determine that the roaming terminal intends to use an AI service. The AI ​​service provider information may include at least one of the AI ​​service provider, the IP address of the AI ​​service provider to be used, or the port of the AI ​​service provider to be used.

[0068] When the resident country communication core recognizes that the roaming terminal is attempting to use the AI ​​service, it can notify the local country communication core that the roaming terminal is attempting to use the AI ​​service (405).

[0069] Upon recognizing that a roaming terminal intends to use an AI service, the resident communication core can search for an available data center that is physically close to the roaming terminal (407). Hereinafter, in this disclosure, a data center that satisfies the conditions of operation (407) is referred to as a "terminal-adjacent data center." A terminal-adjacent data center may be a data center that performs AI inference and computation for AI services.

[0070] In one embodiment, the data center that is physically close to the roaming terminal and is available may be a data center that has the shortest ping RTT (round trip time) among the data centers connected to the resident communication core and that has responded that it is capable of AI inference and computation required for the AI ​​service.

[0071] Specifically, in operation (407), in order to search for a nearby and available data center, the resident communication core pings the surrounding data centers to measure the RTT, and inquires about the top five RTT data centers to determine whether AI inference and computation are possible and receives a response. Among the top five RTT data centers that respond that AI inference and computation are possible, the data center with the fastest RTT can be selected as the terminal-adjacent data center.

[0072] The resident core can search for a terminal-adjacent data center and then transmit the AI ​​model for the AI ​​service to the AI ​​service provider (e.g., a domestic data center or an overseas data center) that the roaming terminal wants to use (409).

[0073] If an AI service provider accepts the AI ​​model transfer request, the AI ​​service provider (e.g., a domestic or overseas data center) can transfer the AI ​​model to a neighboring data center (411). By transferring the AI ​​model to a neighboring data center, the service provider can request that the neighboring data center perform AI inference and other computations.

[0074] In one embodiment, when transmitting an AI model, the AI ​​service provider may transmit at least one of AI service provider information, AI service information, AI model capacity, data information required for AI inference and calculation, or information to be provided to the roaming terminal (e.g., inference or calculation results, etc.) together with the AI ​​model.

[0075] The terminal-adjacent data center can transmit the AI ​​computation results to the roaming terminal or AI service provider (e.g., a domestic data center or an overseas data center) (213). Specifically, in operation (413), the terminal-adjacent data center receives the AI ​​model and performs computations, such as AI inference, on behalf of the data center that originally houses the AI ​​model for the corresponding AI service, thereby providing the corresponding AI service to the roaming terminal. When providing the AI ​​service to the roaming terminal, the AI ​​model can also be transmitted to the AI ​​service provider (e.g., a domestic data center or an overseas data center).

[0076] FIG. 5 illustrates a signal flow diagram for transferring an AI model to a local data center, according to an embodiment of the present disclosure. FIG. 5 illustrates an embodiment for reducing the latency of an AI service by utilizing a data center close to the terminal when the data center where the AI ​​service is stored is physically distant from the terminal or has a latency exceeding a certain time standard (e.g., ping RTT 50 ms, 100 ms, 150 ms, or 200 ms) when the terminal wishes to use an AI service within the local area.

[0077] Referring to FIG. 5, a terminal can request an AI service from its own communication core (501).

[0078] The local communication core can recognize the terminal's AI service request (503) based on the request of the operation (501).

[0079] In one embodiment, in order for the local communication core to recognize the terminal's AI service request, the terminal may transmit a message defined in a separate message format to the local communication core. The message defined in the above-described separate message format may include at least one of the AI ​​service provider to be used, the IP address of the AI ​​service provider to be used, or the port of the AI ​​service provider to be used.

[0080] In another embodiment, in order for the local communication core to recognize the AI ​​service request of the terminal, the local communication core may check the data packets exchanged between the terminal and the local communication core and recognize that the terminal uses the AI ​​service based on information of the AI ​​service provider (e.g., the AI ​​service provider to be used, the IP address of the AI ​​service provider to be used, or the port of the AI ​​service provider to be used).

[0081] The local communication core can search for a nearby data center (505). Recognizing that the terminal is using an AI service, the local communication core can search for an available data center that is physically close to the terminal. A data center that satisfies the conditions of operation (505) can be referred to as a terminal-adjacent data center.

[0082] In one embodiment, the terminal-adjacent data center may be a data center that has the shortest ping RTT among the data centers connected to the local communication core and that has been reported as capable of performing AI inference and computation required for the AI ​​service.

[0083] To discover nearby data centers, the local communication core can ping nearby data centers to measure RTT, and for the top five RTTs, it can inquire and receive a response whether AI inference and computation are possible.

[0084] In one embodiment, among the top five RTT data centers that respond that AI inference and computation are possible, the data center with the fastest RTT may be selected as the terminal-adjacent data center.

[0085] A local communication core may request AI model transmission from an AI service provider (e.g., a local data center or an overseas data center) (507). Specifically, the local communication core may search for an available data center (a terminal-adjacent data center) that is physically close to the terminal, and then request the provider of the AI ​​service the terminal wishes to use to transmit the AI ​​model for that AI service to the terminal-adjacent data center, so that computations such as AI inference can be performed at the terminal-adjacent data center.

[0086] AI service providers (e.g., domestic or overseas data centers) can transfer AI models to adjacent data centers (509). Specifically, if the AI ​​service provider accepts the AI ​​model transfer, the AI ​​model can be transferred from the data center containing the AI ​​model for the AI ​​service to the terminal-adjacent data center.

[0087] According to one embodiment, when transmitting an AI model, the AI ​​service provider may transmit at least one of AI service provider information, AI service information, AI model capacity, data information required for AI inference and calculation, or information to be provided to the terminal (e.g., inference or calculation results, etc.).

[0088] The terminal-adjacent data center can transmit AI computation results to the terminal, the local communication core, or the AI ​​service provider (511). Specifically, the terminal-adjacent data center receives the AI ​​model, performs AI inference and other computations on behalf of the data center that holds the original AI model for the AI ​​service, and uses this to provide the AI ​​service to the terminal. At this time, the AI ​​computation results can be transmitted to the AI ​​service provider that originally held the AI ​​model.

[0089] Figure 6 illustrates a signal flow diagram for deleting an AI model transferred to a terminal-adjacent data center, according to one embodiment of the present disclosure. Figure 6 illustrates a process in which an AI model for an AI service is transferred to a data center near the terminal, and then the AI ​​service is provided. When the AI ​​service is terminated, the AI ​​service provider can request the data center to which the AI ​​model was transferred to delete the AI ​​model.

[0090] The AI ​​service provider may notify the data center (hereinafter referred to as the “terminal adjacent data center”) that received the AI ​​model corresponding to the termination of the AI ​​service (601).

[0091] An AI service provider may request the terminal adjacent data center to delete the AI ​​model (603). Operation (601) may be performed together with or separately from operation (601).

[0092] In one embodiment, a message requesting deletion of an AI model may include at least one of AI service provider information, AI service information, or AI model capacity.

[0093] According to one embodiment, a message requesting deletion of an AI model may be requested after a certain period of time (e.g., 10 minutes, 30 minutes, 1 hour, 2 hours, 4 hours, 6 hours, 12 hours, 18 hours, 24 hours, etc.) has elapsed after the terminal has finished using the AI ​​service.

[0094] The terminal-adjacent data center can delete the AI ​​model and transmit the AI ​​model deletion result to the AI ​​service provider (605). Specifically, the terminal-adjacent data center that received the AI ​​model can receive an AI model deletion request message, perform the AI ​​model deletion, and then transmit a deletion completion message to the AI ​​service provider.

[0095] In one embodiment, if the AI ​​service provider does not receive an AI model deletion completion message from the terminal-adjacent data center within a certain period of time, the terminal-adjacent data center may retransmit an AI model deletion request message. The retransmission cycle for the AI ​​model deletion request message may be set to 10 minutes, 30 minutes, 1 hour, etc., but is not limited thereto.

[0096] In one embodiment, an AI service provider that receives an AI model deletion completion message may terminate the connection with a data center adjacent to the terminal that transmitted the AI ​​model deletion message.

[0097] FIG. 7 illustrates an operation method of a core network of a country of residence in which a roaming terminal is located, according to one embodiment of the present disclosure.

[0098] The operation of Figure 7 allows the resident core network to verify whether the roaming terminal has subscribed to the AI ​​roaming service, collect and process the information necessary to provide the AI ​​service, and transmit the AI ​​model to the appropriate data center, thereby providing the roaming terminal with rapid AI service. The specific operations are as follows.

[0099] Referring to FIG. 7, the core network of a country in which a roaming terminal is located may receive a request for AI roaming from the roaming terminal (701). The request for operation (701) may include information regarding the roaming terminal's desire to use the AI ​​roaming service.

[0100] The resident core network can check whether the roaming terminal is subscribed to the AI ​​roaming service based on the request of the operation (701) (703).

[0101] In one embodiment, the resident core network may request the roaming terminal's home core network for subscription information of the roaming terminal to determine whether the roaming terminal is subscribed to an AI service.

[0102] Additionally, according to one embodiment, the resident core network may determine whether the roaming terminal is subscribed to the AI ​​roaming service based on the subscription information of the roaming terminal to determine whether the roaming terminal is subscribed to the AI ​​service.

[0103] In one embodiment, the subscription information of the roaming terminal may be mapped to the international mobile subscriber identity (IMSI) of the roaming terminal.

[0104] The resident core network may receive a message from a roaming terminal indicating that the terminal wishes to use the AI ​​service if the roaming terminal has subscribed to the AI ​​roaming service (705). In one embodiment, the message of operation (705) may include at least one of the AI ​​service provider, the IP address of the AI ​​service provider to be used, or the port of the AI ​​service provider to be used.

[0105] According to one embodiment, the resident core network may, in order to receive the message of operation (705), recognize that the roaming terminal intends to use the AI ​​service by checking the data packet exchanged with the roaming terminal based on the AI ​​service provider information to which the roaming terminal is connected, or recognize that the roaming terminal intends to use the AI ​​service by receiving a message from the roaming terminal notifying that the roaming terminal intends to use the AI ​​service.

[0106] Additionally, according to one embodiment, the resident core network may notify the roaming terminal's home core network that the roaming terminal intends to use the AI ​​service in order to recognize that the roaming terminal intends to use the AI ​​service.

[0107] Additionally, according to one embodiment, the resident core network may search for an available data center that is physically close to the roaming terminal to recognize that the roaming terminal is attempting to use the AI ​​service.

[0108] In addition, according to one embodiment, the resident core network may measure the round trip time (RTT) by sending a ping to a nearby data center that is connected to the resident core network of the country where the roaming terminal is located in order to search for an available data center that is physically close to the roaming terminal, and may inquire about the possibility of AI inference and calculation to the top five data centers in descending order of the ping round trip time, and may select the data center with the shortest ping round trip time among the data centers that responded that inference and calculation are possible.

[0109] The resident core network may request the AI ​​service provider that the roaming terminal wishes to use to transmit an AI model corresponding to the AI ​​service request to a data center that performs AI inference and calculation for the AI ​​service based on the message of operation (705) (707).

[0110] FIG. 8 illustrates an operation method of a roaming terminal according to one embodiment of the present disclosure.

[0111] Figure 8 describes a method for enabling roaming terminals to efficiently utilize AI services outside their country of residence. This method enables roaming terminals to request AI services through collaboration between their home core network and the core network of their country of residence, and for the necessary AI models and data to be processed at an appropriate data center within their country of residence and quickly provided to the roaming terminal. The specific operations are as follows.

[0112] Referring to FIG. 8, a roaming terminal can subscribe to AI roaming to receive AI services from the roaming terminal's home core network (801). Operation (801) may be an operation that supports the roaming terminal's use of AI services in countries other than its home country.

[0113] A roaming terminal may request AI roaming from the core network of the country in which the roaming terminal is located (803). The request for operation (803) may include information regarding the roaming terminal's desire to use the AI ​​service.

[0114] A roaming terminal may send a message to the core network of the country in which the roaming terminal is located to notify that it wishes to use the AI ​​service (805).

[0115] According to one embodiment, the message of operation (805) may include at least one of an AI service provider, an IP address of an AI service provider to be used, or a port of an AI service provider to be used.

[0116] A roaming terminal may receive an AI computation result corresponding to an AI service from a data center (807). In one embodiment, the AI ​​computation result may be determined based on at least one of an AI model corresponding to the AI ​​service, AI service information, AI model capacity, data information required for AI inference and computation, or information to be provided to the roaming terminal.

[0117] When a roaming terminal cancels AI roaming, it may request the roaming terminal's home core network to cancel AI roaming (not shown).

[0118] FIG. 9 illustrates a configuration diagram of a terminal in a wireless communication system according to various embodiments of the present disclosure. The configuration illustrated in FIG. 7 can be understood as a configuration of a terminal. Terms such as "... unit" and "... unit" used hereinafter refer to a unit that processes at least one function or operation, which can be implemented using hardware, software, or a combination of hardware and software.

[0119] Referring to FIG. 9, the terminal may include a communication unit (910), a storage unit (920), and a control unit (930).

[0120] The communication unit (910) may perform functions for transmitting and receiving signals via a wireless channel. For example, the communication unit (910) may perform a conversion function between a baseband signal and a bit stream according to the physical layer specifications of the system. For example, when transmitting data, the communication unit (910) may generate complex symbols by encoding and modulating a transmission bit stream. When receiving data, the communication unit (910) may restore a reception bit stream by demodulating and decoding the baseband signal. In addition, the communication unit (910) may upconvert a baseband signal to an RF band signal and transmit it through an antenna, and downconvert an RF band signal received through the antenna to a baseband signal. For example, the communication unit (910) may include a transmission filter, a reception filter, an amplifier, a mixer, an oscillator, a DAC, an ADC, etc.

[0121] In addition, the communication unit (910) may include a plurality of transmission and reception paths. Furthermore, the communication unit (910) may include at least one antenna array composed of a plurality of antenna elements. In terms of hardware, the communication unit (910) may be composed of digital circuits and analog circuits (e.g., a radio frequency integrated circuit (RFIC)). Here, the digital circuits and analog circuits may be implemented in a single package. In addition, the communication unit (910) may include a plurality of RF chains. Furthermore, the communication unit (910) may perform beamforming.

[0122] The communication unit (910) transmits and receives signals as described above. Accordingly, all or part of the communication unit (910) may be referred to as a "transmitter," a "receiver," or a "transceiver." Furthermore, in the following description, transmission and reception performed via a wireless channel may be used to mean that the communication unit (910) performs the processing described above.

[0123] The storage unit (920) stores data such as basic programs, application programs, and setting information for the operation of the terminal. The storage unit (920) may be composed of volatile memory, non-volatile memory, or a combination of volatile and non-volatile memory. In addition, the storage unit (920) provides the stored data upon request from the control unit (930).

[0124] The control unit (930) can control the overall operations of the terminal. For example, the control unit (930) can transmit and receive signals through the communication unit (910). In addition, the control unit (930) can record and read data in the storage unit (920). The control unit (930) can perform the functions of the protocol stack required by the communication standard. To this end, the control unit (930) may include at least one processor or microprocessor, or may be a part of a processor. In addition, a part of the communication unit (910) and the control unit (930) may be referred to as a CP (communication processor).

[0125] According to various embodiments, the control unit (930) can control the terminal to perform operations according to various embodiments performed in FIGS. 1 to 8.

[0126] FIG. 10 illustrates a configuration of a core network device in a wireless communication system according to various embodiments of the present disclosure. Terms such as “… unit” and “… device” used hereinafter mean a unit that processes at least one function or operation, which may be implemented by hardware, software, or a combination of hardware and software.

[0127] Referring to FIG. 10, the core network device is configured to include a communication unit (1010), a storage unit (1020), and a control unit (1030).

[0128] The communication unit (1010) provides an interface for performing communication with other devices within the network. That is, the communication unit (1010) converts a bit string transmitted from a core network device to another device into a physical signal, and converts a physical signal received from another device into a bit string. That is, the communication unit (1010) can transmit and receive signals. Accordingly, the communication unit (1010) may be referred to as a modem, a transmitter, a receiver, or a transceiver. At this time, the communication unit (1010) enables the core network device to communicate with other devices or systems via a backhaul connection (e.g., wired backhaul or wireless backhaul) or via a network.

[0129] The storage unit (1020) stores data such as basic programs, applications, and configuration information for the operation of the core network device. The storage unit (1020) may be composed of volatile memory, non-volatile memory, or a combination of volatile and non-volatile memory. In addition, the storage unit (1020) provides stored data upon request from the control unit (1030).

[0130] The control unit (1030) controls the overall operations of the core network device. For example, the control unit (1030) transmits and receives signals through the communication unit (1010). In addition, the control unit (1030) records and reads data from the storage unit (1020). For this purpose, the control unit (1030) may include at least one processor. According to various embodiments, the control unit (1030) may control the core network to perform operations according to various embodiments described above in FIGS. 1 to 8.

[0131] The methods according to the embodiments described in the claims or specification of the present disclosure may be implemented in the form of hardware, software, or a combination of hardware and software.

[0132] When implemented in software, a computer-readable storage medium storing one or more programs (software modules) may be provided. The one or more programs stored in the computer-readable storage medium are configured for execution by one or more processors within an electronic device. The one or more programs include instructions that cause the electronic device to execute methods according to the embodiments described in the claims or specification of the present disclosure.

[0133] These programs (software modules, software) may be stored in random access memory, non-volatile memory including flash memory, read only memory (ROM), electrically erasable programmable read only memory (EEPROM), magnetic disc storage devices, compact disc-ROM (CD-ROM), digital versatile discs (DVDs) or other forms of optical storage devices, magnetic cassettes, or may be stored in memories formed by a combination of some or all of these. In addition, each configuration memory may include multiple copies.

[0134] Additionally, the program may be stored on an attachable storage device that is accessible via a communication network, such as the Internet, an intranet, a local area network (LAN), a wide area network (WAN), a storage area network (SAN), or a combination thereof. Such a storage device may be connected to a device performing an embodiment of the present disclosure via an external port. Additionally, a separate storage device on the communication network may be connected to a device performing an embodiment of the present disclosure.

[0135] In the specific embodiments of the present disclosure described above, components included in the disclosure are expressed in the singular or plural form, depending on the specific embodiment presented. However, the singular or plural expressions are selected to suit the presented situation for convenience of explanation, and the present disclosure is not limited to singular or plural components. Components expressed in the plural form may be composed of singular elements, or components expressed in the singular form may be composed of plural elements.

[0136] While the detailed description of this disclosure has described specific embodiments, it should be understood that various modifications are possible without departing from the scope of this disclosure. Therefore, the scope of this disclosure should not be limited to the described embodiments, but should be defined not only by the scope of the claims described below, but also by equivalents thereof.

Claims

1. A method for providing AI roaming service to a terminal roaming in a country of residence, performed by a core network of the country of residence, A process of obtaining a request for an AI (artificial intelligence) service provided by a designated AI service provider within the country from the terminal located in the country of residence, In accordance with the above request, a process of selecting an AI inference performer required for the AI ​​service from among multiple local data centers provided in the country of residence, A process for controlling the resources required for the above AI inference to be transferred from the above AI service provider to the above selected local data center, A process for controlling the results of AI inference performed by the above-mentioned selected local data center using the above-mentioned resources to be provided to the terminal located in the above-mentioned country of residence. A method for providing AI roaming services.

2. In paragraph 1, The above method, Further comprising a process for checking whether the terminal is subscribed to the AI ​​roaming service, The subject performing the above AI inference is: For terminals subscribed to the above AI roaming service, it is the selected local data center, and for terminals not subscribed to the above AI roaming service, it is the AI ​​service provider. A method for providing AI roaming services.

3. In paragraph 1, In the above selection process, At least one of whether each of the plurality of local data centers can perform the AI ​​inference, the latency until the AI ​​roaming service is provided, the distance between the terminal located in the country of residence and each of the plurality of local data centers, and the amount of resources required for each of the plurality of local data centers to perform the AI ​​inference is taken into consideration. A method for providing AI roaming services.

4. In paragraph 1, Resources transferred from the AI ​​service provider to the selected local data center include: Information about an AI model learned for the AI ​​service, environmental information required for executing the AI ​​model, and at least one of the data formats of the input or output of the AI ​​model are included. A method for providing AI roaming services.

5. In paragraph 1, The types of resources transmitted above are: Depends on the specifications provided in the selected local data center. A method for providing AI roaming services.

6. In paragraph 1, Resources delivered to the selected local data center above are: Under certain conditions, the selected local data center is no longer available. A method for providing AI roaming services.

7. A method for providing AI roaming services performed by a terminal roaming in a country of residence. The process of requesting AI (artificial intelligence) services provided by a designated AI service provider within the country while located in the country of residence; Including a process of obtaining the result of AI inference corresponding to the request from a local data center selected from among multiple local data centers provided in the above-mentioned country of residence, AI inference corresponding to the above request is: The selected local data center that received the resources required for the AI ​​inference from the AI ​​service provider and performed the process using the resources. How to receive AI roaming services.

8. In the core network of the country of residence where the terminal roaming in the country of residence is located, Transmitter and receiver, A control unit operably connected to the above transmitter and receiver, wherein the control unit comprises: Obtaining a request for an AI (artificial intelligence) service provided by a designated AI service provider within the country from the terminal located in the country of residence, In accordance with the above request, the AI ​​inference performer required for the AI ​​service is selected from among multiple local data centers provided in the country of residence, Controlling that the resources required for the above AI inference are transferred from the above AI service provider to the above selected local data center, Controlling the results of AI inference performed by the above-selected local data center using the above-mentioned resources to be provided to the terminal located in the above-mentioned country of residence Resident country core network.

9. In paragraph 8, The above control unit, Check if the above terminal is subscribed to the AI ​​roaming service, The subject performing the above AI inference is: For terminals subscribed to the above AI roaming service, it is the selected local data center, and for terminals not subscribed to the above AI roaming service, it is the AI ​​service provider. Resident country core network.

10. In paragraph 8, In the above selection process, At least one of whether each of the plurality of local data centers can perform the AI ​​inference, the latency until the AI ​​roaming service is provided, the distance between the terminal located in the country of residence and each of the plurality of local data centers, and the amount of resources required for each of the plurality of local data centers to perform the AI ​​inference is taken into consideration. Resident country core network.

11. In paragraph 8, Resources transferred from the AI ​​service provider to the selected local data center include: Information about an AI model learned for the AI ​​service, environmental information required for executing the AI ​​model, and at least one of the data formats of the input or output of the AI ​​model are included. Resident country core network.

12. In paragraph 8, The types of resources transmitted above are: Depends on the specifications provided in the selected local data center. Resident country core network.

13. In paragraph 8, Resources delivered to the selected local data center above are: Under certain conditions, the selected local data center is no longer available. Resident country core network.

Images