Mechanism for accessing knowledge sources

Abstract

Example embodiments of the present disclosure relate to mechanisms for accessing knowledge sources. A method includes transmitting, at a first device and to a second device, a discovery request to access a knowledge source. The method also includes receiving a discovery response from the second device, the discovery response including a list of network function instances, the list of network function instances including a knowledge source and a profile of the list of network function instances. The method also includes transmitting an access token request to the knowledge source to a second device. The method also includes receiving an access token response from the second device, the access token response including at least the identification of the knowledge source.

Description

Technical Field

[0001] Various exemplary embodiments of this disclosure generally relate to the telecommunications field, and more particularly to methods, apparatuses, devices, and computer-readable storage media for accessing knowledge sources for retrieving augmented generation (RAG). Background Technology

[0002] Artificial intelligence (AI) and machine learning (ML) technologies are increasingly being used in communication systems and are considered key enablers of the 5G-Advanced and 6G mobile network generations. Therefore, how to more efficiently adopt AI / ML technologies is a topic worthy of further research. Summary of the Invention

[0003] In a first aspect of the invention, a first apparatus is provided. The first apparatus includes at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the first apparatus to: transmit a discovery request for access to a knowledge source to a second apparatus; receive a discovery response from the second apparatus, the discovery response including a list of network function instances, the list of network function instances including knowledge sources and profiles of the network function instances; transmit an access token request for the knowledge source to the second apparatus; and receive an access token response from the second apparatus, the access token response including at least an identifier of the knowledge source.

[0004] In a second aspect of the invention, a second apparatus is provided. The second apparatus includes at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the second apparatus to: receive from a first apparatus a discovery request for access to a knowledge source; transmit to the first apparatus a discovery response including a list of network function instances, the list including knowledge sources and profiles of the network function instances; receive from the first apparatus an access token request for the knowledge source; and transmit to the first apparatus an access token response, the access token response including at least an identifier of the knowledge source.

[0005] In a third aspect of the invention, a third apparatus is provided. The third apparatus includes at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the third apparatus to: receive a knowledge service request from a first apparatus, the knowledge service request including a token of a knowledge source; determine, based on the token, whether the first apparatus is authorized to access the knowledge source; and transmit a knowledge service response to the first apparatus based on the determination that the first apparatus is authorized to access the knowledge source.

[0006] In a fourth aspect of the invention, a method is provided. The method includes: transmitting a discovery request for access to a knowledge source to a second device; receiving a discovery response from the second device, the discovery response including a list of network function instances, the list of network function instances including knowledge sources and profiles of the network function instances; transmitting an access token request for the knowledge source to the second device; and receiving an access token response from the second device, the access token response including at least an identifier of the knowledge source.

[0007] In a fifth aspect of the invention, a method is provided. The method includes: receiving a discovery request for access to a knowledge source from a first device; transmitting a discovery response to the first device, the discovery response including a list of network function instances, the list of network function instances including knowledge sources and profiles of the network function instances; receiving an access token request for the knowledge source from the first device; and transmitting an access token response to the first device, the access token response including at least an identifier of the knowledge source.

[0008] In a sixth aspect of the invention, a method is provided. The method includes: receiving a knowledge service request from a first device, the knowledge service request including a token of a knowledge source; determining, based on the token, whether the first device is authorized to access the knowledge source; and transmitting a knowledge service response to the first device based on the determination that the first device is authorized to access the knowledge source.

[0009] In a seventh aspect of the invention, a first apparatus is provided. The first apparatus includes means for transmitting a discovery request for access to a knowledge source to a second apparatus; means for receiving a discovery response from the second apparatus, the discovery response including a list of network function instances, the list including knowledge sources and profiles of the network function instances; means for transmitting an access token request for the knowledge source to the second apparatus; and means for receiving an access token response from the second apparatus, the access token response including at least an identifier of the knowledge source.

[0010] In an eighth aspect of the invention, a second apparatus is provided. The second apparatus includes: means for receiving from a first apparatus a discovery request for access to a knowledge source; means for transmitting to the first apparatus a discovery response including a list of network function instances, the list including knowledge sources and profiles of the network function instances; means for receiving from the first apparatus an access token request for the knowledge source; and means for transmitting to the first apparatus an access token response, the access token response including at least an identifier of the knowledge source.

[0011] In a ninth aspect of the invention, a third apparatus is provided. The third apparatus includes: means for receiving a knowledge service request from a first apparatus, the knowledge service request including a token of a knowledge source; means for determining, based on the token, whether the first apparatus is authorized to access the knowledge source; and means for transmitting a knowledge service response to the first apparatus based on the determination that the first apparatus is authorized to access the knowledge source.

[0012] In a tenth aspect of the invention, a computer-readable medium is provided. The computer-readable medium includes instructions stored thereon for causing an apparatus to perform at least the method according to the fourth, fifth, or sixth aspect.

[0013] It should be understood that the summary portion is neither intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to be used to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0014] Now, some exemplary embodiments will be described with reference to the accompanying drawings, in which:

[0015] Figure 1 An example communication environment in which this disclosure can be implemented is shown;

[0016] Figure 2 An example of enhanced user prompts (or queries) via a multi-layered RAG framework is shown;

[0017] Figure 3 Example deployments of primary knowledge / information sources and secondary knowledge / information sources for RAG are shown according to some example embodiments of this disclosure;

[0018] Figure 4 Example deployments of primary knowledge / information sources and secondary knowledge / information sources for RAG are shown according to some other example embodiments of this disclosure;

[0019] Figure 5 Example roles for discovering and selecting relevant primary and secondary RAG instances are shown, respectively, according to some example embodiments of this disclosure;

[0020] Figures 6A to 6D Examples of network RAG consumers according to some exemplary embodiments of this disclosure are shown respectively;

[0021] Figure 7 The signaling flow for accessing knowledge sources is illustrated according to some example embodiments of this disclosure;

[0022] Figure 8 The signaling flow for knowledge source selection according to some example embodiments of this disclosure is shown;

[0023] Figure 9 A flowchart is shown illustrating a method implemented at a first device according to some exemplary embodiments of the present disclosure;

[0024] Figure 10 A flowchart is shown illustrating a method implemented at a second device according to some example embodiments of the present disclosure;

[0025] Figure 11 A flowchart is shown illustrating a method implemented at a third device according to some example embodiments of the present disclosure;

[0026] Figure 12 A simplified block diagram of a device suitable for implementing example embodiments of the present disclosure is shown; and

[0027] Figure 13 A block diagram of an example computer-readable medium according to some example embodiments of the present disclosure is shown.

[0028] In all the accompanying drawings, the same or similar reference numerals denote the same or similar elements. Detailed Implementation

[0029] The principles of this disclosure will now be described with reference to some exemplary embodiments. It should be understood that these embodiments are described for illustrative purposes only and to assist those skilled in the art in understanding and implementing this disclosure, and do not imply any limitation on the scope of this disclosure. The various embodiments described herein can be implemented in various ways other than those described below.

[0030] In the following description and claims, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains.

[0031] In this disclosure, references to "an embodiment," "an example embodiment," "an exemplary embodiment," etc., indicate that the described embodiment may include a particular feature, structure, or characteristic, but each embodiment need not necessarily include that particular feature, structure, or characteristic. Furthermore, these phrases do not necessarily refer to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it should be assumed that its influence in conjunction with other embodiments (whether explicitly described or not) is within the knowledge of those skilled in the art.

[0032] It should be understood that although the terms "first," "second," etc., preceding nouns may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another, and they do not restrict the order of one or more nouns. For example, without departing from the scope of the exemplary embodiments, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element. As used herein, the term "and / or" includes any and all combinations of one or more of the listed terms.

[0033] As used herein, “at least one of the following: ” and “at least one of ” and similar wording (where a list of two or more elements is combined by “and” or “or”) means at least any one of the elements, or at least any two or more of the elements, or at least all of the elements.

[0034] As used herein, unless explicitly stated otherwise, the execution step “in response to A” does not indicate that the step is performed immediately after “A” occurs, and may include one or more intermediate steps.

[0035] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments. As used herein, unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “the” are also intended to include the plural forms. It should also be understood that the terms “comprises,” “comprising,” “has,” “having,” “includes,” and / or “including” as used herein specify the presence of the stated features, elements, and / or components, etc., but do not exclude the presence or addition of one or more other features, elements, components, and / or combinations thereof.

[0036] As used in this application, the term "circuit system" may refer to one or more or all of the following: (a) Hardware-only implementations (such as implementations in analog and / or digital circuit systems only); and (b) A combination of hardware circuitry and software, such as (if applicable): (i) A combination of one or more analog and / or digital hardware circuits with software / firmware, and (ii) Any part of a hardware processor (including one or more digital signal processors), software, and one or more memories that work together to enable a device such as a mobile phone or server to perform various functions; and (c) One or more hardware circuits and / or one or more processors (such as one or more microprocessors or a portion thereof) that require software (e.g., firmware) to operate, but which may not exist when no software is required to operate.

[0037] This definition of "circuit system" applies to all uses of the term in this application (including in any claim). As another example, as used herein, the term "circuit system" also covers implementations of hardware circuitry or processors (or processors) or portions thereof and their accompanying software and / or firmware. The term "circuit system" also covers (e.g., and if applicable to elements of a particular claim) baseband integrated circuits or processor integrated circuits for mobile devices or similar integrated circuits in servers, cellular network devices, or other computing or networking devices.

[0038] As used herein, the term "communication network" refers to a network that conforms to any suitable communication standard, such as New Radio (NR), Long Term Evolution (LTE), LTE-A Advanced (LTE-A), Wideband Code Division Multiple Access (WCDMA), High-Speed ​​Packet Access (HSPA), Narrowband Internet of Things (NB-IoT), etc. Furthermore, communication between terminal devices and network devices in a communication network can be performed according to any suitable generation communication protocol, including but not limited to first-generation (1G), second-generation (2G), 2.5G, 2.75G, third-generation (3G), fourth-generation (4G), 4.5G, fifth-generation (5G), 5.5G, sixth-generation (6G) communication protocols and / or any other currently known or future protocols. The embodiments of this disclosure can be applied to various communication systems. Given the rapid development of communications, and of course, future types of communication technologies and systems, this disclosure can be implemented using these communication technologies and systems. It should not be construed as limiting the scope of this disclosure to the systems described above.

[0039] As used herein, the term "network device" refers to a node in a communication network through which terminal devices access the network and receive services. Depending on the terminology and technology applied, a network device can refer to a base station (BS) or access point (AP), such as a Node B (Node B or NB), an evolved Node B (eNodeB or eNB), an NR NB (also known as a gNB), a Remote Radio Unit (RRU), a Radio Header (RH), a Remote Radio Header (RRH), a relay, an Integrated Access and Backhaul (IAB) node, low-power nodes such as femtoseconds and picoseconds, non-terrestrial network (NTN) or non-terrestrial network equipment (such as satellite network equipment), low Earth orbit (LEO) satellites and geostationary Earth orbit (GEO) satellites, spacecraft network equipment, etc. In some example embodiments, the Radio Access Network (RAN) split architecture includes a centralized unit (CU) and a distributed unit (DU) at the IAB donor node. An IAB node includes a mobile terminal (IAB-MT) portion that behaves similarly to a UE approaching a parent node, and a DU portion of the IAB node that behaves similarly to a base station approaching a next-hop IAB node.

[0040] The term "terminal device" refers to any terminal device capable of wireless communication. By way of example and not limitation, a terminal device may also be referred to as a communication device, user equipment (UE), subscriber station (SS), portable subscriber station, mobile station (MS), or access terminal (AT). Terminal devices may include, but are not limited to, mobile phones, cellular phones, smartphones, Voice over IP (VoIP) phones, wireless local loop phones, tablets, wearable terminal devices, personal digital assistants (PDAs), portable computers, desktop computers, image capture terminal devices such as digital cameras, gaming terminal devices, music storage and playback devices, vehicle-mounted wireless terminal devices, wireless endpoints, mobile stations, laptop embedded devices (LEEs), laptop mounted devices (LMEs), USB dongles, smart devices, wireless customer premises equipment (CPEs), Internet of Things (IoT) devices, watches or other wearable devices, head-mounted displays (HMDs), vehicles, drones, medical devices and applications (e.g., remote surgery), industrial devices and applications (e.g., robots and / or other wireless devices operating in the context of industrial and / or automated processing chains), consumer electronics devices, devices operating on commercial and / or industrial wireless networks, etc. Terminal devices may also correspond to the mobile termination (MT) portion of an IAB node (e.g., a relay node). In the following description, the terms "terminal equipment", "communication equipment", "terminal", "user equipment" and "UE" are used interchangeably.

[0041] As used herein, the terms “resource,” “transmission resource,” “resource block,” “physical resource block” (PRB), “uplink resource,” or “downlink resource” can refer to any resource used to perform communication (e.g., communication between a terminal device and a network device), such as resources in the time domain, resources in the frequency domain, resources in the spatial domain, resources in the code domain, or any other combination of time, frequency, spatial, and / or code domain resources that enable communication. In the following, unless explicitly stated otherwise, resources in both the frequency and time domains will be used as examples of transmission resources used to describe some exemplary embodiments of the invention. It should be noted that the exemplary embodiments of this disclosure are equally applicable to other resources in other domains. The term “source of knowledge” as used herein can refer to one or more entities that implement the RAG functionality.

[0042] Figure 1 An example communication environment 100 in which this disclosure can be implemented is shown. In communication environment 100, a plurality of communication devices, including a first device 110, a second device 120, and a third device 130, can communicate with each other. Figure 1 In the example, the first device 110 may be a Large Language Model (LLM) agent. The first device 110 may be implemented on an end device. The second device 120 may be a network device, such as an NRF. The third device 130 may be another network device, such as an Analysis Data Repository Function (ADRF).

[0043] It should be understood that Figure 1 The number of devices and their connections shown are for illustrative purposes only and do not imply any limitation. Communication environment 100 may include any suitable number of devices configured to implement the exemplary embodiments of this disclosure. Although not shown, it should be understood that one or more additional cells may be deployed in communication environment 100.

[0044] Communication in communication environment 100 can be implemented according to any suitable communication protocol, including but not limited to cellular communication protocols (such as wireless local area network communication protocols like IEEE 802.11) and / or any other currently known or to be developed in the future. Furthermore, the communication can utilize any suitable wireless communication technology, including but not limited to: Code Division Multiple Access (CDMA), Frequency Division Multiple Access (FDMA), Time Division Multiple Access (TDMA), Frequency Division Duplex (FDD), Time Division Duplex (TDD), Multiple Input Multiple Output (MIMO), Orthogonal Frequency Division Multiplexing (OFDM), Discrete Fourier Transform Extended OFDM (DFT-s-OFDM), and / or any other currently known or to be developed in the future.

[0045] As mentioned above, AI / ML technologies are used in communication systems. Large Language Models (LLMs) are pre-trained generative transforms (GPTs) with a large number of trainable parameters, designed to predict and generate natural and logical language. The transformer architecture relies on a deep neural network (DNN) architecture that uses a parallel multi-head attention mechanism to indicate the importance of specific words or parts of words. This allows for amplification of signals for key tokens and reduction of signals for less important tokens, a heuristic particularly prominent in natural language processing.

[0046] In some cases, LLM can be used for simple software code generation, customer care services, creating customer documentation, standardizing document search and generation, and network troubleshooting. Through certain measures, GPT-4's ability to generate information when prompted with natural language is satisfactory.

[0047] It can point to several data science reasons why Chat GPT results might be questionable and should be treated with caution, especially given the unavailability of the training dataset. For example, the most likely explanation for GPT-4's performance is that the training data includes test data. One cannot expect to replicate every piece of information on the internet, including the tests used to evaluate GPT-4's performance. Another explanation might be that the performance is exaggerated. When delving deeper into the performance results, it is often found that they are not presented scientifically but rather based on a few examples extrapolated from the performance. Finally, it is entirely possible that GPT is able to learn language patterns or how to build standardized tests that are not obvious to humans.

[0048] Due to its inherent unreliability, the output of an LLM can always be thoroughly tested or checked by human experts before being applied to any mission-critical system.

[0049] Most publicly available LLMs are trained on large, general-purpose corpora and may perform poorly in specific contexts / domains. To improve the responses generated by LLMs, various techniques exist for adding domain knowledge to them, such as fine-tuning, contextual learning / hinting, and retrieval-enhanced generation (RAG).

[0050] Fine-tuning aims to train some (or all) of the model parameters using a small amount or high-quality (labeled) domain / task-specific data. This approach works even for smaller models, provided that a sufficient quantity and quality of domain-specific training data is available. However, retraining can be costly, especially for larger models, or when large amounts of domain-specific training data are required.

[0051] The goal of contextual learning / hints is to enhance user input using relevant domain-specific knowledge; that is, during inference time, users can provide contextual information (such as a task description) along with a few input / output examples in the hints. This customization is flexible, easily adaptable to changes in domain facts, and performs well even with a small number of input / output examples. However, it is primarily suitable for large models. Further research is needed on the standardized relevance of the format of the hint examples, the ontology of the hints, and the interfaces between the entities involved in the hints.

[0052] Retrieval-Enhanced Generation (RAG) is an AI framework used to improve the quality of LLM-generated responses by supplementing the LLM's internal information representation by placing the model on top of external knowledge sources, enabling it to access information beyond its training data. It comprises two phases: a retrieval phase and a generation phase. In the retrieval phase, the algorithm searches for and retrieves the most relevant information fragments to the user's prompt or question. This external knowledge is attached to the user's prompt and passed to the language model. In the generation phase, the LLM extracts enhanced prompts from its training data and its internal representations to synthesize a tailored answer; the final response is a combination of the retrieved information and the model's own generative capabilities. This approach ensures that the LLM has access to the most relevant and current reliable facts, thereby improving accuracy and reducing illusions. Note: In the context of LLM, an illusion is a phenomenon where the LLM generates content that is meaningless to the source content or unfaithful to the original content. Additionally, the RAG approach reduces the cost of updating the LLM model (e.g., regarding retraining / fine-tuning).

[0053] Regarding the GenAI agent framework envisioned in 6G telecommunications, a mechanism is needed to enable multiple RAGs to interact based on: the type and quantity of available knowledge, the sensitivity of the data being processed, and the trustworthiness of the end GenAI agents acting as consumers who can access these RAGs. Furthermore, RAG systems (single or multiple RAGs) face the following potential challenges, especially considering their use within the telecommunications domain. They may lack the necessary access control mechanisms to allow authorized consumers / agents access to different types of knowledge (e.g., generic (i.e., standardized / shareable knowledge between network vendors and / or mobile operators), vendor / operator sensitive data that network vendors / mobile operators want to keep confidential, and one or more third-party RAGs consisting of information from the outside world). For unauthorized third-party LLM agents wanting to retrieve any operator deployment details, it is necessary to avoid accessing sensitive operator information and to prevent single points of failure.

[0054] These challenges highlight the need for a more sophisticated, multi-layered RAG approach, while also ensuring corresponding security mechanisms are in place. Therefore, new solutions are needed for the RAG portion of the telecom LLM framework; more specifically, architectural enhancements are required to achieve multi-layered RAG and their corresponding security requirements.

[0055] According to some example embodiments of this disclosure, a solution for accessing knowledge sources is provided. Specifically, a first device transmits a discovery request for access to the knowledge source to a second device. The first device receives a discovery response from the second device, which includes a list of network function instances, the list of network function instances including knowledge sources and profiles of the network function instances. The first device transmits an access token request for the knowledge source to the second device. The access token request includes at least an identifier of the knowledge source. The first device receives an access token response from the second device, the access token response including at least an identifier of the knowledge source. Furthermore, a multi-layered knowledge source is proposed. This ensures the security of accessing the knowledge source.

[0056] Referring to the accompanying drawings, exemplary embodiments of this disclosure are described below.

[0057] In some example embodiments, a multi-layered knowledge source that can be used for RAG is applied to the communication environment 100. For example, the multi-layered knowledge / information source of RAG may include at least two layers, namely, a primary layer and a secondary layer. In some example embodiments, the primary knowledge / information source and the secondary knowledge / information source of RAG may reside on the same entity. Alternatively, the primary knowledge source and the secondary knowledge source may reside on different entities.

[0058] The primary knowledge / information sources of a RAG can handle general public information retrieval, which, while effective in a broader context, may not address the security needs of sensitive domains such as telecommunications. Secondary knowledge / information sources of a RAG can act as a dedicated layer, ensuring that only domain-specific / use-case-specific information is processed and retrieved, thus minimizing the risk of opening or mishandling sensitive data. The primary and secondary knowledge / information sources of a RAG can be used in various possible ways (e.g., as a cascaded RAG for multi-level cue enhancement, i.e., firstly through primary knowledge / information sources of the RAG that are openly accessible (such as 3GPP standards or open multimodal information such as open source code) and then through secondary knowledge / information sources of the RAG that are restricted to mobile operators (such as infrastructure, deployment aspects, etc.).

[0059] In a cascaded RAG setup, consider the following: Figure 1The cascaded RAG system in the model, with its multi-layered RAG framework 200, includes a secondary knowledge / information source for RAG 220. This secondary knowledge / information source aims to streamline the retrieval process by narrowing the scope to a smaller space of highly relevant documents or blocks within a large corpus. This reduces computational load and improves the overall efficiency of the system, ensuring faster and more accurate responses, particularly in high-stakes scenarios where time and accuracy are critical. For example, information maintained by the mobile operator regarding the persistent mapping from initial prompts to specific enhanced prompts via the primary knowledge / information source of RAG 210 and then to specific enhanced prompts via the secondary knowledge / information source of RAG 220 can enable bypassing the primary knowledge / information source of RAG 210 and relying solely on the secondary knowledge / information source of RAG 220 to enhance the initial prompts. It also ensures that sensitive data is stored only in the secondary RAGs and controlled by the operator. It should be noted that... Figure 2 The multi-layer RAG architecture 200 addresses two aspects: the concept of cascading multi-layer RAGs or simply enabling multiple RAGs in the system, depending on the type of information stored in the RAGs.

[0060] Furthermore, the multi-tiered knowledge / information sources of a RAG can be viewed from the perspective of the exact location where the information is stored or from the perspective of splitting based on access permissions or information sensitivity. This can even be achieved if a mobile network operator (MNO) wants to have different databases for different types of content or a single database. Access to specific blocks of knowledge from different databases can be granted at the consumer level and at the individual prompt level. In some example embodiments, the splitting of a RAG can be based on information stored by the owner of the RAG (e.g., the MNO or one or more third-party application functions (one or more AFs)). Alternatively or additionally, the splitting of a RAG can be based on information stored by a defined authorization mechanism, either for end users or at the level of the GenAI agent requesting the RAG service.

[0061] In some example implementations, there are different possible deployment implementations for the primary and secondary knowledge / information sources of a RAG. For example, if the guidelines for deploying a multi-tiered RAG are maintaining the confidentiality of sensitive MNO data, one option is that the primary knowledge / information source of the RAG can be found as an entity located outside the MNO's 3GPP management system (e.g., an (E) cloud service provider - (E)CSP), while the secondary knowledge / information source is within the MNO's premises (i.e., part of the MNO's Operations, Administration, and Maintenance (OAM)). For example, the secondary knowledge / information source of the RAG contains deployment and topology-related information private to the MNO. Another option for the secondary knowledge / information source of the RAG is that it may also be outside the MNO's premises, but with a Service Level Agreement (SLA) at an appropriate location between the RAG provider (e.g., (E)CSP) and the MNO, where data confidentiality is part of that SLA. The CSP is then responsible for keeping the MNO's sensitive data confidential and disposing of it securely after use.

[0062] Figure 3 and Figure 4 This illustrates a possible multi-tiered knowledge / information source for RAG deployment options, depending on the existence of an SLA between the MNO and the (E)CSP and the cardinality of the secondary RAGs (i.e., a single secondary RAG or multiple secondary RAGs spanning all network domains managed by the MNO, or one secondary RAG per network domain managed by the MNO). Specifically, based on this deployment, a single secondary RAG can exist hosted on the core network, and an LLM agent can query this single secondary RAG, where data in the RAG can be updated via network elements in different domains. Figure 3 This illustrates a deployment 300 where a single secondary RAG exists on the OAM side. For example, as... Figure 3 As shown, the primary knowledge / information source for RAG 210 can be located at an authorized (E)CSP application server (AS), while the secondary knowledge / information source for RAG 220 can be located at an MNO trusted domain. Alternatively, another deployment can be envisioned where RAGs can be located in each domain, and the corresponding network element in that domain is updating the RAG, and an LLM agent can access the relevant domain-specific RAG based on the required data. Figure 4 This illustrates a deployment 400 where multiple secondary RAGs exist, and these secondary RAGs are network domain specific. For example, as... Figure 3 As shown, the primary knowledge / information source for RAG 210 can be located at the authorized (E)CSP application server (AS), while the secondary knowledge / information source for RAG 220-1 can be located at the OAM domain; the secondary knowledge / information source for RAG 220-2 can be located at the CN domain, while the secondary knowledge / information source for RAG 220-3 can be located at the Radio Access Network (RAN) domain.

[0063] Based on the semantics of the user prompts issued and the availability / capabilities of the platform hosting the primary and secondary RAGs, there is a need for a method and criteria for selecting primary(one) and secondary(one) RAGs to enhance the prompts. Figure 5 The process 500 for selecting a primary RAG and a secondary RAG based on the primary RAG and secondary RAG (latest) profiles is shown. Figure 5 As shown, the operator inputs an initial prompt to the LLM 510, which can then transmit prompt enhancement requests to the NEF 530 and NRF 520. The NEF 530 can perform primary RAG selection, transmitting prompt enhancement requests to selected primary knowledge sources and forwarding external enhancement prompts to the LLM 510. The NRF 520 can perform secondary RAG selection, transmitting prompt enhancement requests to selected secondary RAGs. The LLM 510 can aggregate internal and external enhancement prompts to issue a prompt response to the operator. Figure 5 The approach shown focuses on cue enhancement based on non-cascaded (i.e., parallel) RAGs, which has the benefit of circumventing, for example, RANs (primary or secondary) that are difficult to enhance within the required time budget.

[0064] There may be different use cases involving different entities as RAG consumers. In short, RAG consumers (e.g., third-party LLMs hosted in the cloud, LLMs / Small Language Models (SLMs) at vertical users such as factory machines, SLMs hosted at the UE, GenAI agents on cloud-based (OTT) devices) can be outside the network. Alternatively, RAG consumers can be inside the network; for example, any network function instance can be powered by an LLM / SLM and consume RAGs for cue enhancement, or the LLM / SLM can be hosted in an edge cloud within the mobile network operator's premises. Figures 6A to 6D This illustrates a different scenario. Figure 6A The diagram shows a structure 600 where the RAG consumer is an external RAG consumer to the network. Figure 6B The diagram shows a structure 600' where the RAG consumer is an external RAG consumer and the SLM / LLM is located in a third-party cloud. Figure 6C and Figure 6D The structures 600'' and 600'''', where the RAG consumer is an internal RAG consumer in the network, are shown respectively.

[0065] Figure 7Signaling flow 700 for accessing a knowledge source according to some example embodiments of this disclosure is illustrated. Signaling flow 700 relates to an LLM agent 710, an NRF entity 720, an ADRF entity 730, a NEF entity 740, an OAM entity 750, an AF entity 760, and an LLM prompt indicator 770. For example, the LLM agent 710 may be a first device 110, for example, implemented at a terminal device or network device. The NRF entity 720 may be implemented at a second device 720. The terms “RAG” and “knowledge base” are used interchangeably, and the terms “LLM agent” and “GenAI agent” are used interchangeably. It should be noted that, for illustrative purposes, the knowledge source is merely an example of a knowledge source. The example embodiments of this disclosure can be applied to any suitable kind of knowledge source.

[0066] OAM entity 750 may transmit (7005-1) a profile of a knowledge source, for example, that can be used for RAG, to NRF entity 720. That is, NRF entity 720 may receive (7005-1) a profile of a knowledge source from OAM entity 750. In some example embodiments, the profile of a RAG knowledge source may include an identifier of the knowledge source. Alternatively or additionally, the profile of a RAG knowledge source may include a dataset identifier or a dataset topic description of the knowledge source. In some other example embodiments, the profile of a RAG knowledge source may also include an authorized entity for updating the knowledge source by dataset identifier. In some other example embodiments, the profile of a RAG knowledge source may include a Uniform Resource Identifier (URI) of a third device hosting the knowledge resource. The profile of a RAG knowledge source may also include one or more of the following: a knowledge resource-specific domain, a list of authorized entities for discovering the knowledge source, or dataset identifier-specific information. For example, OAM 750 can register a knowledge source profile (such as Nnrf_NFmanagement_Registration) with NRF entity 720 (e.g., the profile contains sensitive operator data). The profile of a RAG knowledge source can include the RAG identifier, dataset subject / identifier (ID) (such as RAN gNB details, OAM details, CN, subscriber details, etc.), analysis ID, and tracking area identifier (TAI).

[0067] In some example implementations, if a RAG is associated with multiple datasets or data producers, all dataset topics included in the RAG are mentioned. For example, a profile of a knowledge source may include the URI (optional) of the storage entity (e.g., ADRF) in which the RAG is hosted, and the DomainSpecificRAG as one of RAN, CN, OAM, or Unified.

[0068] In some other example embodiments, if RAG is dedicated to a single domain or it contains data from multiple domains (such as...) Figure 3 and Figure 4 (As shown in the deployment), the profile of a knowledge source in a RAG can include a list of authorized consumers (such as including NFs, entities, or collections) to access / discover knowledge sources in the RAG. This can be an internal core NF, a GenAI agent, or one or more external AFs registered in the system. The profile of a knowledge source in a RAG can include a DataSetId-specific parameter. For example, the profile of a knowledge source in a RAG can include information about the extent to which authorized consumers can discover and access data. Such information can indicate that it is only allowed hints or access to the entire knowledge source of the RAG, i.e., the entire knowledge source is allowed. This can be indicated by setting flags to true / false (such as access to the entire RAG is allowed as true / false, hint enrichment is only allowed as true / false, or the knowledge source or only hint enrichment of the knowledge source for RAG access is allowed as a list of authorized consumers (NFs / entities / collections)).

[0069] In some example embodiments, if access to a RAG is only permitted through hint enrichment via a central repository or via an entity hosting the RAG therein, or if the complete RAG itself is accessible, the LLM agent 710 may use it locally for hint enrichment. In the case of full knowledge source access, the LLM agent 710 may download the RAG locally and access it, while in the case of hint enrichment only, the LLM agent 710 may send hints to the entity providing the RAG service and then receive only hints enriched using the RAG data as output. Since there may be some RAGs that, if they consist of extremely sensitive data, will only allow hint enrichment without full knowledge source access.

[0070] In some example embodiments, AF entity 760 may transmit (7005-2) a knowledge source profile to NRF entity 720. That is, NRF entity 720 may receive (7005-2) a knowledge source profile from AF entity 760. For example, AF entity 760 may register a similar knowledge profile with NEF entity 740, and then NEF entity 740 may transmit (7005-3) that profile to NRF entity 720. In some example embodiments, if the knowledge source that can be used for RAG is an additional property / service of the native AI network element (e.g., ADRF entity 730 provides RAG services), then their individual network profiles are enhanced to have additional knowledge source access services and transmitted (7005-4) to NRF entity 720 (or any other registration server).

[0071] In some example embodiments, the LLM agent 710 may select the knowledge sources for the RAG based on a list of knowledge sources or information about available knowledge sources. For example, the discovery of the RAG may also be constrained by a consumer request. As an example, based on a consumer requesting the RAG (e.g., first device 110), it may be provided with full access to the RAG, such as all datasets contained in the RAG and their attributes, and the consumer can then precisely select what to access. Alternatively, it can be mediated by prompting to a RAG mapping service (which may be hosted in one or more external AF NEF entities 740). In some example embodiments, the LLM agent 710 may determine whether it is authorized to access the one or more available knowledge sources based on the profiles of the one or more available knowledge sources and information about trusted consumers of the one or more available knowledge sources. If the LLM agent 710 is a trusted consumer among trusted consumers, it can determine that the LLM agent 710 is authorized to access the one or more available knowledge sources.

[0072] Figure 8 The signaling flow for knowledge sources for RAG selection according to some example embodiments of this disclosure is illustrated. For example, such as... Figure 8 As shown, in the case of a trusted RAG consumer, detailed information about available knowledge sources for RAG can be made available, from which the RAG consumer can directly select one or more knowledge sources for RAG. For example, network RAG source 830 can make RAG available to RAG consumer 810 (e.g., LLM agent 710). By way of example, the information of network RAG source 830 may include one or more of the following: knowledge-based characteristics (e.g., the type of knowledge including network knowledge: static, semi-dynamic, or dynamic), and what content is included (such as which analytics IDs, which geographic regions, UE groups, etc.). RAG consumer 810 can send a request / subscription (8020) to network RAG source 830, which may include knowledge-based characteristics (such as knowledge base characteristics).

[0073] In some example implementations, in the case of an untrusted RAG consumer, it may be necessary to enable an intermediate mapping service in the mobile network. This intermediate service maps RAG consumer cue fragments to available knowledge sources of RAG within the network (since the RAG consumer is unaware of or unable to access these knowledge sources), and only provides the RAG consumer with information about matching knowledge sources. Based on this information, the RAG consumer can request cue enhancement from specific knowledge sources. For example, such as... Figure 8As shown, RAG consumer 810 can send (8010') a request to the prompt-to-RAG mapping service 820 to discover available knowledge sources for RAGs. This request may include requests based on RAG knowledge (such as prompt fragments). The prompt-to-RAG mapping service 820 can map (8020') the prompt fragments to available knowledge sources for RAGs in the network. The prompt-to-RAG mapping service 820 can provide (8030') the RAG consumer 810 with a list of matching knowledge sources for RAGs along with their effects (such as costs). RAG consumer 810 can send (8040') a request / subscription to a network RAG source 830. This request / subscription may include a knowledge base, such as knowledge-based features.

[0074] Return to reference Figure 7 The LLM prompt indicator 770 can transmit a (7010) prompt as a service request. For example, a UE or human operator or LLM agent (or SLM) running on the UE can send an (7010) automation prompt to the LLM agent 710 running in the native AI entity.

[0075] LLM agent 710 transmits (7015) a discovery request for access to a knowledge source to NRF entity 720. That is, NRF entity 720 receives (7015) a discovery request for access to a knowledge source from LLM agent 710. For example, in cases where rich hints are required, LLM agent 710 may send (7015) a discovery request to the NRF to attach access to the knowledge source. The discovery request may include an identifier of the knowledge source. In some example embodiments, the discovery request may include a dataset identifier or dataset topic description associated with the knowledge source. Alternatively or additionally, the discovery request may include a domain associated with the knowledge source. The discovery request may also include data attributes for discovering the knowledge source. For example, parameters in the discovery request may include filter=RAG ID, dataset topic / ID, domain (discovery by specific data attributes to find the correct RAG). Where known, the discovery request may also indicate a RAG identifier.

[0076] NRF entity 720 identifies zero or more network function instances. Upon receiving a discovery request, each network function instance is associated with a knowledge source. NRF entity 720 also determines zero or more network function instance identifiers corresponding to the zero or more network function instances, and then determines zero or more knowledge source profiles. Each knowledge source profile corresponds to a knowledge source associated with one of the zero or more network function instances.

[0077] In some example embodiments, NRF entity 720 stores a list of empty network function instance identifiers for zero or more network function instances. NRF entity 720 can identify at least one network function instance associated with a knowledge source. For each of the at least one network function instance, NRF entity 720 can determine whether the knowledge source is allowed to be discovered. For example, NRF entity 720 can verify (7020) a discovery request and determine whether the knowledge source is allowed to be discovered based on the verification of the discovery request. In this case, if the knowledge source is allowed to be discovered, NRF entity 720 can provide a profile of the knowledge source including the dataset included in the knowledge source. For example, if a requesting user is allowed to discover the corresponding knowledge source, NRF entity 720 can verify the discovery request and provide details of the corresponding knowledge source including details of the dataset included in the knowledge source. For example, if the knowledge source is allowed to be discovered, for each of the at least one network function instance, NRF entity 720 can add the instance identifier of the network function instance to the list of network function instance identifiers.

[0078] NRF entity 720 transmits a discovery response (7025) to LLM agent 710. That is, LLM agent 710 receives a discovery response (7025) from NRF entity 720. The discovery response includes a list of network function instances, which includes knowledge sources and profiles of the network function instances. For example, NRF entity 720 could send a discovery response with a list of Nfprofiles (i.e., RAG profiles).

[0079] LLM agent 710 transmits (7030) an access token request for the knowledge source to NRF entity 720. That is, NRF entity 720 receives the access token request from LLM agent 710. For example, LLM agent 710 may send an enhanced access token request for a RAG service provided by a producer. The access token request includes at least the identifier of the RAG knowledge source. Additionally, the access token request may include consumer information of the knowledge source or information about the target producer providing the RAG service.

[0080] In some example embodiments, the access token request may include consumer information or target producer information for the knowledge source. For example, the access token request may include: (where multiple RAG authorizations are required in a single access token) RAGID=xyz,ijk or (where multiple access tokens are expected for each knowledge source access) a single RAGID, a dataset Id=eNB data (or dataset identifier) ​​precisely indicating which data must be accessed, (where RAG is domain-specific) domain Id:RAN, information requesting the data (e.g., indicating whether rich access is true or false, full access to the RAG for download, and local access is true or false), and consumer information (such as NF ID). Target producer information may provide the RAG service (NF type, instance, region information (if present)) along with the specified RAG service. For example, the consumer information for the knowledge source includes at least one of the following: an authorized entity for updating the knowledge source by dataset identifier, a Uniform Resource Identifier for a third device hosting the knowledge source, a list of authorized entities that discover the knowledge source, an authorized entity for the dataset identifier, or the manner in which the knowledge source is allowed access.

[0081] In some example embodiments, NRF entity 720 may verify (7035) an access token request based on the received knowledge source profile and information indicating which consumer is allowed to access the knowledge source. NRF entity 720 may determine whether the first device is allowed to access the knowledge source based on the verification of the access token request. In this case, if LLM agent 710 is allowed to access the knowledge source, NRF entity 720 may generate (7040) an access token response. For example, NRF entity 720 may generate an access token response based on external information in the knowledge source and the knowledge source profile. For example, NRF entity 720 may verify whether a consumer (such as LLM agent 710) is authorized to access the knowledge source based on the received (7005-1) information. In this case, if LLM agent 710 is authorized to access the knowledge source, NRF entity 720 may generate an access token accordingly and send (7045) an access token response to LLM agent 710. For example, if the requested RAG is a primary knowledge source of a RAG that includes external information, and if NRF entity 720 has RAG details updated by NEF entity 740, then NRF entity 720 can generate an access token accordingly. If the consumer is authorized for both the primary and secondary knowledge sources of the RAG, the generated access token can include information corresponding to both the primary and secondary knowledge sources of the RAG.

[0082] In some other example embodiments, NRF entity 720 may transmit the profile information of LLM agent 710 to NEF entity 740. In this case, NRF entity 720 may receive an authorization response from NEF entity 740 for the first device. If the authorization response indicates that LLM agent 710 is permitted to access the knowledge source, NRF entity 720 may generate an access token response. For example, if NRF entity 720 does not have a corresponding external profile for the AF registered by NEF entity 740, NRF entity 720 may send consumer details to NEF entity 740 for further authorization. NEF entity 740 may then provide NRF entity 720 with an acknowledgment of the authorization response, and NRF entity 720 may directly generate an access token. Alternatively, NEF 740 may verify whether the consumer is authorized to access the knowledge source based on the AF profile. If successful, NEF 740 may generate an access token that includes relevant information about the primary knowledge source. If the RAG is hosted in the AF, the access token may include the AF details in the target NF. Alternatively, if the RAG is hosted within NEF entity 740, the access token may include its own URI. In this case, the generated access token may be sent to NRF 720 for forwarding (7045) to LLM agent 710, or it may be sent directly to LLM agent 710 via NEF entity 740.

[0083] Access token responses may include one or more of the following: a RAG token, RAG consumer information, or target producer information providing the RAG service. For example, an access token response may include one or more of the following parameters: RAG Id=xyz,ijk (which can be either a single RAG ID or multiple RAG IDs), a dataset Id precisely indicating which data must be accessed (eNB data or dataset identifier), a domain Id of RAN (which may also include multiple domains, in the case of generating access tokens for multiple RAGs belonging to different domains), information requesting the data (e.g., whether the prompted access is true or false, or whether knowledge source access is allowed is true or false), and consumer information (such as an NF ID). Target producer information may provide the RAG service (NF type, instance, region information (if present)) along with the specified RAG service. For example, knowledge source consumer information may include at least one of the following: an authorized entity for updating the knowledge source by dataset identifier, a Uniform Resource Identifier of the third device hosting the knowledge source, a list of authorized entities discovering the knowledge source, an authorized entity for the dataset identifier, or the manner in which the knowledge source is allowed access.

[0084] LLM agent 710 may transmit (7050) a knowledge service request (e.g., a RAG service request) including a token to ADRF entity 730. For example, LLM agent 710 sends a new service request for access to a knowledge source to its corresponding entity (such as ADRF entity 730). The service request for RAG may include one or more of the following: RAG ID, dataset ID, knowledge source access, generated hints, and the received (7045) enhanced token. In some example embodiments, if the service is used for hint enrichment, the corresponding hints may also be sent to the RAG service.

[0085] ADRF entity 730 can determine (7055) whether LLM agent 710 is authorized to access a knowledge source based on a token. For example, a RAG service request can be authorized based on a token. After authorizing LLM agent 710, ADRF entity 730 can send (7060) a RAG service response to LLM agent. For example, the RAG service response includes at least one of a rich hint or a Uniform Resource Identifier for accessing the knowledge source. In some example embodiments, ADRF entity 730 can determine whether LLM agent 710 is authorized to access the one or more available knowledge sources based on the profile of the one or more available knowledge sources and information about trusted consumers of the one or more available knowledge sources. If LLM agent 710 is a trusted consumer among trusted consumers, ADRF entity 730 can determine that LLM agent 710 is authorized to access one or more available knowledge sources.

[0086] According to the reference Figures 2 to 8 The described example embodiments allow for more efficient access to knowledge sources. For example, the security of the access to knowledge sources can be ensured.

[0087] Figure 9 A flowchart of an example method 900 implemented at a first device according to some example embodiments of the present disclosure is shown. For example, the method 900 can be implemented in... Figure 1 The first device 110 in the middle is implemented.

[0088] At frame 910, the first device sends a discovery request for access to the knowledge source to the second device.

[0089] At box 920, the first device receives a discovery response from the second device, the discovery response including a list of network function instances, the network function instance list including knowledge sources and profiles of the network function instance list.

[0090] At frame 930, the first device sends an access token request to the second device for the knowledge source.

[0091] At box 940, the first device receives an access token response from the second device, the access token response including at least the identifier of the knowledge source.

[0092] In some example embodiments, the discovery request includes filtering information, which includes at least one of the following: an identifier of the knowledge source, an identifier of the dataset associated with the knowledge source, a topic description of the dataset associated with the knowledge source, a domain associated with the knowledge source, or a data attribute of the knowledge source being discovered.

[0093] In some example embodiments, the access token request includes at least one of the following: an identifier of the knowledge source, consumer information of the knowledge source, or information of the target producer providing the knowledge service.

[0094] In some example embodiments, the access token response includes at least one of the following: a token of the knowledge source, consumer information of the knowledge source, or target producer information providing the knowledge service.

[0095] In some example embodiments, the target producer information includes at least one of the following: at least one identifier of the knowledge source, a dataset identifier or a description of a dataset from the knowledge source (indicating that the data must be accessed), a domain identifier, or the type and quantity of data from the knowledge source that is allowed to be discovered and accessed; and the consumer information of the knowledge source includes at least one of the following: an authorized entity for updating the knowledge source by dataset identifier, a Uniform Resource Identifier of a third device hosting the knowledge source, a list of authorized entities for discovering the knowledge source, an authorized entity for the dataset identifier, or the manner in which the knowledge source is allowed to be accessed.

[0096] In some example embodiments, the method 900 further includes: transmitting a knowledge service request to a third device, the knowledge service request including a token of a knowledge source; and receiving a knowledge service response from the third device, the knowledge service response including at least one of a richly enriched hint or a Uniform Resource Identifier for accessing the knowledge source.

[0097] In some example embodiments, knowledge sources include primary knowledge sources and secondary knowledge sources.

[0098] In some example embodiments, the primary knowledge source and the secondary knowledge source are on the same entity, or the primary knowledge source and the secondary knowledge source are on different entities.

[0099] In some example embodiments, the method 900 further includes: selecting a knowledge source based on a list of knowledge sources or based on information from one or more available knowledge sources, based on determining that the first device is authorized to access one or more available knowledge sources.

[0100] In some example embodiments, the method 900 further includes: determining whether a first device is authorized to access one or more available knowledge sources based on profiles of one or more available knowledge sources and information of trusted consumers of one or more available knowledge sources; and determining that the first device is authorized to access one or more available knowledge sources based on determining that the first device is a trusted consumer among trusted consumers.

[0101] In some example implementations, the profile of the list of network function instances includes detailed information about the knowledge sources.

[0102] In some example embodiments, the first device is implemented at a terminal device or network device, and the second device is implemented on another network device.

[0103] Figure 10 A flowchart is shown of an example method 1000 implemented at a second device according to some example embodiments of the present disclosure. For example, the method 1000 can be implemented in... Figure 1 The second device 120 is implemented.

[0104] At box 1010, the second device receives a discovery request for access to a knowledge source from the first device.

[0105] At box 1020, the second device identifies zero or more network function instances, each of which is associated with a knowledge source.

[0106] At box 1030, the second device determines zero or more network function instance identifiers corresponding to zero or more network function instances.

[0107] At box 1040, the second device determines zero or more knowledge source profiles, wherein each knowledge source profile corresponds to a knowledge source associated with one of the zero or more network function instances.

[0108] At box 1050, the second device transmits a discovery response to the first device, the discovery response including zero or more network function instance identifiers and zero or more knowledge source profiles.

[0109] At box 1060, the second device receives an access token request for the knowledge source from the first device.

[0110] At box 1070, the second device sends an access token response to the first device, the access token response including at least the identifier of the knowledge source.

[0111] In some example embodiments, determining zero or more network function instances further includes: storing an empty network function instance identifier list of zero or more network function instances; determining at least one network function instance associated with a knowledge source; determining, for each of the at least one network function instances, whether the knowledge source is allowed to be discovered based on the verification of the discovery request; and adding the instance identifier of the network function instance to the network function instance identifier list for each of the at least one network function instances based on the determination that the knowledge source is allowed to be discovered.

[0112] In some example embodiments, the method 1000 further includes receiving a profile of a knowledge source from at least one of the following: an Operation, Management and Maintenance (OAM) entity, an Application Function entity, a Third Device or Network Open Function entity.

[0113] In some example embodiments, the profile of the knowledge source includes at least one of the following: an identifier of the knowledge source, a dataset identifier of the knowledge source, a subject description of the dataset associated with the knowledge source, an authorized entity for updating the knowledge source by dataset identifier, a Uniform Resource Identifier of a third device hosting the knowledge source, a domain specific to the knowledge source, a list of authorized entities for discovering the knowledge source, or dataset identifier-specific information.

[0114] In some example embodiments, the method 1000 further includes: verifying an access token request based on a received profile of the knowledge source and information indicating which consumer is allowed to access the knowledge source; determining whether a first device is allowed to access the knowledge source based on the verification of the access token request; and generating an access token response based on determining that the first device is allowed to access the knowledge source.

[0115] In some example embodiments, the method 1000 further includes generating an access token response based on external information in the knowledge source and a profile of the knowledge source.

[0116] In some example embodiments, the method 1000 further includes: transmitting profile information of the first device to the fourth device; receiving an authorization response from the first device from the fourth device; and generating an access token response based on determining that the authorization response indicates that the first device is allowed to access the knowledge source.

[0117] In some example embodiments, the method 1000 further includes: transmitting profile information of the first device to the fourth device; and receiving an access token response from the fourth device, the access token response including at least one of the following: a token of the knowledge source, information relating to an AF entity, or information relating to a NEF entity.

[0118] In some example embodiments, the access token response includes information corresponding to the primary knowledge source and the secondary knowledge source of the knowledge source.

[0119] In some example embodiments, the discovery request includes filtering information, which includes at least one of the following: an identifier of the knowledge source, a dataset identifier of the knowledge source, a topic description of the dataset associated with the knowledge source, a domain associated with the knowledge source, or a data attribute of the discovered knowledge source.

[0120] In some example embodiments, the access token request includes at least one of the following: an identifier of the knowledge source, consumer information of the knowledge source, or information of the target producer providing the knowledge service.

[0121] In some example embodiments, the access token response includes at least one of the following: a token of the knowledge source, consumer information of the knowledge source, or target producer information providing the knowledge service.

[0122] In some example embodiments, the target producer information includes at least one of the following: at least one identifier of the knowledge source, a dataset identifier or a description of the dataset from the knowledge source indicating which data must be accessed, a domain identifier, or the type or quantity of data from the requested knowledge source.

[0123] In some example embodiments, knowledge sources include primary knowledge sources and secondary knowledge sources.

[0124] In some example embodiments, the primary knowledge source and the secondary knowledge source are on the same entity, or the primary knowledge source and the secondary knowledge source are on different entities.

[0125] In some example implementations, the profile of the list of network function instances includes detailed information about the knowledge sources.

[0126] In some example embodiments, the first device is implemented at a terminal device or a network device, and the second device is implemented at another network device.

[0127] Figure 11 A flowchart of an example method 1100 implemented at a third device according to some example embodiments of the present disclosure is shown. For example, the method 1100 can be implemented at a third device. Figure 1 The third device, at point 130, was implemented.

[0128] At box 1110, the third device receives a knowledge service request from the first device, the knowledge service request including a token of the knowledge source.

[0129] At box 1120, the third device uses a token to determine whether the first device is authorized to access the knowledge source.

[0130] At box 1130, based on the determination that the first device is authorized to access the knowledge source, the third device transmits a knowledge service response to the first device.

[0131] In some example embodiments, the method 1100 further includes: determining whether a first device is authorized to access one or more available knowledge sources based on profiles of one or more available knowledge sources and information of trusted consumers of one or more available knowledge sources; and determining that the first device is authorized to access one or more available knowledge sources based on determining that the first device is a trusted consumer among trusted consumers.

[0132] In some example embodiments, the knowledge service response includes at least one of a rich hint or a Uniform Resource Identifier for accessing the knowledge source.

[0133] In some example embodiments, the first device is implemented at a terminal device or a network device, and the third device is implemented at another network device.

[0134] In some example embodiments, a first means capable of performing any of the methods in method 900 (e.g., Figure 1 The first device 110 may include a component for performing the corresponding operation of method 900. This component may be implemented in any suitable form. For example, the component may be implemented in a circuit system or a software module. The first device may be implemented as a terminal device or included in a terminal device.

[0135] In some example embodiments, the first device includes components for transmitting a discovery request for access to a knowledge source to a second device; components for receiving a discovery response from the second device, the discovery response including a list of network function instances, the list of network function instances including knowledge sources and profiles of the network function instances; components for transmitting an access token request for the knowledge source to the second device; and components for receiving an access token response from the second device, the access token response including at least an identifier of the knowledge source.

[0136] In some example embodiments, the discovery request includes filtering information, which includes at least one of the following: an identifier of the knowledge source, an identifier of the dataset associated with the knowledge source, a topic description of the dataset associated with the knowledge source, a domain associated with the knowledge source, or a data attribute of the knowledge source being discovered.

[0137] In some example embodiments, the access token request includes at least one of the following: an identifier of the knowledge source, consumer information of the knowledge source, or information of the target producer providing the knowledge service.

[0138] In some example embodiments, the access token response includes at least one of the following: a token of the knowledge source, consumer information of the knowledge source, or target producer information providing the knowledge service.

[0139] In some example embodiments, the target producer information includes at least one of the following: at least one identifier of the knowledge source, a dataset identifier or a description of a dataset from the knowledge source indicating which data must be accessed, a domain identifier, or the type and quantity of data from the knowledge source that is allowed to be discovered and accessed; and the knowledge source consumer information includes at least one of the following: an authorized entity for updating the knowledge source by dataset identifier, a Uniform Resource Identifier of a third device hosting the knowledge source, a list of authorized entities for discovering the knowledge source, an authorized entity for the dataset identifier, or the manner in which the knowledge source is allowed to be accessed.

[0140] In some example embodiments, the first device further includes: a component for transmitting a knowledge service request including a token of a knowledge source to a third device; and a component for receiving a knowledge service response from the third device, the knowledge service response including at least one of a riched prompt or a Uniform Resource Identifier for accessing the knowledge source.

[0141] In some example embodiments, knowledge sources include primary knowledge sources and secondary knowledge sources.

[0142] In some example embodiments, the primary knowledge source and the secondary knowledge source are on the same entity, or the primary knowledge source and the secondary knowledge source are on different entities.

[0143] In some example embodiments, the first device further includes a component for selecting a knowledge source based on a list of knowledge sources or information about one or more available knowledge sources, based on determining that the first device is authorized to access one or more available knowledge sources.

[0144] In some example embodiments, the first device further includes: a component for determining whether the first device is authorized to access one or more available knowledge sources based on profiles of one or more available knowledge sources and information of trusted consumers of one or more available knowledge sources; and a component for determining that the first device is authorized to access one or more available knowledge sources based on determining that the first device is a trusted consumer among trusted consumers.

[0145] In some example implementations, the profile of the list of network function instances includes detailed information about the knowledge sources.

[0146] In some example embodiments, the first device is implemented at a terminal device or a network device, and the second device is implemented at another network device.

[0147] In some example embodiments, a second means capable of performing any of the methods in method 1000 (e.g., Figure 1The second device (120) may include a component for performing the corresponding operation of method 1000. This component may be implemented in any suitable form. For example, the component may be implemented in a circuit system or a software module. The second device may be implemented as a core network device or included within a core network device.

[0148] In some example embodiments, the second apparatus includes components for receiving a discovery request for access to a knowledge source from the first apparatus; components for determining zero or more network function instances, wherein each network function instance is associated with a knowledge source; components for determining zero or more network function instance identifiers corresponding to the zero or more network function instances; components for determining zero or more knowledge source profiles, wherein each knowledge source profile corresponds to a knowledge source associated with one of the zero or more network function instances; components for transmitting a discovery response to the first apparatus, the discovery response including zero or more network function instance identifiers and zero or more knowledge source profiles; components for receiving an access token request for the knowledge source from the first apparatus; and components for transmitting an access token response to the first apparatus, the access token response including at least an identifier of the knowledge source.

[0149] In some example embodiments, the components for determining zero or more network function instances further include: components for storing an empty list of network function instance identifiers for zero or more network function instances; components for determining at least one network function instance associated with a knowledge source; components for determining, based on verification of a discovery request, whether the knowledge source is allowed to be discovered for each of the at least one network function instances; and components for adding an instance identifier of the network function instance to the list of network function instance identifiers for each of the at least one network function instances based on the determination that the knowledge source is allowed to be discovered.

[0150] In some example embodiments, the second device further includes a component for receiving a profile of a knowledge source from at least one of the following: an Operation, Management and Maintenance (OAM) entity, an application function entity, a third device, or a network open function entity.

[0151] In some example embodiments, the profile of the knowledge source includes at least one of the following: an identifier of the knowledge source, a dataset identifier of the knowledge source, a subject description of the dataset associated with the knowledge source, an authorized entity for updating the knowledge source by dataset identifier, a Uniform Resource Identifier of a third device hosting the knowledge source, a domain specific to the knowledge source, a list of authorized entities for discovering the knowledge source, or dataset identifier-specific information.

[0152] In some example embodiments, the second device further includes: a component for verifying an access token request based on a received profile of the knowledge source and information indicating which consumer is allowed to access the knowledge source; a component for determining whether the first device is allowed to access the knowledge source based on the verification of the access token request; and a component for generating an access token response based on determining that the first device is allowed to access the knowledge source.

[0153] In some example embodiments, the second device further includes a component for generating an access token response based on external information in the knowledge source and a profile of the knowledge source.

[0154] In some example embodiments, the second device further includes: a component for transmitting profile information of the first device to the fourth device; a component for receiving an authorization response from the first device from the fourth device; and a component for generating an access token response based on determining that the authorization response indicates that the first device is allowed to access a knowledge source.

[0155] In some example embodiments, the second device further includes: a component for transmitting profile information of the first device to the fourth device; and a component for receiving an access token response from the fourth device, the access token response including at least one of the following: a token of a knowledge source, information relating to an AF entity, or information relating to a NEF entity.

[0156] In some example embodiments, the access token response includes information corresponding to the primary knowledge source and the secondary knowledge source of the knowledge source.

[0157] In some example embodiments, the discovery request includes filtering information, which includes at least one of the following: an identifier of the knowledge source, a dataset identifier of the knowledge source, a topic description of the dataset associated with the knowledge source, a domain associated with the knowledge source, or a data attribute of the discovered knowledge source.

[0158] In some example embodiments, the access token request includes at least one of the following: an identifier of the knowledge source, consumer information of the knowledge source, or information of the target producer providing the knowledge service.

[0159] In some example embodiments, the access token response includes at least one of the following: a token of the knowledge source, consumer information of the knowledge source, or target producer information providing the knowledge service.

[0160] In some example embodiments, the target producer information includes at least one of the following: at least one identifier of the knowledge source, a dataset identifier or a description of the dataset from the knowledge source indicating which data must be accessed, a domain identifier, or the type or quantity of data from the requested knowledge source.

[0161] In some example embodiments, knowledge sources include primary knowledge sources and secondary knowledge sources.

[0162] In some example embodiments, the primary knowledge source and the secondary knowledge source are on the same entity, or the primary knowledge source and the secondary knowledge source are on different entities.

[0163] In some example implementations, the profile of the list of network function instances includes detailed information about the knowledge sources.

[0164] In some example embodiments, the first device is implemented at a terminal device or a network device, and the second device is implemented at another network device.

[0165] In some example embodiments, a third means capable of performing any of the methods in method 1100 (e.g., Figure 1 The third device (130) may include a component for performing the corresponding operation of method 1100. This component may be implemented in any suitable form. For example, the component may be implemented in a circuit system or a software module. The third device may be implemented as a core network device or included within a core network device.

[0166] In some example embodiments, the third device includes: a component for receiving a knowledge service request from the first device, the knowledge service request including a token of a knowledge source; a component for determining, based on the token, whether the first device is authorized to access the knowledge source; and a component for transmitting a knowledge service response to the first device based on the determination that the first device is authorized to access the knowledge source.

[0167] In some example embodiments, the third device further includes: a component for determining whether the first device is authorized to access one or more available knowledge sources based on profiles of one or more available knowledge sources and information of trusted consumers of one or more available knowledge sources; and a component for determining that the first device is authorized to access one or more available knowledge sources based on determining that the first device is a trusted consumer among trusted consumers.

[0168] In some example embodiments, the knowledge service response includes at least one of a rich hint or a Uniform Resource Identifier for accessing the knowledge source.

[0169] In some example embodiments, the first device is implemented at a terminal device or a network device, and the third device is implemented at another network device.

[0170] Figure 12 This is a simplified block diagram of a device 1200 suitable for implementing an example embodiment of the present disclosure. The device 1200 can be provided to implement a communication device, for example, such as... Figure 1The first device 110 or the second device 120 shown. As shown, the device 1200 includes one or more processors 1210, one or more memories 1220 coupled to the processors 1210, and one or more communication modules 1240 coupled to the processors 1210.

[0171] Communication module 1240 is used for bidirectional communication. Communication module 1240 has one or more communication interfaces to facilitate communication with one or more other modules or devices. The communication interface can represent any interface required for communication with other network elements. In some example embodiments, communication module 1240 may include at least one antenna.

[0172] Processor 1210 can be any type suitable for a local technology network and may include one or more of the following as non-limiting examples: general-purpose computer, special-purpose computer, microprocessor, digital signal processor (DSP), and processor based on a multi-core processor architecture. Device 1200 may have multiple processors, such as application-specific integrated circuit chips that are time-dependent on a clock that synchronizes with the main processor.

[0173] Memory 1220 may include one or more non-volatile memories and one or more volatile memories. Examples of non-volatile memories include, but are not limited to, read-only memory (ROM) 1224, electrically programmable read-only memory (EPROM), flash memory, hard disk, optical disc (CD), digital video disc (DVD), optical disc, laser disc, and other magnetic and / or optical storage devices. Examples of volatile memories include, but are not limited to, random access memory (RAM) 1222 and other volatile memories that do not persist during power-off periods.

[0174] Computer program 1230 includes computer-executable instructions that are executed by a associated processor 1210. The instructions of program 1230 may include instructions for performing operations / actions of some example embodiments of this disclosure. Program 1230 may be stored in memory (e.g., ROM 1224). Processor 1210 can perform any suitable actions and processes by loading program 1230 into RAM 1222.

[0175] The exemplary embodiments of this disclosure can be implemented by means of program 1230, so that device 1200 can perform as described in the reference. Figures 2 to 11 Any process discussed in this disclosure. Exemplary embodiments of this disclosure may also be implemented by hardware or a combination of software and hardware.

[0176] In some example embodiments, program 1230 may be tangibly contained in a computer-readable medium, which may be included in device 1200 (such as in memory 1220) or in other storage devices accessible by device 1200. Device 1200 may load program 1230 from the computer-readable medium into RAM 1222 for execution. In some example embodiments, the computer-readable medium may include any type of non-transitory storage medium, such as ROM, EPROM, flash memory, hard disk, CD, DVD, etc. The term "non-transitory" as used herein is a limitation on the medium itself (i.e., tangible, not tactile), rather than a limitation on the persistence of data storage (e.g., RAM and ROM).

[0177] Figure 13 An example of a computer-readable medium 1300, which may be in the form of a CD, DVD, or other optical storage disc, is shown. The computer-readable medium 1300 has a program 1230 stored thereon.

[0178] Generally, the various embodiments of this disclosure can be implemented in hardware or dedicated circuitry, software, logic, or any combination thereof. Some aspects can be implemented in hardware, while others can be implemented in firmware or software, which can be executed by a controller, microprocessor, or other computing device. Although aspects of the embodiments of this disclosure are illustrated and described using block diagrams, flowcharts, or some other illustrated representation, it should be understood that the blocks, apparatuses, systems, techniques, or methods described herein can be implemented in hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware, or controllers or other computing devices, or some combination thereof, as non-limiting examples.

[0179] Some exemplary embodiments of this disclosure also provide at least one computer program product tangibly stored on a computer-readable medium, such as a non-transitory computer-readable medium. The computer program product includes computer-executable instructions (such as computer-executable instructions included in a program module) that are executed in a device on a target physical or virtual processor to perform any of the methods described above. Generally, a program module includes routines, programs, libraries, objects, classes, components, data structures, etc., that perform specific tasks or implement specific abstract data types. The functionality of a program module can be combined or split among program modules as needed in various embodiments. The machine-executable instructions of a program module can execute within a local device or a distributed device. In a distributed device, a program module can reside on both local storage media and remote storage media.

[0180] The program code used to perform the methods of this disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on a machine, partially on a machine, as a standalone software package, partially on a machine, partially on a remote machine, or entirely on a remote machine or server.

[0181] In the context of this disclosure, computer program code or related data may be carried by any suitable carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, etc.

[0182] Computer-readable media can be computer-readable signal media or computer-readable storage media. Computer-readable media can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any suitable combination thereof. More specific examples of computer-readable storage media may include electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable optical disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0183] Furthermore, although the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in sequential order, or to perform all illustrated operations to achieve the desired result. In some cases, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the foregoing discussion, these details should not be construed as limiting the scope of this disclosure, but rather as a description of features that may be specific to an embodiment. Unless explicitly stated otherwise, certain features described in the context of separate embodiments may also be implemented in combination in a single embodiment. Conversely, unless explicitly stated otherwise, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0184] Although this disclosure has been described in language specific to structural features and / or methodological actions, it should be understood that the disclosure as defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are disclosed as exemplary forms for implementing the claims. Example 1. A first apparatus, comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the first apparatus to: transmit a discovery request for access to a knowledge source to a second apparatus; receive a discovery response from the second apparatus indicating a list of network function instances, the list of network function instances including knowledge sources and profiles of the network function instances; send an access token request to the second apparatus for the knowledge source; and receive an access token response from the second apparatus, the access token response including at least an identifier of the knowledge source. Example 2. According to the first apparatus of Example 1, the discovery request includes filtering information, which includes at least one of the following: an identifier of a knowledge source; an identifier of a dataset associated with the knowledge source; a topic description of a dataset associated with the knowledge source; a domain associated with the knowledge source; or a data attribute for discovering the knowledge source. Example 3. The first apparatus according to Example 1 or 2, wherein the access token request includes at least one of the following: an identifier of the knowledge source, consumer information of the knowledge source, or target producer information providing the knowledge service. Example 4. The first apparatus according to any one of Examples 1 to 3, wherein the access token response includes at least one of the following: a token of the knowledge source, consumer information of the knowledge source, or target producer information providing the knowledge service. Example 5. The first apparatus according to Example 3 or 4, wherein the target producer information includes at least one of the following: at least one identifier of the knowledge source, a dataset identifier or description of a dataset from the knowledge source, the dataset identifier or description indicating which data must be accessed, a domain identifier, or the type and quantity of data of the knowledge source that is allowed to be discovered and accessed, and wherein the consumer information of the knowledge source includes at least one of the following: an authorized entity for updating the knowledge source by dataset identifier, wherein the Uniform Resource Identifier of the third apparatus hosting the knowledge source; a list of authorized entities for discovering the knowledge source, an authorized entity for the dataset identifier, or the manner in which the knowledge source is allowed to be accessed. Example 6. A first apparatus according to any one of Examples 1 to 5, wherein the first apparatus is configured to: transmit a knowledge service request to a third apparatus, the knowledge service request including a token of a knowledge source; and receive a knowledge service response from the third apparatus, the knowledge service response including at least one of an enriched prompt or a Uniform Resource Identifier for accessing the knowledge source. Example 7. The first apparatus according to any one of Examples 1 to 6, wherein the knowledge source includes a primary knowledge source and a secondary knowledge source. Example 8. The first apparatus according to Example 7, wherein the primary knowledge source and the secondary knowledge source are on the same entity, or wherein the primary knowledge source and the secondary knowledge source are on different entities. Example 9. The first apparatus according to Example 1, wherein the first apparatus is configured to: select a knowledge source based on a list of knowledge sources or based on information from one or more available knowledge sources, based on determining that the first apparatus is authorized to access one or more available knowledge sources. Example 10. A first apparatus according to any one of Examples 1 to 9, wherein the first apparatus is configured to: determine whether the first apparatus is authorized to access one or more available knowledge sources based on profiles of one or more available knowledge sources and information of trusted consumers of one or more available knowledge sources; and determine that the first apparatus is authorized to access one or more available knowledge sources based on determining that the first apparatus is a trusted consumer among trusted consumers. Example 11. A first apparatus according to any one of Examples 1 to 10, wherein the profile of the list of network function examples includes detailed information on knowledge sources. Example 12. A first device according to any one of Examples 1 to 11, wherein the first device is implemented at a terminal device or a network device, and the second device is implemented at another network device. Example 13. A second apparatus, comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the second apparatus to: receive from a first apparatus a discovery request for access to a knowledge source; determine zero or more network function instances, wherein each network function instance is associated with a knowledge source; determine zero or more network function instance identifiers corresponding to the zero or more network function instances; determine zero or more knowledge source profiles, wherein each knowledge source profile corresponds to a knowledge source associated with one of the zero or more network function instances; transmit a discovery response to the first apparatus, the discovery response including zero or more network function instance identifiers and zero or more knowledge source profiles; receive from the first apparatus an access token request for the knowledge source; and transmit an access token response to the first apparatus, the access token response including at least an identifier of the knowledge source. Example 14. The second apparatus according to Example 13, wherein the second apparatus is further configured to determine zero or more network function instances by causing the second apparatus to perform the following: storing an empty network function instance identifier list of zero or more network function instances; determining at least one network function instance associated with a knowledge source; determining, for each of the at least one network function instances, whether the knowledge source is allowed to be discovered based on verification of a discovery request; and adding an instance identifier of the network function instance to the network function instance identifier list for each of the at least one network function instances based on the determination that the knowledge source is allowed to be discovered. Example 15. The second apparatus according to Example 13 or 14, wherein the second apparatus is configured to: receive a profile of a knowledge source from at least one of the following: an Operation, Management and Maintenance (OAM) entity, an Application Function entity, a third apparatus or a Network Open Function entity. Example 16. The second apparatus according to Example 15, wherein the profile of the knowledge source includes at least one of the following: an identifier of the knowledge source; a dataset identifier of the knowledge source; a dataset topic description related to the knowledge source for updating the authorized entities of the knowledge source by the dataset identifier, wherein a Uniform Resource Identifier of the third apparatus hosting the knowledge source; a knowledge source-specific domain; a list of authorized entities for discovering the knowledge source; or dataset identifier-specific information. Example 17. A second apparatus according to any one of Examples 13 to 16, wherein the second apparatus is configured to: verify an access token request based on a received profile of a knowledge source and information indicating which consumer is allowed to access the knowledge source; determine whether a first apparatus is allowed to access the knowledge source based on the verification of the access token request; and generate an access token response based on the determination that the first apparatus is allowed to access the knowledge source. Example 18. A second apparatus according to any one of Examples 13 to 17, wherein the second apparatus is configured to generate an access token response based on external information in the knowledge source and a profile of the knowledge source. Example 19. A second device according to any one of Examples 13 to 17, wherein the second device is configured to: transmit profile information of the first device to a fourth device; receive an authorization response from the first device from the fourth device; and generate an access token response based on determining that the authorization response indicates that the first device is permitted to access a knowledge source. Example 20. A second apparatus according to any one of Examples 13 to 17, wherein the second apparatus is configured to: transmit profile information of the first apparatus to a fourth apparatus; and receive an access token response from the fourth apparatus, the access token response including at least one of the following: a token of a knowledge source, information relating to an AF entity, or information relating to a NEF entity. Example 21. A second apparatus according to any one of Examples 18 to 20, wherein the access token response includes information corresponding to the primary knowledge source and the secondary knowledge source of the knowledge source. Example 22. The second apparatus according to Examples 13 to 21, wherein the discovery request includes filtering information, the filtering information including at least one of the following: an identifier of the knowledge source; a dataset identifier of the knowledge source; a dataset topic description related to the knowledge source; a domain related to the knowledge source; or a data attribute for discovering the knowledge source. Example 23. The second apparatus according to any one of Examples 13 to 22, wherein the access token request includes at least one of the following: an identifier of the knowledge source, consumer information of the knowledge source, or target producer information providing the knowledge service. Example 24. The second apparatus according to any one of Examples 13 to 23, wherein the access token response includes at least one of the following: a token of the knowledge source, consumer information of the knowledge source, or target producer information providing the knowledge service. Example 25. The second apparatus according to Example 23 or 24, wherein the target producer information includes at least one of the following: at least one identifier of a knowledge source, a dataset identifier or description of a dataset from the knowledge source, the dataset identifier or description indicating which data must be accessed, a domain identifier, or the type or quantity of data from the knowledge source being requested. Example 26. The second apparatus according to any one of Examples 13 to 25, wherein the knowledge source includes a primary knowledge source and a secondary knowledge source. Example 27. The second apparatus according to Example 26, wherein the primary knowledge source and the secondary knowledge source are on the same entity, or wherein the primary knowledge source and the secondary knowledge source are on different entities. Example 28. A second apparatus according to any one of Examples 13 to 27, wherein the profile of the list of network function examples includes detailed information on knowledge sources. Example 29. A second device according to any one of Examples 13 to 28, wherein the first device is implemented at a terminal device or a network device, and the second device is implemented at another network device. Example 30. A third apparatus, comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the third apparatus to: receive a knowledge service request from a first apparatus, the knowledge service request including a token of a knowledge source; determine, based on the token, whether the first apparatus is authorized to access the knowledge source; and, based on the determination that the first apparatus is authorized to access the knowledge source, transmit a knowledge service response to the first apparatus. Example 31: According to the third apparatus of Example 30, the third apparatus is configured to: determine whether the first apparatus is authorized to access one or more available knowledge sources based on the profiles of one or more available knowledge sources and information of trusted consumers of one or more available knowledge sources; and determine that the first apparatus is authorized to access one or more available knowledge sources based on the determination that the first apparatus is a trusted consumer among trusted consumers. Example 32. The third apparatus according to Example 30, wherein the knowledge service response includes at least one of rich prompts or Uniform Resource Identifiers for accessing the knowledge source. Example 33. A third device according to Example 30 or 32, wherein the first device is implemented at a terminal device or a network device, and the third device is implemented at another network device. Example 34. A method comprising: transmitting a discovery request for access to a knowledge source at a first device and to a second device; receiving a discovery response from the second device, the discovery response including a list of network function instances, the list of network function instances including knowledge sources and profiles of the network function instances; transmitting an access token request for the knowledge source to the second device; and receiving an access token response from the second device, the access token response including at least an identifier of the knowledge source. Example 35. A method comprising: receiving, at a second device and from a first device, a discovery request for access to a knowledge source; transmitting a discovery response to the first device, the discovery response including a list of network function instances, the list of network function instances including knowledge sources and profiles of the network function instances; receiving, from the first device, an access token request for the knowledge source; and transmitting an access token response to the first device, the access token response including at least an identifier of the knowledge source. Example 36. A method comprising: receiving a knowledge service request at a third device and from a first device, the knowledge service request including a token of a knowledge source; determining, based on the token, whether the first device is authorized to access the knowledge source; and, based on determining that the first device is authorized to access the knowledge source, transmitting a knowledge service response to the first device. Example 37. A first apparatus, comprising: a component for transmitting a discovery request for access to a knowledge source to a second apparatus; a component for receiving a discovery response from the second apparatus, the discovery response including a list of network function instances, the list of network function instances including knowledge sources and profiles of the network function instances; a component for transmitting an access token request for the knowledge source to the second apparatus; and a component for receiving an access token response from the second apparatus, the access token response including at least an identifier of the knowledge source. Example 38. A second apparatus, comprising: a component for receiving a discovery request for access to a knowledge source from a first apparatus; a component for transmitting a discovery response to the first apparatus, the discovery response including a list of network function instances, the list of network function instances including knowledge sources and profiles of the network function instances; a component for receiving an access token request for the knowledge source from the first apparatus; and a component for transmitting an access token response to the first apparatus, the access token response including at least an identifier of the knowledge source. Example 39. A third apparatus, comprising: a component for receiving a knowledge service request from a first apparatus, the knowledge service request including a token of a knowledge source; a component for determining, based on the token, whether the first apparatus is authorized to access the knowledge source; and a component for transmitting a knowledge service response to the first apparatus based on the determination that the first apparatus is authorized to access the knowledge source. Example 40. A computer-readable medium including instructions stored on the computer-readable medium for causing a device to perform at least the method according to any one of Examples 34 to 36.

Claims

1. A first means for communication, the first means comprising: At least one processor; as well as At least one memory stores instructions that, when executed by the at least one processor, cause the first device to: Send a discovery request for access to the knowledge source to the second device; The second device receives a discovery response, the discovery response indicating a list of network function instances, the list of network function instances including the knowledge source and profile of the list of network function instances; Send an access token request for the knowledge source to the second device; as well as The second device receives an access token response, which includes at least the identifier of the knowledge source.

2. The first apparatus of claim 1, wherein the discovery request includes filtering information, the filtering information including at least one of the following: The identifier of the knowledge source; Dataset identifiers related to the knowledge source, Dataset topic descriptions related to the knowledge source, The domain related to the knowledge source, or Discover the data attributes of the knowledge source.

3. The first apparatus according to claim 1, wherein the access token request includes at least one of the following: the identifier of the knowledge source, consumer information of the knowledge source, or target producer information providing the knowledge service.

4. The first apparatus according to claim 1, wherein the access token response includes at least one of the following: a token of the knowledge source, consumer information of the knowledge source, or target producer information providing the knowledge service.

5. The first apparatus according to claim 3 or 4, wherein the target producer information includes at least one of the following: At least one identifier of the knowledge source, A dataset identifier or description from the knowledge source, indicating which data must be accessed. Domain identifier, or The type and quantity of data from the knowledge sources that are permitted to be discovered and accessed, and The consumer information of the knowledge source includes at least one of the following: Approved entities used to update knowledge sources based on dataset identifiers. The Uniform Resource Identifier (URI) of the third device hosting the knowledge source; Discover the list of authorized entities for the knowledge source. The dataset identifies the authorized entity, or The knowledge source is allowed access in a certain way.

6. The first device according to claim 1, wherein the first device is configured to: A knowledge service request is transmitted to a third device, the knowledge service request including a token of the knowledge source; and The third device receives a knowledge service response, the knowledge service response including at least one of rich hints or a Uniform Resource Identifier for accessing the knowledge source.

7. The first apparatus according to claim 1, wherein the knowledge source includes a primary knowledge source and a secondary knowledge source.

8. The first device according to claim 1, wherein the first device is configured to: Based on determining that the first device is authorized to access one or more available knowledge sources, the knowledge source is selected based on the list of knowledge sources or based on information from the one or more available knowledge sources.

9. The first apparatus of claim 1, wherein the first apparatus is implemented at a terminal device or a network device, and the second apparatus is implemented at another network device.

10. A second means for communication, the second means comprising: At least one processor; as well as At least one memory stores instructions that, when executed by the at least one processor, cause the second device to: Receive a discovery request for access to a knowledge source from the first device; Identify zero or more network function instances, each of which is associated with a knowledge source; Determine zero or more network function instance identifiers corresponding to the zero or more network function instances; Identify zero or more knowledge source profiles, wherein each knowledge source profile corresponds to a knowledge source associated with one of the zero or more network function instances; A discovery response is transmitted to the first device, the discovery response including the zero or more network function instance identifiers and the zero or more knowledge source profiles; Receive an access token request for the knowledge source from the first device; as well as An access token response is sent to the first device, the access token response including at least the identifier of the knowledge source.

11. The second apparatus of claim 10, wherein the second apparatus is further configured to determine the zero or more network function instances by causing the second apparatus to perform the following: Store a list of empty network function instance identifiers for the zero or more network function instances; Identify at least one instance of a network function associated with the knowledge source; Based on the verification of the discovery request, for each of the at least one network function instance, it is determined whether the knowledge source is allowed to be discovered; as well as Based on the determination that the knowledge source is allowed to be discovered, for each of the at least one network function instance, the instance identifier of the network function instance is added to the network function instance identifier list.

12. The second device according to claim 10, wherein the second device is configured to: The knowledge source profile is received from at least one of the following: an Operation, Administration and Maintenance (OAM) entity, an application function entity, a third device, or a network open function entity; and The profile of the knowledge source said therein includes at least one of the following: The identifier of the knowledge source; The dataset identifier of the knowledge source; Dataset topic descriptions related to the knowledge source, Approved entities used to update knowledge sources based on dataset identifiers. The Uniform Resource Identifier of the third device hosting the knowledge source; Domains specific to the knowledge source Discover the list of authorized entities for the knowledge source, or Data sets identify specific information.

13. The second device according to claim 10, wherein the second device is configured to: The access token request is verified based on the received profile of the knowledge source and information indicating which consumer is allowed to access the knowledge source. The first device is determined to be allowed to access the knowledge source based on the verification of the access token request. as well as The access token response is generated based on the determination that the first device is allowed to access the knowledge source.

14. A third means for communication, the third means comprising: At least one processor; as well as At least one memory stores instructions that, when executed by the at least one processor, cause the third device to: Receive a knowledge service request from the first device, the knowledge service request including a token of the knowledge source; The token is used to determine whether the first device is authorized to access the knowledge source; as well as Based on the determination that the first device is authorized to access the knowledge source, a knowledge service response is transmitted to the first device.

Images