Model reasoning method and device in communication system, equipment and medium

Abstract

The invention discloses a model reasoning method and device in a communication system, equipment and a medium, and belongs to the field of communication, and the method comprises the steps that first equipment receives first information sent by second equipment, the first equipment stores knowledge information and at least one of calling interface information of the knowledge information, and the calling interface information of the knowledge information comprises calling interface information of the second equipment; the knowledge information comprises at least one of a knowledge graph and a knowledge vector library; the first device sends second information to a third device, a first model is deployed on the third device, and the second information is determined according to the first information and knowledge information stored or called on the first device; wherein the first information comprises at least one of the following information: auxiliary information used for reasoning of the first model; the demand information is used for reasoning of the first model; input information of the first model; calling interface information of the knowledge information; indicating a receiving device of the second information; the input information of the first model corresponds to a sample indication.

Description

Technical Field

[0001] This application belongs to the field of communication technologies, and particularly relates to a model inference method, apparatus, device, and medium in a communication system. Background Art

[0002] In mobile communication systems, there are more and more use cases combined with artificial intelligence (AI). For example, at the physical layer, there are AI-based CSI (channel state information) prediction and feedback compression, AI-based beam management, AI-based positioning, etc. To make different communication scenarios universal, it is proposed to apply large models to communication networks to solve problems in communication systems. However, large models are usually superior in solving general problems, but there are many different problems in communication systems, such as beamforming, resource allocation, channel prediction, etc. Therefore, how to ensure the performance of large models in solving specific downstream problems is an urgent problem to be solved. Summary of the Invention

[0003] Embodiments of this application provide a model inference method, apparatus, device, and medium in a communication system, which can improve the performance of the model.

[0004] In a first aspect, a model inference method in a communication system is provided. The method includes:

[0005] A first device receives first information sent by a second device. Among them, at least one of knowledge information and call interface information of the knowledge information is stored on the first device, and the knowledge information includes at least one of the following: knowledge graph, knowledge vector library;

[0006] The first device sends second information to a third device, and a first model is deployed on the third device. The second information is determined according to the first information and the knowledge information stored or called on the first device;

[0007] Among them, the first information includes at least one of the following:

[0008] Auxiliary information for the first model to perform inference;

[0009] Requirement information for the first model to perform inference;

[0010] Input information of the first model;

[0011] Call interface information of the knowledge information;

[0012] Receiving device indication of the second information;

[0013] Sample indication corresponding to the input information of the first model.

[0014] In a second aspect, a method for model inference in a communication system is provided. The method includes:

[0015] A second device sends first information to a first device, where at least one of knowledge information and call interface information of the knowledge information is stored on the first device, and the knowledge information includes at least one of the following: a knowledge graph, a knowledge vector library;

[0016] Wherein, the first information includes at least one of the following:

[0017] Auxiliary information for the first model to perform inference;

[0018] Requirement information for the first model to perform inference;

[0019] Input information of the first model;

[0020] Call interface information of the knowledge information;

[0021] Receiver indication of the second information;

[0022] Sample indication corresponding to the input information of the first model.

[0023] In a third aspect, a method for model inference in a communication system is provided, including:

[0024] A third device obtains second information from the first device or a target device, where the first model is deployed on the third device, and at least one of knowledge information and call interface information of the knowledge information is stored on the first device or the target device, and the target device is the device accessed using the call interface information, and the knowledge information includes at least one of the following: a knowledge graph, a knowledge vector library;

[0025] Wherein, the second information is determined according to the first information and the knowledge information stored or called on the first device;

[0026] Wherein, the first information includes at least one of the following:

[0027] Auxiliary information for the first model to perform inference;

[0028] Requirement information for the first model to perform inference;

[0029] Input information of the first model;

[0030] Call interface information of the knowledge information;

[0031] Receiver indication of the second information;

[0032] Sample indication corresponding to the input information of the first model.

[0033] In a fourth aspect, a communication device is provided, including:

[0034] a communication unit, configured to receive first information sent by a second device, where at least one of knowledge information and call interface information of the knowledge information is stored on the communication device, and the knowledge information includes at least one of the following: a knowledge graph, a knowledge vector library; and

[0035] send second information to a third device, where a first model is deployed on the third device, and the second information is determined according to the first information and the knowledge information stored or called on the first device;

[0036] wherein the first information includes at least one of the following:

[0037] auxiliary information for the first model to perform inference;

[0038] requirement information for the first model to perform inference;

[0039] input information of the first model;

[0040] call interface information of the knowledge information;

[0041] a receiving device indication of the second information;

[0042] a sample indication corresponding to the input information of the first model.

[0043] In a fifth aspect, a communication device is provided, including:

[0044] a communication unit, configured to send first information to a first device, where at least one of knowledge information and call interface information of the knowledge information is stored on the first device, and the knowledge information includes at least one of the following: a knowledge graph, a knowledge vector library;

[0045] wherein the first information includes at least one of the following:

[0046] auxiliary information for the first model to perform inference;

[0047] requirement information for the first model to perform inference;

[0048] input information of the first model;

[0049] call interface information of the knowledge information;

[0050] a receiving device indication of the second information;

[0051] a sample indication corresponding to the input information of the first model.

[0052] In a sixth aspect, a communication device is provided, including:

[0053] a communication unit, configured to obtain second information from a first device or a target device, where a first model is deployed on the third device, and at least one of knowledge information and call interface information of the knowledge information is stored on the first device or the target device, where the target device is a device accessed using the call interface information, and the knowledge information includes at least one of the following: a knowledge graph, a knowledge vector library;

[0054] wherein the second information is determined according to first information and the knowledge information stored or called on the first device;

[0055] wherein the first information includes at least one of the following:

[0056] auxiliary information for inference by the first model;

[0057] requirement information for inference by the first model;

[0058] input information of the first model;

[0059] call interface information of the knowledge information;

[0060] a receiving device indication of the second information;

[0061] a sample indication corresponding to the input information of the first model.

[0062] In a seventh aspect, a communication device is provided. The network-side device includes a processor and a memory. The memory stores a program or instruction that can be run on the processor. When the program or instruction is executed by the processor, the steps of the method described in any one of the first aspect to the third aspect are implemented.

[0063] In an eighth aspect, a readable storage medium is provided. A program or instruction is stored on the readable storage medium. When the program or instruction is executed by a processor, the steps of the method described in any one of the first aspect to the third aspect are implemented.

[0064] In a ninth aspect, a wireless communication system is provided, including: a first device, a second device, and a third device. The first device can be used to execute the steps of the method described in the first aspect, the second device can be used to execute the steps of the method described in the second aspect, and the third device can be used to execute the steps of the method described in the third aspect.

[0065] In a tenth aspect, a chip is provided, which includes a processor and a communication interface. The communication interface is coupled to the processor, and the processor is configured to run programs or instructions to implement the method described in any one of the first to third aspects.

[0066] In an eleventh aspect, a computer program / program product is provided. The computer program / program product is stored in a storage medium and is executed by at least one processor to implement the method described in any one of the first to third aspects.

[0067] In the embodiments of the present application, the second device can assist the third device in obtaining the second information by sending the first information to the first device. Further, the third device can use the second information to assist in the inference of the model, which is beneficial to improving the inference performance of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 FIG. is a schematic diagram of a communication system provided by an embodiment of the present application.

[0069] Figure 2 FIG. is a schematic diagram of a model inference method in a communication system provided by an embodiment of the present application.

[0070] Figure 3 FIG. is a schematic diagram of a hidden variable provided by an embodiment of the present application.

[0071] Figures 4 to 14 FIG. is a schematic interaction diagram of a model inference method provided by an embodiment of the present application.

[0072] Figure 15 FIG. is a schematic diagram of a communication device provided by an embodiment of the present application.

[0073] Figure 16 FIG. is a schematic diagram of another communication device provided by an embodiment of the present application.

[0074] Figure 17 FIG. is a schematic diagram of yet another communication device provided by an embodiment of the present application.

[0075] Figure 18 FIG. is a schematic diagram of a communication device provided by an embodiment of the present application.

[0076] Figure 19 FIG. is a hardware structure diagram of a terminal provided by an embodiment of the present application.

[0077] Figure 20 FIG. is a hardware structure diagram of a network-side device provided by an embodiment of the present application.

[0078] Figure 21It is a hardware structure diagram of another network-side device provided by an embodiment of the present application. Detailed implementation manners

[0079] Next, the technical solutions in the embodiments of the present application will be clearly described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application belong to the scope of protection of the present application.

[0080] The terms "first", "second", etc. in the present application are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such terms can be interchanged under appropriate circumstances so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same type, and the number of objects is not limited. For example, the first object can be one or multiple. In addition, "or" in the present application means at least one of the connected objects. For example, "A or B" covers three scenarios, namely, Scenario 1: including A and not including B; Scenario 2: including B and not including A; Scenario 3: including both A and B. The character " / " generally indicates an "or" relationship between the associated objects before and after.

[0081] The term "indication" in the present application can be either a direct indication (or an explicit indication) or an indirect indication (or an implicit indication). Among them, a direct indication can be understood as that the sender clearly informs the receiver of specific information, operations to be performed, or request results, etc. in the sent indication; an indirect indication can be understood as that the receiver determines the corresponding information according to the indication sent by the sender, or makes a judgment and determines the operations to be performed or request results, etc. according to the judgment result.

[0082] It should be noted that the technology described in the embodiments of this application is not limited to the Long Term Evolution (LTE) / LTE-Advanced (LTE-A) system, and can also be used in other wireless communication systems, such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single-carrier Frequency-Division Multiple Access (SC-FDMA), or other systems. The terms "system" and "network" in the embodiments of this application are often used interchangeably, and the described technology can be used in the above-mentioned systems and radio technologies, as well as in other systems and radio technologies. The following description describes the New Radio (NR) system for example purposes, and the NR terms are used in most of the following descriptions, but these technologies can also be applied to systems other than the NR system, such as the 6th Generation (6 th Generation, 6G) communication system.

[0083] Figure 1A block diagram of a wireless communication system to which embodiments of the present application can be applied is shown. The wireless communication system includes a terminal 11 and a network-side device 12. Among them, the terminal 11 can be a mobile phone, a tablet personal computer, a laptop computer, a notebook computer, a personal digital assistant (PDA), a handheld computer, a netbook, an ultra-mobile personal computer (UMPC), a mobile internet device (MID), an augmented reality (AR), a virtual reality (VR) device, a robot, a wearable device, a flight vehicle, a vehicle user equipment (VUE), a shipborne device, a pedestrian user equipment (PUE), a smart home (home devices with wireless communication functions, such as refrigerators, TVs, washing machines, or furniture, etc.), a game console, a personal computer (PC), a teller machine, or a self-service machine, etc., which are terminal-side devices. Wearable devices include: smart watches, smart bracelets, smart earphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets, smart ankle chains, etc.), smart wristbands, smart clothing, etc. Among them, the vehicle user equipment can also be referred to as a vehicle terminal, a vehicle controller, a vehicle module, a vehicle component, a vehicle chip, or a vehicle unit, etc. It should be noted that the specific type of the terminal 11 is not limited in the embodiments of the present application.

[0084] The terminal can also be referred to as a user equipment (UE), a terminal device, an access terminal, a user unit, a user station, a mobile station, a mobile device, a remote terminal, a mobile device, a user terminal, a terminal, a wireless communication device, a user agent, or a user device, etc.

[0085] The network-side device 12 may include an access network device or a core network device. Among them, the access network device may also be referred to as a Radio Access Network (RAN) device, a radio access network function, or a radio access network unit. The access network device may include a base station, a Wireless Local Area Network (WLAN) Access Point (AP), or a Wireless Fidelity (WiFi) node, etc. Among them, the base station may be referred to as Node B (NB), Evolved Node B (eNB), the next generation Node B (gNB), New Radio Node B (NR Node B), access point, Relay Base Station (RBS), Serving Base Station (SBS), Base Transceiver Station (BTS), radio base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (ESS), home Node B (HNB), home evolved Node B, Transmission Reception Point (TRP), or some other suitable term in the art. As long as the same technical effect is achieved, the base station is not limited to specific technical terms. It should be noted that in the embodiments of this application, only the base station in the NR system is used as an example for introduction, and the specific type of the base station is not limited.

[0086] The core network device may include, but is not limited to, at least one of the following: core network nodes, core network functions, Mobility Management Entity (MME), Access and Mobility Management Function (AMF), Session Management Function (SMF), User Plane Function (UPF), Policy Control Function (PCF), Policy and Charging Rules Function (PCRF), Edge Application Server Discovery Function (EASDF), Unified Data Management (UDM), Unified Data Repository (UDR), Home Subscriber Server (HSS), Centralized network configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), Local NEF (L-NEF), Binding Support Function (BSF), Application Function (AF), etc. It should be noted that in the embodiments of this application, only the core network devices in the NR system are taken as examples for introduction, and the specific types of core network devices are not limited.but not limited to at least one of the following: core network node, core network function, Mobility Management Entity (MME), Access and Mobility Management Function (AMF), Session Management Function (SMF), User Plane Function (UPF), Policy Control Function (PCF), Policy and Charging Rules Function (PCRF), Edge Application Server Discovery Function (EASDF), Unified Data Management (UDM), Unified Data Repository (UDR), Home Subscriber Server (HSS), Centralized network configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), Local NEF (L-NEF), Binding Support Function (BSF), Application Function (AF), etc. It should be noted that in the embodiments of this application, only the core network devices in the NR system are taken as examples for introduction, and the specific types of core network devices are not limited.

[0087] (Network Exposure Function, NEF), Local NEF (L-NEF), Binding Support Function (BSF), Application Function (AF), etc. It should be noted that in the embodiments of this application, only the core network devices in the NR system are taken as examples for introduction, and the specific types of core network devices are not limited.

[0088] To facilitate the understanding of the embodiments of this application, the related technologies of this application are described.

[0089] I. Knowledge Graph

[0090] A knowledge graph is a knowledge base that uses graph structures or topologies to represent and integrate data. It can store the associated descriptions between entities (objects, events, situations, or abstract concepts), and at the same time, it can also encode the semantic relationships between entities. The application prospects of knowledge graphs are very broad. They can not only improve the effectiveness of data services such as information retrieval, search engines, and recommendation systems, but also support intelligent interactions such as natural language question answering, dialogue, and reasoning, with characteristics such as strong representation ability, high flexibility, strong computability, and strong cross-domain nature.

[0091] Knowledge graphs store a large amount of knowledge in a clear and structured way and can be used to enhance the knowledge awareness of large models. Incorporating knowledge graphs into large models during the reasoning stage can significantly improve the performance of large models in accessing domain-specific knowledge by retrieving knowledge from the knowledge graph.

[0092] In the communication field, network data knowledge graphs are mainly used for knowledge representation, association relationship analysis, and in-depth mining, providing effective knowledge rules and knowledge computing support for the intelligence of communication systems.

[0093] The network structure, terminal types, terminal behaviors, data service requirements, and system resources of communication systems all have characteristics such as high dynamism, strong timeliness, and mutual coupling. Mobile communication data faces many challenges, such as difficult decentralized data acquisition, a wide variety of complex structures, and difficult complex association mining, etc.

[0094] Using knowledge graphs can effectively clarify various relationships between data fields and communication network metrics, and further conduct in-depth mining based on the established relationships, such as quantifying the degree of association of relationships, characterizing the characteristic attributes of data fields and metrics, etc.

[0095] In some implementation methods, a knowledge graph can be represented by the correlation between entities. One representation method can be:

[0096] (Entity 1, correlation coefficient between Entity 1 and Entity 2, Entity 2).

[0097] For example, Entity 1 is the cell throughput, Entity 2 is the L1-RSRP of the strongest beam, and the correlation coefficient is 0.5.

[0098] In other implementation methods, a knowledge graph can be represented by Subject, Predicate, Object (SPO) triples. For example, the SPO triple of (cell id, shaping codebook, codebook indication).

[0099] II. Knowledge Vector Library

[0100] A knowledge vector library is a database used to store, retrieve, and analyze vectors. The reason it is called a database is that it has the following characteristics:

[0101] a) Provide a standard access interface to lower the user's usage threshold;

[0102] b) Provide the ability for efficient data organization, retrieval, and analysis. When general users store and retrieve vectors, they also need to manage structured data, that is, support the management ability of traditional databases for structured data.

[0103] Therefore, the knowledge vector library can be simply understood as a database for storing the input feature vectors of the model.

[0104] For example, when using image search for images or voice search for voice, what is stored and compared in the knowledge vector library is not the images and voice segments, but the "features" extracted through algorithms such as deep learning, such as arrays of 256 or 512 floating-point numbers (float), which can be represented by vectors in mathematics.

[0105] In some cases, the knowledge vector library can be a model, with an image or text as the input. For example, in the communication field, the input of the model can be the measurement results of wireless signals, etc., and the output is a feature vector used to characterize the measurement results.

[0106] The following methods can be adopted locally to use the knowledge graph or knowledge vector library:

[0107] 1) Directly as the input of the model;

[0108] 2) After processing the model input with the knowledge graph, then input the processed data and the original model input into the model together.

[0109] 3) After processing the model input with the knowledge vector library, then input the processed data and the original model input into the model together, or directly use the processed data as the input to the model.

[0110] III. Prompt Engineering

[0111] Prompt engineering is a technique in Natural Language Processing (NLP) that can create text fragments, prompts, or templates to guide pre-trained large language models to generate high-quality outputs for specific tasks or applications. It is widely used in fields such as question answering, summarization, translation, sentiment analysis, and text generation. Prompt engineering can utilize the powerful capabilities of pre-trained large language models to achieve various complex natural language processing tasks, reduce the dependence on labeled data and model fine-tuning, lower development costs and time, as well as improve the interpretability and controllability of pre-trained large language models, increasing user trust and satisfaction. Common prompts can include Zero-Shot Prompting, Few-Shot Prompting, Chain of Thought Prompting, etc.

[0112] In prompt engineering, the description of the task is embedded in the input to guide the generative AI solution to generate the required output.

[0113] In mobile communication systems, there are an increasing number of use cases that combine Artificial Intelligence (AI). For example, at the physical layer, there are AI-based Channel State Information (CSI) prediction and feedback compression, AI-based beam management, AI-based positioning, etc. In some scenarios, the introduction of AI-based energy saving, AI-based load balancing, etc. is also considered. In the future, there will be more use cases that combine AI in mobile communication systems.

[0114] In recent years, large language models have received extensive attention in fields such as chatting and image generation. Therefore, it is considered to apply large models to communication networks as a tool to optimize the performance of communication networks. However, there are two problems with communication large models.

[0115] Problem 1: The size of large models is very large. How to deploy large models to devices with limited storage resources such as base stations or terminals for inference? A typical method is to process the large model through compression methods such as quantization and pruning, and then deploy it to communication devices.

[0116] Problem 2: Large models often solve general problems, while there are many different problems in communication, such as beamforming, resource allocation, and channel prediction. How to match large models to specific downstream problems? There is currently no mature solution in communication systems.

[0117] The following will combine the accompanying drawings to elaborate in detail on the inference method based on the artificial intelligence (AI) model in the communication system provided by the embodiments of the present application through some embodiments and their application scenarios.

[0118] Figure 2 The schematic diagram of the model inference method in the communication system according to the embodiments of the present application is shown. As Figure 2 shown, the method 200 includes:

[0119] S201, the first device receives the first information sent by the second device, where at least one of the knowledge information and the call interface information of the knowledge information is stored on the first device;

[0120] S202, the first device sends the second information to the third device, and the first model is deployed on the third device, and the second information is determined according to the first information and the knowledge information.

[0121] In some embodiments, the first model is a compressed AI model. For example, the first model is an AI model compressed from a large model. Here, the compression may include, but is not limited to, quantization, pruning (for example, deleting some layers in the model, or deleting some nodes in a certain layer), etc.

[0122] It should be noted that in the embodiments of the present application, the AI model may also be referred to as an AI unit, an ML (machine learning) model, an ML unit, an AI structure, an AI function, an AI feature, a machine learning model, a neural network, a neural network function, a neural network function, etc. Or, the AI model may also refer to a processing unit capable of implementing specific algorithms, formulas, processing flows, capabilities, etc. related to AI, or the AI model may be a processing method, algorithm, function, module or unit for a specific data set, or the AI model may be a processing method, algorithm, function, module or unit running on AI / ML-related hardware such as a Graphics Processing Unit (GPU), a Neural Processing Unit (NPU), a Tensor Processing Unit (TPU), an Application Specific Integrated Circuit (ASIC), etc. The present application does not make specific limitations in this regard.

[0123] In the embodiments of the present application, the first device can be regarded as a device for storing knowledge-related information, or a knowledge storage device. For example, knowledge information, call interface information of knowledge information, etc. are stored on the first device. The second device can be regarded as a device that initiates a demand, that is, a device that initiates an inference demand. The third device can be regarded as an inference device, or an AI model deployment device. Among them, the second device and the third device can be the same device, or they can also be different devices.

[0124] In some embodiments, the knowledge information includes, but is not limited to, at least one of a knowledge graph and a knowledge vector library.

[0125] In some embodiments, the third device can use the second information to expand the input of the first model and assist the first model in performing accurate inference.

[0126] Therefore, in the embodiments of the present application, by obtaining the second information from the knowledge storage device and further using the model for inference, the second information and the input information of the first model can be used, so that the model can obtain more input information and improve the inference performance of the model.

[0127] In some embodiments, the first device is an access network device (such as a base station) or a core network function, such as a core network function with a data storage function, such as a database, a data function, or a Network Repository Function (NRF) or UDM, or an AI model library, or an AI model management function, or a third-party server of a terminal, etc. The second device is a terminal, and the third device is a terminal.

[0128] In other embodiments, the first device is a core network function, such as a core network function with a data storage function, such as a database, a data function, or a Network Repository Function (NRF) or UDM, or an AI model library, or an AI model management function, or a third-party server of a terminal, etc. The second device is a terminal or an access network device (such as a base station), and the third device is an access network device (such as a base station).

[0129] In still other embodiments, the first device is a core network function, such as a core network function with a data storage function, such as a database, a data function, or a Network Repository Function (NRF) or UDM, or an AI model library, or an AI model management function, or a third-party server of a terminal, etc. The second device is a terminal or an access network device (such as a base station), a third-party server, a core network function (such as an AI control function, a task control function, a collaborative control function, or other core network functions, etc.), and the third device is an inference function of the core network.

[0130] In an embodiment of the present application, the AI control node can be used for controlling AI-related functions, and the task control node can be used for controlling service-related functions or task-related control functions. For example, the task may include, but is not limited to, the coordination and allocation of connection, computing, data, and algorithm resources in a multi-node scenario involved in network new capabilities, for jointly achieving a specific goal. The coordination control node can be used for the collaborative management between multiple functions (such as communication function, data function, computing power function, algorithm function, model function).

[0131] In an embodiment of the present application, the information interaction between the terminal and the access network device can be through at least one of the following signaling: Layer 1 signaling, Layer 2 signaling, Layer 3 signaling, data plane signaling, AI layer signaling. The information interaction between the terminal and the core network device can be through at least one of the following signaling: NAS signaling, data plane signaling, AI layer signaling. The terminal and the core network device can directly perform information interaction, or alternatively, can also forward information through other devices, such as through communication control functions (such as access mobility management function AMF), AI control functions, task control functions, coordination control functions, network openness functions, etc.

[0132] It should be understood that the access network device and the core network device, as well as between core network devices, can directly perform information interaction, or alternatively, can also forward information through other devices, such as through communication control functions (such as access mobility management function AMF), AI control functions, task control functions, coordination control functions, network openness functions, etc.

[0133] In some embodiments, if the first device is an access network device and the second device is a terminal, the first information is carried in at least one of the following signaling: Layer 1 signaling, Layer 2 signaling, Layer 3 signaling, data plane signaling, AI layer signaling. Exemplarily, the Layer 1 signaling may include, but is not limited to, PDCCH. The Layer 2 signaling may include, for example, but is not limited to, downlink MAC CE, and the Layer 3 signaling may include, for example, but is not limited to, RRC signaling.

[0134] In other embodiments, if the first device is a core network function and the second device is a terminal, the first information is carried in at least one of the following signaling: NAS signaling, data plane signaling, AI layer signaling.

[0135] In some embodiments, if the first device is an access network device and the third device is a terminal, the second information is carried in at least one of the following signaling: layer 1 signaling, layer 2 signaling, layer 3 signaling, data plane signaling, and AI layer signaling. Exemplarily, the layer 1 signaling may include, but is not limited to, PDCCH. The layer 2 signaling may include, for example, but is not limited to, downlink MAC CE, and the layer 3 signaling may include, for example, but is not limited to, RRC signaling.

[0136] In some other embodiments, if the first device is a core network function and the third device is a terminal, the second information is carried in at least one of the following signaling: NAS signaling, data plane signaling, and AI layer signaling.

[0137] In some embodiments of the present application, the first device may also send the first information to the third device to assist the third device in performing model inference. For example, the first information may be carried in the second information, that is, the second information includes the content of the first information.

[0138] In some embodiments, the first information includes at least one of the following:

[0139] Auxiliary information for the first model to perform inference;

[0140] Requirement information for the first model to perform inference;

[0141] Input information of the first model;

[0142] Invocation interface information of knowledge information;

[0143] Receiving device indication of the second information;

[0144] Sample indication corresponding to the input information of the first model.

[0145] Therefore, by the requirement initiating device indicating the relevant information (i.e., the first information) for model inference to the knowledge storage device, the knowledge storage device can determine the second information based on the first information and the knowledge information. For example, the knowledge information in the professional field is obtained according to the first information, so as to obtain the second information, and then the second information is further sent to the inference device. In this way, when the inference device uses the second information to assist in model inference, it can achieve accurate inference of professional field tasks, reduce the hallucination problem in model inference, and improve the inference performance of the model.

[0146] In some embodiments, the receiving device indication may be the identification information of the receiving device of the second device, which is used for the first device to determine to which device the second information is to be sent.

[0147] In some embodiments, the sample indication may be which sample the input information of the first model in the first information corresponds to, so as to prevent the inference device from using the first information and the second information corresponding to different samples for model inference, which may affect the inference result. For example, the sample indication may be a sample identifier.

[0148] In some embodiments, the auxiliary information for inference by the first model includes at least one of the following:

[0149] The identification information of the AI model, the function indication of the AI model, the characteristic indication of the AI model, the target knowledge graph indication, the target knowledge vector library indication, the cell identifier, the Bandwidth Part (BWP) identifier.

[0150] In some embodiments, the identification information of the AI model may be, for example, the model identifier (modelID) of the AI model, which can be used for AI structure identification, AI algorithm identification, or the identifier of a specific dataset associated with the AI model, or the identifier of a specific scenario, environment, channel characteristic, or device related to AI / ML, or the identifier of a function, characteristic, capability, or module related to AI / ML. This application does not make specific limitations on this.

[0151] In some embodiments, the characteristic indication of the AI model may explicitly or implicitly indicate the characteristics supported by the AI model, such as supporting CSI prediction and compression feedback, beam prediction, positioning, load balancing, resource allocation, etc. By indicating the characteristics of the AI model to the first device, the first device can be assisted in obtaining knowledge information related to the characteristics, and then obtaining the corresponding second information according to the knowledge information for assisting the third device in model inference.

[0152] In the embodiments of this application, it can be considered that the characteristics of the AI model under specific configurations are the functions of the AI model.

[0153] For example, if the characteristic of the AI model is beam prediction, the function of the AI model is time-domain beam prediction with 32 transmission beams configured for the base station.

[0154] In some embodiments, the function indication of the AI model may explicitly or implicitly indicate the functions supported by the AI model, such as supporting CSI prediction and compression feedback, beam prediction, positioning, load balancing, resource allocation, etc. under specific configurations. By indicating the function of the AI model to the first device, the first device can be assisted in obtaining knowledge information related to the function, and then obtaining the corresponding second information according to the knowledge information for assisting the third device in model inference.

[0155] In some embodiments, the target knowledge graph may be used to indicate the knowledge graph that the second device expects to use. By indicating the expected target knowledge graph to the first device, the first device can be assisted in using the target knowledge graph to obtain corresponding second information, which is used to assist the third device in model inference.

[0156] In some implementation manners, the target knowledge graph indication may explicitly indicate the target knowledge graph.

[0157] Exemplarily, the target knowledge graph indication is used for at least one of the following:

[0158] The relationships between entities in the target knowledge graph;

[0159] The attribute information of entities in the target knowledge graph;

[0160] The data structure of the target knowledge graph;

[0161] The correlation between entities in the target knowledge graph;

[0162] The Subject, Predicate, Object (SPO) triples of statements in the target knowledge graph;

[0163] The target knowledge graph file.

[0164] In some embodiments, the relationships between entities in the target knowledge graph may be represented in the following format:

[0165] (Entity 1, relationship, Entity 2).

[0166] Optionally, the relationships between entities may include but are not limited to the hyponymy relationship, inheritance relationship, etc.

[0167] In some embodiments, the attribute information of entities in the target knowledge graph may be represented in the following format:

[0168] (Entity, attribute name, attribute value).

[0169] In some embodiments, the relationships between nodes may be represented by triples, for example, a relationship may be (Node 1, edge, Node 2).

[0170] In some embodiments, the correlation between entities in the target knowledge graph may be represented in the following format:

[0171] (Entity 1, correlation coefficient between Entity 1 and Entity 2, Entity 2).

[0172] For example, Entity 1 is the cell throughput, Entity 2 is the L1-RSRP of the strongest beam, and the correlation coefficient is 0.8.

[0173] In some embodiments, the SPO triples of the statements in the target knowledge graph are used to describe the subject, predicate, and object of the statements.

[0174] In some other implementation manners, the target knowledge graph indication may implicitly indicate the target knowledge graph.

[0175] Exemplarily, the target knowledge graph indication includes, but is not limited to, at least one of the following:

[0176] Knowledge graph identifier;

[0177] Knowledge graph file address.

[0178] In some embodiments, multiple knowledge graph identifiers may be predefined or preconfigured (e.g., pre-indicated by a network-side device), and each knowledge graph identifier corresponds to a knowledge graph. For example, the second device may select one or more knowledge graph identifiers from the multiple knowledge graph identifiers and indicate them to the first device. Further, the first device may determine the corresponding knowledge graph according to the indicated knowledge graph identifier, and further determine the second information based on the knowledge graph.

[0179] In some embodiments, multiple knowledge graph file addresses may be predefined or preconfigured (e.g., pre-indicated by a network-side device), and each knowledge graph file address corresponds to a knowledge graph file. For example, the second device may select one or more knowledge graph file addresses from the multiple knowledge graph file addresses and indicate them to the first device. Further, the first device may obtain the corresponding knowledge graph file according to the indicated knowledge graph file address, and further determine the second information based on the knowledge graph file.

[0180] In some embodiments, taking the task of the AI model as beam prediction as an example, the usage mode of the knowledge graph may be:

[0181] According to the cell identifier and the target task, the cell transmission beam configuration, such as the pointing of the transmission beam or the shaping codebook, can be retrieved using the knowledge graph.

[0182] In some implementation manners, the target knowledge vector library indication may implicitly indicate the target knowledge vector library expected to be used by the second device. By indicating the expected target knowledge vector library to the first device, the first device can be assisted in using the target knowledge vector library to obtain the corresponding second information for assisting the third device in model inference.

[0183] Exemplarily, the target knowledge vector library indication includes, but is not limited to, at least one of the following:

[0184] Knowledge vector library identifier;

[0185] Knowledge vector library file address.

[0186] In some embodiments, multiple knowledge vector library identifiers may be predefined or preconfigured (e.g., pre-indicated by a network-side device), and each knowledge vector library identifier corresponds to a knowledge vector library. For example, the second device may select one or more knowledge vector library identifiers from the multiple knowledge vector library identifiers and indicate them to the first device. Further, the first device may determine the corresponding knowledge vector library according to the indicated knowledge vector library identifier, and further determine the second information based on the knowledge vector library.

[0187] In some embodiments, multiple knowledge vector library file addresses may be predefined or preconfigured (e.g., pre-indicated by a network-side device), and each knowledge vector library file address corresponds to a knowledge vector library file. For example, the second device may select one or more knowledge vector library file addresses from the multiple knowledge vector library file addresses and indicate them to the first device. Further, the first device may obtain the corresponding knowledge vector library file according to the indicated knowledge vector library file address, and further determine the second information based on the knowledge vector library file.

[0188] In some embodiments, taking the task of the AI model being beam prediction as an example, the usage method of the knowledge vector library may be:

[0189] According to the cell identifier and the target task, obtain the cell transmission beam configuration by using the knowledge vector library, such as the pointing of the transmission beam or the feature vector related to the shaping codebook.

[0190] In some embodiments, the cell identifier may be used for the first device to obtain the knowledge information of the cell corresponding to the cell identifier, such as cell configuration, etc. Further, the second information may be determined according to the knowledge information.

[0191] In some embodiments, the BWP identifier may be used for the first device to obtain the knowledge information of the BWP corresponding to the BWP identifier, such as BWP configuration, etc. Further, the second information may be determined according to the knowledge information.

[0192] In some embodiments, the requirement information for the first model to perform inference includes at least one of the following:

[0193] The target task indication required for the AI model to be adapted;

[0194] The quality of experience (QoE) requirement of the AI model;

[0195] The service processing delay of the AI model;

[0196] The service processing accuracy of the AI model;

[0197] The service calculation amount of the AI model;

[0198] The data processing scale of the AI model.

[0199] For example, the first device may select knowledge information related to the target task according to the target task instruction, and then determine the second information according to the knowledge information.

[0200] For another example, the second device may select knowledge information that meets the QoE requirements, or processing delay, processing accuracy, or processing scale, etc., according to the QoE requirements, or processing delay, processing accuracy, or processing scale, and then determine the second information according to the knowledge information.

[0201] In some embodiments, the input information of the first model is related to the target task that the first model needs to adapt to. Taking beam prediction as an example of the target task, the input information may include the cell identifier and the measurement quality of the cell transmission beam.

[0202] In some embodiments, the call interface information of the knowledge information includes but is not limited to at least one of the following:

[0203] Call interface name, input data format of the call interface, output data format of the call interface, input data of the call interface. Among them, the input data format, output data format, and input data are related to the field, task, scenario, etc.

[0204] In some embodiments, the call interface may include but is not limited to interfaces for accessing external devices such as Application Programming Interface (API).

[0205] The communication system will continuously iterate and update, and the knowledge of the communication system will also be updated. After network deployment, the updated system knowledge, such as the latest knowledge from organizations such as 3GPP, can be stored in an external database. In this case, by calling the API interface, services are provided for the communication system, improving the applicability of the technical solution of the present application.

[0206] In some embodiments, both the second device and the third device are terminals, and the first information may be sent under specific circumstances, such as:

[0207] When at least one of the following events occurs, the second device sends the first information to the first device:

[0208] Event 1: Receiving a first indication for instructing the terminal to activate or switch the model;

[0209] Event 2: The terminal determines to activate or switch the model;

[0210] Event 3: The terminal determines that the currently used model does not meet the requirements;

[0211] Event 4: A second indication is received, and the second indication is used to indicate that the model used by the current terminal does not meet the requirements.

[0212] In some embodiments, the event that triggers the second device to send the first information to the first device may be predefined or configured by the network-side device. When the event that triggers the first information is satisfied, the second device may send the identification information of the event to the network-side device. For example, it is sent to the network-side device carried in the first information, so that the network-side device can learn about the events occurring on the second device.

[0213] In some embodiments, the first indication may be sent by the network-side device. For example, the network-side device can control the activation or switching of the AI model. For example, when the currently used model does not meet the requirements, the network-side device can send the first indication to the terminal, so that the second device can send the first information according to the first indication.

[0214] In some embodiments, when the performance of the currently used model by the terminal does not meet the requirements, the terminal can independently determine to activate or switch the model.

[0215] In some embodiments, the second indication may be sent by the network-side device. For example, when the currently used model does not meet the requirements, the network-side device can send the second indication to the terminal to indicate that the model currently used by the terminal does not meet the requirements. Further, the terminal can independently determine whether to activate or switch the model according to the second indication.

[0216] In some embodiments, the model not meeting the requirements may mean that the model does not meet the requirement information, for example, it may include at least one of the following: the inference of the model does not meet the Quality of Experience (QoE) requirement, the business processing delay of the model does not meet the delay requirement, and the business processing accuracy of the model does not meet the accuracy requirement.

[0217] In some embodiments of the present application, the second information includes at least one of the following:

[0218] Hidden variable, where the hidden variable is a feature vector obtained by mapping the input information of the first model;

[0219] Prompt information;

[0220] Configuration information of the network-side device;

[0221] Scene type information;

[0222] The sample indication corresponding to the input information of the first model, or the sample indication in the first information used to determine the second information.

[0223] In some embodiments, the sample indication can be used to indicate which sample the second information is determined based on, and the sample indication can be a sample identifier. In this way, when the inference device obtains the first information and the second information, the samples corresponding to the first information and the second information can be determined through the sample indication. Further, when performing model inference, the first information and the second information corresponding to the same sample can be used, avoiding using the first information and the second information corresponding to different samples for model inference, which may affect the inference result.

[0224] In some embodiments, the latent variable can refer to the low-dimensional feature obtained by mapping the input information with physical meaning through an encoder, or the feature vector obtained by removing redundancy from the original input.

[0225] For example, as Figure 3 shown, the autoencoder includes an encoder and a decoder. The encoder maps the input image to a low-dimensional feature ( Figure 3 the latent variable in it), and the decoder restores the input image based on the latent variable. The latent variable contains the feature information of the input image.

[0226] In some embodiments, the prompt information can be used to prompt or guide the first model to infer an output that meets expectations, improving the inference performance of the model.

[0227] Exemplarily, the prompt information includes at least one of the following:

[0228] The thought chain related to the target task adapted to the first model;

[0229] The prompt text related to the target task;

[0230] The prompt file related to the target task;

[0231] The prompt code related to the target task;

[0232] The thought chain identifier related to the target task, used to identify a thought chain;

[0233] The prompt text identifier related to the target task, used to identify a piece of prompt text;

[0234] The prompt file identifier related to the target task, used to identify a prompt file;

[0235] The prompt code identifier related to the target task, used to identify a piece of prompt code;

[0236] The prompt file address related to the target task.

[0237] In some embodiments, the configuration information of the network-side device includes, but is not limited to, at least one of the following:

[0238] The number of transmit antennas of the network-side device;

[0239] The number of transmit beams of the network-side device;

[0240] The number of antenna ports of the network-side device;

[0241] The transmit power of the network-side device;

[0242] The antenna gain of the network-side device;

[0243] The 3dB bandwidth of the beam of the network-side device;

[0244] The spacing of the network-side device;

[0245] The frequency of the network-side device;

[0246] The system bandwidth;

[0247] The terminal distribution characteristics under the coverage of the network-side device.

[0248] In some embodiments, the scenario type information includes at least one of the following:

[0249] Line of sight (LOS), non-line of sight (NLOS), outdoor, indoor, urban microcell (UMi), urban macrocell (UMa), rural macrocell (RMa), rural microcell (RMi), high speed, low speed.

[0250] In some embodiments, the second information may be determined by the first device. For example, the first device determines it according to the first information and the knowledge information stored on the first device, or according to the first information and the invoked knowledge information. For example, the first device may use the invoked interface information to obtain knowledge information from the target device, and further determine the second information according to this knowledge information, or the first information and this knowledge information.

[0251] In other embodiments, the second information may also be determined by the target device that needs to be accessed using the invoked interface information. Optionally, after the target device determines the second information, it may directly send it to the third device, or it may also be sent to the third device through other devices, such as forwarding through the network development function, or sent to the third device through the first device.

[0252] The following describes the specific implementation of the third device obtaining the second information in combination with specific embodiments.

[0253] Embodiment 1: Knowledge information for assisting the first model in reasoning is stored on the first device.

[0254] In this case, after the first device receives the first information from the second device, the first device can determine the second information based on the first information and the knowledge information stored on the first device. Further, the first device can send the second information to the third device, so that the third device can use the second information as auxiliary information when using the first model for reasoning, which can improve the reasoning performance of the model.

[0255] Embodiment 2: Knowledge information for assisting the first model in reasoning is not stored on the first device, and the first information includes the call interface information of the knowledge information. In this case, as Figure 4 shown, it can be implemented through the following steps:

[0256] S211, the second device sends the first information to the first device. The first information includes the call interface information of the knowledge information, and the call interface information can be the call interface information of the knowledge information that the second device expects to use. At this time, the call interface information of the knowledge information includes the call interface name and the input data of the call interface.

[0257] S212, the first device sends the first information to the target device according to the call interface information of the knowledge information. The target device is the device that needs to be accessed using the call interface information;

[0258] S213, the target device determines the second information according to the first information;

[0259] S214, the target device sends the second information to the first device;

[0260] S215, the first device sends the second information to the third device.

[0261] It should be understood that the interaction of the first information and the second information between the first device and the target device can be forwarded by other nodes (such as network open functions), or the information can be directly interacted.

[0262] Embodiment 3: Knowledge information for assisting the first model in reasoning is not stored on the first device, and the first information does not include the call interface information of the knowledge information.

[0263] In this case, as an embodiment, as Figure 5 shown, it can be implemented through the following steps:

[0264] S221, The second device sends the first information to the first device, and the first information does not include the call interface information of the knowledge information.

[0265] S222, The first device determines the call interface information of the knowledge information according to the first information.

[0266] For example, the first device can determine the target call interface among multiple call interfaces according to the requirement information for the first model inference in the first information, so as to obtain information such as the call interface name of the target call interface, the input data format of the call interface, the output data format of the call interface, and the input data of the call interface.

[0267] S223, The first device sends the third information to the target device, and the third information includes the call interface information of the knowledge information. Wherein, the target device is the device to be accessed using the call interface information. The call interface information of the knowledge information includes the call interface name and the input data of the call interface. For example, the call interface name is determined according to the requirement information or auxiliary information or call interface information in the first information, and the input data of the call interface is determined based on the input information of the first model in the first information.

[0268] S224, The target device determines the second information according to the third information;

[0269] S225, The target device sends the second information to the first device;

[0270] S226, The first device sends the second information to the third device.

[0271] That is, in the case where the second device does not indicate the call interface information, the first device can independently determine the call interface information of the knowledge information, and further send the call interface information, or the call interface information and the first information to the target device, so that the target device can obtain the corresponding knowledge information based on the call interface information, and thus obtain the second information. Then the target device can send the determined second information to the first device, and the first device sends the second information to the third device.

[0272] It should be understood that the interaction of the third information and the second information between the first device and the target device can be forwarded by other nodes (such as network open functions), or the information can be directly interacted.

[0273] In this case, as another embodiment, as Figure 6 shown, it can be implemented through the following steps:

[0274] S231, The second device sends the first information to the first device, and the first information does not include the call interface information of the knowledge information.

[0275] S232. The first device determines the call interface information of the knowledge information according to the first information.

[0276] For example, the first device may determine a target call interface from multiple call interfaces according to the requirement information for the first model inference in the first information, so as to obtain information such as the call interface name of the target call interface, the input data format of the call interface, the output data format of the call interface, and the input data of the call interface.

[0277] S233. The first device sends fourth information to the third device. The fourth information includes the call interface information of the knowledge information, or the fourth information includes the first information and the call interface information of the knowledge information. Wherein, the call interface information of the knowledge information includes the call interface name of the target call interface, or the call interface name of the target call interface and the input data format of the call interface, or the call interface name of the target call interface, the input data format of the call interface, and the output data format of the call interface.

[0278] S234. The third device sends seventh information to the target device according to the fourth information. Wherein, the seventh information includes the call interface information of the knowledge information; wherein, the call interface information of the knowledge information includes the call interface name and the input data of the call interface. For example, the call interface name is included in the call interface information of the knowledge information in the fourth information, and the input data of the call interface is determined based on the input data format of the call interface of the knowledge information in the fourth information and the input information of the first model that the third device can provide. Or, the call interface name is determined according to the requirement information or auxiliary information or call interface information in the first information (included in the fourth information), and the input data of the call interface is determined based on the input information of the first model in the first information (included in the fourth information).

[0279] S235. The target device determines the second information according to the seventh information and sends the second information to the third device.

[0280] That is, in the case where the second device does not indicate the call interface information, the first device can independently determine the call interface information of the knowledge information, and further send the call interface information, or the call interface information and the first information to the third device. The third device independently requests the second information from the target device, and then the target device can send the determined second information to the third device.

[0281] It should be understood that the interaction of the seventh information and the second information between the third device and the target device can be forwarded by other nodes (such as network open functions), or the information can be directly interacted.

[0282] In some embodiments of the present application, the method 200 further includes at least one of the following steps:

[0283] The first device receives the fifth information sent by the target device;

[0284] The first device registers at least one call interface information according to the fifth information;

[0285] The first device sends the sixth information to the target device.

[0286] In some embodiments, the fifth information includes, but is not limited to, at least one of the following:

[0287] Call interface name, call interface parameter list, call interface description, usage example of the call interface, call interface instruction set; wherein, the call interface parameter list includes input parameters and output parameters of the call interface, and each parameter includes at least one of parameter name, parameter description, data type, and default value. The call interface description is used to explain the function of the call interface.

[0288] In some embodiments, the sixth information includes, but is not limited to, at least one of the following:

[0289] Whether the call interface is successfully registered, refusal to register the call interface, list of registered call interfaces, list of call interfaces with failed registration.

[0290] Further, when the first information does not include call interface information subsequently, the first device can determine the target call interface information from the registered call interface information according to the first information, and use the target call interface information to obtain the corresponding knowledge information.

[0291] In some embodiments of the present application, if the third device is a terminal, the method 200 further includes:

[0292] The third device sends at least one of the eighth information and the first capability information to the network-side device;

[0293] Wherein, the first model is deployed on the third device, the eighth information is used to indicate the auxiliary information required for the third device to perform inference based on the first model, and the first capability information is used to indicate the capability information of the third device to perform inference based on the first model.

[0294] In some embodiments, the eighth information or the first capability information can be used by the network-side device to determine the second information to be indicated to the terminal.

[0295] In some embodiments, the eighth information includes at least one of the following:

[0296] Indication of the network element where the knowledge information required for the third device to perform inference based on the first model is located;

[0297] Indication of knowledge information required for the third device to perform inference based on the first model;

[0298] Indication of the call interface for the knowledge information required for the third device to perform inference based on the first model;

[0299] Indication of the preprocessing method for the second information;

[0300] Indication of the format of the second information.

[0301] Optionally, the indication of the network element where the knowledge information required for the third device to perform inference based on the first model is located may be a network element identifier. For example, if there are multiple network elements for storing knowledge information and the knowledge required for the third device to perform model inference is stored on the first network element, the third device may indicate the network element identifier of the first network element to the network-side device.

[0302] In some embodiments, the indication of the knowledge information required for the third device to perform inference based on the first model may include indication information of the knowledge graph and / or indication information of the knowledge vector library required for the third device to perform inference based on the first model. For example, the indication information of the knowledge graph is one or more knowledge graph identifiers, or a knowledge graph group identifier, a knowledge graph file address, etc., or explicitly indicates one or more knowledge graphs. For example, the indication information of the knowledge vector library is used to indicate one or more knowledge vector library identifiers, or explicitly indicates one or more knowledge vector libraries.

[0303] Exemplarily, the indication information of the knowledge graph includes at least one of the following:

[0304] Knowledge graph identifier;

[0305] Knowledge graph file address.

[0306] In some embodiments, the indication of the call interface for the knowledge information required for the third device to perform inference based on the first model may explicitly indicate one or more call interface information, or may also implicitly indicate one or more call interface identifiers, and the call interface identifier corresponds to a set of call interface information.

[0307] In some embodiments, the indication of the format of the second information may be used to indicate the model input format of the second information. For example, the model input format may include the content included in the model input (such as including prompt information, including prompt information and call interface information, including prompt information and configuration information of the network-side device, etc.), as well as the dimension, quantization precision, size limit, etc. of the second information.

[0308] In some embodiments, the preprocessing method indication of the second information can be used to indicate the method for preprocessing the second information. For example, offsetting a numerical value, scaling a numerical value, applying a polynomial function transformation to a numerical value, performing a multi-dimensional mapping on a single numerical value, etc.

[0309] In some embodiments, the first capability information includes at least one of the following:

[0310] Indication of the knowledge information supported by the third device for inference based on the first model;

[0311] Indication of the call interface for the knowledge information supported by the third device for inference based on the first model;

[0312] The preprocessing method of the second information supported;

[0313] Format indication of the second information supported.

[0314] In some embodiments, the indication of the knowledge information supported by the third device for inference based on the first model may include indication information of the knowledge graph and / or indication information of the knowledge vector library supported by the third device for inference based on the first model. For example, the indication information of the knowledge graph is one or more knowledge graph identifiers, or a knowledge graph group identifier, a knowledge graph file address, etc., or explicitly indicates one or more knowledge graphs. For example, the indication information of the knowledge vector library is used to indicate one or more knowledge vector library identifiers, or explicitly indicates one or more knowledge vector libraries.

[0315] In some embodiments, the indication of the call interface for the knowledge information supported by the third device for inference based on the first model may explicitly indicate one or more call interface information, or may also implicitly indicate one or more call interface identifiers, and the call interface identifier corresponds to a set of call interface information.

[0316] In some embodiments, the format indication of the second information supported by the third device can be used to indicate the model input format of the second information supported by the third device. For example, the model input format may include what contents are included in the mode input (such as including prompt information, including prompt information and call interface information, including prompt information and configuration information of the network-side device, etc.), as well as the dimension, quantization precision, size limit, etc. of this information.

[0317] In some embodiments, the preprocessing method of the second information supported by the third device can be used to indicate the processing methods supported by the third device. These processing methods are the ways for the network-side device to preprocess the second information, such as offsetting a numerical value, scaling a numerical value, applying a polynomial function transformation to a numerical value, performing a multi-dimensional mapping on a single numerical value, etc.

[0318] In some embodiments, during the model registration (or model identification) process or the capability reporting process, the third device sends the eighth information and / or the first capability information to the network-side device.

[0319] It should be understood that the eighth information and the first capability information may be reported to the network-side device through one signaling, or may also be reported to the network-side device through different signaling. This application does not make any limitations in this regard.

[0320] Hereinafter, in combination with Figures 7 to 14 the specific embodiments shown, the model inference method in the communication system provided by this application will be described.

[0321] Embodiment 1: The inference device is a terminal

[0322] As Figure 7 shown, the first device is an access network device or a core network function, the second device and the third device are the same device and are terminals, and the first model is deployed on this terminal. The inference method may include the following steps:

[0323] Step 1: The terminal sends the first information to the access network device or the core network device, for requesting to obtain the second information.

[0324] Step 2: The access network device or the core network device sends the second information to the terminal according to the first information.

[0325] Thus, the inference device can learn the second information. Further, the inference device can use the second information to assist in the inference of the model, and can improve the inference performance of the model.

[0326] Embodiment 2: The inference device is an access network device, such as a base station.

[0327] For example, in Figure 8 the example of, the first device is a core network function, the second device and the third device are the same device and are access network devices, and the first model is deployed on this access network device.

[0328] As Figure 8 shown, the inference method may include the following steps:

[0329] Step 1: The access network device sends the first information to the core network device, for requesting to obtain the second information.

[0330] Step 2: The core network device sends the second information to the access network device according to the first information.

[0331] Thus, the inference device can learn the second information. Further, the inference device can use the second information to assist in the inference of the model, and can improve the inference performance of the model.

[0332] For another example, in Figure 9 the example, the first device is a core network function, the second device is a terminal, and the third device is an access network device. The first model is deployed on the access network device, that is, the inference device is the access network device, such as a base station.

[0333] As Figure 9 shown, the inference method may include the following steps:

[0334] Step 1: The terminal sends the first information to the core network function to request to obtain the second information.

[0335] Optionally, in Step 1, the terminal may also send the first information to the access network device.

[0336] Optionally, the first information sent by the terminal to the access network device may be different from the first information sent to the core network function.

[0337] For example, the first information sent by the terminal to the core network may include the auxiliary information for inference by the first model and the input information of the first model; while the first information sent by the terminal to the access network device only includes the input information of the first model. In addition, in order for the core network device to know who to send the second information to, the first information may further include a receiving device indication for indicating the receiving device of the second information, that is, the third device.

[0338] Step 2: The core network device sends the second information to the access network device according to the first information.

[0339] At this time, the input for the first model by the access network device includes two parts. One part comes from the first information sent by the terminal, and the other part comes from the second information sent by the core network. Therefore, the access network device needs to synchronize the two parts of information to generate the complete input information of the first model. Thus, it is prevented from combining the first information of sample 1 and the second information of sample 2 to generate an incorrect sample. Therefore, at this time, the first information and the second information need to include a sample indication, such as a sample identifier.

[0340] Optionally, when the core network device sends the second information to the access network device, it may also send the first information to the access network device together. Thus, the inference device can know the first information and the second information. Further, the inference device can use the first information and the second information to assist in model inference, which can improve the inference performance of the model.

[0341] For another example, in Figure 10 the example, the first device is a core network function, the second device is a terminal, and the third device is an access network device. The first model is deployed on the access network device, that is, the inference device is the access network device, such as a base station.

[0342] As Figure 10 shown, the inference method may include the following steps:

[0343] Step 1: The terminal sends a first piece of information to the core network device to request obtaining a second piece of information.

[0344] Step 2: The core network device sends the second piece of information to the terminal according to the first piece of information.

[0345] Step 3: The terminal sends the second piece of information to the access network device.

[0346] Optionally, the terminal may also send the first piece of information to the access network device together.

[0347] That is, in this example, the second piece of information is not directly sent from the knowledge storage device to the inference device, but is forwarded by the demand initiating device to the inference device.

[0348] Thus, the inference device can obtain the first piece of information and the second piece of information. Further, the inference device can use the first piece of information and the second piece of information to assist in model inference, which can improve the inference performance of the model.

[0349] For another example, in the Figure 11 example, the first device is the core network function, the second device is the terminal, and the third device is the access network device. The first model is deployed on the access network device, that is, the inference device is the access network device, such as a base station.

[0350] As Figure 11 shown, the inference method may include the following steps:

[0351] Step 1: The terminal sends a first piece of information to the core network device to request obtaining a second piece of information.

[0352] Among them, the first piece of information includes at least one of the following:

[0353] A receiving device indication for indicating the receiving device of the second piece of information.

[0354] Auxiliary information for inference of the first model;

[0355] Requirement information for inference of the first model;

[0356] The input information of the first model. Step 2: The core network device sends the first piece of information and the second piece of information to the access network device according to the first piece of information.

[0357] Thus, the inference device can obtain the first piece of information and the second piece of information. Further, the inference device can use the first piece of information and the second piece of information to assist in model inference, which can improve the inference performance of the model.

[0358] Embodiment 3: The inference device is a core network function.

[0359] For example, in Figure 12 the example, the first device is a core network function, the second device is a terminal or an access network device or a core network device, and the third device is a core network function.

[0360] As Figure 12 shown, the inference method may include the following steps:

[0361] Step 1: The second device sends first information to the first device to request to obtain second information.

[0362] Among them, the first information includes at least one of the following:

[0363] A receiving device indication for indicating the receiving device of the second information;

[0364] Auxiliary information for inference using the first model;

[0365] Requirement information for inference using the first model;

[0366] The input information of the first model.

[0367] Optionally, in Step 1, the second device may also send the first information to the third device to assist the third device in model inference. At this time, the input for the first model used by the third device includes two parts, one part from the first information sent by the second device and one part from the second information sent by the first device. Therefore, the third device needs to synchronize the two parts of information to generate the complete input information of the first model. Thus, it is prevented from combining the first information of sample 1 and the second information of sample 2 to generate an incorrect sample. Therefore, at this time, the first information and the second information need to include a sample indication, such as a sample identifier.

[0368] Step 2: The first device sends second information to the third device according to the first information.

[0369] Thus, the inference device can learn the first information and the second information. Further, the inference device can use the first information and the second information to assist in model inference, which can improve the inference performance of the model.

[0370] For another example, in Figure 13 the example, the first device is a core network function, the second device is a terminal or an access network device or a core network device, and the third device is a core network function.

[0371] As Figure 13 shown, the inference method may include the following steps:

[0372] Step 1: The second device sends the first information to the first device to request to obtain the second information.

[0373] Step 2: The first device sends the second information to the second device according to the first information.

[0374] Step 3: The second device sends the second information to the third device. Optionally, the first information can also be sent to the third device.

[0375] Thus, the inference device can learn the second information, or the first information and the second information. Further, the inference device can use the second information, or the first information and the second information to assist in the inference of the model, which can improve the inference performance of the model.

[0376] For another example, in Figure 14 the example of, the first device is a core network function, the second device is a terminal or an access network device or a core network device, and the third device is a core network function.

[0377] As Figure 14 shown, the inference method may include the following steps:

[0378] Step 1: The second device sends the first information to the first device to request to obtain the second information.

[0379] Wherein, the first information includes at least one of the following:

[0380] A receiving device indication for indicating the receiving device of the second information;

[0381] Auxiliary information for inference by the first model;

[0382] Requirement information for inference by the first model;

[0383] Input information of the first model.

[0384] Step 2: The first device sends the first information and the second information to the third device according to the first information.

[0385] Thus, the inference device can learn the second information, or the first information and the second information. Further, the inference device can use the second information, or the first information and the second information to assist in the inference of the model, which can improve the inference performance of the model.

[0386] It should be understood that in the embodiments of the present application, when the inference device learns the first information and the second information, the first information and the second information include a sample indication to prevent the inference device from using the first information and the second information of different samples for inference and affecting the inference result.

[0387] In summary, in the embodiments of the present application, by obtaining the second information from the device storing the knowledge information, when further using the model for inference, more information can be obtained as the input of the model, thereby improving the inference performance of the model.

[0388] As described above in conjunction with Figures 2 to 14 , the method embodiments of the present application have been described in detail. Below in conjunction with Figures 15 to 21 , the apparatus embodiments of the present application will be described in detail. It should be understood that the apparatus embodiments and the method embodiments correspond to each other, and similar descriptions can refer to the method embodiments.

[0389] In the model inference method in the communication system provided by the embodiments of the present application, the execution subject can be a communication device. In the embodiments of the present application, taking the communication device executing the model inference method in the communication system as an example, the communication device provided by the embodiments of the present application is described.

[0390] Figure 15 FIG. shows a schematic block diagram of a communication device 500 according to an embodiment of the present application. As Figure 15 shown, the device 500 includes:

[0391] A communication unit 510, configured to receive first information sent by a second device, where at least one of knowledge information and call interface information of the knowledge information is stored on the communication device 500; where the knowledge information includes at least one of the following: a knowledge graph, a knowledge vector library; and

[0392] Send second information to a third device, where a first model is deployed on the third device, and the second information is determined according to the first information and the knowledge information stored or called on the communication device 500;

[0393] Wherein, the first information includes at least one of the following:

[0394] Auxiliary information for inference by the first model;

[0395] Requirement information for inference by the first model;

[0396] Input information of the first model;

[0397] Call interface information of the knowledge information;

[0398] Receiver indication of the second information;

[0399] Sample indication corresponding to the input information of the first model.

[0400] In some embodiments, the auxiliary information includes at least one of the following:

[0401] Identification information of the artificial intelligence (AI) model, function indication of the AI model, characteristic indication of the AI model, target knowledge graph indication, cell identification, bandwidth part (BWP) identification.

[0402] In some embodiments, the requirement information includes at least one of the following:

[0403] Target task indication that the AI model needs to adapt to;

[0404] Quality of Experience (QoE) requirement of the AI model for services;

[0405] Service processing delay of the AI model;

[0406] Service processing accuracy of the AI model;

[0407] Service computing volume of the AI model;

[0408] Data processing scale of the AI model.

[0409] In some embodiments, the second information includes at least one of the following:

[0410] Hidden variable, which is a feature vector obtained by mapping the input information of the first model;

[0411] Hint information;

[0412] Configuration information of the network-side device;

[0413] Scenario type information;

[0414] Sample indication corresponding to the input information of the first model.

[0415] In some embodiments, the hint information includes at least one of the following:

[0416] Thought chain related to the target task adapted to the first model;

[0417] Hint text related to the target task;

[0418] Hint file related to the target task;

[0419] Hint code related to the target task;

[0420] Thought chain identifier related to the target task;

[0421] Hint text identifier related to the target task;

[0422] Hint file identifier related to the target task;

[0423] Hint code identifier related to the target task;

[0424] The address of the hint file related to the target task.

[0425] In some embodiments, the configuration information of the network-side device includes at least one of the following:

[0426] The number of transmitting antennas of the network-side device;

[0427] The number of transmitting beams of the network-side device;

[0428] The number of antenna ports of the network-side device;

[0429] The transmitting power of the network-side device;

[0430] The antenna gain of the network-side device;

[0431] The 3dB bandwidth of the beam of the network-side device;

[0432] The spacing of the network-side device;

[0433] The frequency of the network-side device;

[0434] The system bandwidth;

[0435] The terminal distribution characteristics under the coverage of the network-side device.

[0436] In some embodiments, the scenario type information includes at least one of the following:

[0437] Line of Sight (LOS), Non-Line of Sight (NLOS), indoor, outdoor, urban microcell, urban macrocell, suburban microcell, high speed, low speed.

[0438] In some embodiments, the communication device 500 further includes:

[0439] A processing unit, configured to determine the second information according to the first information and the knowledge information stored on the communication device 500.

[0440] In some embodiments, the communication unit 510 is further configured to:

[0441] Receive the second information from the target device, where the target device is the device accessed using the call interface information.

[0442] In some embodiments, the first information includes the call interface information of the knowledge information, and the communication device 500 further includes: a processing unit, configured to determine the second information according to the first information and the knowledge information stored on the communication device 500.

[0443] In some embodiments, the communication unit 510 is further configured to: send the first information to a target device according to the call interface information of the knowledge information, where the target device is the device accessed using the call interface information; and receive the second information from the target device.

[0444] In some embodiments, the first information does not include the call interface information of the knowledge information, and the communication device 500 further includes: a processing unit, configured to determine the call interface information of the knowledge information according to the first information;

[0445] The communication device 500 is further configured to: send third information to a target device, where the target device is the device accessed using the call interface information, and the third information includes the call interface information of the knowledge information; and receive the second information from the target device.

[0446] In some embodiments, the first information does not include the call interface information of the knowledge information, and the communication device 500 further includes: a processing unit, configured to determine the call interface information of the knowledge information according to the first information;

[0447] The communication unit 510 is further configured to: send fourth information to the third device, where the fourth information includes the call interface information of the knowledge information, or the fourth information includes the first information and the call interface information of the knowledge information, and the fourth information is used for the third device to request the second information from the target device, and the target device is the device accessed using the call interface information.

[0448] In some embodiments, the communication unit 510 is further configured to:

[0449] receive fifth information from the target device;

[0450] The communication device 500 further includes a processing unit, configured to register at least one call interface information according to the fifth information;

[0451] The communication unit 510 is further configured to: send sixth information to the target device;

[0452] where the fifth information includes at least one of the following: call interface name, call interface parameter list, call interface description, usage example of the call interface, call interface instruction set;

[0453] The sixth information includes at least one of the following:

[0454] whether the call interface is successfully registered, rejection of call interface registration, list of registered call interfaces, list of failed call interface registrations.

[0455] In some embodiments, the communication unit 510 is further configured to:

[0456] Send the first information to the third device.

[0457] In some embodiments, the communication device 500 is an access network device or a core network function, the second device and the third device are terminals; or

[0458] The communication device 500 is a core network function, the second device is a terminal or an access network device, and the third device is an access network device; or

[0459] The communication device 500 is a core network function, the second device is a terminal or an access network device or a core network function, and the third device is a core network function.

[0460] In some embodiments, the communication device 500 is an access network device, the second device is a terminal, and the first information is carried in at least one of the following signaling: layer 1 signaling, layer 2 signaling, layer 3 signaling, AI layer signaling, data plane signaling; or

[0461] The communication device 500 is a core network function, the third device is a terminal, and the first information is carried in at least one of the following signaling: non-access stratum (NAS) signaling, AI layer signaling, data plane signaling.

[0462] In some embodiments, the call interface information includes at least one of the following:

[0463] Call interface name, input data format of the call interface, output data format of the call interface, input data of the call interface.

[0464] Optionally, in some embodiments, the above communication unit may be a communication interface or a transceiver, or an input / output interface of a communication chip or a system-on-chip.

[0465] It should be understood that the communication device 500 according to the embodiments of the present application may correspond to the first device in the method embodiments of the present application, and the above and other operations and / or functions of each unit in the communication device 500 are respectively for implementing Figures 2 to 14 the corresponding processes of the first device in the method embodiments shown, and achieve the same technical effects. To avoid repetition, they are not described herein again.

[0466] Figure 16 Fig. shows a schematic block diagram of a communication device 600 according to an embodiment of the present application. As Figure 16 shown, the device 600 includes:

[0467] A communication unit 610, configured to send first information to a first device, where at least one of knowledge information and call interface information of the knowledge information is stored on the first device, and the knowledge information includes at least one of the following: a knowledge graph, a knowledge vector library;

[0468] Wherein, the first information includes at least one of the following:

[0469] Auxiliary information for a first model to perform inference;

[0470] Requirement information for a first model to perform inference;

[0471] Input information of the first model;

[0472] Call interface information of the knowledge information;

[0473] An indication of the receiving device of the second information;

[0474] An indication of a sample corresponding to the input information of the first model.

[0475] In some embodiments, the auxiliary information includes at least one of the following:

[0476] Identification information of an artificial intelligence (AI) model, function indication of the AI model, characteristic indication of the AI model, target knowledge graph indication, cell identification, bandwidth part (BWP) identification.

[0477] In some embodiments, the requirement information includes at least one of the following:

[0478] Target task indication that an AI model needs to adapt to;

[0479] Quality of Experience (QoE) requirement of an AI model for services;

[0480] Service processing latency of an AI model;

[0481] Service processing accuracy of an AI model;

[0482] Service computing volume of an AI model;

[0483] Data processing scale of an AI model.

[0484] In some embodiments, the second information includes at least one of the following:

[0485] A hidden variable, which is a feature vector obtained by mapping the input information of the first model;

[0486] Hint information;

[0487] Configuration information of a network-side device;

[0488] Scene type information;

[0489] Sample indication corresponding to the input information of the first model.

[0490] In some embodiments, the prompt information includes at least one of the following:

[0491] Thought chain related to the target task adapted to the first model;

[0492] Prompt text related to the target task;

[0493] Prompt file related to the target task;

[0494] Prompt code related to the target task;

[0495] Thought chain identifier related to the target task;

[0496] Prompt text identifier related to the target task;

[0497] Prompt file identifier related to the target task;

[0498] Prompt code identifier related to the target task;

[0499] Prompt file address related to the target task.

[0500] In some embodiments, the configuration information of the network-side device includes at least one of the following:

[0501] Number of transmitting antennas of the network-side device;

[0502] Number of transmitting beams of the network-side device;

[0503] Number of antenna ports of the network-side device;

[0504] Transmitting power of the network-side device;

[0505] Antenna gain of the network-side device;

[0506] 3dB bandwidth of the beam of the network-side device;

[0507] Spacing of the network-side device;

[0508] Frequency of the network-side device;

[0509] System bandwidth;

[0510] Terminal distribution characteristics covered by the network-side device.

[0511] In some embodiments, the scene type information includes at least one of the following:

[0512] Line-of-sight (LOS), non-line-of-sight (NLOS), indoor, outdoor, urban microcell, urban macrocell, suburban microcell, high speed, low speed.

[0513] In some embodiments, the call interface information includes at least one of the following:

[0514] Call interface name, input data format of the call interface, output data format of the call interface, input data of the call interface.

[0515] In some embodiments, the communication device 600 is a terminal, and the communication unit 610 is specifically configured to:

[0516] When at least one of the following events occurs, send the first information to the first device:

[0517] Receiving a first indication for indicating activation or switching to the first model;

[0518] The communication device 600 determines to activate or switch to the first model;

[0519] The communication device 600 determines that the first model does not meet the requirements;

[0520] Receiving a second indication for indicating that the first model does not meet the requirements.

[0521] In some embodiments, the communication unit 610 is further configured to:

[0522] Send the first information to a third device on which the first model is deployed.

[0523] Optionally, in some embodiments, the above communication unit may be a communication interface or a transceiver, or an input / output interface of a communication chip or a system-on-chip.

[0524] It should be understood that the communication device 600 according to the embodiments of the present application may correspond to the second device in the method embodiments of the present application, and the above and other operations and / or functions of the respective units in the communication device 600 are respectively for implementing Figures 2 to 14 the corresponding processes of the second device in the method embodiments shown, and achieve the same technical effects. To avoid repetition, they are not described herein again.

[0525] Figure 17 Shows a schematic block diagram of a communication device 700 according to an embodiment of the present application. As Figure 17 shown, the device 700 includes:

[0526] A communication unit 710, configured to obtain second information from a first device or a target device, where a first model is deployed on the communication device 700, and at least one of knowledge information and call interface information of the knowledge information is stored on the first device or the target device, where the target device is a device accessed using the call interface information, and the knowledge information includes at least one of the following: a knowledge graph, a knowledge vector library;

[0527] Wherein, the second information is determined according to first information and the knowledge information stored or called on the first device;

[0528] Wherein, the first information includes at least one of the following:

[0529] Auxiliary information for inference by the first model;

[0530] Requirement information for inference by the first model;

[0531] Input information of the first model;

[0532] Call interface information of the knowledge information;

[0533] Receiver indication of the second information;

[0534] Sample indication corresponding to the input information of the first model.

[0535] In some embodiments, the auxiliary information includes at least one of the following:

[0536] Identification information of an artificial intelligence (AI) model, function indication of the AI model, characteristic indication of the AI model, target knowledge graph indication, cell identification, bandwidth part (BWP) identification.

[0537] In some embodiments, the requirement information includes at least one of the following:

[0538] Target task indication to be adapted by the AI model;

[0539] Quality of experience (QoE) requirement of the AI model for the service;

[0540] Service processing delay of the AI model;

[0541] Service processing accuracy of the AI model;

[0542] Service computing amount of the AI model;

[0543] Data processing scale of the AI model.

[0544] In some embodiments, the second information includes at least one of the following:

[0545] Hidden variable, which is a feature vector obtained by mapping the input information of the first model;

[0546] Prompt information;

[0547] Configuration information of the network-side device;

[0548] Scene type information;

[0549] Sample indication corresponding to the input information of the first model.

[0550] In some embodiments, the prompt information includes at least one of the following:

[0551] Thought chain related to the target task adapted to the first model;

[0552] Prompt text related to the target task;

[0553] Prompt file related to the target task;

[0554] Prompt code related to the target task;

[0555] Thought chain identifier related to the target task;

[0556] Prompt text identifier related to the target task;

[0557] Prompt file identifier related to the target task;

[0558] Prompt code identifier related to the target task;

[0559] Prompt file address related to the target task.

[0560] In some embodiments, the configuration information of the network-side device includes at least one of the following:

[0561] Number of transmit antennas of the network-side device;

[0562] Number of transmit beams of the network-side device;

[0563] Number of antenna ports of the network-side device;

[0564] Transmit power of the network-side device;

[0565] Antenna gain of the network-side device;

[0566] 3dB bandwidth of the beam of the network-side device;

[0567] Spacing of the network-side device;

[0568] Frequency of the network-side device;

[0569] System bandwidth;

[0570] Terminal distribution characteristics under the coverage of network-side devices.

[0571] In some embodiments, the scenario type information includes at least one of the following:

[0572] Line of Sight (LOS), Non-Line of Sight (NLOS), indoor, outdoor, urban microcell, urban macrocell, suburban microcell, high speed, low speed.

[0573] In some embodiments, the communication unit 710 is further configured to:

[0574] Receive fourth information from the first device, where the fourth information includes the call interface information of the knowledge information, or the fourth information includes the first information and the call interface information of the knowledge information;

[0575] Send seventh information to the target device, where the seventh information includes the call interface information of the knowledge information;

[0576] Receive the second information from the target device.

[0577] In some embodiments, the communication device 700 is a terminal, and the communication unit 710 is further configured to:

[0578] Send at least one of eighth information and first capability information to the network-side device, where the first model is deployed on the communication device 700, the eighth information is used to indicate the auxiliary information required for the communication device 700 to perform inference based on the first model, and the first capability information is used to indicate the capability information of the communication device 700 to perform inference based on the first model.

[0579] In some embodiments, the eighth information includes at least one of the following:

[0580] The network element indication where the knowledge information required for the communication device 700 to perform inference based on the first model is located;

[0581] The knowledge information indication required for the communication device 700 to perform inference based on the first model;

[0582] The call interface indication of the knowledge information required for the communication device 700 to perform inference based on the first model;

[0583] The format indication of the second information.

[0584] In some embodiments, the first capability information includes at least one of the following:

[0585] The knowledge information indication supported by the communication device 700 for performing inference based on the first model;

[0586] The call interface indication of the knowledge information supported by the communication device 700 for inference based on the first model;

[0587] The format indication of the second information supported.

[0588] In some embodiments, the communication unit 710 is further configured to:

[0589] During the model registration process or the capability reporting process, send at least one of the eighth information and the first capability information to the network-side device.

[0590] In some embodiments, the communication unit 710 is further configured to:

[0591] Receive the first information from a second device, and the first model is deployed on the communication device 700.

[0592] In some embodiments, the communication device 700 further includes:

[0593] A processing unit, configured to perform inference based on the second information and the input information of the first model by using the first model.

[0594] In some embodiments, the call interface information includes at least one of the following:

[0595] The call interface name, the input data format of the call interface, the output data format of the call interface, the input data of the call interface.

[0596] Optionally, in some embodiments, the above communication unit may be a communication interface or a transceiver, or an input / output interface of a communication chip or a system-on-chip.

[0597] It should be understood that the communication device 700 according to the embodiments of the present application may correspond to the third device in the method embodiments of the present application, and the above and other operations and / or functions of each unit in the communication device 700 are respectively for implementing Figures 2 to 14 the corresponding processes of the third device in the method embodiments shown, and achieve the same technical effects. To avoid repetition, details are not described here again.

[0598] In some embodiments, the devices 500, 600, and 700 in the embodiments of the present application may be electronic devices, such as electronic devices with an operating system, or components in an electronic device, such as an integrated circuit or a chip. The electronic device may be a terminal or other devices other than a terminal. Exemplarily, the terminal may include, but is not limited to, the types of the terminal 11 listed above, and other devices may be a server, a Network Attached Storage (NAS), etc. The embodiments of the present application do not make specific limitations.

[0599] As Figure 18 shown, an embodiment of the present application further provides a communication device 1000, including a processor 1001 and a memory 1002. A program or instruction that can run on the processor 1001 is stored on the memory 1002. For example, when the communication device 1000 is a first device, when the program or instruction is executed by the processor 1001, it implements the steps executed by the first device in the above-mentioned inference method embodiment and can achieve the same technical effect. For example, when the communication device 1000 is a second device, when the program or instruction is executed by the processor 1001, it implements the steps executed by the second device in the above-mentioned inference method embodiment and can achieve the same technical effect. For example, when the communication device 1000 is a third device, when the program or instruction is executed by the processor 1001, it implements the steps executed by the third device in the above-mentioned inference method embodiment and can achieve the same technical effect. To avoid repetition, details are not described herein again.

[0600] An embodiment of the present application further provides a terminal, including a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run a program or instruction to implement the steps in the method embodiment as Figures 2 to 14 shown. This terminal embodiment corresponds to the above-mentioned terminal-side method embodiment. Each implementation process and implementation manner of the above method embodiment can be applied to this terminal embodiment and can achieve the same technical effect. Specifically, Figure 19 FIG. is a schematic hardware structure diagram of a terminal for implementing an embodiment of the present application.

[0601] The terminal 1100 includes, but is not limited to: at least some components such as a radio frequency unit 1101, a network module 1102, an audio output unit 1103, an input unit 1104, a sensor 1105, a display unit 1106, a user input unit 1107, an interface unit 1108, a memory 1109, and a processor 1110.

[0602] Those skilled in the art can understand that the terminal 1100 may further include a power supply (such as a battery) for supplying power to each component. The power supply can be logically connected to the processor 1110 through a power management system, so as to implement functions such as management of charging, discharging, and power consumption management through the power management system. Figure 19 The terminal structure shown in FIG. does not limit the terminal. The terminal may include more or fewer components than shown in the figure, or combine some components, or have different component arrangements, which are not described herein again.

[0603] It should be understood that in the embodiments of the present application, the input unit 1104 may include a Graphics Processing Unit (GPU) 11041 and a microphone 11042. The graphics processor 11041 processes the image data of static pictures or videos obtained by an image capturing device (such as a camera) in the video capture mode or the image capture mode. The display unit 1106 may include a display panel 11061, and the display panel 11061 may be configured in the form of, for example, a liquid crystal display, an organic light emitting diode, etc. The user input unit 1107 includes at least one of a touch panel 11071 and other input devices 11072. The touch panel 11071 is also referred to as a touch screen. The touch panel 11071 may include two parts: a touch detection device and a touch controller. The other input devices 11072 may include, but are not limited to, a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, which will not be elaborated herein.

[0604] In the embodiments of the present application, after receiving downlink data from a network-side device, the radio frequency unit 1101 may transmit it to the processor 1110 for processing; in addition, the radio frequency unit 1101 may send uplink data to the network-side device. Generally, the radio frequency unit 1101 includes, but is not limited to, an antenna, an amplifier, a transceiver, a coupler, a low-noise amplifier, a duplexer, etc.

[0605] The memory 1109 can be used to store software programs or instructions as well as various data. The memory 1109 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data. Among them, the first storage area may store an operating system, application programs or instructions required for at least one function (such as a sound playback function, an image playback function, etc.). In addition, the memory 1109 may include volatile memory or non-volatile memory. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDR SDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synchronous link dynamic random access memory (SLDRAM), and a direct rambus random access memory (DRRAM). The memory 1109 in the embodiments of the present application includes but is not limited to these and any other suitable types of memory.

[0606] The processor 1110 may include one or more processing units; optionally, the processor 1110 integrates an application processor and a modem processor. Among them, the application processor mainly processes operations related to the operating system, user interface, and application programs, etc., and the modem processor mainly processes wireless communication signals, such as a baseband processor. It can be understood that the above-mentioned modem processor may not be integrated into the processor 1110 either.

[0607] It can be understood that the implementation processes of the various implementation manners mentioned in this embodiment can refer to Figures 2 to 14 the relevant descriptions of the method embodiments shown, and achieve the same or corresponding technical effects. To avoid repetition, they will not be elaborated here.

[0608] The embodiments of the present application further provide a network-side device, including a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement as Figures 2 to 14The steps of the method embodiments shown. This network-side device embodiment corresponds to the above access network device side or core network function side method embodiments. Each implementation process and realization manner of the above method embodiments can be applied to this network-side device embodiment and can achieve the same technical effects.

[0609] Specifically, an embodiment of the present application also provides a network-side device. As Figure 20 shown, the network-side device 1200 includes: an antenna 1201, a radio frequency device 1202, a baseband device 1203, a processor 1204, and a memory 1205. The antenna 1201 is connected to the radio frequency device 1202. In the uplink direction, the radio frequency device 1202 receives information through the antenna 1201 and sends the received information to the baseband device 1203 for processing. In the downlink direction, the baseband device 1203 processes the information to be sent and sends it to the radio frequency device 1202. After processing the received information, the radio frequency device 1202 sends it out through the antenna 1201.

[0610] The method executed by the network-side device in the above embodiments can be implemented in the baseband device 1203, and the baseband device 1203 includes a baseband processor.

[0611] The baseband device 1203 may include, for example, at least one baseband board, and a plurality of chips are arranged on the baseband board. As Figure 20 shown, one of the chips is, for example, a baseband processor, which is connected to the memory 1205 through a bus interface to call the program in the memory 1205 and execute the network device operations shown in the above method embodiments.

[0612] The network-side device may further include a network interface 1206, and this interface is, for example, a Common Public Radio Interface (CPRI).

[0613] Specifically, the network-side device 1200 in the embodiment of the present application further includes: instructions or programs stored on the memory 1205 and executable on the processor 1204. The processor 1204 calls the instructions or programs in the memory 1205 to execute Figures 15 to 17 the methods executed by the modules shown and achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0614] Specifically, an embodiment of the present application also provides a network-side device. As Figure 21 shown, the network-side device 1300 includes: a processor 1301, a network interface 1302, and a memory 1303. Among them, the network interface 1302 is, for example, a Common Public Radio Interface (CPRI).

[0615] Specifically, the network-side device 1300 in the embodiments of the present application further includes: instructions or programs stored in the memory 1303 and executable on the processor 1301. The processor 1301 invokes the instructions or programs in the memory 1303 to execute Figures 15 to 17 the methods executed by the modules shown, and achieves the same technical effects. To avoid repetition, they will not be elaborated here.

[0616] The embodiments of the present application further provide a readable storage medium, on which programs or instructions are stored. When the programs or instructions are executed by a processor, the various processes in the embodiments of the model inference method in the above communication system are implemented, and the same technical effects can be achieved. To avoid repetition, they will not be elaborated here.

[0617] Wherein, the processor is the processor in the communication device, communication equipment, terminal or network-side device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory ROM, random access memory RAM, magnetic disks or optical discs, etc. In some examples, the readable storage medium may be a non-transitory readable storage medium.

[0618] The embodiments of the present application further provide a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes in the embodiments of the model inference method in the above communication system, and the same technical effects can be achieved. To avoid repetition, they will not be elaborated here.

[0619] It should be understood that the chip mentioned in the embodiments of the present application may also be referred to as a system-on-chip, system chip, chip system or system-on-chip, etc.

[0620] The embodiments of the present application further provide a computer program / program product, which is stored in a storage medium. The computer program / program product is executed by at least one processor to implement the various processes in the embodiments of the model inference method in the above communication system, and the same technical effects can be achieved. To avoid repetition, they will not be elaborated here.

[0621] The embodiments of the present application further provide a communication system, including: a first device, a second device and a third device. The first device can be used to execute the steps executed by the first device in the model inference method in the communication system as described above. The second device can be used to execute the steps executed by the second device in the model inference method in the communication system as described above. The third device can be used to execute the steps executed by the third device in the model inference method in the communication system as described above.

[0622] It should be noted that in this text, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements not only includes those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising a..." does not exclude the presence of additional identical elements in the process, method, article or device comprising such element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the reverse order according to the functions involved. For example, the described methods may be performed in an order different from that described, and various steps may be added, omitted or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0623] Through the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of a computer software product plus a necessary general hardware platform, and of course, can also be implemented by hardware. The computer software product is stored in a storage medium (such as ROM, RAM, magnetic disk, optical disk, etc.) and includes several instructions for causing a terminal or a network-side device to execute the methods described in various embodiments of the present application.

[0624] The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make many forms of embodiments without departing from the purpose of the present application and the scope protected by the claims. These embodiments are all within the protection scope of the present application.

Claims

1. A model reasoning method in a communication system, characterized in that: include: A first device receives first information sent by a second device, wherein at least one of knowledge information and calling interface information of knowledge information is stored on the first device; wherein the knowledge information includes at least one of the following: a knowledge graph and a knowledge vector library; The first device sends second information to a third device, the first model is deployed on the third device, and the second information is determined according to the first information and knowledge information stored or called on the first device; The first information includes at least one of the following: Auxiliary information used for reasoning with the first model; Requirement information for performing reasoning on the first model; input information of the first model; Calling interface information of knowledge information; an indication of a receiving device of the second information; An indication of samples corresponding to the input information of the first model.

2. The method according to claim 1, characterized in that The auxiliary information includes at least one of the following: Artificial intelligence AI model identification information, AI model function indication, AI model feature indication, target knowledge graph indication, cell identification, bandwidth part BWP identification.

3. The method according to claim 1 or 2, characterized in that: The demand information includes at least one of the following: Indication of target tasks that the AI ​​model needs to adapt to; Service quality of experience (QoE) requirements for AI models; Business processing latency of AI models; The accuracy of AI model’s business processing; The business computing volume of AI models; The data processing scale of AI models.

4. The method according to any one of claims 1 to 3, characterized in that The second information includes at least one of the following: Hidden variables, where the hidden variables are feature vectors after mapping the input information of the first model; Prompt information; Configuration information of network-side devices; Scene type information; A sample indication corresponding to the input information of the first model; The prompt information includes at least one of the following: A chain of thoughts related to the target task adapted by the first model; Prompt text related to the target task; Prompt files related to the target task; a prompt code associated with the target task; Identification of thought chains related to the target task; Prompt text labels related to the target task; The reminder file identifier related to the target task; A prompt code identifier related to the target task; The address of the prompt file related to the target task; The configuration information of the network side device includes at least one of the following: The number of transmitting antennas of the network-side equipment; The number of transmit beams of the network-side device; Number of antenna ports on network-side equipment; The transmit power of the network-side device; Antenna gain of network-side equipment; The beam 3dB bandwidth of the network-side equipment; The distance between network-side devices; Frequency of network-side equipment; System bandwidth; Terminal distribution characteristics under the coverage of network-side equipment; The scene type information includes at least one of the following: Line-of-sight LOS, non-line-of-sight NLOS, indoor, outdoor, urban microcell, urban macrocell, suburban microcell, high speed, low speed.

5. The method according to any one of claims 1 to 4, characterized in that The method further comprises: The first device determines the second information according to the first information and knowledge information stored on the first device; or, The first device receives the second information from a target device, where the target device is a device accessed using the calling interface information.

6. The method according to claim 5, characterized in that The method further comprises: The first device receives fifth information sent by the target device; The first device registers at least one calling interface information according to the fifth information; The first device sends sixth information to the target device; The fifth information includes at least one of the following: a calling interface name, a calling interface parameter list, a calling interface description, a calling interface usage example, and a calling interface instruction set; The sixth information includes at least one of the following: Whether the calling interface is successfully registered, whether the calling interface is refused to be registered, the list of registered calling interfaces, and the list of failed registration calling interfaces.

7. The method according to any one of claims 1 to 6, characterized in that The method further comprises: The first device sends the first information to the third device.

8. The method according to any one of claims 1 to 7, characterized in that The calling interface information includes at least one of the following: Calling interface name, calling interface input data format, calling interface output data format, calling interface input data.

9. A model reasoning method in a communication system, characterized in that: include: The second device sends the first information to the first device, wherein the first device stores at least one of knowledge information and calling interface information of the knowledge information, wherein the knowledge information includes at least one of the following: a knowledge graph and a knowledge vector library; The first information includes at least one of the following: Auxiliary information used for reasoning by the first model; Requirement information for the first model to perform reasoning; Input information of the first model; Calling interface information of knowledge information; a receiving device indication of the second information; An indication of samples corresponding to the input information of the first model.

10. The method according to claim 9, characterized in that The second device is a terminal, and the second device sends first information to the first device, including: When at least one of the following events occurs, sending the first information to the first device: receiving a first indication, wherein the first indication is used to indicate activation or switching to the first model; The second device determines to activate or switch to the first model; The second device determines that the first model does not meet the requirements; A second indication is received, where the second indication is used to indicate that the first model does not meet the requirements.

11. The method according to claim 9 or 10, characterized in that: The method further comprises: The second device sends the first information to a third device, and the first model is deployed on the third device.

12. A model reasoning method in a communication system, characterized in that: include: The third device obtains the second information from the first device or the target device, wherein the first model is deployed on the third device, and at least one of knowledge information and calling interface information of knowledge information is stored on the first device or the target device, wherein the target device is a device accessed using the calling interface information, and the knowledge information includes at least one of the following: a knowledge graph and a knowledge vector library; Wherein, the second information is determined according to the first information and the knowledge information stored or called on the first device; The first information includes at least one of the following: Auxiliary information used for reasoning with the first model; Requirement information for performing reasoning on the first model; input information of the first model; Calling interface information of knowledge information; an indication of a receiving device of the second information; An indication of samples corresponding to the input information of the first model.

13. The method according to claim 12, characterized in that The third device obtains second information from the target device, including: The third device receives fourth information sent by the first device, where the fourth information includes calling interface information of the knowledge information, or the fourth information includes the first information and calling interface information of the knowledge information; The third device sends seventh information to the target device, where the seventh information includes calling interface information of the knowledge information; The third device receives the second information sent by the target device.

14. The method according to claim 12 or 13, characterized in that The third device is a terminal, and the method further includes: The third device sends at least one of the eighth information and the first capability information to the network side device, wherein the first model is deployed on the third device, the eighth information is used to indicate the auxiliary information required for the third device to perform reasoning based on the first model, and the first capability information is used to indicate the capability information of the third device to perform reasoning based on the first model.

15. The method according to claim 14, characterized in that The eighth information includes at least one of the following: The third device indicates a network element where knowledge information required for reasoning based on the first model is located; The third device indicates the knowledge information required for reasoning based on the first model; A calling interface indication of knowledge information required for the third device to perform reasoning based on the first model; an indication of a preprocessing method of the second information; The format of the second information is indicated.

16. The method according to claim 14 or 15, characterized in that The first capability information includes at least one of the following: The third device indicates the knowledge information supported by reasoning based on the first model; A calling interface indication of knowledge information supported by reasoning performed by a third device based on the first model; an indication of a supported preprocessing method of the second information; An indication of a format of the second information supported.

17. The method according to any one of claims 14 to 16, characterized in that: The third device sends at least one of the eighth information and the first capability information to the network side device, including: The third device sends at least one of the eighth information and the first capability information to the network side device during the model registration process or the capability reporting process.

18. The method according to any one of claims 12 to 17, characterized in that: The method further comprises: The third device receives the first information sent by the second device, and deploys the first model on the third device.

19. The method according to any one of claims 12 to 18, characterized in that: The method further comprises: The third device uses the first model to perform inference based on the second information and input information of the first model.

20. The method according to any one of claims 12 to 19, characterized in that The calling interface information includes at least one of the following: Calling interface name, calling interface input data format, calling interface output data format, calling interface input data.

21. A communication device, characterized in that: include: A communication unit, configured to receive first information sent by a second device, wherein the communication device stores at least one of knowledge information and calling interface information of the knowledge information; wherein the knowledge information includes at least one of the following: a knowledge graph, a knowledge vector library; and Sending second information to a third device, on which the first model is deployed, wherein the second information is determined according to the first information and knowledge information stored or called on the communication device; The first information includes at least one of the following: Auxiliary information used for reasoning with the first model; Requirement information for performing reasoning on the first model; input information of the first model; Calling interface information of knowledge information; an indication of a receiving device of the second information; An indication of samples corresponding to the input information of the first model.

22. A communication device, characterized in that: include: A communication unit, configured to send first information to a first device, wherein the first device stores at least one of knowledge information and calling interface information of the knowledge information, wherein the knowledge information includes at least one of the following: a knowledge graph and a knowledge vector library; The first information includes at least one of the following: Auxiliary information used for reasoning by the first model; Requirement information for the first model to perform reasoning; Input information of the first model; Calling interface information of knowledge information; a receiving device indication of the second information; An indication of samples corresponding to the input information of the first model.

23. A communication device, characterized in that: include: A communication unit, configured to obtain second information from a first device or a target device, wherein a first model is deployed on the communication apparatus 700, and at least one of knowledge information and calling interface information of knowledge information is stored on the first device or the target device, wherein the target device is a device accessed using the calling interface information, and the knowledge information includes at least one of the following: a knowledge graph and a knowledge vector library; Wherein, the second information is determined according to the first information and the knowledge information stored or called on the first device; The first information includes at least one of the following: Auxiliary information used for reasoning with the first model; Requirement information for performing reasoning on the first model; input information of the first model; Calling interface information of knowledge information; an indication of a receiving device of the second information; An indication of samples corresponding to the input information of the first model.

24. A communication device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the steps in the method according to any one of claims 1 to 8, or the steps in the method according to any one of claims 9 to 11, or the steps in the method according to any one of claims 12 to 20 are implemented.

25. A readable storage medium, characterized in that: The readable storage medium stores programs or instructions, and when the programs or instructions are executed by the processor, the steps in the method as described in any one of claims 1 to 8, or the steps in the method as described in any one of claims 9 to 11, or the steps in the method as described in any one of claims 12 to 20 are implemented.

Images