Retrieval augmentation method in communication system and related apparatus

Abstract

A retrieval augmentation method in a communication system and a related apparatus. In the method, a server receives a first task request from a terminal device, and then the server sends a retrieval augmentation request to a network device. The network device sends first information to the server, the first information being used for indicating a retrieval result corresponding to the retrieval augmentation request. The server acquires a first inference result of the first task request on the basis of the retrieval result and an AI model. In the present application, a network device provides an RAG retrieval result to a server, and the retrieval result is used for assisting an AI model of the server in performing inference, thereby improving the accuracy of AI model inference, and improving the performance of the AI model.

Description

A retrieval enhancement method and related apparatus in a communication system

[0001] This application claims priority to Chinese Patent Application No. 202411018642.4, filed with the State Intellectual Property Office of China on July 26, 2024, entitled "A Search Enhancement Method and Related Device in a Communication System", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of communications, and more particularly to a retrieval enhancement method and related apparatus in a communication system. Background Technology

[0003] With the development of communication technology, AI models may be introduced into wireless networks to support artificial intelligence (AI) technology. AI models can be deployed in one or more of the following: access network nodes, terminal devices, core network devices, or servers.

[0004] Generally, terminal devices can send task requests to servers that have deployed AI models. The server then processes the task request using the AI ​​model to obtain the inference result. Currently, improving the performance of AI models in networks is a pressing issue that needs to be addressed. Summary of the Invention

[0005] This application provides a communication method and related apparatus for improving the performance of AI models.

[0006] Firstly, this application provides a communication method. This method can be applied to the server side, such as a server or a communication / processing module within a server, or circuits or chips within a server responsible for communication functions (such as a modem chip, also known as a baseband chip, or a system-on-a-chip (SoC) chip containing a modem core, or a system-in-package (SIP) chip), or circuits or chips within a server responsible for processing functions (such as a graphics processing unit (GPU)). Taking the application of this method to a server as an example, in this method, the server receives a first task request from a terminal device, and then the server sends a retrieval enhancement request to a network device. The network device sends first information to the server, the first information indicating the retrieval result corresponding to the retrieval enhancement request. Based on the retrieval result and an AI model, the server obtains a first inference result for the first task request.

[0007] In this application, the network device provides the server with retrieval augmented generation (RAG) retrieval results, which are used to assist the server's AI model in reasoning, thereby improving the accuracy of the AI ​​model's reasoning and improving the performance of the AI ​​model.

[0008] On the other hand, since the search results are obtained by network devices, they belong to network-side data. The server uses the search results to assist the server's AI model in reasoning, enabling the AI ​​model to reason based on network-side data. This enriches the information content of the AI ​​model's input data, improves the accuracy of the AI ​​model's reasoning, and enhances the AI ​​model's performance.

[0009] On the other hand, network-side data is generally private data. However, in this application, the network device sends the search results obtained based on network-side data to the server, enabling the AI ​​model to effectively utilize the originally private network-side data, improving data utilization efficiency, and enriching the information content of the input data for the AI ​​model.

[0010] Optionally, in this application, an AI module is deployed in the server to implement AI processing capabilities, AI-related functions, and / or AI-related services. The terminal device sends a first task request to the server, which is used to access the server's AI processing capabilities, AI-related functions, and / or AI-related services.

[0011] Optionally, the retrieval enhancement request is used to instruct the network device to perform a RAG task. Specifically, the network device has a database deployed for performing RAG. Based on the retrieval enhancement request, the network device retrieves the search results corresponding to the retrieval enhancement request from the database, thereby completing the RAG task.

[0012] Optionally, the AI ​​model in the server includes, but is not limited to, neural networks, neural network models, AI neural network models, machine learning models, mathematical models, or AI processing models. The examples provided in this application are merely illustrative; this application still applies to other implementations of AI models, and no specific limitations are imposed.

[0013] Based on the first aspect, in one optional implementation, the server sends a retrieval enhancement request to the network device when a first condition is met. The first condition includes one or more of the following:

[0014] 1. The accuracy of the second inference result obtained from the first task request based on the AI ​​model is less than a first threshold. Specifically, after receiving the first task request from the terminal device, the server first determines whether the accuracy of the second inference result is less than the first threshold. This second inference result is obtained by the server processing the first task request using the AI ​​model, without relying on the network device to perform RAG retrieval. If the accuracy of the second inference result is less than the first threshold, it indicates that the inference result obtained solely by processing the first task request using the AI ​​model is not accurate enough, and the server sends a retrieval enhancement request to the network device. If the accuracy of the second inference result is greater than or equal to the first threshold, it indicates that the inference result obtained solely by processing the first task request using the AI ​​model is accurate enough, and the server does not need to send a retrieval enhancement request to the network device. Therefore, when the accuracy of the inference result independently completed by the AI ​​model is sufficient to meet the accuracy requirements, the server can reduce unnecessary RAG processes and improve the processing efficiency of the AI ​​model.

[0015] 2. The first inference time is less than the second threshold. The first inference time is at least the sum of the time required for the network device to obtain the search results and the time required for the AI ​​model to output the inference results. The second threshold is a threshold for the processing time of the first task request that the terminal device can accept. The time required for the network device to obtain the search results is the time required for the network device to execute RAG to obtain the search results. The time required for the AI ​​model to output the inference results is the time required for the AI ​​model to obtain the first inference result based on the search results and the AI ​​model. Specifically, after receiving the first task request from the terminal device, the server first checks whether the first inference time is less than the second threshold. If the first inference time is less than the second threshold, it means that the first inference time does not exceed the threshold for the processing time of the first task request that the terminal device can accept, and the server sends a search enhancement request to the network device. If the first inference time is greater than or equal to the second threshold, it means that the first inference time will exceed the threshold for the processing time of the first task request that the terminal device can accept, and the server should skip the process of the network device obtaining the search results and reduce the first inference time to meet the second threshold, which is the threshold for the waiting time that the terminal device can accept. This reduces the processing time of the first task request, thus improving the processing efficiency of the AI ​​model.

[0016] Based on the first aspect, in one optional implementation, the first task request sent by the terminal device to the server carries first indication information, which is used to indicate a second threshold. Therefore, the server determines whether the aforementioned first condition is met based on the second threshold indicated by the first indication information.

[0017] Based on the first aspect, in one optional implementation, the network device may send a second indication message to the server, the second indication message being used to indicate the time required for the network device to obtain the search results.

[0018] Based on the first aspect, in an optional implementation, the retrieval enhancement request sent by the server to the network device includes a first sub-request and a second sub-request, with the first sub-request associated with the second sub-request. After receiving the first and second sub-requests from the retrieval enhancement request, the network device performs a Retrieval Augmentation (RAG) association between the first and second sub-requests to obtain the associated retrieval result. Therefore, in this application, the retrieval result of the retrieval enhancement request indicated by the first information is the associated retrieval result of the first and second sub-requests. Thus, the network device can perform a RAG association between the first and second sub-requests, improving the quality of the retrieval results.

[0019] Based on the first aspect, in one optional implementation, the server sends third indication information to the network device. This third indication information indicates a correlation between the first sub-request and the second sub-request. Upon receiving this third indication information, the network device determines that the first sub-request is associated with the second sub-request based on this information. The network device then performs a Relationship Analysis (RAG) between the first and second sub-requests to obtain the association retrieval result for the first and second sub-requests.

[0020] Secondly, this application provides a communication device that has the functions of the first aspect described above. For example, the communication device includes modules, units, or means that perform the operations involved in the first aspect. These modules, units, or means can be implemented by software, hardware, or a combination of software and hardware.

[0021] Thirdly, this application provides a communication device including an interface circuit and one or more processors. The one or more processors are coupled to a memory. The memory stores part or all of the necessary computer program or instructions for implementing the functions described in the first aspect. The one or more processors can execute the computer program or instructions, causing the communication device to implement the methods in any possible design or implementation of the first aspect. The interface circuit is used to implement the communication functions within the communication device and / or the communication functions between the communication device and other devices or components.

[0022] In one possible design, the processor is used to communicate with other devices or components through the interface circuit.

[0023] In one possible design, the communication device may also include the memory.

[0024] The aforementioned communication device may be a server, or a communication / processing module in a server, or a chip in a server responsible for communication functions such as a modem chip (also known as a baseband chip) or a SoC or SIP chip containing a modem module, or a circuit or chip in a server responsible for processing functions (such as a GPU).

[0025] Fourthly, this application provides a communication system, including the server, network device, and terminal device mentioned in the first aspect.

[0026] Fifthly, this application provides a computer-readable storage medium storing computer-readable instructions that, when read and executed by a computer, cause the computer to perform any of the possible designs in the first aspect described above.

[0027] In a sixth aspect, this application provides a computer program product that, when read and executed by a computer, causes the computer to perform any of the possible design methods of the first aspect described above.

[0028] The technical effects of any of the design methods in aspects two through six can be found in the technical effects of the different design methods in aspect one above, and will not be repeated here. Attached Figure Description

[0029] Figures 1 and 2 are schematic diagrams of possible, non-limiting systems used in the communication methods and related devices of this application;

[0030] Figures 3 and 4 are schematic diagrams of possible application frameworks in a communication system;

[0031] Figures 5–7 are schematic diagrams illustrating possible implementations of the communication method in this application;

[0032] Figure 8 is a schematic diagram of a communication device provided in an embodiment of this application. Detailed Implementation

[0033] The present application will now be described with reference to the accompanying drawings. The terminology used in the embodiments section is for illustrative purposes only and is not intended to limit the scope of the application. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided in this application are equally applicable to similar technical problems.

[0034] First, some of the nouns or terms used in this application will be explained, and these nouns or terms are also part of the content of the invention.

[0035] (1) The terms “system” and “network” in this application are used interchangeably. “Multiple” refers to two or more. “And / or” describes the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, or B exists alone, where A and B can be singular or plural. The character “ / ” generally indicates that the related objects before and after are in an “or” relationship. “At least one of the following” or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, “at least one of A, B and C” includes A, B, C, AB, AC, BC or ABC. Unless otherwise specified, the ordinal numbers such as “first” and “second” mentioned in this application are used to distinguish multiple objects and are not used to limit the order, sequence, priority or importance of multiple objects. Furthermore, the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, such that a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such process, method, product, or apparatus.

[0036] (2) Configuration and Predefined: In this application, both configuration and predefined are used. Configuration refers to the process by which a network device or server sends configuration information or parameter values ​​to a terminal device via messages or signaling, so that the terminal device can determine communication parameters or resources for transmission based on these values ​​or information. Predefined is similar to configuration; it can be parameter information or values ​​pre-negotiated between the network device / server and the terminal device, parameter information or values ​​specified by standard protocols for use by the base station / network device or terminal device, or parameter information or values ​​pre-stored in the base station / server or terminal device. This application does not limit the scope of these terms.

[0037] It should be understood that these values ​​and parameters can change or be updated.

[0038] (3) In this application, “instruction” may include direct instruction and indirect instruction, and may also include explicit instruction and implicit instruction. When a certain instruction information is used to instruct A, it can be understood that the instruction information carries A, directly instructs A, or indirectly instructs A.

[0039] In this application, the information indicated by the instruction information is called the information to be instructed. In specific implementations, there are many ways to indicate the information to be instructed, such as, but not limited to, directly indicating the information to be instructed, such as the information to be instructed itself or its index. It can also indirectly indicate the information to be instructed by indicating other information, where there is a relationship between the other information and the information to be instructed; or it can indicate only a part of the information to be instructed, while the other parts are known or pre-agreed upon, for example, by using a pre-agreed (e.g., protocol-predefined) arrangement of various information to indicate specific information, thereby reducing instruction overhead to some extent. This application does not limit the specific method of instruction. It is understood that for the sender of the instruction information, the instruction information can be used to indicate the information to be instructed, and for the receiver of the instruction information, the instruction information can be used to determine the information to be instructed.

[0040] (4) Artificial intelligence (AI) enables machines to possess human-like intelligence, such as allowing machines to use computer hardware and software to simulate certain intelligent human behaviors. To achieve AI, machine learning methods can be employed. In machine learning, machines learn (or train) a model using training data. This model represents the mapping between input and output. The learned model can be used for reasoning (or prediction), that is, it can be used to predict the output corresponding to a given input. This output can also be called the reasoning result (or prediction result).

[0041] Machine learning can include supervised learning, unsupervised learning, and reinforcement learning. Unsupervised learning can also be called learning without supervision.

[0042] Supervised learning, based on collected sample values ​​and labels, uses machine learning algorithms to learn the mapping relationship between sample values ​​and labels, and then expresses this learned mapping relationship using an AI model. The process of training the machine learning model is the process of learning this mapping relationship. During training, sample values ​​are input into the model to obtain the model's predicted values, and the model parameters are optimized by calculating the error between the model's predicted values ​​and the sample labels (ideal values). After the mapping relationship is learned, it can be used to predict new sample labels. The mapping relationship learned in supervised learning can include linear or non-linear mappings. Based on the type of label, the learning task can be divided into classification tasks and regression tasks.

[0043] Unsupervised learning relies on collected sample values ​​to discover inherent patterns within the samples themselves. One type of unsupervised learning algorithm uses the samples themselves as supervisory signals, meaning the model learns the mapping relationship from sample to sample; this is called self-supervised learning. During training, model parameters are optimized by calculating the error between the model's predictions and the samples themselves. Self-supervised learning can be used for signal compression and decompression recovery applications; common algorithms include autoencoders and generative adversarial networks.

[0044] Reinforcement learning, unlike supervised learning, is a type of algorithm that learns problem-solving strategies through interaction with the environment. Unlike supervised and unsupervised learning, reinforcement learning problems do not have explicit "correct" action labels. The algorithm needs to interact with the environment to obtain reward signals from the environment, and then adjust its decision actions to obtain a larger reward signal value. For example, in downlink power control, the reinforcement learning model adjusts the downlink transmission power of each user based on the total system throughput feedback from the wireless network, aiming to achieve a higher system throughput. The goal of reinforcement learning is also to learn the mapping relationship between the environment state and a better (e.g., optimal) decision action. However, because the label of the "correct action" cannot be obtained in advance, the network cannot be optimized by calculating the error between the action and the "correct action." Reinforcement learning training is achieved through iterative interaction with the environment.

[0045] A model can also be called an AI model, a rule, or other names. An AI model can be considered a specific method for implementing AI functions. An AI model represents the mapping relationship or function between the model's input and output. AI functions can include one or more of the following: data collection, model training (or model learning), model information dissemination, model inference (or model reasoning, inference, or prediction, etc.), model monitoring or model validation, or inference result publication, etc. AI functions can also be called AI (related) operations or AI-related functions.

[0046] (5) Retrieval Augmented Generation (RAG): RAG is one of the most popular cutting-edge technologies for large-scale modeling. RAG combines language modeling and information retrieval techniques. Specifically, RAG works by retrieving relevant information from a large collection of documents and then using this information to guide text generation, thereby improving the quality and accuracy of predictions. RAG operates through three key parts: retrieval, utilization, and generation. In the retrieval phase, the system retrieves relevant information from the document collection; in the utilization phase, the system uses this retrieved information to populate text or answer questions; and finally, in the generation phase, the system generates the final text content based on the retrieved knowledge. Through this process, RAG technology can play a role in various natural language processing tasks, such as question answering systems, document generation and automatic summarization, intelligent assistants and virtual agents, information retrieval, and knowledge graph filling. Furthermore, RAG technology offers advantages such as timely updates, strong interpretability, high customization capabilities, secure privacy management, and reduced training costs.

[0047] For example, let's take a large language model (LLM) as an example of an AI model. Suppose we input a question into the LLM, then the RAG can retrieve relevant information from various data sources and input the retrieved information into the LLM. Based on the question and the retrieved information, the LLM outputs the answer to the question.

[0048] In summary, RAG retrieves relevant knowledge through retrieval and uses this knowledge as prompts, enabling AI models to refer to this knowledge and provide reasonable answers. Therefore, the core of RAG can be understood as "retrieval + generation." The former mainly utilizes the efficient storage and retrieval capabilities of vector databases to recall knowledge; the latter uses large models and prompt engineering to rationally utilize the recalled knowledge and generate answers.

[0049] The use of RAG involves the following steps:

[0050] 1. Data preparation.

[0051] Data preparation is generally an offline process, which mainly involves vectorizing private domain data, building an index, and storing it in a database.

[0052] 2. Data extraction.

[0053] Data extraction is equivalent to data retrieval. RAG performs a RAG search within the aforementioned database based on the questions provided by the user. Common data retrieval methods include similarity search and full-text search.

[0054] Similarity retrieval involves calculating the similarity score between the query vector and all stored vectors, and returning the records with the highest scores. Common similarity calculation methods include cosine similarity, Euclidean distance, or Manhattan distance.

[0055] Full-text search: Full-text search is a classic search method. When data is stored, an inverted index is built using keywords; during retrieval, a full-text search is performed using keywords to find the corresponding records.

[0056] Optionally, the above-mentioned multiple search methods can be combined and executed.

[0057] 3. Data usage.

[0058] Based on RAG (Retrieval and Analysis), search results can be obtained and provided to the AI ​​model as prompts. These prompts, as direct input to the AI ​​model, are a key factor influencing the model's output accuracy. In RAG scenarios, prompts typically include task descriptions, background knowledge (obtained through RAG retrieval), and / or task instructions (usually user questions). Depending on the task scenario and AI model performance, other instructions can be appropriately added to the prompts to optimize the output of larger models.

[0059] Next, we will introduce the possible, non-limiting scenarios involved in this application.

[0060] With the development of communication technology, AI models may be introduced into wireless networks to support AI technology. These AI models can be deployed in one or more of the following: access network nodes, terminal devices, core network devices, or servers.

[0061] Generally, terminal devices can send task requests to servers deployed with AI models. The server processes the task request using the AI ​​model to obtain the inference result. However, in practical applications, general-purpose basic AI models often fail to meet actual business needs, mainly for the following reasons:

[0062] Reason 1: Limitations of Knowledge: The knowledge of an AI model originates entirely from its training data. However, the training sets of most mainstream AI models are built from publicly available online data. Real-time, non-public, or offline data is unavailable, leaving this knowledge unavailable. This results in insufficient training samples for AI models, thus reducing their inference accuracy.

[0063] Reason 2: Inference Accuracy: The underlying principle of AI models is based on mathematical probability, and their output is essentially a series of numerical calculations. Sometimes, when an AI model cannot obtain strongly relevant information but still needs to provide feedback to the user, it may provide inaccurate inference results. This problem is particularly likely to occur in scenarios where the AI ​​model itself lacks knowledge in a certain area or is not proficient in a certain domain.

[0064] Therefore, improving the performance of AI models in the network is an urgent problem to be solved.

[0065] To address the aforementioned problems, this application provides a communication method and related apparatus for improving the performance of AI models. The communication method and related apparatus provided in this application can be applied to various communication systems. For example, 5th generation (5G) mobile communication systems, new radio (NR) systems, long term evolution (LTE) systems, LTE frequency division duplex (FDD) systems, LTE time division duplex (TDD) systems, future communication systems, vehicle-to-everything (V2X) communication systems, device-to-device (D2D) communication systems, Internet of Things (IoT) communication systems, industrial internet communication systems, or satellite communication systems, etc. The wireless communication systems involved in this application also include, but are not limited to, narrowband Internet of Things (NB-IoT) systems.

[0066] For example, please refer to Figure 1, which is a possible, non-limiting system diagram of the communication method and related apparatus used in this application. As shown in Figure 1, the communication system 10 includes a radio access network (RAN) 100 and a core network (CN) 200. Optionally, the communication system 10 may also include an Internet 300. The RAN 100 includes at least one RAN node (110a and 110b in Figure 1, collectively referred to as 110) and at least one terminal device (120a-120j in Figure 1, collectively referred to as 120). The RAN 100 may also include other RAN nodes, such as wireless relay devices and / or wireless backhaul devices (not shown in Figure 1). The terminal device 120 is wirelessly connected to the RAN node 110. The RAN node 110 is connected to the core network 200 wirelessly or via a wired connection. The core network equipment in core network 200 and RAN node 110 in RAN 100 can be different physical devices, or they can be the same physical device integrating core network logical functions and radio access network logical functions. Terminal devices and RAN nodes can be interconnected via wired or wireless means.

[0067] RAN 100 can be a cellular system related to the 3rd Generation Partnership Project (3GPP), such as a fourth-generation (4G) mobile communication system, a fifth-generation (5G) mobile communication system, or a future communication system. RAN 100 can also be an open RAN (O-RAN or ORAN), a cloud radio access network (CRAN), an evolved universal terrestrial radio access (E-UTRA) system, or a wireless fidelity (WiFi) system. RAN 100 can also be a communication system that integrates two or more of the above systems.

[0068] RAN node 110, sometimes also referred to as network equipment, access network equipment, RAN device, RAN entity, or access node, constitutes part of the communication system and is used to help terminal equipment achieve wireless access. Multiple RAN nodes 110 in communication system 10 can be of the same type or different types. In some scenarios, the roles of RAN node 110 and terminal equipment 120 are relative. For example, network element 120i in Figure 1 can be a helicopter or drone, which can be configured as a mobile base station. For terminal equipment 120j accessing RAN 100 through network element 120i, network element 120i is a base station; but for base station 110a, network element 120i is a terminal equipment. RAN node 110 and terminal equipment 120 are sometimes both referred to as communication devices. For example, network elements 110a and 110b in Figure 1 can be understood as communication devices with base station functions, and network elements 120a-120j can be understood as communication devices with terminal equipment functions.

[0069] In one possible scenario, RAN node 110 can be a base station, an evolved NodeB (eNodeB), an access point (AP), a transmission reception point (TRP), a next-generation NodeB (gNB), a base station in a future mobile communication system, or an access node in a WiFi system. Optionally, RAN node 110 can also be a macro base station (as shown in Figure 1, 110a), a micro base station or indoor station (as shown in Figure 1, 110b), a relay node or donor node, or a radio controller in a CRAN scenario. Optionally, RAN node 110 can also be a server, a wearable device, a vehicle, or in-vehicle equipment. For example, the access network equipment in vehicle-to-everything (V2X) technology can be a roadside unit (RSU). All or part of the functions of RAN node 110 in this application can also be implemented through software functions running on hardware, or through virtualization functions instantiated on a platform (e.g., a cloud platform). The RAN node 110 may also be equipped with communication modules, circuits, or chips that perform corresponding communication functions. The RAN node 110 may also be configured with program instructions for performing corresponding communication functions, as well as corresponding program instructions. The RAN node 110 in this application may also be a logic node, logic module, or software capable of implementing all or part of the functions of the RAN node 110.

[0070] In another possible scenario, multiple RAN nodes collaborate to assist terminal devices in achieving wireless access, with different RAN nodes each implementing a portion of the base station's functions. For example, RAN nodes can be central units (CUs), distributed units (DUs), CU-control plane (CPs), CU-user plane (UPs), or radio units (RUs), etc. CUs and DUs can be set up separately or included in the same network element, such as a baseband unit (BBU). RUs can be included in radio frequency equipment or radio frequency units, such as remote radio units (RRUs), active antenna units (AAUs), or remote radio heads (RRHs).

[0071] In different systems, CU (or CU-CP and CU-UP), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an ORAN system, CU can also be called O-CU (open CU), DU can also be called O-DU, CU-CP can also be called O-CU-CP, CU-UP can also be called O-CU-UP, and RU can also be called O-RU. For ease of description, this application uses CU, CU-CP, CU-UP, DU, and RU as examples. Any of the units among CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through software modules, hardware modules, or a combination of software and hardware modules.

[0072] Terminal equipment can be any device or module that connects to the communication system shown above and has corresponding communication functions. Terminal equipment can also be referred to as a terminal, user equipment (UE), mobile station (MS), mobile terminal (MT), fixed wireless access (FWA), or customer premises equipment (CPE), etc. Terminal equipment includes wireless communication functions (providing voice / data connectivity to users). Examples include handheld devices with wireless connectivity, in-vehicle devices, and machine-type communication (MTC) terminals. Currently, terminal devices can include: mobile phones, tablets, laptops, PDAs, mobile internet devices (MIDs), wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, wireless terminals in industrial control, wireless terminals in self-driving (e.g., drones, vehicles), wireless terminals in remote medical surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, and wireless terminals in smart homes. For example, wireless terminals in self-driving can be drones, helicopters, or airplanes. For example, wireless terminals in vehicle-to-everything (V2X) can be in-vehicle equipment, vehicle-mounted equipment, in-vehicle modules, vehicles, or ships. Wireless terminals in industrial control can be cameras, robots, or robotic arms. Wireless terminals in smart homes can be televisions, air conditioners, robot vacuums, speakers, or set-top boxes. Terminal devices typically contain communication modules, circuits, or chips that perform corresponding communication functions, and they also contain program instructions for performing those functions.

[0073] Please refer to Figure 2, which is a schematic diagram of another possible, non-limiting system applied to the communication method and related apparatus in this application. The communication system adopts a server-network-terminal device network architecture. As shown in Figure 2, the communication system includes one or more of the following:

[0074] 1. UE (also known as terminal device): For a detailed description of the terminal device, please refer to the description in the embodiment corresponding to Figure 1 above, which will not be repeated here.

[0075] 2. A data network (DN) is used to provide services such as carrier services, internet access, or third-party services. A server (as shown in Figure 2) can be considered part of the DN. Optionally, in this application, an AI model is deployed on the server to implement AI processing capabilities, AI-related functions, and / or AI-related services.

[0076] 3. Access Network (AN): Includes RAN node 100 as shown in Figure 1. For a detailed description of RAN node 100, please refer to the description in the corresponding embodiment of Figure 1 above; it will not be repeated here.

[0077] 4. Core Network: Used to perform three main functions: registration, connection, and session management. The core network mainly includes Network Exposure Function (NEF) network elements, Policy Control Function (PCF) network elements, Application Function (AF) network elements, Access and Mobility Management Function (AMF) network elements, Session Management Function (SMF) network elements, and User Plane Function (UPF) network elements, etc.

[0078] The following section introduces the various network elements in the core network.

[0079] NEF network element: Used to expose the services and capabilities of 3GPP network functions to AF network element, and at the same time, it can also allow AF network element to provide information to 3GPP network functions. The corresponding interface is N33 interface.

[0080] PCF network element: Used for policy management of charging and QoS policies;

[0081] AF (AF Element): Used to transmit the application's requirements to the network side;

[0082] AMF network elements are primarily used for mobility management, access authentication / authorization, and other functions. They are also responsible for transmitting user policies between the UE and PCF network elements. The N1 interface is the signaling plane interface between the UE and AMF network elements; since the UE cannot directly interact with the core network, it needs to pass NAS (non-access stratum) information through the AN. The N2 interface is the signaling plane interface through which the AMF network element requests resources from the AN for Protocol Data Unit (PDU) sessions.

[0083] SMF network element: Used to complete session management functions such as Internet Protocol (IP) address allocation for UE, UPF selection, and billing and QoS policy control;

[0084] UPF network elements: As the interface with the data network, they are used to perform functions such as user plane data forwarding, session / flow-based billing statistics, and bandwidth limiting. The N3 interface is the interface between the radio access network (RAN) and UPF network elements, mainly used to transmit uplink and downlink user plane data between the 5G RAN and UPF network elements.

[0085] Optionally, AI nodes may be introduced into the network to support artificial intelligence (AI) technology.

[0086] AI nodes can be deployed in one or more of the following locations within the communication system: access network nodes (RAN nodes), terminal devices, or core network devices. Alternatively, AI nodes can be deployed independently, for example, in a location other than any of the aforementioned devices, such as in the host or cloud server of an over-the-top (OTT) system. AI nodes can communicate with other devices in the communication system, which can be one or more of the following: network devices, terminal devices, or core network elements.

[0087] It is understood that this application does not limit the number of AI nodes. For example, when there are multiple AI nodes, they can be divided based on function, such as different AI nodes being responsible for different functions.

[0088] It can also be understood that AI nodes can be independent devices, or they can be integrated into the same device to achieve different functions. Alternatively, they can be network elements in hardware devices, software functions running on dedicated hardware, or virtualization functions instantiated on a platform (e.g., a cloud platform). This application does not limit the specific form of the aforementioned AI nodes.

[0089] AI nodes can be AI network elements or AI modules.

[0090] Figure 3 illustrates a possible application framework in a communication system. As shown in Figure 3, network elements in the communication system are connected via interfaces (e.g., NG, Xn) or air interfaces. These network element nodes, such as core network equipment, access network nodes (RAN nodes), terminals, or one or more devices in operations administration and maintenance (OAM), are equipped with one or more AI modules (only one is shown in Figure 3 for clarity). An access network node can be a single RAN node or can include multiple RAN nodes, for example, including CU and DU. The CU and / or DU can also be equipped with one or more AI modules. A CU can also be split into CU-CP and CU-UP, with one or more AI modules configured in the CU-CP and / or CU-UP.

[0091] AI modules are used to implement corresponding AI functions. AI modules deployed in different network elements can be the same or different. The models of AI modules can achieve different functions depending on the parameter configurations. The models of AI modules can be configured based on one or more of the following parameters: structural parameters (e.g., at least one of the following: number of neural network layers, neural network width, inter-layer connections, neuron weights, neuron activation function, or biases in the activation function), input parameters (e.g., the type and / or dimension of the input parameters), or output parameters (e.g., the type and / or dimension of the output parameters). The biases in the activation function can also be referred to as the biases of the neural network.

[0092] In one example, the neural network mentioned above can be a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), or a generative adversarial network (GAN).

[0093] Deep Neural Networks (DNNs) are artificial neural network architectures with multiple layers of nonlinear transformation units stacked in a hierarchical structure to form deep computational models. Compared to shallow neural networks, deep neural networks have more hidden layers, allowing the network model to capture more complex data structures and higher-level abstract features.

[0094] A CNN is a deep neural network with a convolutional structure. A CNN contains a feature extractor consisting of convolutional layers and subsampling layers. This feature extractor can be viewed as a filter, and the convolution process can be seen as performing convolution between a trainable filter and an input image or a convolutional feature map.

[0095] RNN is a type of recursive neural network that takes sequence data as input, recursively moves along the direction of sequence evolution, and connects all nodes (recurrent units) in a chain-like manner.

[0096] GAN is a deep learning model. It consists of a generator and a discriminator, and is trained through adversarial learning. Its purpose is to estimate the potential distribution of data samples and generate new data samples.

[0097] An AI module can have one or more models. A model can infer an output, which includes one or more parameters. The learning, training, or inference processes of different models can be deployed on different nodes or devices, or they can be deployed on the same node or device.

[0098] Figure 4 illustrates another possible application framework in a communication system. As shown in Figure 4, the communication system includes a RAN intelligent controller (RIC). For example, the RIC can be the AI ​​module shown in Figure 3, used to implement AI-related functions. RICs include near-real-time RICs (near-RT RICs) and non-real-time RICs (non-RT RICs). Non-real-time RICs primarily process non-real-time information, such as data that is not sensitive to latency, with latency in the order of seconds. Real-time RICs primarily process near-real-time information, such as data that is relatively sensitive to latency, with latency in the order of tens of milliseconds.

[0099] Near real-time (NRT) RICs are used for model training and inference. For example, they are used to train AI models and then use those models for inference. NRT RICs can obtain network-side and / or terminal-side information from RAN nodes (e.g., CUs, CU-CPs, CU-UPs, DUs, and / or RUs) and / or terminals. This information can be used as training data or inference data. NRT RICs can deliver inference results to RAN nodes and / or terminals. Inference results can be exchanged between CUs and DUs, and / or between DUs and RUs. For example, a NRT RIC delivers an inference result to a DU, which then forwards it to an RU.

[0100] Non-real-time RICs are also used for model training and inference. For example, they are used to train AI models and then use those models for inference. Non-real-time RICs can obtain network-side and / or terminal-side information from RAN nodes (e.g., CUs, CU-CPs, CU-UPs, DUs, and / or RUs) and / or terminals. This information can be used as training data or inference data, and the inference results can be delivered to RAN nodes and / or terminals. Inference results can be exchanged between CUs and DUs, and / or between DUs and RUs; for example, a non-real-time RIC delivers inference results to a DU, which then forwards them to an RU.

[0101] Near real-time RICs and non-real-time RICs can also be configured as separate network elements. Near real-time RICs and non-real-time RICs can also be part of other devices. For example, near real-time RICs can be set in RAN nodes (e.g., CU, DU), while non-real-time RICs can be set in OAM, cloud servers, core network devices, or other network devices.

[0102] The communication method and related apparatus of this application will be further described below with reference to the accompanying drawings.

[0103] In this application, the RAN nodes shown in Figures 1 to 4 can be replaced with other terms, such as "network device". For ease of description, unless otherwise specified, "network device" will be used throughout this application. It should be understood that the technical solutions provided in this application are also applicable to other different expressions or types of "network devices".

[0104] In this application, the server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), big data or artificial intelligence platforms, etc. Specific limitations are not specified here. Terminal devices and servers can be directly or indirectly connected via wired or wireless communication, and terminal devices and servers can be connected to form a blockchain network; this application does not impose any restrictions.

[0105] In this application, "sending information" can be understood as one device sending information to another device, or it can also be understood as one logical module within a device sending information to another logical module. For example, "server sending information" can be understood as the server sending information to another device (such as a terminal device or a network device), or it can be understood as logical module 1 in the server sending information to logical module 2 in the server.

[0106] In this application, "receiving information" can be understood as one device receiving information from another device, or it can also be understood as a logical module within a device receiving information from another logical module. For example, "the server receiving information" can be understood as the server receiving information from another device (such as a terminal device or a network device), or it can be understood as logical module 1 in the server receiving information from logical module 2 in the server.

[0107] In this application, "sending information to... (e.g., a network device)" or the relevant illustrations in the accompanying drawings can be understood as the destination of the information being a network device. This can include sending information directly or indirectly to a network device. "Receiving information from... (e.g., a network device)," "receiving information from... (e.g., a network device)," or "receiving information sent (e.g., by a network device)," or the relevant illustrations in the accompanying drawings, can be understood as the source of the information being a network device. This can include receiving information directly or indirectly from a network device. Information may undergo necessary processing between the source and destination, such as format changes, but the destination can understand the valid information from the source. Similar expressions in this application can be understood in a similar way, and will not be elaborated further here.

[0108] Please refer to Figure 5, which is a schematic diagram of a possible implementation of the communication method in this application. It should be understood that this application uses network devices, terminal devices, and servers as examples of the execution entities in this interactive illustration to illustrate the method, but this application does not limit the execution entities of this interactive illustration. For example, the method executed by the network device in Figure 5 can also be implemented by chips, baseband chips, modem chips, system-on-chip (SoC) chips containing modem cores, system-in-package (SIP) chips, communication modules, chip systems, processors, logic modules, or software within the network device. In this application, when referring to a network device, it can refer to the network device itself, or to chips, communication modules, integrated circuits, processors, logic modules, or software within the network device used to implement the communication method provided in this application; no specific limitation is made in this application. Similarly, the method executed by the terminal device in Figure 5 can also be implemented by chips, baseband chips, modem chips, SoC chips containing modem cores, SIP chips, communication modules, chip systems, processors, logic modules, or software within the terminal device. In this application, the term "terminal device" can refer to either the terminal device itself or the chips, communication modules, integrated circuits, processors, logic modules, or software within the terminal device used to implement the communication methods provided in this application; no specific limitation is made in this application. Similarly, the method executed by the server in Figure 5 can also be implemented by chips, baseband chips, modem chips, SoC chips containing modem cores, SIP chips, communication modules, chip systems, processors, logic modules, or software within the server. In this application, the term "server" can refer to either the server itself or the chips, communication modules, integrated circuits, processors, logic modules, or software within the server used to implement the communication methods provided in this application; no specific limitation is made in this application.

[0109] For example, in the embodiment corresponding to Figure 5 shown below, if steps 501 to 504 are executed by the server, they can be understood as the server sending and receiving information with the network device and the terminal device; if steps 501 to 504 are executed by the graphics processing unit (GPU) in the server, in step 501, "the server receives a first task request from the terminal device," it can be understood that the communication module in the terminal device sends a first task request to the communication module in the server, and then the communication module in the server forwards the first task request to the GPU in the server; in step 502, "the server sends a retrieval enhancement request to the network device," it can be understood that the GPU in the server sends a retrieval enhancement request to the communication module in the server, and then the communication module in the server sends a retrieval enhancement request to the communication module in the network device; in step 503, "the network device sends first information to the server," it can be understood that the communication module in the network device sends first information to the communication module in the server, and then the communication module in the server sends first information to the GPU in the server.

[0110] As shown in Figure 5, in Embodiment 1, the communication method of this application includes, but is not limited to, steps 501 to 504.

[0111] 501. The terminal device sends a first task request to the server, and the server receives the first task request from the terminal device.

[0112] In this application, an AI module is deployed in the server to implement AI processing capabilities, AI-related functions, and / or AI-related services. The terminal device sends a first task request to the server, which is used to access the server's AI processing capabilities, AI-related functions, and / or AI-related services.

[0113] Optionally, the AI ​​model in the server includes, but is not limited to, neural networks, neural network models, AI neural network models, machine learning models, mathematical models, or AI processing models. The examples provided in this application are merely illustrative; this application still applies to other implementations of AI models, and no specific limitations are imposed.

[0114] 502. The server sends a retrieval enhancement request to the network device, and the network device receives the retrieval enhancement request from the server.

[0115] After receiving the first task request from the terminal device, the server generates a search enhancement request based on the first task request. Then, the server sends the search enhancement request to the network device. Accordingly, the network device receives the search enhancement request from the server.

[0116] In this application, a retrieval augmentation request is used to instruct a network device to perform a retrieval augmented generation (RAG) task. Specifically, the network device deploys a database for performing RAG. Based on the retrieval augmentation request, the network device retrieves the search results corresponding to the retrieval augmentation request from the database, thereby completing the RAG task.

[0117] Optionally, after receiving the first task request, the server generates a search enhancement request based on the RAG rules in the network device's database and sends the search enhancement request to the network device. Since the search enhancement request is generated based on the RAG rules in the network device's database, it reduces the likelihood of the network device being unable to parse or process the search enhancement request, thus improving the efficiency of the network device in executing RAG rules. Specifically, the network device can send its database's RAG rules to the server before step 501, so that the server can generate the search enhancement request in step 502 based on the network device's database's RAG rules.

[0118] Optionally, in step 502, the server can first pass the retrieval enhancement request to the core network via the N33 interface, and the core network will then pass the GPRS tunneling protocol-control (GTP-C) signaling carrying the retrieval enhancement request to the network device.

[0119] 503. The network device sends the first information to the server, and the server receives the first information from the network device accordingly.

[0120] Based on the search enhancement request, the network device performs a search algorithm (RAG) to obtain the search results. The network device sends first information to the server, and the server receives the first information from the network device. The first information indicates the search results corresponding to the search enhancement request.

[0121] 504. Based on the search results and the AI ​​model, the server obtains the first inference result of the first task request.

[0122] Specifically, the server inputs the search results and the first task request into the AI ​​model to obtain the first inference result corresponding to the first task request.

[0123] In this application, the network device provides the server with the search results for RAG, which are used to assist the server's AI model in reasoning, thereby improving the accuracy of the AI ​​model's reasoning and improving the performance of the AI ​​model.

[0124] On the other hand, since the search results are obtained by network devices, they belong to network-side data. The server uses the search results to assist the server's AI model in reasoning, enabling the AI ​​model to reason based on network-side data. This enriches the information content of the AI ​​model's input data, improves the accuracy of the AI ​​model's reasoning, and enhances the AI ​​model's performance.

[0125] On the other hand, network-side data is generally private data. However, in this application, the network device sends the search results obtained based on network-side data to the server, enabling the AI ​​model to effectively utilize the originally private network-side data, improving data utilization efficiency, and enriching the information content of the input data for the AI ​​model.

[0126] Optionally, in the embodiment shown in Figure 5, the communication method of this application further includes step 505: the server sends the processing result corresponding to the first task request to the terminal device. The processing result includes a first inference result, or data obtained by processing based on the first processing result. For example, the data includes one or more of business data, signals, information, and signaling.

[0127] In one possible implementation, before step 502 above, the server first determines whether the network device needs to perform RAG first. Specifically, if the first condition is met, it means that the network device needs to perform RAG first, then the server sends a retrieval enhancement request to the network device, that is, the server performs step 502;

[0128] The first condition includes one or more of the following:

[0129] 1. The accuracy of the second inference result obtained from the first task request based on the AI ​​model is less than the first threshold. Specifically, after receiving the first task request from the terminal device, the server first determines whether the accuracy of the second inference result is less than the first threshold. This second inference result is obtained by the server processing the first task request using the AI ​​model, without relying on the network device to perform RAG retrieval. If the accuracy of the second inference result is less than the first threshold, it indicates that the inference result obtained solely by processing the first task request using the AI ​​model is not accurate enough, and the server sends a retrieval enhancement request to the network device (i.e., executes step 502). If the accuracy of the second inference result is greater than or equal to the first threshold, it indicates that the inference result obtained solely by processing the first task request using the AI ​​model is accurate enough, and the server does not need to send a retrieval enhancement request to the network device (i.e., does not execute step 502). Thus, when the accuracy of the inference result independently completed by the AI ​​model is sufficient to meet the accuracy requirements, the server can reduce unnecessary RAG processes and improve the processing efficiency of the AI ​​model.

[0130] 2. The first inference time is less than the second threshold, wherein the first inference time is at least the sum of the time required for the network device to obtain the search results and the time required for the AI ​​model to output the inference results, and the second threshold is the threshold for the processing time of the first task request that the terminal device can accept. The time required for the network device to obtain the search results is the time required for the network device to execute RAG to obtain the search results, and the time required for the AI ​​model to output the inference results is the time required for the AI ​​model to obtain the first inference result based on the search results and the AI ​​model. Specifically, after receiving the first task request from the terminal device, the server first checks whether the first inference time is less than the second threshold. If the first inference time is less than the second threshold, it means that the first inference time does not exceed the threshold for the processing time of the first task request that the terminal device can accept, and the server executes step 502; if the first inference time is greater than or equal to the second threshold, it means that the first inference time will exceed the threshold for the processing time of the first task request that the terminal device can accept, and the server should skip the process of the network device obtaining the search results and reduce the first inference time in order to meet the second threshold being the threshold for the waiting time that the terminal device can accept. This reduces the processing time of the first task request, thus improving the processing efficiency of the AI ​​model.

[0131] In this application, "greater than" can specifically mean "higher than", and "less than" can specifically mean "lower than". Furthermore, "less than" in the examples can be replaced with "less than or equal to", and "greater than" can be replaced with "greater than or equal to", without any specific limitation in this application. It should be understood that the examples provided in this application are merely illustrative and do not constitute a limitation on this application.

[0132] In one possible implementation, in step 501, the first task request sent by the terminal device to the server carries first indication information, which indicates a second threshold. The server then determines whether the first condition is met based on the second threshold indicated by the first indication information. Specifically, the first indication information can be carried in user assistance information within radio resource control (RRC) signaling, or it can be carried in a newly added Medium Access Control (MAC) Element. For example, the first indication information can indicate the latest time point for completing the first task request. After receiving the first indication information, the server determines the second threshold based on the current time point and the latest time point indicated by the first indication information; or, the first indication information can indicate the longest waiting time for the terminal device. After receiving the first indication information, the server determines the second threshold based on the longest waiting time indicated by the first indication information. It should be understood that the above implementation of the first indication information is merely an example and does not limit this application. Other implementations of the first indication information are still applicable to this application, and are not specifically limited here.

[0133] In one possible implementation, the network device may send a second indication message to the server, the second indication message indicating the time required for the network device to obtain the search results. Please refer to Figure 6, which is a schematic diagram of another possible implementation of the communication method in this application. As shown in Figure 6, the communication method in this application includes steps 501 to 504 as shown in Figure 5, and further includes step 500a. In step 500a, the network device may send the second indication message to the server, and correspondingly, the server receives the second indication message from the network device.

[0134] Optionally, step 500a shown in Figure 5 can be performed before step 501. Specifically, after the network device establishes a connection with the server, the network device sends second indication information to the server. For example, since the network device has not yet received the retrieval enhancement request in step 502, the network device determines the time required for the network device to obtain the retrieval results based on one or more of the following: device capabilities, network environment, historical RAG execution records, or the amount of data in the database, and obtains the second indication information.

[0135] Furthermore, since the server has received the second instruction, it determines the time required for the network device to obtain the search results. During this waiting period, the server can release resources used to run the AI ​​model, thereby improving resource utilization.

[0136] Furthermore, since the server determines the first processing time, it can notify the terminal device of this first processing time, allowing the terminal device to be aware of it and perform other tasks while waiting for the first processing time. Alternatively, the server can send a wait instruction to the terminal device, indicating that it is currently waiting for the processing of the first task request, so that the terminal device can perform other tasks while waiting.

[0137] Optionally, step 500a shown in Figure 5 can be performed after step 502. Specifically, since the network device receives a search enhancement request from the server in step 502, the network device can determine the time required for the network device to obtain the search results corresponding to the search enhancement request based on the search enhancement request, and obtain the second indication information.

[0138] In one possible implementation, the retrieval enhancement request sent by the server to the network device includes a first sub-request and a second sub-request, with the first sub-request associated with the second sub-request. Upon receiving the first and second sub-requests from the retrieval enhancement request, the network device performs a Retrieval Augmentation (RAG) association between the first and second sub-requests to obtain the associated retrieval result. Therefore, in this application, the retrieval result of the retrieval enhancement request indicated by the first information is the associated retrieval result of the first and second sub-requests. Thus, the network device can perform a RAG association between the first and second sub-requests, improving the quality of the retrieval results.

[0139] Optionally, the first task request may be multiple task requests sent by the terminal device to the server at different times, and the server may obtain a retrieval enhancement request (including a first sub-request and a second sub-request) based on the multiple task requests; or, the first task request may be a single task request sent by the terminal device to the server at a certain time, and the server may obtain a retrieval enhancement request (including a first sub-request and a second sub-request) based on the task request.

[0140] Optionally, the first sub-request and the second sub-request can be carried in the same message / signaling / signal; or, the first sub-request and the second sub-request can be carried in different messages / signaling / signals respectively.

[0141] In one possible implementation, the server sends a third indication message to the network device. This third indication message indicates that there is an association between the first sub-request and the second sub-request. Please refer to Figure 7, which is a schematic diagram of another possible implementation of the communication method in this application. As shown in Figure 7, the communication method in this application includes steps 501 to 504 as shown in Figure 5, and further includes step 500b, which is executed before step 503. In step 500b, the server sends the third indication message to the network device. After receiving the third indication message, the network device determines, based on the third indication message, that the first sub-request is associated with the second sub-request. The network device then performs a Relationship Association (RAG) between the first and second sub-requests to obtain the association retrieval result for the first and second sub-requests.

[0142] Optionally, the network device may carry the third indication information and the first sub-request (or the second sub-request) in the same message / signaling / signal; or, the network device may carry the third indication information and the first sub-request (or the second sub-request) in different messages / signaling / signals.

[0143] Optionally, the third indication information may be carried in user assistance information in radio resource control (RRC) signaling, or it may be carried in a newly added medium access control element (MAC CE).

[0144] In one possible implementation, the communication method in this application may include steps 501 to 504, and further include steps 500a and 500b.

[0145] In one possible implementation, a context length threshold K is predefined in the network device. If the sum of the context lengths of the first sub-request and the second sub-request is less than or equal to K, the network device determines that there is a relationship between the first sub-request and the second sub-request; or, if the network device receives N sub-requests, and the sum of the context lengths of the N sub-requests is less than or equal to K, the network device determines that there is a relationship between the N sub-requests.

[0146] Optionally, the context length threshold K can be configured by the server for the network device, or it can be configured by the network device based on one or more of the device capabilities, network environment, or the amount of data in the database.

[0147] Optionally, in this application, the search enhancement request also includes one or more sub-requests (e.g., a third sub-request and / or a fourth sub-request) associated with the first sub-request and / or the second sub-request. Similarly, the network device can also associate RAGs based on the aforementioned associated requests (e.g., one or more of the first, second, third, or fourth sub-requests) to obtain search results. It should be understood that the various sub-requests included in the search enhancement request mentioned above are merely examples and do not constitute a limitation on this application. This application still applies to other sub-requests included in the search enhancement request, and specific limitations are not made here.

[0148] Accordingly, this application also provides related apparatus for implementing the above-described scheme. Please refer to Figure 8, which is a schematic diagram of a communication apparatus 800 provided in an embodiment of this application.

[0149] As shown in Figure 8, the communication device 800 includes a processing unit 802 and a communication unit 803. Optionally, the communication device 800 may further include a storage unit 801 for storing device program code and / or data.

[0150] The communication device 800 can realize the functions of the server in the above method embodiments, and therefore can also achieve the beneficial effects of the above method embodiments. In this application, the communication device 800 can be a server, or an integrated circuit or component inside the server, such as a chip, baseband chip, modem chip, SoC chip containing a modem core, system-in-package (SIP) chip, communication module, chip system, processor, etc.

[0151] For example, in one embodiment, the communication unit 803 is configured to: receive a first task request from a terminal device; send a retrieval enhancement request to a network device, the retrieval enhancement request being obtained based on the first task request; and receive first information from the network device, the first information indicating the retrieval result of the retrieval enhancement request. The processing unit 802 is configured to: obtain a first inference result of the first task request based on the retrieval result and an AI model.

[0152] In one possible design, the communication unit is used to send a retrieval enhancement request to the network device, including:

[0153] If the first condition is met, the communication unit is used to send a retrieval enhancement request to the network device; wherein the first condition includes one or more of the following:

[0154] The accuracy of the second inference result obtained from the first task request based on the AI ​​model is less than the first threshold.

[0155] The first inference time is less than the second threshold, wherein the first inference time is at least the sum of the time required for the network device to obtain the retrieval results and the time required for the AI ​​model to output the inference results.

[0156] In one possible design, the first task request includes first indication information, which is used to indicate a second threshold.

[0157] In one possible design, the communication unit is also used to receive second indication information from the network device, the second indication information being used to indicate the time required for the network device to obtain the retrieval results.

[0158] In one possible design, the search enhancement request includes a first sub-request and a second sub-request, the first sub-request being associated with the second sub-request, and the search result of the search enhancement request indicated by the first information being the associated search result of the first sub-request and the second sub-request.

[0159] In one possible design, the communication unit is also used to send a third indication message to the network device, the third indication message being used to indicate that there is a correlation between the first sub-request and the second sub-request.

[0160] In one possible design, when the communication device 800 is a server or a communication module within a server, the functionality of the processing unit 802 can be implemented by one or more processors. Specifically, the processor may include a modem chip, or a system-on-a-chip (SoC) or SIP chip containing a modem core. The functionality of the communication unit 803 can be implemented by transceiver circuitry.

[0161] In one possible design, when the communication device 800 is a circuit or chip in a server responsible for communication functions, such as a modem chip or a system-on-a-chip (SoC) or SIP chip containing a modem core, the function of the processing unit 802 can be implemented by a circuit system in the aforementioned chip that includes one or more processors or processor cores. The function of the communication unit 803 can be implemented by interface circuits or data transceiver circuits on the aforementioned chip.

[0162] In one possible design, when the communication device 800 is a server or a processing module within a server, the functionality of the processing unit 802 can be implemented by one or more processors. Specifically, the processor may include a GPU, or a system-on-a-chip (SoC) or SIP chip containing a GPU. The functionality of the communication unit 803 can be implemented by transceiver circuitry.

[0163] In one possible design, when the communication device 800 is a circuit or chip in a server responsible for processing functions, such as a GPU or a system-on-a-chip (SoC) or SIP chip containing a GPU, the function of the processing unit 802 can be implemented by a circuit system in the aforementioned chip that includes one or more processors or processor cores. The function of the communication unit 803 can be implemented by interface circuitry or data transceiver circuitry on the aforementioned chip.

[0164] It is understood that the division of units in the above-described device is merely a logical functional division. One function can correspond to one functional unit, or two or more functions can be integrated into one functional unit. In actual implementation, all or some units can be integrated onto a single physical entity, or distributed across different physical entities. Furthermore, the aforementioned functional units can be implemented in hardware, software, or a combination of both. Whether a function is executed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for specific applications, but such implementations should not be considered beyond the scope of this application.

[0165] In one example, the functional unit in any of the above devices may be one or more integrated circuits configured to implement the above methods, such as: one or more application-specific integrated circuits (ASICs), or one or more central processing units (CPUs), one or more microcontroller units (MCUs), one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0166] In one example, storage unit 801 may include random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory and / or registers, etc.

[0167] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, etc.) containing computer-usable program code.

[0168] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0169] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0170] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0171] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between devices or units, and may be electrical, mechanical, or other forms. Whether a function is implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0172] It should be understood that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided in this application, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines.

[0173] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0174] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to it, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0175] In the various embodiments of this application, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions between different embodiments are consistent and can be referenced by each other. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0176] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A communication method characterized by comprising: The method comprises: receiving a first task request from a terminal device; sending a retrieval enhancement request to a network device, the retrieval enhancement request being obtained based on the first task request; receiving first information from the network device, the first information indicating a retrieval result of the retrieval enhancement request; obtaining a first inference result of the first task request based on the retrieval result and an artificial intelligence (AI) model.

2. The method of claim 1, wherein, The communication unit is configured to send the retrieval enhancement request to the network device, including: in a case where a first condition is met, the communication unit is configured to send the retrieval enhancement request to the network device; the first condition includes one or more of the following: an accuracy rate of a second inference result of the first task request obtained based on the AI model is less than a first threshold value; a first inference time length is less than a second threshold value, where the first inference time length is at least a sum of a time length required by the network device to obtain a retrieval result and a time length required by the AI model to output an inference result.

3. The method of claim 2, wherein, The first task request includes first indication information, and the first indication information is used to indicate the second threshold value.

4. The method according to claim 2 or 3, characterized in that, The method further comprises: receiving second indication information from the network device, the second indication information being used to indicate a time length required by the network device to obtain a retrieval result.

5. The method according to any one of claims 1 to 4, characterized in that, The retrieval enhancement request includes a first sub-request and a second sub-request, the first sub-request being associated with the second sub-request, and the retrieval result of the retrieval enhancement request indicated by the first information being an associated retrieval result of the first sub-request and the second sub-request.

6. The method of claim 5, wherein, The method further comprises: sending third indication information to the network device, the third indication information being used to indicate that there is an association between the first sub-request and the second sub-request.

7. A communication device, characterized by The apparatus comprises a communication unit and a processing unit; the communication unit is configured to receive a first task request from a terminal device; the communication unit is further configured to send a retrieval enhancement request to a network device, the retrieval enhancement request being obtained based on the first task request; the communication unit is further configured to receive first information from the network device, the first information indicating a retrieval result of the retrieval enhancement request; the processing unit is configured to obtain a first inference result of the first task request based on the retrieval result and an artificial intelligence (AI) model.

8. The apparatus of claim 7, wherein, The communication unit is configured to send the retrieval enhancement request to the network device, including: in a case where a first condition is met, the communication unit is configured to send the retrieval enhancement request to the network device; the first condition includes one or more of the following: an accuracy rate of a second inference result of the first task request obtained based on the AI model is less than a first threshold value; a first inference time length is less than a second threshold value, where the first inference time length is at least a sum of a time length required by the network device to obtain a retrieval result and a time length required by the AI model to output an inference result.

9. The apparatus of claim 8, wherein, The first task request includes first indication information, and the first indication information is used to indicate the second threshold value.

10. The apparatus of claim 8 or 9, wherein, The communication unit is further configured to receive second indication information from the network device, the second indication information being used to indicate a time length required by the network device to obtain a retrieval result.

11. The apparatus of any one of claims 7 to 10, wherein, The search enhancement request comprises a first sub-request and a second sub-request, the first sub-request is associated with the second sub-request, and the first information indicates that a search result of the search enhancement request is an associated search result of the first sub-request and the second sub-request.

12. The apparatus of claim 11, wherein, The communication unit is further configured to send third indication information to the network device, the third indication information being used to indicate that there is an association between the first sub-request and the second sub-request.

13. A readable storage medium, characterized by, The storage medium stores a computer program or instructions, and when the computer program or instructions are executed by the communication device, the method in any one of claims 1 to 6 is implemented.

14. A computer program product, characterised in that, When the computer program product is run on the computer, the computer is caused to perform the method in any one of claims 1 to 6.

15. A communications device, characterized by The apparatus comprises a unit for performing the method in any one of claims 1 to 6.

16. A communications device, characterized by The apparatus comprises an interface circuit and one or more processors, the one or more processors being coupled with a memory, the memory being configured to store a computer program or instructions, and when the computer program or instructions are executed by the one or more processors, the apparatus is caused to perform the method in any one of claims 1 to 6.

17. The apparatus of claim 16, wherein, The interface circuit is configured to implement a communication function within the apparatus and / or a communication function of the apparatus with other devices or components.

Images