Model processing method and apparatus, and terminal and network-side device

By acquiring and processing the information used for the target model and determining its input data, the problem of poor flexibility in selecting input data of AI model is solved, and the optimization capability and information acquisition capability of AI model in mobile communication systems are improved.

WO2024235080A1PCT designated stage expired Publication Date: 2024-11-21VIVO MOBILE COMM CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
PCT/CN2024/091815
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-16
Filing Date
2024-05-09
Publication Date
2024-11-21

AI Technical Summary

Technical Problem

In the prior art, the selection flexibility of AI model input data is poor, and it is difficult to adapt to the diverse AI use cases and new use cases that emerge in the self-evolution of the system in mobile communication systems.

Method used

By acquiring and processing the first and second information, the mapping relationship between the target data of the target model inference and its input data is determined, and the input data of the target model is then collected and determined to improve the flexibility of the input data of the AI ​​model.

Benefits of technology

It improves the ability of AI models to optimize communication networks and enhances the flexibility and reliability of obtaining effective information in mobile communication systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2024091815_21112024_PF_FP_ABST
    Figure CN2024091815_21112024_PF_FP_ABST
Patent Text Reader

Abstract

The present application belongs to the technical field of communications. Disclosed are a model processing method and apparatus, and a terminal and a network-side device. The model processing method in the embodiments of the present application comprises: a first device acquiring first information and second information, wherein the first information is used for determining target data for reasoning a target model, and the second information is used for representing a mapping relationship between the target data for reasoning the target model and an input of the target model; the first device performing data collection according to the first information, so as to obtain the target data; and on the basis of the second information and the target data, the first device determining input data for the target model.
Need to check novelty before this filing date? Find Prior Art

Description

Model processing method, device, terminal and network side equipment

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to Chinese Patent Application No. 202310551011.8 filed on May 16, 2023, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present application belongs to the field of communication technology, and specifically relates to a model processing method, apparatus, terminal and network-side equipment. Background Art

[0004] In mobile communication systems, there are an increasing number of use cases incorporating artificial intelligence (AI). Examples include AI-based channel state information (CSI) feedback compression, AI-based beam management, AI-based positioning, AI-based energy saving, and AI-based load balancing. Currently, AI use cases are typically predefined, and the input data for the AI ​​model is also predefined. The inference device obtains the model input data required for inference based on a predefined process. Therefore, the existing technology suffers from poor flexibility in selecting AI model input data.

[0005] Summary of the Invention

[0006] The embodiments of the present application provide a model processing method, apparatus, terminal, and network-side equipment, which can solve the problem of poor flexibility in selecting input data for AI models.

[0007] In a first aspect, a model processing method is provided, comprising:

[0008] The first device acquires first information and second information, wherein the first information is used to determine target data for target model inference, and the second information is used to represent a mapping relationship between the target data for the target model inference and an input of the target model;

[0009] The first device collects data according to the first information to obtain the target data;

[0010] The first device determines input data of the target model based on the second information and the target data.

[0011] In a second aspect, a model processing method is provided, comprising:

[0012] The second device sends the first information and the second information to the first device;

[0013] The first information is used to determine target data for target model reasoning, and the second information is used to represent a mapping relationship between the target data for target model reasoning and the input of the target model.

[0014] In a third aspect, a model processing device is provided, comprising:

[0015] an acquisition module, configured to acquire first information and second information, wherein the first information is used to determine target data for target model reasoning, and the second information is used to represent a mapping relationship between the target data for target model reasoning and an input of the target model;

[0016] an acquisition module, configured to acquire data according to the first information and obtain the target data;

[0017] A determination module is used to determine input data of the target model based on the second information and the target data.

[0018] In a fourth aspect, a model processing device is provided, comprising:

[0019] A first sending module, configured to send first information and second information to a first device;

[0020] The first information is used to determine target data for target model reasoning, and the second information is used to represent a mapping relationship between the target data for target model reasoning and the input of the target model.

[0021] In a fifth aspect, a terminal is provided, which includes a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the method described in the first aspect are implemented, or the steps of the method described in the first aspect are implemented.

[0022] In a sixth aspect, a terminal is provided, comprising a processor and a communication interface, wherein:

[0023] When the terminal is a first device, the communication interface is used to obtain first information and second information, the first information being used to determine target data for target model inference, and the second information being used to represent a mapping relationship between the target data for target model inference and an input of the target model; data collection is performed according to the first information to obtain the target data; and the processor is used to determine input data of the target model based on the second information and the target data;

[0024] When the terminal is a second device, the communication interface is used to send first information and second information to the first device; wherein, the first information is used to determine the target data used for target model reasoning, and the second information is used to represent the mapping relationship between the target data used for the target model reasoning and the input of the target model.

[0025] In the seventh aspect, a network side device is provided, which includes a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the program or instructions are executed by the processor, the steps of the method described in the first aspect are implemented, or the steps of the method described in the first aspect are implemented.

[0026] In an eighth aspect, a network side device is provided, including a processor and a communication interface, wherein:

[0027] When the network-side device is a first device, the communication interface is used to obtain first information and second information, the first information being used to determine target data for target model inference, and the second information being used to represent a mapping relationship between the target data for target model inference and an input of the target model; data collection is performed according to the first information to obtain the target data; and the processor is used to determine input data of the target model based on the second information and the target data;

[0028] When the network side device is a second device, the communication interface is used to send first information and second information to the first device; wherein, the first information is used to determine the target data used for target model inference, and the second information is used to represent the mapping relationship between the target data used for the target model inference and the input of the target model.

[0029] In a ninth aspect, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented, or the steps of the method described in the first aspect are implemented.

[0030] In the tenth aspect, a wireless communication system is provided, comprising: a first device and a second device, wherein the first device can be used to execute the steps of the method described in the first aspect, and the second device can be used to execute the steps of the method described in the second aspect.

[0031] In the eleventh aspect, a chip is provided, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the method as described in the first aspect, or to implement the method as described in the second aspect.

[0032] In the twelfth aspect, a computer program / program product is provided, which is stored in a storage medium and is executed by at least one processor to implement the method as described in the first aspect, or to implement the method as described in the second aspect.

[0033] In an embodiment of the present application, a first device obtains first information and second information, wherein the first information is used to determine target data for target model inference, and the second information is used to represent the mapping relationship between the target data for target model inference and the input of the target model; the first device collects data based on the first information to obtain the target data; and the first device determines the input data of the target model based on the second information and the target data. In this way, since the terminal determines the input data of the target model based on the mapping relationship between the target data inferred by the target model and the input of the target model, the flexibility of the AI ​​model input data selection can be improved, thereby better obtaining effective information of the communication network and improving the AI ​​model's optimization capability for the communication network. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] FIG1 is a schematic diagram of a network structure applicable to the present application;

[0035] FIG2 is a flow chart of a model processing method according to an embodiment of the present invention;

[0036] FIG3 is a second flow chart of the model processing method provided in an embodiment of the present application;

[0037] FIG4 is a third flow chart of the model processing method provided in an embodiment of the present application;

[0038] FIG5 is an example diagram of resource measurement in the model processing method provided in an embodiment of the present application;

[0039] FIG6 is a fourth flow chart of the model processing method provided in an embodiment of the present application;

[0040] FIG7 is a structural diagram of a model processing device provided in an embodiment of the present application;

[0041] FIG8 is a structural diagram of another model processing device provided in an embodiment of the present application;

[0042] FIG9 is a structural diagram of a communication device provided in an embodiment of the present application;

[0043] FIG10 is a structural diagram of a terminal provided in an embodiment of the present application;

[0044] FIG11 is a structural diagram of a network-side device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] The terms "first", "second", etc. in this application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way are interchangeable where appropriate, 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 do not limit the number of objects, for example, the first object can be one or more. In addition, "or" in this application represents at least one of the connected objects. For example, "A or B" covers three options, namely, Option 1: including A but not including B; Option 2: including B but not including A; Option 3: including both A and B. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.

[0046] The term "indication" in this application can be either a direct indication (or explicit indication) or an indirect indication (or implicit indication). A direct indication can be understood as the sender explicitly informing the receiver of specific information, the operation to be performed, or the requested result, etc. in the instruction sent; an indirect indication can be understood as the receiver determining the corresponding information based on the instruction sent by the sender, or making a judgment and determining the operation to be performed or the requested result, etc. based on the judgment result.

[0047] It is worth noting that the technology described in the embodiments of the present application is not limited to the Long Term Evolution (LTE) / LTE-Advanced (LTE-A) system, but 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 the present application are often used interchangeably, and the technology described can be used for the systems and radio technologies mentioned above, as well as for other systems and radio technologies. The following description describes a New Radio (NR) system for illustrative purposes, and NR terminology is used in most of the following description, but these technologies can also be applied to systems other than NR systems, such as 6th generation (6G) systems. thGeneration, 6G) communication system.

[0048] FIG1 is a block diagram of a wireless communication system applicable to an embodiment of the present application. The wireless communication system includes a terminal 11 and a network-side device 12. The terminal 11 may be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (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 (Wearable Device), an aircraft (Flight Vehicle), a vehicle-mounted device (VUE), a ship-mounted device, a pedestrian user equipment (PUE), a smart home (home appliances with wireless communication capabilities, such as refrigerators, televisions, washing machines, or furniture), a game console, a personal computer (PC), an ATM, or a self-service machine, or other terminal-side devices. Wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc. Among them, the vehicle-mounted device can also be called a vehicle-mounted terminal, a vehicle-mounted controller, a vehicle-mounted module, a vehicle-mounted component, a vehicle-mounted chip or a vehicle-mounted unit, etc. It should be noted that the specific type of the terminal 11 is not limited in the embodiment of the present application. The network side device 12 may include an access network device or a core network device, wherein the access network device may also be called a radio access network (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 (AS) or a wireless fidelity (WiFi) node, etc.Among them, the base station can 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 (home evolved Node B), Transmission Reception Point (TRP) or other appropriate terms in the field. As long as the same technical effect is achieved, the base station is not limited to specific technical vocabulary. It should be noted that in the embodiment of the present 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.

[0049] The core network equipment may include but is not limited to at least one of the following: core network node, core network function, mobility management entity (MME), access mobility management function (AMF), session management function (SMF), user plane function (UPF), policy control function (PCF), policy and charging rules function unit (PCRF), edge application service discovery function (EASDF), unified data management (UDM), unified data repository (UDR), home user server (HSS), centralized network configuration (CNC), network storage function (NRF), network exposure function (NEF), local NEF (L-NEF), binding support function (BSF), application function ( It should be noted that in the embodiments of the present application, only the core network device in the NR system is introduced as an example, and the specific type of the core network device is not limited.

[0050] At present, AI use cases are usually predefined, and the input data of the AI ​​model is also predefined. The inference device can obtain the model input data required for inference based on the predefined process. However, in the future, more AI use cases will appear in mobile communication systems, especially AI use cases that appear during the self-evolution of the system. For example, assuming that the AI ​​model is sent from the network side to the terminal for inference, the terminal does not know how to generate input data that matches the model inference through data collection. Therefore, it is necessary to provide a universal input data determination solution, and for this purpose, the model processing method of this application is proposed.

[0051] The model processing method provided in the embodiments of the present application is described in detail below through some embodiments and their application scenarios in conjunction with the accompanying drawings.

[0052] 2 , an embodiment of the present application provides a model processing method. As shown in FIG2 , the model processing method includes:

[0053] Step 201: A first device obtains first information and second information, wherein the first information is used to determine target data for target model inference, and the second information is used to represent a mapping relationship between the target data for target model inference and an input of the target model.

[0054] Step 202: The first device collects data according to the first information to obtain the target data;

[0055] Step 203: The first device determines input data of the target model based on the second information and the target data.

[0056] In the embodiments of the present application, the first device can be understood as a model inference device. Specifically, the first device can be a terminal or a network-side device. The first information is information received by the first device from the second device. The second information can be information agreed upon in a protocol or information received from the second device. The second device can be a network-side device or a terminal. For example, the first device is a terminal and the second device is a network-side device; or the first device is a network-side device and the second device is a terminal.

[0057] It should be understood that when the above-mentioned first device is a terminal, the above-mentioned first information can be understood as configuration information, and when the above-mentioned first device is a network side device, the above-mentioned first information can be understood as terminal demand information.

[0058] Optionally, the data acquisition to obtain the target data may include performing resource measurements to obtain measurement results and querying existing data. After obtaining the target data, the association between the target data and the input data may be determined based on the second information, thereby obtaining the input data of the target model. Finally, the target model is inferred based on the obtained input data of the target model to obtain corresponding inference results. That is, in some embodiments, after the first device determines the input data of the target model based on the second information and the target data, the method further includes:

[0059] The first device inputs the input data into the target model for inference to obtain an inference result;

[0060] The first device sends the inference result to the second device.

[0061] It should be noted that the above-mentioned AI features can be understood or replaced by AI use cases, and AI models refer to models used to implement AI features or AI use cases.

[0062] In an embodiment of the present application, a first device obtains first information and second information, wherein the first information is used to determine target data for target model inference, and the second information is used to represent the mapping relationship between the target data for target model inference and the input of the target model; the first device collects data based on the first information to obtain the target data; and the first device determines the input data of the target model based on the second information and the target data. In this way, since the terminal determines the input data of the target model based on the mapping relationship between the target data inferred by the target model and the input of the target model, the flexibility of the AI ​​model input data selection can be improved, thereby better obtaining effective information of the communication network and improving the AI ​​model's optimization capability for the communication network.

[0063] Optionally, in some embodiments, the first information includes at least one of measurement resource information, measurement time, and measurement amount;

[0064] The measurement resource information includes a measurement resource indication and an order of measurement resources.

[0065] Optionally, in some embodiments, the first information further includes the inference delay of the target model and the feedback interval of the inference result of the target model.

[0066] Optionally, the measurement resource indication may include at least one of the following: the entire measurement set, a subset of the measurement set, and the number of measurement resources of the measurement set.

[0067] In an embodiment of the present application, the first device may provide feedback on the model inference result based on the above-mentioned feedback interval after performing model inference.

[0068] Optionally, the data content included in the above-mentioned target data can be set according to actual needs. For example, in some embodiments, the target data includes the measurement result of the measurement quantity of the measurement resource, or the target data includes the measurement result of the measurement quantity of the measurement resource and the identification information of the measurement resource.

[0069] Optionally, in some embodiments, the second information includes a first mapping relationship between the target data and the dataset, and a second mapping relationship between the dataset and the model input, and the first device determines the input data of the target model based on the second information and the target data, including:

[0070] The first device generates a first data set according to the target data and the first mapping relationship;

[0071] The first device determines a second data set based on a valid measurement time window associated with the first data set and the target model;

[0072] The first device determines input data of the target model based on the second data set and the second mapping relationship.

[0073] In an embodiment of the present application, the first data set may include target data, element identifiers, and a time interval between the measurement time and the current time. Optionally, the first device may obtain corresponding elements from the first data set based on a valid measurement time window associated with the target model to obtain a second data set. Alternatively, the first device may delete elements from the first data set based on the valid measurement time window associated with the target model to obtain a second data set.

[0074] It should be noted that, in the embodiment of the present application, the above-mentioned first mapping relationship can be specifically understood as the mapping relationship between the target data and the first data set, and the above-mentioned second mapping relationship can be specifically understood as the mapping relationship between the above-mentioned second data set and the model input of the target model.

[0075] Optionally, the valid measurement time window is used to determine valid measurement data in the first data set. For example, assuming a time measurement period is 40ms, the current time is t=k+2 (2 represents 2x40ms=80ms, k represents 40*N ms, and N is a natural number), and the valid measurement time window is 45ms. The first data set may include target data measured at time k, target data measured at time k+1, and target data measured at time k+2, wherein the target data measured at time k+1 and the target data measured at time k+2 can be understood as valid measurement data. The second data set obtained in this case includes the target data measured at time k+1 and the target data measured at time k+2.

[0076] It should be understood that in the embodiments of the present application, the second data set can be directly used as the input data of the target model, or the data in the second data set can be preprocessed to obtain the input data of the target model. For example, in some embodiments, the second mapping relationship includes any of the following:

[0077] A mapping relationship between each element in the data set and a model input identifier of the target model;

[0078] a mapping relationship between each element in the data set and a model input identifier of the target model, and pre-processing instructions for the elements in the data set;

[0079] A mapping relationship between each element in the data set and a model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on target information in the data set.

[0080] In the embodiment of the present application, the above-mentioned data set can be understood as the above-mentioned first data set or the above-mentioned second data set. For example, when the element identifiers in the first data set and the second data set are the same, the first data set can be equivalent to the second data set. When the above-mentioned data set is understood as the second data set, the above-mentioned second mapping relationship can include any of the following:

[0081] a mapping relationship between each element in the second data set and a model input identifier of the target model;

[0082] a mapping relationship between each element in the second data set and a model input identifier of the target model, and a preprocessing instruction for the elements in the second data set;

[0083] A mapping relationship between each element in the second data set and a model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on the target information in the second data set.

[0084] In the embodiment of the present application, when the second mapping relationship includes a mapping relationship between each element in the data set and the model input identifier of the target model, it can be understood as: directly mapping the elements in the data set to the input data of the target model.

[0085] When the second mapping relationship includes the mapping relationship between each element in the data set and the model input identifier of the target model, as well as the preprocessing indication of the elements in the data set, it can be understood that: after the elements in the data set are preprocessed accordingly, they are mapped as input data of the target model.

[0086] When the second mapping relationship includes a mapping relationship between each element in the data set and a model input identifier, and an indirect mapping method, and the indirect mapping method is used to indicate that the model input identifier is determined based on the target information in the data set, it can be understood that: the model input has a default value (or can be understood as the default value of the model input), and the model input identifier for inputting a new value is determined by the target information. For example, the input data of the above-mentioned target model is obtained by a form filling method. At this time, each element in the data set can also be further preprocessed. Optionally, the default value can be agreed upon by a protocol or indicated by a network-side device. For example, the above-mentioned second information also includes the default value or indication information of the default value.

[0087] Optionally, in some embodiments, the first device determining the input data of the target model based on the second information and the target data includes:

[0088] The first device determines input data of the target model based on the second information, the target reference time and the target data;

[0089] The target reference time is used to represent the reference time for model inference. The target reference time can be indicated by the second device. For example, when the terminal acts as an inference device, the target reference time can be included in the first information, or carried by a separate message, such as a Radio Resource Control (RRC) message, a Medium Access Control Control Element (MAC CE), a Downlink Control Information (DCI), or a newly defined AI message.

[0090] In the embodiment of the present application, there is no need to predetermine the corresponding data set, and the input data of the target model is directly determined in the target data based on the second information and the target reference time.

[0091] For example, assuming the target reference time is t=k+2, a time measurement period is 40ms, and the valid measurement time window is 45ms. Assume that the currently acquired target data may include target data measured at time k, target data measured at time k+1, and target data measured at time k+2, wherein the target data measured at time k+1 and the target data measured at time k+2 can be understood as valid measurement data. In this case, the input data of the target model can be determined based on the target data measured at time k+1 and the target data measured at time k+2.

[0092] It should be understood that in the embodiment of the present application, the second information includes any one of the following:

[0093] A mapping relationship between the target data and a model input identifier of the target model;

[0094] a mapping relationship between the target data and a model input identifier of the target model, and a preprocessing instruction for the target data;

[0095] The mapping relationship between the target data and the model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on the target information in the target data.

[0096] In the embodiment of the present application, when the second information includes a mapping relationship between the target data and the model input identifier of the target model, it can be understood that the target data is directly mapped to the input data of the target model.

[0097] When the second information includes a mapping relationship between the target data and the model input identifier of the target model, and a preprocessing instruction of the target data, it can be understood that: after the target data is preprocessed accordingly, it is mapped to the input data of the target model.

[0098] When the second information includes a mapping relationship between the target data and a model input identifier, and an indirect mapping method indicating that the model input identifier is determined based on target information in the target data, it can be understood that the model input has a default value (or can be understood as a default value for the model input), and the model input identifier for inputting a new value is determined by the target information. For example, a table-filling method is used to obtain the input data for the target model. At this point, further preprocessing operations can also be performed on the target.

[0099] Optionally, in some embodiments, the first information further includes a virtual identifier, and the target information includes at least one of identification information of the measurement resource, the measurement time, and the virtual identifier.

[0100] In an embodiment of the present application, the virtual identifier is associated with at least one of the identification information of the measurement resource and the measurement time. For example, in some embodiments, the virtual identifier is an identifier corresponding to the measurement time and the current time interval.

[0101] Optionally, in some embodiments, the preprocessing indication includes at least one of the following: a standardization indication, a normalization indication, an eigenvalue decomposition indication, a differential privacy indication, and a one-hot code indication.

[0102] Optionally, in some embodiments, the first device acquiring the first information and the second information includes:

[0103] The first device receives the first information and the second information from a network-side device.

[0104] In an embodiment of the present application, since the second device indicates the first information and the second information to the first device, it can be determined that the input data is more suitable for the current network environment, thereby further improving the reliability of model reasoning.

[0105] In order to better understand the present application, some examples are provided below for illustration.

[0106] In some embodiments, the terminal may obtain target data based on the first information, and then obtain input data of the target model based on the second information. As shown in FIG3 , the process includes the following steps:

[0107] 31. The network-side device sends the first information and the second information to the terminal.

[0108] 32. The terminal collects data according to the first information to obtain target data, and then stores the target data as a model input data set based on the second information, that is, a set of input data for the target model;

[0109] 33. The terminal inputs the model input data set into the target model, performs model inference, and obtains the inference result.

[0110] 34. The terminal feeds back the inference result to the network side device.

[0111] Optionally, the first information is used to determine the target data for target model reasoning, for example, the first information includes measurement resource information and measurement time and measurement amount; the second information is used to represent the mapping relationship between the target data for target model reasoning and the input of the target model

[0112] Optionally, the first information may further include the inference delay of the target model and the feedback interval of the inference result of the target model.

[0113] In some embodiments, the network side device can obtain target data based on the first information, and then obtain input data of the target model based on the second information. As shown in Figure 4, the process includes the following:

[0114] 41. The terminal sends the first information and the second information to the network side device.

[0115] 42. The network-side device collects data based on the first information to obtain target data, and then stores the target data as a model input data set based on the second information, i.e., a set of input data for the target model;

[0116] 43. The network side device inputs the model input data set into the target model, performs model inference, and obtains the inference result.

[0117] 44. The network side device feeds back the inference result to the terminal.

[0118] Optionally, the first information is used to determine the target data used for target model reasoning, for example, the first information includes measurement resource information, measurement time and measurement amount; the second information is used to represent the mapping relationship between the target data used for the target model reasoning and the input of the target model.

[0119] Optionally, the first information may further include the inference delay of the target model and the feedback interval of the inference result of the target model.

[0120] Optionally, in some embodiments, the terminal may collect data based on the first information and then store the data.

[0121] Assume that the terminal performs resource measurement in the resource measurement sequence shown in Figure 5 and obtains corresponding measurement results. The measurement results obtained by performing the resource measurement at time k include Layer 1 Reference Signal Receiving Power 0 (L1-RSRP 0), L1-RSRP 2, L1-RSRP 4, and L1-RSRP 6. The measurement results obtained by performing the resource measurement at time k+1 include L1-RSRP 1, L1-RSRP 3, L1-RSRP 5, and L1-RSRP 7. The measurement results obtained by performing the resource measurement at time k+2 include L1-RSRP 8, L1-RSRP 10, L1-RSRP 12, and L1-RSRP 14. The measurement results obtained by performing the resource measurement at time k+3 include L1-RSRP 9, L1-RSRP 11, L1-RSRP 13, and L1-RSRP 15.

[0122] Optionally, in some embodiments, data may be stored in multiple dimensions. Assuming that the current time is t=k+2, a time measurement period is 40ms, and the effective measurement time window is 45ms, the obtained second data set is shown in Table 1.

[0123] Table 1:

[0124] Optionally, assuming that the current time is t=k+3, a time measurement period is 40 ms, and the effective measurement time window is 45 ms, the obtained second data set is shown in Table 2.

[0125] Table 2:

[0126] It should be noted that, in the embodiment of the present application, the terminal may perform the following process:

[0127] The terminal collects data according to the first information (specifically including performing measurement at the configured measurement resources and measurement time), obtains a measurement result of the corresponding measurement quantity, and generates a first data set according to the first mapping relationship in the second information;

[0128] The terminal deletes elements of the first data set according to the effective measurement time window of the target model to obtain a second data set.

[0129] Optionally, the terminal may adjust the element identifiers in the data set according to the measurement time and measurement sequence.

[0130] Optionally, in some embodiments, the data can be stored in one dimension. Assuming the current time is t = k + 2, a time measurement period is 40 ms, and the effective measurement time window is 45 ms, the parameters related to the measurement quantity indicated by the first information (such as the resource identifier of the measurement resource) are used as inputs to the model. In this case, the obtained second data set is shown in Table 3.

[0131] Table 3:

[0132] It should be noted that, in the embodiment of the present application, the terminal may perform the following process:

[0133] The terminal collects data according to the first information (specifically, performing measurement at the configured measurement resource and measurement time) to obtain a measurement result of the corresponding measurement quantity;

[0134] The terminal generates a first data set according to the first mapping relationship in the measurement quantity, the parameter associated with the measurement quantity, and the second information;

[0135] The terminal deletes elements of the first data set according to the effective measurement time window of the target model to obtain a second data set.

[0136] Optionally, in some embodiments, the terminal obtains input data of the model according to a second mapping relationship between the second data set and the second information.

[0137] Taking Table 3 as an example, assuming that the current time is t=k+2, the corresponding input data is determined based on the following Table 4.

[0138] Table 4:

[0139] Taking Table 2 as an example, assuming that the current time is t=k+2, the corresponding input data at this time is determined based on the following Table 5.

[0140] Table 5:

[0141] The index (2, 1) represents the first row of data in the second column of valid data in the second data set, namely L1-RSRP 9. In Table 1 and Table 2, the second data set includes two columns of valid data, such as the resource identifier and the measurement value of the measured resource.

[0142] The valid data of Table 2 are shown in the following table:

[0143] It should be noted that, in the embodiment of the present application, the second mapping relationship in the above-mentioned second information includes the mapping relationship between each element in the second data set and the model input identifier of the target model, or in other words, the second mapping relationship includes the corresponding relationship between each AI unit input (i.e., the input of one dimension corresponding to each model input identifier of the model) and the element in the second data set. Specifically, the corresponding relationship may include:

[0144] The association relationship between each AI unit input and each element in the column where the valid value of the first data set is located;

[0145] Alternatively, the column containing valid values ​​in the second data set that can be used as model input and the association between the valid values ​​and the model input identifiers.

[0146] It should be understood that the second dataset needs to include time-related information for dataset updates. However, time information does not need to be directly input into the model. Therefore, it is not a valid value.

[0147] Optionally, in some embodiments, the terminal obtains input data of the model according to the second mapping relationship between the second data set and the second information, including a preprocessing scheme.

[0148] In an embodiment of the present application, the above-mentioned second information includes a second mapping relationship, that is, a mapping relationship between each element in the second data set and the model input identifier of the target model, as well as a preprocessing indication and a preprocessing object indication of the elements in the second data set.

[0149] Optionally, the preprocessing indication may include at least one of the following: a standardization indication, a normalization indication, an eigenvalue decomposition indication, a differential privacy indication, and a one-hot coding indication.

[0150] Assume that the preprocessing object is a measurement resource identifier, the preprocessing indicator is a one-hot code indicator, and the length of the one-hot code indicator is 16. Taking Table 1 as an example, assuming that the current time is t=k+2, the corresponding input data at this time is shown in the preprocessed values ​​in Table 6 below.

[0151] Table 6:

[0152] Optionally, in some embodiments, the terminal obtains input data of the model according to the second data set and the second information, including a form-filling method.

[0153] In an embodiment of the present application, the second information includes a second mapping relationship, i.e., a mapping relationship between each element in the second data set and a model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on the target information in the second data set. Furthermore, the second information also includes a default value (i.e., a default value for the model input), such as 0.

[0154] It is assumed that the model input identity is determined by the resource identity of the measurement resource and the time interval between the measurement time and the target reference time.

[0155] Taking Table 1 as an example, assuming that the current time is t=k+2, the corresponding input data at this time is shown as the pre-processed values ​​in Table 7 below.

[0156] Table 7:

[0157] The model input identifier (0, 8) is represented by the first dimension, 0, which indicates the virtual identifier corresponding to the measurement time and the current time interval. The second dimension, 8, is determined based on the resource identifier of the measurement resource. Assuming the default value is 0, the final model input is as shown in the following table:

[0158] Optionally, in some embodiments, the input of the model may be obtained based on the second information and the target data.

[0159] Assuming that the target reference time is t=k+2, the second information includes a mapping relationship between the target data and the model input identifier of the target model. The input data of the target model is shown in Table 8.

[0160] Table 8:

[0161] In an embodiment of the present application, the input data of the target model can be determined based on the first information, the second information, and the target reference time. Specifically, the process may include: the terminal collects data based on the first information; and then determines the input data of the target model based on the second information and the target reference time.

[0162] Optionally, the target reference time may be sent by the network side device to the terminal.

[0163] Optionally, when the terminal acts as an inference device, the target reference time may be included in the first information, or carried by a separate message, such as carried in an RRC message, MAC CE, DCI, or a newly defined AI message.

[0164] Optionally, in some embodiments, the terminal obtains input data of the model according to the target data and the second information, including a preprocessing scheme.

[0165] In this embodiment of the present application, it is assumed that the second information also includes a preprocessing instruction. The preprocessing object is the measurement resource identifier, and the processing instruction is a one-hot code instruction. The length of the one-hot code instruction is 16. Assuming that the current time is t = k + 2, the corresponding input data at this time is shown as the preprocessed values ​​in Table 9 below.

[0166] Table 9:

[0167] In the embodiment of the present application, the terminal may collect data according to the first information; and then determine the input data of the target model according to the second information and the target reference time.

[0168] Optionally, the target reference time may be sent by the network side device to the terminal.

[0169] Optionally, when the terminal acts as an inference device, the target reference time may be included in the first information, or carried by a separate message, such as carried in an RRC message, MAC CE, DCI, or a newly defined AI message.

[0170] Optionally, in some embodiments, the terminal obtains input data of the model according to the target data and the second information, including a form-filling method.

[0171] The mapping relationship between the target data and the model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on the target information in the target data.

[0172] Assuming that the model input identifier is determined by the resource identifier of the measurement resource and the time interval between the measurement time and the target reference time, and the current time is t=k+2, the corresponding input data at this time is shown in Table 10 below.

[0173] Table 10:

[0174] Among them, 0 in the model input identifier (0, 8) is the identifier of the first dimension, which represents the virtual identifier corresponding to the measurement time and the target reference time interval, and 8 is the identifier of the second dimension, which is determined based on the resource identifier of the measurement resource.

[0175] Assuming the default value is 0, the final model input is as follows:

[0176] 6 , an embodiment of the present application further provides a model processing method. As shown in FIG6 , the model processing method includes:

[0177] Step 601: The second device sends first information and second information to the first device;

[0178] The first information is used to determine target data for target model reasoning, and the second information is used to represent a mapping relationship between the target data for target model reasoning and the input of the target model.

[0179] Optionally, the first information includes at least one of measurement resource information, measurement time, and measurement amount;

[0180] The measurement resource information includes a measurement resource indication and an order of measurement resources.

[0181] Optionally, the first information also includes the inference delay of the target model and the feedback interval of the inference result of the target model.

[0182] Optionally, the target data includes a measurement result of a measurement quantity of the measurement resource, or the target data includes a measurement result of the measurement quantity of the measurement resource and identification information of the measurement resource.

[0183] Optionally, the second information includes a first mapping relationship between the target data and the dataset, and a second mapping relationship between the dataset and the model input, where the second mapping relationship includes any one of the following:

[0184] A mapping relationship between each element in the data set and a model input identifier of the target model;

[0185] a mapping relationship between each element in the data set and a model input identifier of the target model, and pre-processing instructions for the elements in the data set;

[0186] A mapping relationship between each element in the data set and a model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on target information in the data set.

[0187] Optionally, the second information includes any one of the following:

[0188] A mapping relationship between the target data and a model input identifier of the target model;

[0189] a mapping relationship between the target data and a model input identifier of the target model, and a preprocessing instruction for the target data;

[0190] The mapping relationship between the target data and the model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on the target information in the target data.

[0191] Optionally, the first information further includes a virtual identifier, and the target information includes at least one of identification information of the measurement resource, a measurement time, and the virtual identifier.

[0192] Optionally, the preprocessing indication includes at least one of the following: a standardization indication, a normalization indication, an eigenvalue decomposition indication, a differential privacy indication, and a one-hot code indication.

[0193] The model processing method provided in the embodiment of the present application can be executed by a model processing device. In the embodiment of the present application, the model processing device provided in the embodiment of the present application is described by taking the execution of the model processing method by the model processing device as an example.

[0194] 7 , an embodiment of the present application further provides a model processing device. As shown in FIG7 , the model processing device 700 includes:

[0195] An acquisition module 701 is configured to acquire first information and second information, wherein the first information is used to determine target data for target model inference, and the second information is used to represent a mapping relationship between the target data for target model inference and an input of the target model;

[0196] A collection module 702 is configured to collect data based on the first information to obtain the target data;

[0197] The determination module 703 is configured to determine input data of the target model based on the second information and the target data.

[0198] Optionally, the first information includes at least one of measurement resource information, measurement time, and measurement amount;

[0199] The measurement resource information includes a measurement resource indication and an order of measurement resources.

[0200] Optionally, the first information also includes the inference delay of the target model and the feedback interval of the inference result of the target model.

[0201] Optionally, the target data includes a measurement result of a measurement quantity of the measurement resource, or the target data includes a measurement result of the measurement quantity of the measurement resource and identification information of the measurement resource.

[0202] Optionally, the second information includes a first mapping relationship between the target data and the data set, and a second mapping relationship between the data set and the model input. The determining module 703 is specifically configured to perform the following operations:

[0203] generating a first data set according to the target data and the first mapping relationship;

[0204] determining a second data set according to a valid measurement time window associated with the first data set and the target model;

[0205] The input data of the target model is determined based on the second data set and the second mapping relationship.

[0206] Optionally, the second mapping relationship includes any one of the following:

[0207] A mapping relationship between each element in the data set and a model input identifier of the target model;

[0208] a mapping relationship between each element in the data set and a model input identifier of the target model, and pre-processing instructions for the elements in the data set;

[0209] A mapping relationship between each element in the data set and a model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on target information in the data set.

[0210] Optionally, the determining module 703 is specifically configured to determine input data of the target model based on the second information, the target reference time and the target data;

[0211] The target reference time is used to represent the reference time for model inference.

[0212] Optionally, the second information includes any one of the following:

[0213] A mapping relationship between the target data and a model input identifier of the target model;

[0214] a mapping relationship between the target data and a model input identifier of the target model, and a preprocessing instruction for the target data;

[0215] The mapping relationship between the target data and the model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on the target information in the target data.

[0216] Optionally, the first information further includes a virtual identifier, and the target information includes at least one of identification information of the measurement resource, a measurement time, and the virtual identifier.

[0217] Optionally, the preprocessing indication includes at least one of the following: a standardization indication, a normalization indication, an eigenvalue decomposition indication, a differential privacy indication, and a one-hot code indication.

[0218] Optionally, the acquisition module 701 is specifically configured to receive the first information and the second information from a network-side device.

[0219] Optionally, the model processing device 700 further includes:

[0220] An inference module, configured to input the input data into the target model for inference to obtain an inference result;

[0221] The second sending module is used to send the inference result to the second device.

[0222] 8 , an embodiment of the present application further provides a model processing device. As shown in FIG8 , the model processing device 800 includes:

[0223] A first sending module 801 is configured to send first information and second information to a first device;

[0224] The first information is used to determine target data for target model reasoning, and the second information is used to represent a mapping relationship between the target data for target model reasoning and the input of the target model.

[0225] Optionally, the first information includes at least one of measurement resource information, measurement time, and measurement amount;

[0226] The measurement resource information includes a measurement resource indication and an order of measurement resources.

[0227] Optionally, the first information also includes the inference delay of the target model and the feedback interval of the inference result of the target model.

[0228] Optionally, the target data includes a measurement result of a measurement quantity of the measurement resource, or the target data includes a measurement result of the measurement quantity of the measurement resource and identification information of the measurement resource.

[0229] Optionally, the second information includes a first mapping relationship between the target data and the dataset, and a second mapping relationship between the dataset and the model input, where the second mapping relationship includes any one of the following:

[0230] A mapping relationship between each element in the data set and a model input identifier of the target model;

[0231] a mapping relationship between each element in the data set and a model input identifier of the target model, and pre-processing instructions for the elements in the data set;

[0232] A mapping relationship between each element in the data set and a model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on target information in the data set.

[0233] Optionally, the second information includes any one of the following:

[0234] A mapping relationship between the target data and a model input identifier of the target model;

[0235] a mapping relationship between the target data and a model input identifier of the target model, and a preprocessing instruction for the target data;

[0236] The mapping relationship between the target data and the model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on the target information in the target data.

[0237] Optionally, the first information further includes a virtual identifier, and the target information includes at least one of identification information of the measurement resource, a measurement time, and the virtual identifier.

[0238] Optionally, the preprocessing indication includes at least one of the following: a standardization indication, a normalization indication, an eigenvalue decomposition indication, a differential privacy indication, and a one-hot code indication.

[0239] The model processing device in the embodiment of the present application can be an electronic device, such as an electronic device with an operating system, or a component in an electronic device, such as an integrated circuit or chip. The electronic device can be a terminal, or it can be other devices other than a terminal. For example, the terminal can include but is not limited to the types of terminal 11 listed above, and other devices can be servers, network attached storage (NAS), etc., which are not specifically limited in the embodiment of the present application.

[0240] The model processing device provided in the embodiment of the present application can implement the various processes implemented in the method embodiments of Figures 2 to 6 and achieve the same technical effects. To avoid repetition, they will not be described here.

[0241] As shown in Figure 9, an embodiment of the present application also provides a communication device 900, including a processor 901 and a memory 902, and the memory 902 stores a program or instruction that can be run on the processor 901. When the program or instruction is executed by the processor 901, the various steps of the above-mentioned model processing method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0242] The present application also provides a terminal including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is configured to execute a program or instruction to implement the steps of the method embodiment shown in Figure 2 or Figure 6. This terminal embodiment corresponds to the above-mentioned terminal-side method embodiment, and each implementation process and implementation method of the above-mentioned method embodiment can be applied to this terminal embodiment and can achieve the same technical effect. Specifically, Figure 10 is a schematic diagram of the hardware structure of a terminal implementing an embodiment of the present application.

[0243] The terminal 1000 includes but is not limited to: a radio frequency unit 1001, a network module 1002, an audio output unit 1003, an input unit 1004, a sensor 1005, a display unit 1006, a user input unit 1007, an interface unit 1008, a memory 1009 and at least some of the components of the processor 1010.

[0244] Those skilled in the art will appreciate that the terminal 1000 may also include a power supply (such as a battery) to power various components. The power supply may be logically connected to the processor 1010 via a power management system, thereby enabling the power management system to manage charging, discharging, and power consumption. The terminal structure shown in FIG10 does not limit the terminal. The terminal may include more or fewer components than shown, or may combine certain components, or have different component arrangements, which will not be described in detail here.

[0245] It should be understood that in an embodiment of the present application, the input unit 1004 may include a graphics processing unit (GPU) 10041 and a microphone 10042, and the graphics processing unit 10041 processes the image data of a static picture or video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The display unit 1006 may include a display panel 10061, and the display panel 10061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 1007 includes a touch panel 10071 and at least one of other input devices 10072. The touch panel 10071 is also called a touch screen. The touch panel 10071 may include two parts: a touch detection device and a touch controller. Other input devices 10072 may include, but are not limited to, a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and an operating stick, which will not be repeated here.

[0246] In the embodiment of the present application, after receiving downlink data from a network-side device, the RF unit 1001 may transmit the data to the processor 1010 for processing. Furthermore, the RF unit 1001 may send uplink data to the network-side device. Typically, the RF unit 1001 includes, but is not limited to, an antenna, an amplifier, a transceiver, a coupler, a low-noise amplifier, a duplexer, and the like.

[0247] The memory 1009 can be used to store software programs or instructions and various data. The memory 1009 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, applications or instructions required for at least one function (such as a sound playback function, an image playback function, etc.). In addition, the memory 1009 may include a volatile memory or a 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 (DDRSDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synchronous link dynamic random access memory (SLDRAM), and a direct memory bus random access memory (DRRAM). The memory 1009 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.

[0248] Processor 1010 may include one or more processing units. Optionally, processor 1010 integrates an application processor and a modem processor. The application processor primarily handles operations related to the operating system, user interface, and application programs, while the modem processor primarily processes wireless communication signals, such as a baseband processor. It is understood that the modem processor may not be integrated into processor 1010.

[0249] When the terminal is a first device, the radio frequency unit 1001 is used to obtain first information and second information, where the first information is used to determine target data for target model inference, and the second information is used to represent a mapping relationship between the target data for target model inference and an input of the target model; and data collection is performed according to the first information to obtain the target data.

[0250] The processor 1010 is configured to determine input data of the target model based on the second information and the target data.

[0251] Optionally, the first information includes at least one of measurement resource information, measurement time, and measurement amount;

[0252] The measurement resource information includes a measurement resource indication and an order of measurement resources.

[0253] Optionally, the first information also includes the inference delay of the target model and the feedback interval of the inference result of the target model.

[0254] Optionally, the target data includes a measurement result of a measurement quantity of the measurement resource, or the target data includes a measurement result of the measurement quantity of the measurement resource and identification information of the measurement resource.

[0255] Optionally, the second information includes a first mapping relationship between the target data and the data set, and a second mapping relationship between the data set and the model input, and the processor 1010 is specifically configured to perform the following operations:

[0256] generating a first data set according to the target data and the first mapping relationship;

[0257] determining a second data set according to a valid measurement time window associated with the first data set and the target model;

[0258] The input data of the target model is determined based on the second data set and the second mapping relationship.

[0259] Optionally, the second information includes a first mapping relationship between the target data and the dataset, and a second mapping relationship between the dataset and the model input, where the second mapping relationship includes any one of the following:

[0260] A mapping relationship between each element in the data set and a model input identifier of the target model;

[0261] a mapping relationship between each element in the data set and a model input identifier of the target model, and pre-processing instructions for the elements in the data set;

[0262] A mapping relationship between each element in the data set and a model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on target information in the data set.

[0263] Optionally, the processor 1010 is specifically configured to determine input data of the target model based on the second information, the target reference time and the target data;

[0264] The target reference time is used to represent the reference time for model inference.

[0265] Optionally, the second information includes any one of the following:

[0266] A mapping relationship between the target data and a model input identifier of the target model;

[0267] a mapping relationship between the target data and a model input identifier of the target model, and a preprocessing instruction for the target data;

[0268] The mapping relationship between the target data and the model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on the target information in the target data.

[0269] Optionally, the first information further includes a virtual identifier, and the target information includes at least one of identification information of the measurement resource, a measurement time, and the virtual identifier.

[0270] Optionally, the preprocessing indication includes at least one of the following: a standardization indication, a normalization indication, an eigenvalue decomposition indication, a differential privacy indication, and a one-hot code indication.

[0271] Optionally, the radio frequency unit 1001 is specifically configured to receive the first information and the second information from a network side device.

[0272] Optionally, the processor 1010 is further configured to input the input data into the target model for inference to obtain an inference result;

[0273] The radio frequency unit 1001 is further configured to send the inference result to the second device.

[0274] When the terminal is a second device, the radio frequency unit 1001 is used to send first information and second information to the first device; wherein, the first information is used to determine the target data used for target model inference, and the second information is used to represent the mapping relationship between the target data used for the target model inference and the input of the target model.

[0275] Optionally, the first information includes at least one of measurement resource information, measurement time, and measurement amount;

[0276] The measurement resource information includes a measurement resource indication and an order of measurement resources.

[0277] Optionally, the first information also includes the inference delay of the target model and the feedback interval of the inference result of the target model.

[0278] Optionally, the target data includes a measurement result of a measurement quantity of the measurement resource, or the target data includes a measurement result of the measurement quantity of the measurement resource and identification information of the measurement resource.

[0279] Optionally, the second information includes any one of the following:

[0280] A mapping relationship between each element in the second data set and a model input identifier of the target model;

[0281] a mapping relationship between each element in the second data set and a model input identifier of the target model, and a preprocessing instruction for the elements in the second data set;

[0282] A mapping relationship between each element in the second data set and a model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on the target information in the second data set;

[0283] The second data set is determined based on a valid measurement time window associated with the target data and the target model.

[0284] Optionally, the second information includes any one of the following:

[0285] A mapping relationship between the target data and a model input identifier of the target model;

[0286] a mapping relationship between the target data and a model input identifier of the target model, and a preprocessing instruction for the target data;

[0287] The mapping relationship between the target data and the model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on the target information in the target data.

[0288] Optionally, the first information further includes a virtual identifier, and the target information includes at least one of identification information of the measurement resource, a measurement time, and the virtual identifier.

[0289] Optionally, the preprocessing indication includes at least one of the following: a standardization indication, a normalization indication, an eigenvalue decomposition indication, a differential privacy indication, and a one-hot code indication.

[0290] It can be understood that the implementation process of each implementation method mentioned in this embodiment can refer to the relevant description of the model processing method embodiment and achieve the same or corresponding technical effects. To avoid repetition, it will not be repeated here.

[0291] The present application also provides a network-side device, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is configured to execute a program or instruction to implement the steps of the method embodiment shown in Figure 2 or Figure 6. This network-side device embodiment corresponds to the aforementioned network-side device method embodiment, and each implementation process and implementation method of the aforementioned method embodiment is applicable to this network-side device embodiment and can achieve the same technical effects.

[0292] Specifically, embodiments of the present application also provide a network-side device. As shown in Figure 11, the network-side device 1100 includes an antenna 111, a radio frequency device 112, a baseband device 113, a processor 114, and a memory 115. Antenna 111 is connected to radio frequency device 112. In the uplink direction, radio frequency device 112 receives information via antenna 111 and sends the received information to baseband device 113 for processing. In the downlink direction, baseband device 113 processes the information to be transmitted and sends it to radio frequency device 112. Radio frequency device 112 processes the received information and then sends it through antenna 111.

[0293] The method executed by the network-side device in the above embodiment may be implemented in the baseband device 113 , which includes a baseband processor.

[0294] The baseband device 113 may include, for example, at least one baseband board, on which multiple chips are arranged, as shown in Figure 11, one of the chips is, for example, a baseband processor, which is connected to the memory 115 through a bus interface to call the program in the memory 115 to execute the network device operations shown in the above method embodiment.

[0295] The network side device may further include a network interface 116, which is, for example, a Common Public Radio Interface (CPRI).

[0296] Specifically, the network side device 1100 of the embodiment of the present application also includes: instructions or programs stored in the memory 115 and executable on the processor 114. The processor 114 calls the instructions or programs in the memory 115 to execute the methods executed by the modules shown in FIG. 7 or FIG. 8 and achieve the same technical effect. To avoid repetition, they will not be elaborated here.

[0297] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the above-mentioned model processing method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0298] The processor is the processor in the terminal described in the above embodiment. The readable storage medium includes a computer-readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk. In some examples, the readable storage medium may be a non-transitory readable storage medium.

[0299] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned model processing method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0300] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0301] An embodiment of the present application further provides 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 of the above-mentioned model processing method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0302] An embodiment of the present application also provides a wireless communication system, including: a first device and a second device, wherein the first device can be used to execute the steps of the method described on the first device side, and the second device can be used to execute the steps of the method described on the second device side.

[0303] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the 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 opposite order according to the functions involved. For example, the described method may be performed in an order different from that described, and various steps may also be added, omitted or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0304] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of a computer software product plus a necessary general-purpose hardware platform, or of course, by hardware. The computer software product is stored in a storage medium (such as ROM, RAM, magnetic disk, optical disk, etc.) and includes a number of instructions for enabling a terminal or network-side device to execute the methods described in each embodiment of the present application.

[0305] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms of implementation methods without departing from the purpose of this application and the scope of protection of the claims. These implementation methods are all within the protection of this application.

Claims

1. A model processing method, comprising: The first device acquires first information and second information, wherein the first information is used to determine target data for target model reasoning, and the second information is used to represent a mapping relationship between the target data for target model reasoning and an input of the target model; The first device collects data according to the first information to obtain the target data; The first device determines input data of the target model based on the second information and the target data.

2. The method according to claim 1, wherein: The first information includes at least one of measurement resource information, measurement time and measurement amount; The measurement resource information includes a measurement resource indication and an order of measurement resources.

3. The method according to claim 2, wherein: The first information also includes the inference delay of the target model and the feedback interval of the inference result of the target model.

4. The method according to claim 2, wherein: The target data includes a measurement result of a measurement quantity of the measurement resource, or the target data includes a measurement result of the measurement quantity of the measurement resource and identification information of the measurement resource.

5. The method according to any one of claims 1 to 4, wherein: The second information includes a first mapping relationship between the target data and the data set, and a second mapping relationship between the data set and the model input, and the first device determines the input data of the target model based on the second information and the target data, including: The first device generates a first data set according to the target data and the first mapping relationship; The first device determines a second data set according to a valid measurement time window associated with the first data set and the target model; The first device determines input data of the target model based on the second data set and the second mapping relationship.

6. The method according to claim 5, wherein: The second mapping relationship includes any one of the following: A mapping relationship between each element in the data set and a model input identifier of the target model; a mapping relationship between each element in the data set and a model input identifier of the target model, and a preprocessing indication of the elements in the data set; A mapping relationship between each element in the data set and a model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on target information in the data set.

7. The method according to any one of claims 1 to 4, wherein: The first device determines the input data of the target model based on the second information and the target data, including: The first device determines input data of the target model based on the second information, the target reference time and the target data; The target reference time is used to represent the reference time for model reasoning.

8. The method according to claim 7, wherein: The second information includes any one of the following: A mapping relationship between the target data and a model input identifier of the target model; a mapping relationship between the target data and a model input identifier of the target model, and a preprocessing instruction for the target data; The mapping relationship between the target data and the model input identifier, and the indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on the target information in the target data.

9. The method according to claim 6 or 8, wherein: The first information further includes a virtual identifier, and the target information includes at least one of identification information of a measurement resource, a measurement time, and the virtual identifier.

10. The method according to claim 6 or 8, wherein: The preprocessing indication includes at least one of the following: a standardization indication, a normalization indication, an eigenvalue decomposition indication, a differential privacy indication, and a one-hot code indication.

11. A model processing method, comprising: The second device sends the first information and the second information to the first device; The first information is used to determine target data for target model reasoning, and the second information is used to represent a mapping relationship between the target data for target model reasoning and the input of the target model.

12. The method according to claim 11, wherein: The first information includes at least one of measurement resource information, measurement time and measurement amount; The measurement resource information includes a measurement resource indication and an order of measurement resources.

13. The method according to claim 12, wherein: The first information also includes the inference delay of the target model and the feedback interval of the inference result of the target model.

14. The method according to claim 12, wherein: The target data includes a measurement result of a measurement quantity of the measurement resource, or the target data includes a measurement result of the measurement quantity of the measurement resource and identification information of the measurement resource.

15. The method according to any one of claims 11 to 14, wherein: The second information includes a first mapping relationship between the target data and the data set, and a second mapping relationship between the data set and the model input, wherein the second mapping relationship includes any one of the following: A mapping relationship between each element in the data set and a model input identifier of the target model; a mapping relationship between each element in the data set and a model input identifier of the target model, and a preprocessing indication of the elements in the data set; A mapping relationship between each element in the data set and a model input identifier, and an indirect mapping method, wherein the indirect mapping method is used to indicate that the model input identifier is determined based on target information in the data set.

16. The method according to any one of claims 11 to 14, wherein: The second information includes any one of the following: A mapping relationship between the target data and a model input identifier of the target model; a mapping relationship between the target data and a model input identifier of the target model, and a preprocessing instruction for the target data; The mapping relationship between the target data and the model input identifier, and the indirect mapping method, the indirect mapping method is used The method is used to indicate that the model input identifier is determined based on the target information in the target data.

17. The method according to claim 15 or 16, wherein: The first information further includes a virtual identifier, and the target information includes at least one of identification information of a measurement resource, a measurement time, and the virtual identifier.

18. The method according to claim 15 or 16, wherein: The preprocessing indication includes at least one of the following: a standardization indication, a normalization indication, an eigenvalue decomposition indication, a differential privacy indication, and a one-hot code indication.

19. A model processing device, comprising: An acquisition module, configured to acquire first information and second information, wherein the first information is used to determine target data for target model reasoning, and the second information is used to represent a mapping relationship between the target data for target model reasoning and an input of the target model; A collection module, used for collecting data according to the first information to obtain the target data; A determination module is used to determine input data of the target model based on the second information and the target data.

20. A model processing device, comprising: A first sending module, used to send first information and second information to a first device; The first information is used to determine target data for target model reasoning, and the second information is used to represent a mapping relationship between the target data for target model reasoning and the input of the target model.

21. A terminal comprising a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the model processing method according to any one of claims 1 to 19 are implemented.

22. A network side device, comprising a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the programs or instructions are executed by the processor, the steps of the model processing method according to any one of claims 1 to 19 are implemented.

Citation Information

Patent Citations

  • Neural network reasoning method and device, computer equipment and storage medium

    CN114398040A

  • Deep learning model conversion and inference method and system

    CN114897130A

  • Systems and methods for automated data analysis and customer relationship management

    US20170220943A1

  • Deep learning model reasoning method and apparatus, computer device, and storage medium

    WO2022141489A1