Communication network prediction method, terminal and network side device

Abstract

The application discloses a communication network prediction method, a terminal and a network side device, and belongs to the technical field of communication. The communication network prediction method comprises the following steps: a terminal uses L models to respectively perform a target task, and obtains a first result of L model outputs; wherein L is a positive integer; and the terminal performs any one of the following operations: the terminal determines a prediction result of the target task based on the first result; the terminal sends the first result to a network side device; the terminal receives a second result sent by the network side device, and determines the prediction result of the target task based on the first result and the second result; wherein the second result is obtained by using M models by the network side device to respectively perform the target task; and M is a positive integer.

Description

Technical Field

[0001] This application belongs to the field of communication technology, specifically relating to a communication network prediction method, a terminal, and network-side equipment. Background Technology

[0002] Location information is a crucial type of sensing information. It not only meets diverse business needs, such as sidelinks or smart factories, but also serves as prior information for communication links, helping to improve the overall performance and service experience of communication systems. Wireless communication network positioning involves a terminal estimating its current geographical location by measuring reference signals. Specifically, the terminal measures positioning reference signals from multiple positioning base stations and reports this measurement information to the core network through the serving base station. The core network's positioning management function then performs location estimation. Finally, the core network distributes the terminal's geographical location information back to the terminal through the serving base station, completing the terminal's positioning.

[0003] In related technologies, wireless communication network positioning methods mainly rely on the measurement of the direct path for positioning. When a line-of-sight (LOS) path exists, it can achieve high positioning accuracy with relatively low implementation complexity. However, it is easily affected by non-line-of-sight (NLOS) conditions, especially when no direct path exists between the terminal and the positioning base station, resulting in a significant drop in positioning accuracy and a high probability of positioning failure. In practice, while artificial intelligence (AI) or machine learning (ML)-based positioning methods can solve the positioning problem in NLOS situations, the stability, robustness, and generalization ability of AI-based positioning methods remain important factors restricting their large-scale application. Therefore, how to effectively improve positioning accuracy and reduce the failure probability is an urgent problem to be solved. Summary of the Invention

[0004] This application provides a communication network prediction method, terminal, and network-side device, which can solve the problem of high positioning failure probability.

[0005] Firstly, a method for predicting communication networks is provided, the method comprising:

[0006] The terminal uses L models to execute the target task respectively, and obtains the first result output by the L models; where L is a positive integer;

[0007] The terminal performs any of the following operations:

[0008] Based on the first result, the terminal determines the prediction result of the target task;

[0009] The terminal sends the first result to the network-side device;

[0010] The terminal receives a second result sent by the network-side device; the terminal determines the prediction result of the target task based on the first result and the second result; wherein, the second result is obtained by the network-side device after executing the target task using M models respectively; M is a positive integer.

[0011] Secondly, a method for predicting communication networks is provided, the method comprising:

[0012] The network-side device uses M models to execute the target task respectively, and obtains the second result output by the M models; where M is a positive integer;

[0013] The network-side device performs any of the following operations:

[0014] Based on the second result, the network-side device determines the prediction result of the target task;

[0015] The network-side device sends the second result to the terminal;

[0016] The network-side device receives a first result sent by the terminal; the network-side device determines the prediction result of the target task based on the first result and the second result; wherein, the first result is obtained by the terminal after executing the target task using L models respectively; L is a positive integer.

[0017] Thirdly, a communication network prediction device is provided, the device comprising:

[0018] The first execution module is used to execute the target task using L models respectively, and obtain the first result output by the L models; where L is a positive integer;

[0019] The first prediction module is used to perform any of the following operations:

[0020] Based on the first result, the terminal determines the prediction result of the target task;

[0021] The terminal sends the first result to the network-side device;

[0022] The terminal receives a second result sent by the network-side device; the terminal determines the prediction result of the target task based on the first result and the second result; wherein, the second result is obtained by the network-side device after executing the target task using M models respectively; M is a positive integer.

[0023] Fourthly, a communication network prediction device is provided, the device comprising:

[0024] The second execution module is used to execute the target task using M models respectively, and obtain the second result output by the M models; where M is a positive integer;

[0025] The second prediction module is used to perform any of the following operations:

[0026] Based on the second result, the network-side device determines the prediction result of the target task;

[0027] The network-side device sends the second result to the terminal;

[0028] The network-side device receives a first result sent by the terminal; the network-side device determines the prediction result of the target task based on the first result and the second result; wherein, the first result is obtained by the terminal after executing the target task using L models respectively; L is a positive integer.

[0029] Fifthly, a terminal is provided, the terminal including a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the method as described in the first aspect.

[0030] Sixthly, a terminal is provided, including a processor and a communication interface; wherein the processor is used for:

[0031] Use L models to execute the target task respectively, and obtain the first result output by the L models; where L is a positive integer; perform any of the following operations:

[0032] Based on the first result, the terminal determines the prediction result of the target task;

[0033] The terminal sends the first result to the network-side device;

[0034] The terminal receives a second result sent by the network-side device; the terminal determines the prediction result of the target task based on the first result and the second result; wherein, the second result is obtained by the network-side device after executing the target task using M models respectively; M is a positive integer.

[0035] In a seventh aspect, a network-side device is provided, the network-side device including a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the method as described in the second aspect.

[0036] Eighthly, a network-side device is provided, including a processor and a communication interface; wherein the processor is used for:

[0037] Use M models to execute the target task respectively, and obtain the second result output by the M models; where M is a positive integer; perform any of the following operations:

[0038] Based on the second result, the network-side device determines the prediction result of the target task;

[0039] The network-side device sends the second result to the terminal;

[0040] The network-side device receives a first result sent by the terminal; the network-side device determines the prediction result of the target task based on the first result and the second result; wherein, the first result is obtained by the terminal after executing the target task using L models respectively; L is a positive integer.

[0041] A ninth aspect provides a communication network prediction system, comprising: a terminal and a network-side device, wherein the terminal is configured to perform the steps of the method described in the first aspect, and the network-side device is configured to perform the steps of the method described in the second aspect.

[0042] In a tenth aspect, a readable storage medium is provided, on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect, or implement the steps of the method described in the second aspect.

[0043] Eleventhly, a chip is provided, the chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being 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.

[0044] In a 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 steps of the method as described in the first aspect, or to implement the steps of the method as described in the second aspect.

[0045] In this embodiment of the application, by using multiple models to execute the same task in the communication network, after obtaining the results corresponding to each model, the results corresponding to multiple models are fused to obtain the prediction result of the task, and then a decision is made. This realizes the comprehensive consideration of the results of multiple models / algorithms to make a decision, which can effectively improve the decision accuracy, reduce the failure probability, and reduce the frequency of model switching, and realize the compatibility and deployment of different models. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of a wireless communication system to which embodiments of this application can be applied;

[0047] Figure 2 This is one of the flowcharts illustrating the communication network prediction method provided in the embodiments of this application;

[0048] Figure 3 This is a second schematic flowchart of the communication network prediction method provided in the embodiments of this application;

[0049] Figure 4 This is one of the structural schematic diagrams of the communication network prediction device provided in the embodiments of this application;

[0050] Figure 5 This is a second schematic diagram of the communication network prediction device provided in the embodiments of this application;

[0051] Figure 6 This is a schematic diagram of the structure of the communication device provided in the embodiments of this application;

[0052] Figure 7 This is a schematic diagram of the terminal structure provided in the embodiments of this application;

[0053] Figure 8 This is one of the structural schematic diagrams of the network-side device provided in the embodiments of this application;

[0054] Figure 9 This is the second schematic diagram of the network-side device provided in the embodiments of this application. Detailed Implementation

[0055] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0056] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0057] It is worth noting that the technologies described in this application are not limited to Long Term Evolution (LTE) / LTE-Advanced (LTE-A) systems, 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), and other systems. The terms "system" and "network" in this application are often used interchangeably, and the described technologies can be used with the systems and radio technologies mentioned above, as well as with other systems and radio technologies. The following description describes New Radio (NR) systems for illustrative purposes, and the term NR is used in most of the following description; however, these technologies can also be applied to communication systems other than NR system applications, such as 6th generation (6G) radio communication systems. th Generation 6G communication system.

[0058] Figure 1 This is a schematic diagram of a wireless communication system to which the embodiments of this application can be applied. Figure 1The wireless communication system shown includes a terminal 11 and a network-side device 12. The terminal 11 can be a mobile phone, tablet computer, laptop computer, personal digital assistant (PDA), handheld computer, netbook, ultra-mobile personal computer (UMPC), mobile internet device (MID), augmented reality (AR) / virtual reality (VR) device, robot, wearable device, vehicle-mounted device (VUE), pedestrian terminal (PUE), smart home (home appliances with wireless communication capabilities, such as refrigerators, televisions, washing machines, or furniture), game console, personal computer (PC), ATM, or self-service machine, etc. Wearable devices include: smartwatches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart chains, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc. It should be noted that the specific type of terminal 11 is not limited in this application embodiment. The network-side device 12 may include access network devices, core network devices, and / or neural network processing nodes. The access network device 12 may also be referred to as a radio access network device, radio access network (RAN), radio access network function, or radio access network unit. The access network device 12 may include a base station, WLAN access point, or WiFi node, etc. The base station may be referred to as a node B, evolved Node B (eNB), access point, base transceiver station (BTS), radio base station, radio transceiver, basic service set (BSS), extended service set (ESS), home B node, home evolved B node, transmitting and receiving point (TRP), or any other suitable term in the field. As long as the same technical effect is achieved, the base station is not limited to specific technical terms. It should be noted that this application embodiment only uses a base station in an NR system as an example for introduction and does not limit the specific type of base station.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 and Mobility Management Function (AMF), Session Management Function (SMF), User Plane Function (UPF), Policy Control Function (PCF), Policy and Charging Rules Function (PCRF), Edge Application Server Discovery Function (EASDF), Unified Data Management (UDM), Unified Data Repository (UDR), Home Subscriber Server (HSS), Centralized network configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), Local NEF (or L-NEF), and Binding Support. Core network functions (BSF), application functions (AF), etc. It should be noted that this application embodiment only uses core network equipment in the NR system as an example for description, and does not limit the specific type of core network equipment.

[0059] The communication network prediction method provided in this application will be described in detail below with reference to the accompanying drawings, through some embodiments and application scenarios.

[0060] This application provides a communication network prediction method. By using multiple models to perform the same task in the communication network, and after obtaining the results corresponding to each model, the results corresponding to multiple models are fused to obtain the prediction result of the task, and then a decision is made. This method can make decisions by comprehensively considering the results of multiple models / algorithms, which can effectively improve the decision accuracy, reduce the failure probability, and reduce the frequency of model switching, and achieve compatibility and deployment of different models.

[0061] Figure 2 This is one of the flowcharts illustrating the communication network prediction method provided in the embodiments of this application, such as... Figure 2 As shown, the method includes steps 201-202; wherein:

[0062] Step 201: The terminal uses L models to execute the target task respectively and obtains the first result output by the L models; where L is a positive integer.

[0063] Step 202: The terminal performs any of the following operations:

[0064] Based on the first result, the terminal determines the prediction result of the target task;

[0065] The terminal sends the first result to the network-side device;

[0066] The terminal receives a second result sent by the network-side device; the terminal determines the prediction result of the target task based on the first result and the second result; wherein, the second result is obtained by the network-side device after executing the target task using M models respectively; M is a positive integer.

[0067] It should be noted that the embodiments of this application can be applied to scenarios in communication networks where prediction is based on models. Terminals include, but are not limited to, the types of terminals 11 listed above; network-side devices include, but are not limited to, the types of network-side devices 12 listed above. For example, network-side devices include at least one of the following: core network nodes; access network nodes (such as base stations); neural network processing nodes. Core network nodes include, for example, Network Data Analytics Function (NWDAF) network elements and / or Location Management Function (LMF) network elements.

[0068] It is understood that the target task performed by the model in the embodiments of this application may include tasks such as positioning and / or Channel State Information (CSI) estimation.

[0069] Optionally, any one of the L models can be either an AI-based model or a non-AI-based model. Any one of the M models can also be either an AI-based model or a non-AI-based model. Specifically, non-AI-based models may include wireless communication network positioning methods based on measurements of the direct path. AI-based models may include at least one of the following: an attention-based convolutional neural network (CNN) model; or a vision transformer model.

[0070] In practice, the terminal uses L models to execute the target task, and each model outputs its own result. Optionally, the first result may include: the L output results of the L models respectively; or, the fusion result of the L output results.

[0071] Optionally, the second result sent by the network-side device may include: M output results output by the M models respectively; or, a fusion result of the M output results.

[0072] Optionally, after determining the prediction result of the target task based on the first result, the terminal makes a decision on the association of the target task based on the prediction result of the target task.

[0073] In the communication network prediction method provided in this application embodiment, multiple models are used to perform the same task in the communication network. After obtaining the results corresponding to each model, the results corresponding to multiple models are fused to obtain the prediction result of the task, and then a decision is made. This method can make a decision by comprehensively considering the results of multiple models / algorithms, which can effectively improve the decision accuracy, reduce the failure probability, and reduce the frequency of model switching, and achieve compatibility and deployment of different models.

[0074] The specific implementation methods of the model deployment method and the task prediction method in the embodiments of this application are described below. The specific implementation methods of the model deployment method and the task prediction method may include any of the following methods:

[0075] Method 1: The terminal determines the prediction result of the target task based on multiple models deployed on the terminal.

[0076] Method 2: The network-side device determines the prediction result of the target task based on multiple models deployed on the network-side device.

[0077] Method 3: The terminal obtains a first result based on L models deployed on the terminal; the network-side device obtains a second result based on M models deployed on the network-side device and sends the second result to the terminal; the terminal determines the prediction result of the target task based on the first result and the second result.

[0078] Method 4: The terminal obtains the first result of the L model outputs based on the L models deployed on the terminal, and sends the first result to the network-side device; the network-side device obtains the second result of the M model outputs based on the M models deployed on the network-side device; then, the network-side device determines the prediction result of the target task based on the first result and the second result.

[0079] It is understandable that the number and type of models used by the terminal may differ for different tasks and methods; similarly, the number and type of models used by the network-side devices may also differ for different tasks and methods.

[0080] Here, we will explain methods 1 through 4 respectively:

[0081] Method 1: The terminal determines the prediction result of the target task based on multiple models deployed on the terminal.

[0082] Specifically, the terminal determines L models; then, the terminal uses the L models to execute the target task respectively, obtaining a first result output by the L models; based on the first result, the terminal determines the prediction result of the target task. Further, based on the prediction result of the target task, the terminal makes a decision regarding the association of the target task.

[0083] Optionally, the terminal may determine the implementation method of the L models in any of the following ways:

[0084] (1) Network-side devices directly configure multiple models for terminals.

[0085] For example, the network-side device actively sends first information to the terminal, the first information including configuration information of L models; the terminal receives the first information sent by the network-side device and determines the L models based on the configuration information of the L models.

[0086] Optionally, after receiving the first information sent by the network-side device, the terminal sends feedback information to the network-side device; wherein, the feedback information is used to indicate whether the terminal supports the model corresponding to the model configuration information. Based on the feedback information reported by the terminal, the network-side device determines whether the terminal has indicated the model corresponding to the model configuration information.

[0087] Optionally, the first information includes at least one of the following:

[0088] a) Model quantity information;

[0089] b) Model category information;

[0090] c) Model identification (ID) information;

[0091] d) Model priority information;

[0092] e) Model attribute information;

[0093] Specifically, model attribute information may include at least one of the following: parametric model, nonparametric model, AI model, non-AI model, model error, and whether the model needs fine-tuning.

[0094] f) Model accuracy information;

[0095] g) Model error information;

[0096] h) Information on model computational capability requirements;

[0097] i) Model storage capacity requirements;

[0098] j) Model feature information;

[0099] For example, model feature information may include at least one of the following: model structure; model parameters (e.g., model weights and / or model biases); model configuration (e.g., the optimizer used by the model and / or the loss function used); and model optimizer state.

[0100] k) Adaptation environment information;

[0101] For example, the adaptation environment information includes at least one of the following: LOS or NLOS scenarios; dense urban or rural scenarios; Internet of Things (IoT) scenarios; sidelink scenarios.

[0102] l) Processing latency information;

[0103] m) Information on the fusion method of the output results of each model;

[0104] For example, the information on how the outputs of each model are fused includes the post-processing methods. In practice, post-processing methods may include further data processing methods such as filtering and weighting.

[0105] n) Model lifecycle information;

[0106] For example, model lifecycle information may include at least one of the following: model activation time; model expiration time; model update cycle; and the relationship between the model and the target task.

[0107] o) Measurement information input for each type of model;

[0108] Specifically, the measurement information input to each type of model may include at least one of the following: channel state information; received signal information; historical state information; sensor information.

[0109] For example, channel state information may include at least one of the following: time domain information; frequency domain information; spatial domain information; time delay domain information; and Doppler domain information.

[0110] For example, received signal information is used to describe signal quality and / or characteristics; received signal information may include at least one of the following: RSRP, RSRQ, SINR, bit error rate, and frame error rate.

[0111] For example, historical status information can include historical location information from the past N time points.

[0112] p) Output information of each category model.

[0113] Specifically, the output information of each type of model may include at least one of the following: direct target parameters; intermediate quantities; and soft information about the direct target parameters or intermediate quantities. Soft information includes, for example, the probability distribution of terminal locations.

[0114] For example, direct target parameters may include relative position information and / or absolute position information.

[0115] For example, intermediate quantities are used to derive target parameters. Intermediate quantities may include at least one of the following: Time of Arrival (TOA), Angle Information (such as AOA, AOD), and RSRP.

[0116] (2) The terminal requests the network-side device to configure multiple models for the terminal.

[0117] Specifically, the terminal sends a request message to the network-side device, the request message being used to request the network-side device to configure L models;

[0118] The network-side device receives a request message sent by the terminal; the network-side device configures the L models for the terminal based on the request message and / or third information, and the network-side device sends first information to the terminal;

[0119] The terminal receives the first information sent by the network-side device and determines the L models based on the configuration information of the L models.

[0120] Optionally, the request information may include second information; wherein the second information includes at least one of the following:

[0121] a) Terminal mobility information;

[0122] Specifically, the terminal's mobility information may include: movement speed and / or beam switching information. For example, the terminal's movement speed is greater than a first threshold; or, for another example, the beam switching frequency is greater than a second threshold.

[0123] b) Terminal environmental information;

[0124] Specifically, the terminal's environmental information may include at least one of the following: LOS / NLOS scenario; dense urban / rural scenario; number of surrounding base stations; base station ID; and measurement information of the reference signal.

[0125] c) Accuracy requirements information;

[0126] Specifically, the accuracy requirement information may include at least one of the following: absolute accuracy requirement; quantification level of accuracy. The absolute accuracy requirement may include: horizontal direction error and / or vertical height direction error.

[0127] d) Task information.

[0128] Specifically, task information is used to indicate the tasks that the model performs. Task information may include localization tasks or CSI estimation tasks.

[0129] Optionally, the terminal may report its capability information to the network-side device.

[0130] Specifically, the terminal sends third information to the network-side device; wherein the third information is used to indicate the terminal's capability information. The third information includes at least one of the following:

[0131] a) Sensor configuration information of the terminal;

[0132] b) Data types available to the terminal;

[0133] For example, the data type may include at least one of the following: data format information; data preprocessing method.

[0134] The data that the terminal can acquire may include: data associated with sensor configuration and / or data associated with reference signals. Data associated with sensor configuration includes: image information captured by a visual sensor, distance information captured by a radar sensor, etc. Data associated with reference signals includes channel state information.

[0135] c) Terminal hardware capability information.

[0136] For example, the hardware capability information of the terminal includes at least one of the following: CPU and / or GPU capabilities (e.g., computing power); space for storing datasets or the amount of data that can be stored; CPU and GPU resources for performing tasks.

[0137] It should be noted that the third information can be reported before the second information, for example, the terminal reports the third information during the initial access phase; the third information can also be reported at the same time as the second information; or, the third information can be reported later than the second information.

[0138] (3) Terminal can independently configure multiple models

[0139] Specifically, the terminal configures the L models based on at least one of the following methods: autonomously determining and informing the network-side device, predefining the protocol, or preconfiguring the higher layer. For example, the terminal autonomously configures the L models; or, the protocol predefines or the higher layer preconfigures the L models.

[0140] (4) The terminal selects multiple models from the pre-configured model pool.

[0141] Specifically, the model pool includes at least one model. For example, the terminal selects L models from the model pool based on target information; wherein the model pool includes K models; K is greater than or equal to L; and K is a positive integer.

[0142] Optionally, the target information may include at least one of the following:

[0143] a) Statistical information on the output results of each model;

[0144] Specifically, the statistical information of the output results of each model includes: statistical information of the results of the same model over a period of time. For example, calculating the statistical information of the input and output of different models over a period of time. Statistical information includes, for example, the mean and variance.

[0145] b) Statistical information on the output results of multiple models;

[0146] Specifically, the statistical information of the output results of multiple models includes: statistical information of the statistical quantities of different models output over a period of time; statistical information such as mean, variance, etc.

[0147] c) Model error information for each model;

[0148] For example, the model error information for each model includes: instantaneous error information and / or statistical error information over a period of time.

[0149] d) Terminal mobility information;

[0150] Specifically, the terminal's mobility information may include: movement speed and / or beam switching information. For example, the terminal's movement speed is greater than a first threshold; or, for another example, the beam switching frequency is greater than a second threshold.

[0151] e) Terminal environmental information;

[0152] Specifically, the terminal's environmental information may include at least one of the following: LOS / NLOS scenario; dense urban / rural scenario; number of surrounding base stations; base station ID; and measurement information of the reference signal.

[0153] f) Accuracy requirements information;

[0154] Specifically, the accuracy requirement information may include at least one of the following: absolute accuracy requirement; quantification level of accuracy. The absolute accuracy requirement may include: horizontal direction error and / or vertical height direction error.

[0155] g) Task information;

[0156] Specifically, task information is used to indicate the tasks that the model performs. Task information may include localization tasks or CSI estimation tasks.

[0157] h) Measurement information input for each type of model;

[0158] Specifically, the measurement information input to each type of model may include at least one of the following: channel state information; received signal information; historical state information; sensor information.

[0159] For example, channel state information may include at least one of the following: time domain information; frequency domain information; spatial domain information; time delay domain information; and Doppler domain information.

[0160] For example, received signal information is used to describe signal quality and / or characteristics; received signal information may include at least one of the following: RSRP, RSRQ, SINR, bit error rate, and frame error rate.

[0161] For example, historical status information can include historical location information from the past N time points.

[0162] i) Model priority information;

[0163] j) Measurement information of the reference signal of the current terminal;

[0164] Specifically, the measurement information of the reference signal of the current terminal may include at least one of the following: SSB measurement information; SRS measurement information; CSI-RS measurement information; PRS measurement information.

[0165] k) Reference terminal model configuration information;

[0166] Optionally, the reference terminal includes surrounding terminals of the current terminal, such as the transmitting or receiving terminal in a sidelink scenario.

[0167] l) Measurement information of the reference signal of the reference terminal.

[0168] Specifically, the measurement information of the reference signal of the reference terminal may include at least one of the following: SSB measurement information; SRS measurement information; CSI-RS measurement information; PRS measurement information.

[0169] Method 2: The network-side device determines the prediction result of the target task based on multiple models deployed on the network-side device.

[0170] Specifically, the network-side device determines M models; then, the network-side device uses the M models to execute the target task respectively, obtaining a second result output by the M models; based on the second result, the network-side device determines the prediction result of the target task. Further, based on the prediction result of the target task, the network-side device makes a decision regarding the association of the target task.

[0171] Optionally, the network-side device may determine the implementation method of the M models in any of the following ways:

[0172] (1) Multiple models are configured on the network side devices.

[0173] For example, the network-side device configures the M models based on at least one of the following methods: autonomous determination, protocol predefinition, or preconfiguration.

[0174] Optionally, in a scenario where the network-side device determines M models, the network-side device can obtain fifth and / or sixth information from the terminal. The fifth and / or sixth information can be used to assist the network-side device in model selection and configuration.

[0175] Specifically, the terminal sends fifth information to the network-side device; the network-side device receives the fifth information sent by the terminal; wherein the fifth information includes at least one of the following:

[0176] a) Terminal mobility information;

[0177] Specifically, the terminal's mobility information may include: movement speed and / or beam switching information. For example, the terminal's movement speed is greater than a first threshold; or, for another example, the beam switching frequency is greater than a second threshold.

[0178] b) Terminal environmental information;

[0179] Specifically, the terminal's environmental information may include at least one of the following: LOS / NLOS scenario; dense urban / rural scenario; number of surrounding base stations; base station ID; and measurement information of the reference signal.

[0180] c) Accuracy requirements information;

[0181] Specifically, the accuracy requirement information may include at least one of the following: absolute accuracy requirement; quantification level of accuracy. The absolute accuracy requirement may include: horizontal direction error and / or vertical height direction error.

[0182] d) Task information.

[0183] Specifically, task information is used to indicate the tasks that the model performs. Task information may include localization tasks or CSI estimation tasks.

[0184] Optionally, the terminal may report its capability information to the network-side device. For example, the terminal sends sixth information to the network-side device; the network-side device receives the sixth information sent by the terminal; wherein the sixth information is used to indicate the terminal's capability information. The sixth information may include at least one of the following:

[0185] a) Sensor configuration information of the terminal;

[0186] b) Data types available to the terminal;

[0187] For example, the data type may include at least one of the following: data format information; data preprocessing method.

[0188] The data that the terminal can acquire may include: data associated with sensor configuration and / or data associated with reference signals. Data associated with sensor configuration includes: image information captured by a visual sensor, distance information captured by a radar sensor, etc. Data associated with reference signals includes channel state information.

[0189] c) Terminal hardware capability information.

[0190] For example, the hardware capability information of the terminal includes at least one of the following: CPU and / or GPU capabilities (e.g., computing power); space for storing datasets or the amount of data that can be stored; CPU and GPU resources for performing tasks.

[0191] It should be noted that the sixth piece of information can be reported before the fifth piece of information, for example, the terminal reports the sixth piece of information during the initial access phase; the sixth piece of information can also be reported at the same time as the fifth piece of information; or, the sixth piece of information can be reported later than the fifth piece of information.

[0192] (2) The network-side device selects multiple models from the pre-configured model pool.

[0193] Specifically, the model pool includes at least one model. For example, the network-side device selects M models from the model pool based on target information; wherein the model pool includes P models; P is greater than or equal to M; and P is a positive integer.

[0194] Optionally, the target information may include at least one of the following:

[0195] a) Statistical information on the output results of each model;

[0196] Specifically, the statistical information of the output results of each model includes: statistical information of the results of the same model over a period of time. For example, calculating the statistical information of the input and output of different models over a period of time. Statistical information includes, for example, the mean and variance.

[0197] b) Statistical information on the output results of multiple models;

[0198] Specifically, the statistical information of the output results of multiple models includes: statistical information of the statistical quantities of different models output over a period of time; statistical information such as mean, variance, etc.

[0199] c) Model error information for each model;

[0200] For example, the model error information for each model includes: instantaneous error information and / or statistical error information over a period of time.

[0201] d) Terminal mobility information;

[0202] Specifically, the terminal's mobility information may include: movement speed and / or beam switching information. For example, the terminal's movement speed is greater than a first threshold; or, for another example, the beam switching frequency is greater than a second threshold.

[0203] e) Terminal environmental information;

[0204] Specifically, the terminal's environmental information may include at least one of the following: LOS / NLOS scenario; dense urban / rural scenario; number of surrounding base stations; base station ID; and measurement information of the reference signal.

[0205] f) Accuracy requirements information;

[0206] Specifically, the accuracy requirement information may include at least one of the following: absolute accuracy requirement; quantification level of accuracy. The absolute accuracy requirement may include: horizontal direction error and / or vertical height direction error.

[0207] g) Task information;

[0208] Specifically, task information is used to indicate the tasks that the model performs. Task information may include localization tasks or CSI estimation tasks.

[0209] h) Measurement information input for each type of model;

[0210] Specifically, the measurement information input to each type of model may include at least one of the following: channel state information; received signal information; historical state information; sensor information.

[0211] For example, channel state information may include at least one of the following: time domain information; frequency domain information; spatial domain information; time delay domain information; and Doppler domain information.

[0212] For example, received signal information is used to describe signal quality and / or characteristics; received signal information may include at least one of the following: RSRP, RSRQ, SINR, bit error rate, and frame error rate.

[0213] For example, historical status information can include historical location information from the past N time points.

[0214] i) Model priority information;

[0215] j) Measurement information of the reference signal of the current terminal;

[0216] Specifically, the measurement information of the reference signal of the current terminal may include at least one of the following: SSB measurement information; SRS measurement information; CSI-RS measurement information; PRS measurement information.

[0217] k) Reference terminal model configuration information;

[0218] Optionally, the reference terminal includes surrounding terminals of the current terminal, such as the transmitting or receiving terminal in a sidelink scenario.

[0219] l) Measurement information of the reference signal of the reference terminal.

[0220] Specifically, the measurement information of the reference signal of the reference terminal may include at least one of the following: SSB measurement information; SRS measurement information; CSI-RS measurement information; PRS measurement information.

[0221] Optionally, after determining the M models, the network-side device sends a seventh piece of information to the terminal; wherein the seventh piece of information includes the configuration information of the M models.

[0222] The terminal receives the seventh information sent by the network-side device, and based on the configuration information of the M models included in the seventh information, obtains the measurement quantities corresponding to the M models; the terminal sends the measurement quantity information to the network-side device.

[0223] The network-side device receives the measurement information sent by the terminal, and based on the measurement information sent by the terminal, performs the target task using the M models respectively, and obtains the second result output by the M models.

[0224] Specifically, the seventh information may include at least one of the following:

[0225] [1] Input requirements for each model;

[0226] In practice, the input requirements for each model may include at least one of the following: measurement information for each type of model input; data type for each type of model input;

[0227] For example, the measurement information input to each category of model may include at least one of the following: channel state information; received signal information; historical state information; sensor information.

[0228] The channel state information may include at least one of the following: time domain information; frequency domain information; spatial domain information; time delay domain information; and Doppler domain information.

[0229] Received signal information is used to describe signal quality and / or characteristics; received signal information may include at least one of the following: RSRP, RSRQ, SINR, bit error rate, and frame error rate.

[0230] Historical status information can include historical location information from the past N time points.

[0231] For example, the data type of input for each type of model may include at least one of the following: data format information; data preprocessing method. Data obtainable by the terminal may include: data associated with sensor configuration and / or data associated with reference signals; wherein, data associated with sensor configuration includes: image information captured by a visual sensor, distance information captured by a radar sensor, etc. Data associated with reference signals includes channel state information.

[0232] [2] Model accuracy information;

[0233] [3] Processing latency information;

[0234] [4] Model lifecycle information.

[0235] In practice, model lifecycle information may include at least one of the following: model activation time; model expiration time; model update cycle; and the relationship between the model and the target task.

[0236] Optionally, the implementation method of the network-side device determining the prediction result of the target task based on the second result may include: the network-side device determining the prediction result of the target task based on the second fusion method and the second result.

[0237] Specifically, the second fusion method may include at least one of the following:

[0238] 1) By filtering the output results of each model, the fusion method of the prediction results is obtained;

[0239] 2) Determine the fusion method of the prediction results based on the weights and output results of each model.

[0240] Optionally, the network-side device determines the second fusion method based on the target information.

[0241] Specifically, the target information may include at least one of the following:

[0242] a) Statistical information on the output results of each model;

[0243] Specifically, the statistical information of the output results of each model includes: statistical information of the results of the same model over a period of time. For example, calculating the statistical information of the input and output of different models over a period of time. Statistical information includes, for example, the mean and variance.

[0244] b) Statistical information on the output results of multiple models;

[0245] Specifically, the statistical information of the output results of multiple models includes: statistical information of the statistical quantities of different models output over a period of time; statistical information such as mean, variance, etc.

[0246] c) Model error information for each model;

[0247] For example, the model error information for each model includes: instantaneous error information and / or statistical error information over a period of time.

[0248] d) Terminal mobility information;

[0249] Specifically, the terminal's mobility information may include: movement speed and / or beam switching information. For example, the terminal's movement speed is greater than a first threshold; or, for another example, the beam switching frequency is greater than a second threshold.

[0250] e) Terminal environmental information;

[0251] Specifically, the terminal's environmental information may include at least one of the following: LOS / NLOS scenario; dense urban / rural scenario; number of surrounding base stations; base station ID; and measurement information of the reference signal.

[0252] f) Accuracy requirements information;

[0253] Specifically, the accuracy requirement information may include at least one of the following: absolute accuracy requirement; quantification level of accuracy. The absolute accuracy requirement may include: horizontal direction error and / or vertical height direction error.

[0254] g) Task information;

[0255] Specifically, task information is used to indicate the tasks that the model performs. Task information may include localization tasks or CSI estimation tasks.

[0256] h) Measurement information input for each type of model;

[0257] Specifically, the measurement information input to each type of model may include at least one of the following: channel state information; received signal information; historical state information; sensor information.

[0258] For example, channel state information may include at least one of the following: time domain information; frequency domain information; spatial domain information; time delay domain information; and Doppler domain information.

[0259] For example, received signal information is used to describe signal quality and / or characteristics; received signal information may include at least one of the following: RSRP, RSRQ, SINR, bit error rate, and frame error rate.

[0260] For example, historical status information can include historical location information from the past N time points.

[0261] i) Model priority information;

[0262] j) Measurement information of the reference signal;

[0263] Specifically, the measurement information of the reference signal may include at least one of the following: SSB measurement information; SRS measurement information; CSI-RS measurement information; PRS measurement information.

[0264] Method 3: Deploy multiple models on the terminal side and the network side respectively, and let the terminal determine the prediction result of the target task.

[0265] Specifically, the terminal determines L models; the network-side device determines M models. The terminal obtains a first result based on the L models deployed on the terminal. The network-side device obtains a second result based on the M models deployed on the network-side device and sends the second result to the terminal. Based on the first and second results, the terminal determines the prediction result for the target task. Further, based on the prediction result for the target task, the terminal makes a decision regarding the association of the target task.

[0266] Optionally, the network-side device may send an eleventh message to the terminal; wherein the eleventh message is used to indicate the prediction method based on the first result and the second result. The terminal receives the eleventh message sent by the network-side device.

[0267] The prediction method includes any one of the following:

[0268] 1) A terminal-centric approach, supplemented by network-side devices:

[0269] Based on the first result and the second result, the terminal determines the prediction result of the target task.

[0270] 2) A network-side device-centric approach with terminal assistance:

[0271] Based on the first result and the second result, the network-side device determines the prediction result of the target task;

[0272] It is understandable that the prediction method corresponding to Method 3 is a terminal-centric approach with network-side devices as auxiliary tools.

[0273] Optionally, the terminal may determine the prediction result of the target task based on the first result and the second result in the following ways: the terminal determines the prediction result of the target task based on a first fusion method, the first result and the second result.

[0274] Specifically, the first fusion method may include at least one of the following:

[0275] 1) By filtering the output results of each model, the fusion method of the prediction results is obtained;

[0276] 2) Determine the fusion method of the prediction results based on the weights and output results of each model.

[0277] Optionally, the terminal determines the first fusion method based on the target information.

[0278] Optionally, the target information may include at least one of the following:

[0279] a) Statistical information on the output results of each model;

[0280] Specifically, the statistical information of the output results of each model includes: statistical information of the results of the same model over a period of time. For example, calculating the statistical information of the input and output of different models over a period of time. Statistical information includes, for example, the mean and variance.

[0281] b) Statistical information on the output results of multiple models;

[0282] Specifically, the statistical information of the output results of multiple models includes: statistical information of the statistical quantities of different models output over a period of time; statistical information such as mean, variance, etc.

[0283] c) Model error information for each model;

[0284] For example, the model error information for each model includes: instantaneous error information and / or statistical error information over a period of time.

[0285] d) Terminal mobility information;

[0286] Specifically, the terminal's mobility information may include: movement speed and / or beam switching information. For example, the terminal's movement speed is greater than a first threshold; or, for another example, the beam switching frequency is greater than a second threshold.

[0287] e) Terminal environmental information;

[0288] Specifically, the terminal's environmental information may include at least one of the following: LOS / NLOS scenario; dense urban / rural scenario; number of surrounding base stations; base station ID; and measurement information of the reference signal.

[0289] f) Accuracy requirements information;

[0290] Specifically, the accuracy requirement information may include at least one of the following: absolute accuracy requirement; quantification level of accuracy. The absolute accuracy requirement may include: horizontal direction error and / or vertical height direction error.

[0291] g) Task information;

[0292] Specifically, task information is used to indicate the tasks that the model performs. Task information may include localization tasks or CSI estimation tasks.

[0293] h) Measurement information input for each type of model;

[0294] Specifically, the measurement information input to each type of model may include at least one of the following: channel state information; received signal information; historical state information; sensor information.

[0295] For example, channel state information may include at least one of the following: time domain information; frequency domain information; spatial domain information; time delay domain information; and Doppler domain information.

[0296] For example, received signal information is used to describe signal quality and / or characteristics; received signal information may include at least one of the following: RSRP, RSRQ, SINR, bit error rate, and frame error rate.

[0297] For example, historical status information can include historical location information from the past N time points.

[0298] i) Model priority information;

[0299] j) Measurement information of the reference signal of the current terminal;

[0300] Specifically, the measurement information of the reference signal of the current terminal may include at least one of the following: SSB measurement information; SRS measurement information; CSI-RS measurement information; PRS measurement information.

[0301] k) Reference terminal model configuration information;

[0302] Optionally, the reference terminal includes surrounding terminals of the current terminal, such as the transmitting or receiving terminal in a sidelink scenario.

[0303] l) Measurement information of the reference signal of the reference terminal.

[0304] Specifically, the measurement information of the reference signal of the reference terminal may include at least one of the following: SSB measurement information; SRS measurement information; CSI-RS measurement information; PRS measurement information.

[0305] Method 4: Deploy multiple models on both the terminal and network devices, and let the network device determine the prediction results for the target task.

[0306] Specifically, the terminal determines L models; the network-side device determines M models. The terminal obtains a first result from the L models deployed on the terminal and sends the first result to the network-side device. The network-side device obtains a second result from the M models deployed on the network-side device; based on the first result and the second result, the network-side device determines the prediction result for the target task. Further, based on the prediction result for the target task, the network-side device makes a decision regarding the association of the target task.

[0307] Optionally, the network-side device may send an eleventh message to the terminal; wherein the eleventh message is used to indicate the prediction method based on the first result and the second result. The terminal receives the eleventh message sent by the network-side device.

[0308] The prediction method includes any one of the following:

[0309] 1) A terminal-centric approach, supplemented by network-side devices:

[0310] Based on the first result and the second result, the terminal determines the prediction result of the target task.

[0311] 2) A network-side device-centric approach with terminal assistance:

[0312] Based on the first result and the second result, the network-side device determines the prediction result of the target task;

[0313] It is understandable that the prediction method corresponding to Method 4 is a network-side device-centric, terminal-assisted approach.

[0314] Optionally, the implementation method of the network-side device determining the prediction result of the target task based on the first result and the second result may include: the network-side device determining the prediction result of the target task based on a second fusion method, the first result and the second result.

[0315] Specifically, the second fusion method may include at least one of the following:

[0316] 1) By filtering the output results of each model, the fusion method of the prediction results is obtained;

[0317] 2) Determine the fusion method of the prediction results based on the weights and output results of each model.

[0318] Optionally, the network-side device determines the second fusion method based on the target information.

[0319] Specifically, the target information may include at least one of the following:

[0320] a) Statistical information on the output results of each model;

[0321] Specifically, the statistical information of the output results of each model includes: statistical information of the results of the same model over a period of time. For example, calculating the statistical information of the input and output of different models over a period of time. Statistical information includes, for example, the mean and variance.

[0322] b) Statistical information on the output results of multiple models;

[0323] Specifically, the statistical information of the output results of multiple models includes: statistical information of the statistical quantities of different models output over a period of time; statistical information such as mean, variance, etc.

[0324] c) Model error information for each model;

[0325] For example, the model error information for each model includes: instantaneous error information and / or statistical error information over a period of time.

[0326] d) Terminal mobility information;

[0327] Specifically, the terminal's mobility information may include: movement speed and / or beam switching information. For example, the terminal's movement speed is greater than a first threshold; or, for another example, the beam switching frequency is greater than a second threshold.

[0328] e) Terminal environmental information;

[0329] Specifically, the terminal's environmental information may include at least one of the following: LOS / NLOS scenario; dense urban / rural scenario; number of surrounding base stations; base station ID; and measurement information of the reference signal.

[0330] f) Accuracy requirements information;

[0331] Specifically, the accuracy requirement information may include at least one of the following: absolute accuracy requirement; quantification level of accuracy. The absolute accuracy requirement may include: horizontal direction error and / or vertical height direction error.

[0332] g) Task information;

[0333] Specifically, task information is used to indicate the tasks that the model performs. Task information may include localization tasks or CSI estimation tasks.

[0334] h) Measurement information input for each type of model;

[0335] Specifically, the measurement information input to each type of model may include at least one of the following: channel state information; received signal information; historical state information; sensor information.

[0336] For example, channel state information may include at least one of the following: time domain information; frequency domain information; spatial domain information; time delay domain information; and Doppler domain information.

[0337] For example, received signal information is used to describe signal quality and / or characteristics; received signal information may include at least one of the following: RSRP, RSRQ, SINR, bit error rate, and frame error rate.

[0338] For example, historical status information can include historical location information from the past N time points.

[0339] i) Model priority information;

[0340] j) Measurement information of the reference signal;

[0341] Specifically, the measurement information of the reference signal may include at least one of the following: SSB measurement information; SRS measurement information; CSI-RS measurement information; PRS measurement information.

[0342] Figure 3 This is a second flowchart illustrating the communication network prediction method provided in this application embodiment, as shown below. Figure 3 As shown, the method includes steps 301-302; wherein:

[0343] Step 301: The network-side device uses M models to execute the target task respectively and obtains the second result output by the M models; where M is a positive integer.

[0344] Step 302: The network-side device performs any of the following operations:

[0345] Based on the second result, the network-side device determines the prediction result of the target task;

[0346] The network-side device sends the second result to the terminal;

[0347] The network-side device receives a first result sent by the terminal; the network-side device determines the prediction result of the target task based on the first result and the second result; wherein, the first result is obtained by the terminal after executing the target task using L models respectively; L is a positive integer.

[0348] It should be noted that the embodiments of this application can be applied to scenarios in communication networks where prediction is based on models. Terminals include, but are not limited to, the types of terminals 11 listed above; network-side devices include, but are not limited to, the types of network-side devices 12 listed above. For example, network-side devices include at least one of the following: core network nodes; access network nodes (such as base stations); neural network processing nodes. Core network nodes include, for example, Network Data Analytics Function (NWDAF) network elements and / or Location Management Function (LMF) network elements.

[0349] It is understood that the target task performed by the model in the embodiments of this application may include tasks such as positioning and / or Channel State Information (CSI) estimation.

[0350] Optionally, any one of the L models can be either an AI-based model or a non-AI-based model. Any one of the M models can also be either an AI-based model or a non-AI-based model. Specifically, non-AI-based models may include wireless communication network positioning methods based on measurements of the direct path. AI-based models may include at least one of the following: an attention-based convolutional neural network (CNN) model; or a vision transformer model.

[0351] In practice, the terminal uses L models to execute the target task, and each model outputs its own result. Optionally, the first result may include: the L output results of the L models respectively; or, the fusion result of the L output results.

[0352] Optionally, the second result sent by the network-side device may include: M output results output by the M models respectively; or, a fusion result of the M output results.

[0353] Optionally, after determining the prediction result of the target task based on the first result, the terminal makes a decision on the association of the target task based on the prediction result of the target task.

[0354] In the communication network prediction method provided in this application embodiment, multiple models are used to perform the same task in the communication network. After obtaining the results corresponding to each model, the results corresponding to multiple models are fused to obtain the prediction result of the task, and then a decision is made. This method can make a decision by comprehensively considering the results of multiple models / algorithms, which can effectively improve the decision accuracy, reduce the failure probability, and reduce the frequency of model switching, and achieve compatibility and deployment of different models.

[0355] Optionally, the method further includes:

[0356] The network-side device sends first information to the terminal; wherein, the first information includes configuration information of the L models; the configuration information of the L models is used by the terminal to determine the L models;

[0357] or,

[0358] The network-side device receives a request message sent by the terminal, the request message being used to request the network-side device to configure the L models; the network-side device configures the L models based on the request message and / or third information; the network-side device sends first information to the terminal; wherein, the first information includes configuration information of the L models; the configuration information of the L models is used by the terminal to determine the L models.

[0359] Optionally, the first information includes at least one of the following:

[0360] Model quantity information;

[0361] Model category information;

[0362] Model identification ID information;

[0363] Model priority information;

[0364] Model attribute information;

[0365] Model accuracy information;

[0366] Model error information;

[0367] Information regarding the model's computational requirements;

[0368] Model storage capacity requirements;

[0369] Model feature information;

[0370] Adaptation environment information;

[0371] Processing latency information;

[0372] Information on the fusion method of the output results of each model;

[0373] Model lifecycle information;

[0374] Measurement information input for each type of model;

[0375] Output information for each category of model.

[0376] Optionally, the measurement information input to each category of model includes at least one of the following:

[0377] Channel state information;

[0378] Receive signal information;

[0379] Historical status information;

[0380] Sensor information.

[0381] Optionally, the output information of each category model includes at least one of the following:

[0382] Direct target parameters;

[0383] intermediate quantity;

[0384] Soft information about direct target parameters.

[0385] Optionally, the request information includes second information; wherein the second information includes at least one of the following:

[0386] Terminal mobility information;

[0387] Terminal environmental information;

[0388] Accuracy requirements information;

[0389] Task information.

[0390] Optionally, the method further includes:

[0391] The network-side device receives third information sent by the terminal; wherein the third information is used to indicate the terminal's capability information.

[0392] Optionally, the third information includes at least one of the following:

[0393] The terminal's sensor configuration information;

[0394] Data types that the terminal can access;

[0395] Terminal hardware capability information.

[0396] Optionally, the data type includes at least one of the following:

[0397] Data format information;

[0398] Data preprocessing methods.

[0399] Optionally, after the network-side device sends the first information to the terminal, the method further includes:

[0400] The network-side device receives feedback information sent by the terminal; wherein the feedback information is used to indicate whether the terminal supports the model corresponding to the model configuration information.

[0401] Optionally, the method further includes:

[0402] The network-side device determines the M models.

[0403] Optionally, the network-side device determines the M models, including:

[0404] The network-side device configures the M models based on at least one of the following methods: autonomous determination, protocol predefinition, or preconfiguration.

[0405] or,

[0406] The network-side device selects M models from the model pool based on the target information; wherein the model pool includes P models; P is greater than or equal to M; and P is a positive integer.

[0407] Optionally, the method further includes:

[0408] The network-side device receives a fifth message sent by the terminal; wherein the fifth message includes at least one of the following:

[0409] Terminal mobility information;

[0410] Terminal environmental information;

[0411] Accuracy requirements information;

[0412] Task information.

[0413] Optionally, the method further includes:

[0414] The network-side device receives the sixth information sent by the terminal; wherein the sixth information is used to indicate the terminal's capability information.

[0415] Optionally, the sixth information includes at least one of the following:

[0416] The terminal's sensor configuration information;

[0417] Data types that the terminal can access;

[0418] Terminal hardware capability information.

[0419] Optionally, the data type includes at least one of the following:

[0420] Data format information;

[0421] Data preprocessing methods.

[0422] Optionally, the method further includes:

[0423] The network-side device sends a seventh message to the terminal; wherein the seventh message includes the configuration information of the M models.

[0424] Optionally, the seventh information includes at least one of the following:

[0425] Input requirements for each model;

[0426] Model accuracy information;

[0427] Processing latency information;

[0428] Model lifecycle information.

[0429] Optionally, the method further includes:

[0430] The network-side device receives the measurement information sent by the terminal.

[0431] Optionally, the network-side device uses M models to execute the target task respectively, and obtains a second result output by the M models, including:

[0432] The network-side device, based on the measurement information sent by the terminal, uses the M models to perform the target task respectively, and obtains the second result output by the M models.

[0433] Optionally, the network-side device determines the prediction result of the target task based on the second result, including: the network-side device determines the prediction result of the target task based on the second fusion method and the second result;

[0434] Alternatively, the network-side device determines the prediction result of the target task based on the first result and the second result, including: the network-side device determines the prediction result of the target task based on the second fusion method, the first result and the second result.

[0435] Optionally, the second fusion method includes:

[0436] The outputs of each model are filtered to obtain the prediction results;

[0437] And / or, based on the weights and outputs of each model, determine the prediction results.

[0438] Optionally, the method further includes:

[0439] The network-side device determines the second fusion method based on the target information.

[0440] Optionally, the target information includes at least one of the following:

[0441] Statistical information on the output results of each model;

[0442] Statistical information on the output results of multiple models;

[0443] Model error information for each model;

[0444] Terminal mobility information;

[0445] Terminal environmental information;

[0446] Accuracy requirements information;

[0447] Task information;

[0448] Measurement information input for each type of model;

[0449] Model priority information;

[0450] Measurement information of the reference signal.

[0451] Optionally, the method further includes:

[0452] The network-side device makes a decision on the association of the target task based on the prediction result of the target task.

[0453] Optionally, the second result includes:

[0454] The M outputs of the M models respectively; or the fusion result of the M outputs.

[0455] Optionally, the method further includes:

[0456] The network-side device sends an eleventh message to the terminal; wherein the eleventh message is used to indicate the prediction method based on the first result and the second result.

[0457] Optionally, the prediction method includes any of the following:

[0458] Based on the first result and the second result, the network-side device determines the prediction result of the target task;

[0459] Based on the first result and the second result, the terminal determines the prediction result of the target task.

[0460] The communication network prediction method provided in this application can be executed by a communication network prediction device. This application uses the example of a communication network prediction device executing the communication network prediction method to illustrate the communication network prediction device provided in this application.

[0461] Figure 4 This is one of the structural schematic diagrams of the communication network prediction device provided in the embodiments of this application, such as... Figure 4 As shown, the communication network prediction device 400, applied to a terminal, includes:

[0462] The first execution module 401 is used to execute the target task using L models respectively, and obtain the first result output by the L models; where L is a positive integer;

[0463] The first prediction module 402 is used to perform any of the following operations:

[0464] Based on the first result, the terminal determines the prediction result of the target task;

[0465] The terminal sends the first result to the network-side device;

[0466] The terminal receives a second result sent by the network-side device; the terminal determines the prediction result of the target task based on the first result and the second result; wherein, the second result is obtained by the network-side device after executing the target task using M models respectively; M is a positive integer.

[0467] The communication network prediction device provided in this application uses multiple models to perform the same task in the communication network. After obtaining the results corresponding to each model, the results corresponding to multiple models are fused to obtain the prediction result of the task, and then a decision is made. This realizes the comprehensive consideration of the results of multiple models / algorithms to make a decision, which can effectively improve the decision accuracy, reduce the failure probability, and reduce the frequency of model switching, and realize the compatibility and deployment of different models.

[0468] Optionally, the device further includes:

[0469] The first determining module is used to determine the L models.

[0470] Optionally, the first determining module is specifically used for any of the following:

[0471] Receive first information sent by a network-side device; wherein the first information includes configuration information of the L models; determine the L models based on the configuration information of the L models;

[0472] Send a request message to the network-side device, the request message being used to request the network-side device to configure the L models; receive first information sent by the network-side device; wherein the first information includes configuration information of the L models; determine the L models based on the configuration information of the L models;

[0473] Configure the L models based on at least one of the following methods: autonomously determining and informing network-side devices, predefined protocols, or pre-configured higher layers;

[0474] Based on the target information, select L models from the model pool; wherein, the model pool includes K models; K is greater than or equal to L; K is a positive integer.

[0475] Optionally, the first information includes at least one of the following:

[0476] Model quantity information;

[0477] Model category information;

[0478] Model identification ID information;

[0479] Model priority information;

[0480] Model attribute information;

[0481] Model accuracy information;

[0482] Model error information;

[0483] Information regarding the model's computational requirements;

[0484] Model storage capacity requirements;

[0485] Model feature information;

[0486] Adaptation environment information;

[0487] Processing latency information;

[0488] Information on the fusion method of the output results of each model;

[0489] Model lifecycle information;

[0490] Measurement information input for each type of model;

[0491] Output information for each category of model.

[0492] Optionally, the measurement information input to each category of model includes at least one of the following:

[0493] Channel state information;

[0494] Receive signal information;

[0495] Historical status information;

[0496] Sensor information.

[0497] Optionally, the output information of each category model includes at least one of the following:

[0498] Direct target parameters;

[0499] intermediate quantity;

[0500] Soft information about direct target parameters or intermediate quantities.

[0501] Optionally, the request information includes second information; wherein the second information includes at least one of the following:

[0502] Terminal mobility

[0503] information;

[0504] Terminal environmental information;

[0505] Accuracy requirements information;

[0506] Task information.

[0507] Optionally, the device further includes:

[0508] The first sending module is used to send third information to the network-side device; wherein the third information is used to indicate the capability information of the terminal.

[0509] Optionally, the third information includes at least one of the following:

[0510] The terminal's sensor configuration information;

[0511] Data types that the terminal can access;

[0512] Terminal hardware capability information.

[0513] Optionally, the data type includes at least one of the following:

[0514] Data format information;

[0515] Data preprocessing methods.

[0516] Optionally, the device further includes:

[0517] The second sending module is used to send feedback information to the network-side device; wherein the feedback information is used to indicate whether the terminal supports the model corresponding to the model configuration information.

[0518] Optionally, the device further includes:

[0519] The third sending module is used to send fifth information to the network-side device; wherein the fifth information includes at least one of the following:

[0520] Terminal mobility information;

[0521] Terminal environmental information;

[0522] Accuracy requirements information;

[0523] Task information.

[0524] Optionally, the device further includes:

[0525] The fourth sending module is used to send sixth information to the network-side device; wherein the sixth information is used to indicate the capability information of the terminal.

[0526] Optionally, the sixth information includes at least one of the following:

[0527] The terminal's sensor configuration information;

[0528] Data types that the terminal can access;

[0529] Terminal hardware capability information.

[0530] Optionally, the data type includes at least one of the following:

[0531] Data format information;

[0532] Data preprocessing methods.

[0533] Optionally, the device further includes:

[0534] The first receiving module is used to receive the seventh information sent by the network-side device; wherein the seventh information includes the configuration information of the M models.

[0535] Optionally, the seventh information includes at least one of the following:

[0536] Input requirements for each model;

[0537] Model accuracy information;

[0538] Processing latency information;

[0539] Model lifecycle information.

[0540] Optionally, the device further includes:

[0541] The fifth sending module is used to send measurement information to the network-side device.

[0542] Optionally, the first prediction module 402 is used for:

[0543] Based on the first fusion method and the first result, the prediction result of the target task is determined;

[0544] Alternatively, the prediction result of the target task can be determined based on the first fusion method, the first result, and the second result.

[0545] Optionally, the first fusion method includes:

[0546] The outputs of each model are filtered to obtain the prediction results;

[0547] And / or, based on the weights and outputs of each model, determine the prediction results.

[0548] Optionally, the device further includes:

[0549] The second determining module is used to determine the first fusion method based on the target information.

[0550] Optionally, the target information includes at least one of the following:

[0551] Statistical information on the output results of each model;

[0552] Statistical information on the output results of multiple models;

[0553] Model error information for each model;

[0554] Terminal mobility information;

[0555] Terminal environmental information;

[0556] Accuracy requirements information;

[0557] Task information;

[0558] Measurement information input for each type of model;

[0559] Model priority information;

[0560] Measurement information of the reference signal of the current terminal;

[0561] Refer to the model configuration information of the terminal;

[0562] Measurement information of the reference signal of the reference terminal.

[0563] Optionally, the device further includes:

[0564] The first decision module is used to make decisions related to the target task based on the prediction results of the target task.

[0565] Optionally, the first result includes:

[0566] The L outputs of the L models respectively; or the fusion result of the L outputs.

[0567] Optionally, the device further includes:

[0568] The second receiving module is used to receive the eleventh information sent by the network-side device; wherein the eleventh information is used to indicate the prediction method based on the first result and the second result.

[0569] Optionally, the prediction method includes any of the following:

[0570] Based on the first result and the second result, the network-side device determines the prediction result of the target task;

[0571] Based on the first result and the second result, the terminal determines the prediction result of the target task.

[0572] Figure 5 This is a second schematic diagram of the communication network prediction device provided in the embodiments of this application, as shown below. Figure 5 As shown, the communication network prediction device 500 is applied to network-side equipment and includes:

[0573] The second execution module 501 is used to execute the target task using M models respectively, and obtain the second result output by the M models; where M is a positive integer;

[0574] The second prediction module 502 is used to perform any of the following operations:

[0575] Based on the second result, the network-side device determines the prediction result of the target task;

[0576] The network-side device sends the second result to the terminal;

[0577] The network-side device receives a first result sent by the terminal; the network-side device determines the prediction result of the target task based on the first result and the second result; wherein, the first result is obtained by the terminal after executing the target task using L models respectively; L is a positive integer.

[0578] The communication network prediction device provided in this application uses multiple models to perform the same task in the communication network. After obtaining the results corresponding to each model, the results corresponding to multiple models are fused to obtain the prediction result of the task, and then a decision is made. This realizes the comprehensive consideration of the results of multiple models / algorithms to make a decision, which can effectively improve the decision accuracy, reduce the failure probability, and reduce the frequency of model switching, and realize the compatibility and deployment of different models.

[0579] Optionally, the device further includes:

[0580] The sixth sending module is used to send first information to the terminal; wherein the first information includes configuration information of the L models; the configuration information of the L models is used by the terminal to determine the L models;

[0581] Alternatively, the processing module is configured to receive request information sent by the terminal from the network-side device, the request information being used to request the network-side device to configure the L models; the network-side device configuring the L models based on the request information and / or third information; the network-side device sending first information to the terminal; wherein the first information includes configuration information of the L models; the configuration information of the L models being used by the terminal to determine the L models.

[0582] Optionally, the first information includes at least one of the following:

[0583] Model quantity information;

[0584] Model category information;

[0585] Model identification ID information;

[0586] Model priority information;

[0587] Model attribute information;

[0588] Model accuracy information;

[0589] Model error information;

[0590] Information regarding the model's computational requirements;

[0591] Model storage capacity requirements;

[0592] Model feature information;

[0593] Adaptation environment information;

[0594] Processing latency information;

[0595] Information on the fusion method of the output results of each model;

[0596] Model lifecycle information;

[0597] Measurement information input for each type of model;

[0598] Output information for each category of model.

[0599] Optionally, the measurement information input to each category of model includes at least one of the following:

[0600] Channel state information;

[0601] Receive signal information;

[0602] Historical status information;

[0603] Sensor information.

[0604] Optionally, the output information of each category model includes at least one of the following:

[0605] Direct target parameters;

[0606] intermediate quantity;

[0607] Soft information about direct target parameters.

[0608] Optionally, the request information includes second information; wherein the second information includes at least one of the following:

[0609] Terminal mobility information;

[0610] Terminal environmental information;

[0611] Accuracy requirements information;

[0612] Task information.

[0613] Optionally, the device further includes:

[0614] The third receiving module is used to receive third information sent by the terminal; wherein the third information is used to indicate the capability information of the terminal.

[0615] Optionally, the third information includes at least one of the following:

[0616] The terminal's sensor configuration information;

[0617] Data types that the terminal can access;

[0618] Terminal hardware capability information.

[0619] Optionally, the data type includes at least one of the following:

[0620] Data format information;

[0621] Data preprocessing methods.

[0622] Optionally, the device further includes:

[0623] The fourth receiving module is used to receive feedback information sent by the terminal; wherein the feedback information is used to indicate whether the terminal supports the model corresponding to the model configuration information.

[0624] Optionally, the device further includes:

[0625] The second determining module is used to determine the M models.

[0626] Optionally, the second determining module is used for:

[0627] Configure the M models based on at least one of the following methods: autonomous determination, protocol predefinition, or preconfiguration;

[0628] or,

[0629] Based on the target information, select M models from the model pool; wherein, the model pool includes P models; P is greater than or equal to M; P is a positive integer.

[0630] Optionally, the device further includes:

[0631] The fifth receiving module is used to receive fifth information sent by the terminal; wherein the fifth information includes at least one of the following:

[0632] Terminal mobility information;

[0633] Terminal environmental information;

[0634] Accuracy requirements information;

[0635] Task information.

[0636] Optionally, the device further includes:

[0637] The sixth receiving module is used to receive the sixth information sent by the terminal; wherein the sixth information is used to indicate the capability information of the terminal.

[0638] Optionally, the sixth information includes at least one of the following:

[0639] The terminal's sensor configuration information;

[0640] Data types that the terminal can access;

[0641] Terminal hardware capability information.

[0642] Optionally, the data type includes at least one of the following:

[0643] Data format information;

[0644] Data preprocessing methods.

[0645] Optionally, the device further includes:

[0646] The seventh sending module is used to send seventh information to the terminal; wherein the seventh information includes the configuration information of the M models.

[0647] Optionally, the seventh information includes at least one of the following:

[0648] Input requirements for each model;

[0649] Model accuracy information;

[0650] Processing latency information;

[0651] Model lifecycle information.

[0652] Optionally, the device further includes:

[0653] The seventh receiving module is used to receive measurement information sent by the terminal.

[0654] Optionally, the second execution module is used to perform the target task using the M models respectively based on the measurement information sent by the terminal, and obtain the second result output by the M models.

[0655] Optionally, the second prediction module is used for:

[0656] Based on the second fusion method and the second result, the prediction result of the target task is determined;

[0657] Alternatively, the prediction result of the target task can be determined based on the second fusion method, the first result, and the second result.

[0658] Optionally, the second fusion method includes:

[0659] The outputs of each model are filtered to obtain the prediction results;

[0660] And / or, based on the weights and outputs of each model, determine the prediction results.

[0661] Optionally, the device further includes:

[0662] The third determining module is used to determine the second fusion method based on the target information.

[0663] Optionally, the target information includes at least one of the following:

[0664] Statistical information on the output results of each model;

[0665] Statistical information on the output results of multiple models;

[0666] Model error information for each model;

[0667] Terminal mobility information;

[0668] Terminal environmental information;

[0669] Accuracy requirements information;

[0670] Task information;

[0671] Measurement information input for each type of model;

[0672] Model priority information;

[0673] Measurement information of the reference signal.

[0674] Optionally, the device further includes:

[0675] The second decision module is used by the network-side device to make a decision on the association of the target task based on the prediction result of the target task.

[0676] Optionally, the second result includes:

[0677] The M outputs of the M models respectively; or the fusion result of the M outputs.

[0678] Optionally, the device further includes:

[0679] The eighth sending module is used to send eleventh information to the terminal; wherein the eleventh information is used to indicate the prediction method based on the first result and the second result.

[0680] Optionally, the prediction method includes any of the following:

[0681] Based on the first result and the second result, the network-side device determines the prediction result of the target task;

[0682] Based on the first result and the second result, the terminal determines the prediction result of the target task.

[0683] The communication network prediction device in this application embodiment 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 a chip. The electronic device can be a terminal, or other devices besides a terminal. For example, the terminal can include, but is not limited to, the type of terminal 11 listed above; other devices can be servers, network attached storage (NAS), etc., and this application embodiment does not specifically limit the type.

[0684] The communication network prediction device provided in this application embodiment can achieve... Figures 1 to 3 The various processes implemented in the method embodiments achieve the same technical effect, and will not be described again here to avoid repetition.

[0685] Figure 6 This is a schematic diagram of the structure of the communication device provided in the embodiments of this application, such as... Figure 6As shown, the communication device 600 includes a processor 601 and a memory 602. The memory 602 stores programs or instructions that can run on the processor 601. For example, when the communication device 600 is a terminal, the program or instructions executed by the processor 601 implement the various steps of the above-described communication network prediction method embodiment and achieve the same technical effect. When the communication device 600 is a network-side device, the program or instructions executed by the processor 601 implement the various steps of the above-described communication network prediction method embodiment and achieve the same technical effect. To avoid repetition, this will not be described again here.

[0686] This application embodiment also provides a terminal, including a processor and a communication interface; wherein, the processor is used for:

[0687] Use L models to execute the target task respectively, and obtain the first result output by the L models; where L is a positive integer; perform any of the following operations:

[0688] Based on the first result, the terminal determines the prediction result of the target task;

[0689] The terminal sends the first result to the network-side device;

[0690] The terminal receives a second result sent by the network-side device; the terminal determines the prediction result of the target task based on the first result and the second result; wherein, the second result is obtained by the network-side device after executing the target task using M models respectively; M is a positive integer.

[0691] This terminal embodiment corresponds to the above-described terminal-side method embodiment. All implementation processes and methods of the above-described method embodiment can be applied to this terminal embodiment and can achieve the same technical effect.

[0692] Figure 7 This is a schematic diagram of the terminal structure provided in the embodiments of this application, such as... Figure 7 As shown, the terminal 700 includes, but is not limited to, at least some of the following components: radio frequency unit 701, network module 702, audio output unit 703, input unit 704, sensor 705, display unit 706, user input unit 707, interface unit 708, memory 709, and processor 710.

[0693] Those skilled in the art will understand that the terminal 700 may also include a power supply (such as a battery) for supplying power to various components. The power supply may be logically connected to the processor 710 through a power management system, thereby enabling functions such as managing charging, discharging, and power consumption through the power management system. Figure 7The terminal structure shown does not constitute a limitation on the terminal. The terminal may include more or fewer components than shown, or combine certain components, or have different component arrangements, which will not be elaborated here.

[0694] It should be understood that, in this embodiment, the input unit 704 may include a graphics processing unit (GPU) 7041 and a microphone 7042. The GPU 7041 processes image data of still images or videos obtained by an image capture device (such as a camera) in video capture mode or image capture mode. The display unit 706 may include a display panel 7061, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, or the like. The user input unit 707 includes at least one of a touch panel 7071 and other input devices 7072. The touch panel 7071 is also called a touch screen. The touch panel 7071 may include a touch detection device and a touch controller. Other input devices 7072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, and joysticks, which will not be described in detail here.

[0695] In this embodiment, after receiving downlink data from the network-side device, the radio frequency unit 701 can transmit it to the processor 710 for processing; in addition, the radio frequency unit 701 can send uplink data to the network-side device. Typically, the radio frequency unit 701 includes, but is not limited to, an antenna, amplifier, transceiver, coupler, low-noise amplifier, duplexer, etc.

[0696] The memory 709 can be used to store software programs or instructions, as well as various data. The memory 709 may primarily include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area may store the operating system, application programs or instructions required for at least one function (such as sound playback, image playback, etc.). Furthermore, the memory 709 may include volatile memory or non-volatile memory, or both. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus RAM (DRRAM). The memory 709 in the embodiments of this application includes, but is not limited to, these and any other suitable types of memory.

[0697] Processor 710 may include one or more processing units; optionally, processor 710 integrates an application processor and a modem processor, wherein the application processor mainly handles operations involving the operating system, user interface, and applications, and the modem processor mainly handles wireless communication signals, such as a baseband processor. It is understood that the aforementioned modem processor may also not be integrated into processor 710.

[0698] The processor 710 is configured to execute the target task using L models respectively, and obtain the first result output by the L models; where L is a positive integer; and to perform any of the following operations:

[0699] Based on the first result, the terminal determines the prediction result of the target task;

[0700] The terminal sends the first result to the network-side device;

[0701] The terminal receives a second result sent by the network-side device; the terminal determines the prediction result of the target task based on the first result and the second result; wherein, the second result is obtained by the network-side device after executing the target task using M models respectively; M is a positive integer.

[0702] The terminal provided in this application embodiment executes the same task using multiple models in a communication network. After obtaining the results corresponding to each model, the results corresponding to multiple models are fused to obtain the prediction result of the task, and then a decision is made. This realizes the ability to make decisions by comprehensively considering the results of multiple models / algorithms, which can effectively improve decision accuracy, reduce the probability of failure, and reduce the frequency of model switching, and achieve compatibility and deployment of different models.

[0703] This application embodiment also provides a network-side device, including a processor and a communication interface; wherein, the processor is used for:

[0704] Use M models to execute the target task respectively, and obtain the second result output by the M models; where M is a positive integer; perform any of the following operations:

[0705] Based on the second result, the network-side device determines the prediction result of the target task;

[0706] The network-side device sends the second result to the terminal;

[0707] The network-side device receives a first result sent by the terminal; the network-side device determines the prediction result of the target task based on the first result and the second result; wherein, the first result is obtained by the terminal after executing the target task using L models respectively; L is a positive integer.

[0708] This network-side device embodiment corresponds to the above-described network-side device method embodiment. All implementation processes and methods of the above-described method embodiment can be applied to this network-side device embodiment and can achieve the same technical effect.

[0709] Figure 8 This is one of the structural schematic diagrams of the network-side device provided in the embodiments of this application, such as... Figure 8As shown, the network-side device 800 includes: an antenna 801, a radio frequency (RF) device 802, a baseband device 803, a processor 804, and a memory 805. The antenna 801 is connected to the RF device 802. In the uplink direction, the RF device 802 receives information through the antenna 801 and transmits the received information to the baseband device 803 for processing. In the downlink direction, the baseband device 803 processes the information to be transmitted and sends it to the RF device 802. The RF device 802 processes the received information and transmits it through the antenna 801.

[0710] The method executed by the network-side device in the above embodiments can be implemented in the baseband device 803, which includes a baseband processor.

[0711] The baseband device 803 may, for example, include at least one baseband board on which multiple chips are disposed, such as... Figure 8 As shown, one of the chips is, for example, a baseband processor, which is connected to the memory 805 via a bus interface to call the program in the memory 805 and execute the network device operation shown in the above method embodiment.

[0712] The network-side device may also include a network interface 806, such as a common public radio interface (CPRI).

[0713] Specifically, the network-side device 800 of this embodiment of the invention further includes: instructions or programs stored in memory 805 and executable on processor 804. The processor 804 calls the instructions or programs in memory 805 to execute the steps of the communication network prediction method on the network-side device side and achieve the same technical effect. To avoid repetition, it will not be described in detail here.

[0714] Figure 9 This is a second schematic diagram of the network-side device provided in the embodiments of this application, as shown below. Figure 9 As shown, the network-side device 900 includes a processor 901, a network interface 902, and a memory 903. The network interface 902 is, for example, a common public radio interface (CPRI).

[0715] Specifically, the network-side device 900 of this embodiment of the invention further includes: instructions or programs stored in memory 903 and executable on processor 901. The processor 901 calls the instructions or programs in memory 903 to execute the steps of the communication network prediction method on the network-side device side and achieve the same technical effect. To avoid repetition, it will not be described in detail here.

[0716] This application also provides a communication network prediction system, including: a terminal and a network-side device. The terminal can be used to perform the steps of the communication network prediction method described above, and the network-side device can be used to perform the steps of the communication network prediction method described above.

[0717] This application also provides a readable storage medium, which can be volatile or non-volatile. The readable storage medium stores a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described communication network prediction method embodiments and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0718] The processor is the processor in the terminal described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0719] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described communication network prediction method embodiment and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0720] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0721] This application also provides a computer program / program product, which is stored in a storage medium and executed by at least one processor to implement the various processes of the above-described communication network prediction method embodiments, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0722] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0723] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0724] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for predicting communication networks, characterized in that, include: The terminal uses L models to execute the target task respectively and obtains the first result output by the L models; where L is a positive integer greater than 1; The terminal performs any of the following operations: The terminal determines a first fusion method based on the target information, and determines the prediction result of the target task based on the first fusion method and the first result; The terminal sends the first result to the network-side device, so that the network-side device determines the second fusion method based on the target information, and determines the prediction result of the target task based on the second fusion method, the first result, and the second result; wherein, the first result includes: L output results output by the L models respectively; or the fusion result of the L output results; The terminal receives the second result sent by the network-side device, the terminal determines the first fusion method based on the target information, and the terminal determines the prediction result of the target task based on the first fusion method, the first result and the second result. Wherein, the second result is obtained by the network-side device after executing the target task using M models respectively; M is a positive integer greater than 1; The first fusion method includes: The outputs of each model are filtered to obtain the prediction results; And / or, based on the weights and outputs of each model, determine the prediction results; The second fusion method includes: The outputs of each model are filtered to obtain the prediction results; And / or, based on the weights and outputs of each model, determine the prediction results.

2. The method according to claim 1, characterized in that, The method further includes: The terminal determines the L models.

3. The method according to claim 2, characterized in that, The terminal determines the L models, including any one of the following: The terminal receives first information sent by the network-side device; wherein, the first information includes configuration information of the L models; the terminal determines the L models based on the configuration information of the L models; The terminal sends a request message to the network-side device, the request message being used to request the network-side device to configure the L models; the terminal receives first information sent by the network-side device; wherein, the first information includes configuration information of the L models; the terminal determines the L models based on the configuration information of the L models; The terminal configures the L models based on at least one of the following methods: autonomously determining and informing network-side devices, protocol predefinition, or higher-layer preconfiguration; The terminal selects L models from the model pool based on the target information; wherein the model pool includes K models; K is greater than or equal to L; K is a positive integer.

4. The method according to claim 3, characterized in that, The first information includes at least one of the following: Model quantity information; Model category information; Model identification ID information; Model priority information; Model attribute information; Model accuracy information; Model error information; Information regarding the model's computational requirements; Model storage capacity requirements; Model feature information; Adaptation environment information; Processing latency information; Information on the fusion method of the output results of each model; Model lifecycle information; Measurement information input for each type of model; Output information for each category of model.

5. The method according to claim 4, characterized in that, The measurement information input to each category of model includes at least one of the following: Channel state information; Receive signal information; Historical status information; Sensor information.

6. The method according to claim 4, characterized in that, The output information of each category of model includes at least one of the following: Direct target parameters; intermediate quantity; Soft information about direct target parameters or intermediate quantities.

7. The method according to claim 3, characterized in that, The request information includes second information; wherein the second information includes at least one of the following: Terminal mobility information; Terminal environmental information; Accuracy requirements information; Task information.

8. The method according to claim 3, characterized in that, The method further includes: The terminal sends third information to the network-side device; wherein the third information is used to indicate the terminal's capability information.

9. The method according to claim 8, characterized in that, The third information includes at least one of the following: The terminal's sensor configuration information; Data types that the terminal can access; Terminal hardware capability information.

10. The method according to claim 9, characterized in that, The data type includes at least one of the following: Data format information; Data preprocessing methods.

11. The method according to claim 3, characterized in that, After the terminal receives the first information sent by the network-side device, the method further includes: The terminal sends feedback information to the network-side device; wherein, the feedback information is used to indicate whether the terminal supports the model corresponding to the model configuration information.

12. The method according to claim 1, characterized in that, The method further includes: The terminal sends a fifth piece of information to the network-side device; wherein the fifth piece of information includes at least one of the following: Terminal mobility information; Terminal environmental information; Accuracy requirements information; Task information.

13. The method according to claim 1, characterized in that, The method further includes: The terminal sends a sixth piece of information to the network-side device; wherein the sixth piece of information is used to indicate the terminal's capability information.

14. The method according to claim 13, characterized in that, The sixth piece of information includes at least one of the following: The terminal's sensor configuration information; Data types that the terminal can access; Terminal hardware capability information.

15. The method according to claim 14, characterized in that, The data type includes at least one of the following: Data format information; Data preprocessing methods.

16. The method according to claim 1, characterized in that, The method further includes: The terminal receives the seventh information sent by the network-side device; wherein the seventh information includes the configuration information of the M models.

17. The method according to claim 16, characterized in that, The seventh piece of information includes at least one of the following: Input requirements for each model; Model accuracy information; Processing latency information; Model lifecycle information.

18. The method according to claim 17, characterized in that, The method further includes: The terminal sends measurement information to the network-side device.

19. The method according to claim 1 or 3, characterized in that, The target information includes at least one of the following: Statistical information on the output results of each model; Statistical information on the output results of multiple models; Model error information for each model; Terminal mobility information; Terminal environmental information; Accuracy requirements information; Task information; Measurement information input for each type of model; Model priority information; Measurement information of the reference signal of the current terminal; Refer to the model configuration information of the terminal; Measurement information of the reference signal of the reference terminal.

20. The method according to claim 1, characterized in that, The method further includes: The terminal makes a decision on the association of the target task based on the prediction result of the target task.

21. The method according to claim 1, characterized in that, The method further includes: The terminal receives the eleventh information sent by the network-side device; wherein the eleventh information is used to indicate the prediction method based on the first result and the second result.

22. The method according to claim 21, characterized in that, The prediction method includes any of the following: Based on the first result and the second result, the network-side device determines the prediction result of the target task; Based on the first result and the second result, the terminal determines the prediction result of the target task.

23. A method for predicting communication networks, characterized in that, include: The network-side device uses M models to execute the target task respectively, and obtains the second result output by the M models; where M is a positive integer greater than 1; The network-side device performs any of the following operations: The network-side device determines the second fusion method based on the target information, and determines the prediction result of the target task based on the second fusion method and the second result; The network-side device sends the second result to the terminal, so that the terminal determines the first fusion method based on the target information, and determines the prediction result of the target task based on the first fusion method, the first result, and the second result; wherein, the second result includes: M output results output by the M models respectively; or the fusion result of the M output results; The network-side device receives the first result sent by the terminal, determines the second fusion method based on the target information, and determines the prediction result of the target task based on the second fusion method, the first result, and the second result. Wherein, the first result is obtained by the terminal using L models to execute the target task respectively; L is a positive integer greater than 1; The second fusion method includes: The outputs of each model are filtered to obtain the prediction results; And / or, based on the weights and outputs of each model, determine the prediction results; The first fusion method includes: The outputs of each model are filtered to obtain the prediction results; And / or, based on the weights and outputs of each model, determine the prediction results.

24. The method according to claim 23, characterized in that, The method further includes: The network-side device sends first information to the terminal; wherein, the first information includes configuration information of the L models; the configuration information of the L models is used by the terminal to determine the L models; or, The network-side device receives a request message sent by the terminal, the request message being used to request the network-side device to configure the L models; the network-side device configures the L models based on the request message and / or third information; the network-side device sends first information to the terminal; wherein, the first information includes configuration information of the L models; the configuration information of the L models is used by the terminal to determine the L models.

25. The method according to claim 24, characterized in that, The first information includes at least one of the following: Model quantity information; Model category information; Model identification ID information; Model priority information; Model attribute information; Model accuracy information; Model error information; Information regarding the model's computational requirements; Model storage capacity requirements; Model feature information; Adaptation environment information; Processing latency information; Information on the fusion method of the output results of each model; Model lifecycle information; Measurement information input for each type of model; Output information for each category of model.

26. The method according to claim 23, characterized in that, The method further includes: The network-side device determines the M models.

27. The method according to claim 26, characterized in that, The network-side device determines the M models, including: The network-side device configures the M models based on at least one of the following methods: autonomous determination, protocol predefinition, or preconfiguration. or, The network-side device selects M models from the model pool based on the target information; wherein the model pool includes P models; P is greater than or equal to M; and P is a positive integer.

28. The method according to claim 26, characterized in that, The method further includes: The network-side device receives a fifth message sent by the terminal; wherein the fifth message includes at least one of the following: Terminal mobility information; Terminal environmental information; Accuracy requirements information; Task information.

29. The method according to claim 26, characterized in that, The method further includes: The network-side device receives the sixth information sent by the terminal; wherein the sixth information is used to indicate the terminal's capability information.

30. The method according to claim 26, characterized in that, The method further includes: The network-side device sends a seventh message to the terminal; wherein the seventh message includes the configuration information of the M models.

31. The method according to claim 30, characterized in that, The method further includes: The network-side device receives the measurement information sent by the terminal.

32. The method according to claim 31, characterized in that, The network-side device uses M models to execute the target task respectively, and obtains a second result output by the M models, including: The network-side device, based on the measurement information sent by the terminal, uses the M models to perform the target task respectively, and obtains the second result output by the M models.

33. The method according to claim 23 or 27, characterized in that, The target information includes at least one of the following: Statistical information on the output results of each model; Statistical information on the output results of multiple models; Model error information for each model; Terminal mobility information; Terminal environmental information; Accuracy requirements information; Task information; Measurement information input for each type of model; Model priority information; Measurement information of the reference signal.

34. The method according to claim 23, characterized in that, The method further includes: The network-side device makes a decision on the association of the target task based on the prediction result of the target task.

35. The method according to claim 23, characterized in that, The method further includes: The network-side device sends an eleventh message to the terminal; wherein the eleventh message is used to indicate the prediction method based on the first result and the second result.

36. The method according to claim 35, characterized in that, The prediction method includes any of the following: Based on the first result and the second result, the network-side device determines the prediction result of the target task; Based on the first result and the second result, the terminal determines the prediction result of the target task.

37. A communication network prediction device, characterized in that, include: The first execution module is used to execute the target task using L models respectively, and obtain the first result output by the L models; where L is a positive integer greater than 1; The first prediction module is used to perform any of the following operations: A first fusion method is determined based on the target information, and the prediction result of the target task is determined based on the first fusion method and the first result; The first result is sent to the network-side device so that the network-side device determines a second fusion method based on the target information, and determines the prediction result of the target task based on the second fusion method, the first result, and the second result; wherein, the first result includes: L output results output by the L models respectively; or the fusion result of the L output results; The system receives the second result sent by the network-side device, determines the first fusion method based on the target information, and determines the prediction result of the target task based on the first fusion method, the first result, and the second result; wherein the second result is obtained by the network-side device after executing the target task using M models respectively; M is a positive integer greater than 1. The first fusion method includes: The outputs of each model are filtered to obtain the prediction results; And / or, based on the weights and outputs of each model, determine the prediction results; The second fusion method includes: The outputs of each model are filtered to obtain the prediction results; And / or, based on the weights and outputs of each model, determine the prediction results.

38. A communication network prediction device, characterized in that, include: The second execution module is used to execute the target task using M models respectively, and obtain the second result output by the M models; where M is a positive integer greater than 1; The second prediction module is used to perform any of the following operations: A second fusion method is determined based on the target information, and the prediction result of the target task is determined based on the second fusion method and the second result; The second result is sent to the terminal so that the terminal determines a first fusion method based on the target information, and determines the prediction result of the target task based on the first fusion method, the first result, and the second result; wherein the second result includes: M output results output by the M models respectively; or a fusion result of the M output results; The receiving terminal sends the first result, determines the second fusion method based on the target information, and determines the prediction result of the target task based on the second fusion method, the first result, and the second result; wherein, the first result is obtained by the terminal after executing the target task using L models respectively; L is a positive integer greater than 1; The second fusion method includes: The outputs of each model are filtered to obtain the prediction results; And / or, based on the weights and outputs of each model, determine the prediction results; The first fusion method includes: The outputs of each model are filtered to obtain the prediction results; And / or, based on the weights and outputs of each model, determine the prediction results.

39. A terminal, characterized in that, It includes a processor and a memory, the memory storing a program or instructions that can run on the processor, the program or instructions being executed by the processor to implement the steps of the communication network prediction method as described in any one of claims 1 to 22.

40. A network-side device, characterized in that, It includes a processor and a memory, the memory storing a program or instructions that can run on the processor, the program or instructions being executed by the processor to implement the steps of the communication network prediction method as described in any one of claims 23 to 36.

41. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the communication network prediction method as described in any one of claims 1 to 22, or implement the steps of the communication network prediction method as described in any one of claims 23 to 36.

Images