Data set determination method and apparatus, device, chip and storage medium

Abstract

Embodiments of the present application provide a data set determination method and apparatus, a communication device, a chip and a storage medium. The method comprises: a terminal sends a data set to a network device, wherein the data set comprises at least one input instance and at least one tag instance; the input instance is obtained by the terminal measuring a downlink reference signal in a first downlink reference signal set or a second downlink reference signal set, and the tag instance is obtained by the terminal measuring a downlink reference signal in a third downlink reference signal set; and the data set is used for training a first model on the network device, and the first model is used for predicting a target downlink reference signal and / or target link quality.

Description

A method for determining a data set, a device, a chip, and a storage medium Technical Field

[0001] The embodiments of the present application relate to the field of communication technology, and in particular to a method for determining a data set, and an apparatus, device, chip, and storage medium. Background Art

[0002] During downlink beam (pair) scanning, a large number of reference signal measurements are required, resulting in significant downlink resource overhead and measurement latency. For example, each Transmission Reception Point (TRP) in Frequency Range 2 (FR2) can deploy 64 different downlink transmission directions (carried by 64 Synchronization Signal and PBCH Blocks (SSBs)). When the terminal uses multi-beam reception (assuming each antenna panel has four different receive beams), the terminal needs to measure at least 64×4=256 beam pairs, which means that 256 resources of downlink resource overhead are required.

[0003] Related technologies have explored model-based beam (pair) prediction mechanisms, primarily aimed at reducing downlink resource overhead and measurement latency. However, model training requires the collection of large amounts of data. Therefore, relevant technologies are needed to support the collection of large amounts of data for model training.

[0004] Summary of the Invention

[0005] Embodiments of the present application provide a method, apparatus, device, chip, and storage medium for determining a data set.

[0006] In a first aspect, an embodiment of the present application provides a method for determining a data set, the method comprising:

[0007] A terminal sends a data set to a network device, where the data set includes at least one input instance and at least one label instance; the input instance is obtained based on the terminal measuring a downlink reference signal in a first downlink reference signal set or a second downlink reference signal set, and the label instance is obtained based on the terminal measuring a downlink reference signal in a third downlink reference signal set; wherein the data set is used to train a first model on the network device side, and the first model is used to predict a target downlink reference signal and / or a target link quality.

[0008] In a second aspect, an embodiment of the present application provides a method for determining a data set, the method comprising:

[0009] The terminal obtains a data set, where the data set includes at least one input instance and at least one label instance; the input instance is obtained based on measuring a downlink reference signal in a first downlink reference signal set or a second downlink reference signal set, and the label instance is obtained based on measuring a downlink reference signal in a third downlink reference signal set; wherein the data set is used to adjust a second model on the terminal side, and the second model is used to predict a target downlink reference signal and / or a target link quality.

[0010] In a third aspect, an embodiment of the present application provides a method for determining a data set, the method comprising:

[0011] A network device receives a data set sent by a terminal, where the data set includes at least one input instance and at least one label instance; the input instance is obtained based on the terminal measuring a downlink reference signal in a first downlink reference signal set or a second downlink reference signal set, and the label instance is obtained based on the terminal measuring a downlink reference signal in a third downlink reference signal set; wherein the data set is used to train a first model on the network device side, and the first model is used to predict a target downlink reference signal and / or a target link quality.

[0012] In a fourth aspect, an embodiment of the present application provides a device for determining a data set, applied to a terminal, the device comprising:

[0013] A sending unit is used to send a data set to a network device, where the data set includes at least one input instance and at least one label instance; the input instance is obtained based on the terminal measuring the downlink reference signals in the first downlink reference signal set or the second downlink reference signal set, and the label instance is obtained based on the terminal measuring the downlink reference signals in the third downlink reference signal set; wherein the data set is used to train a first model on the network device side, and the first model is used to predict a target downlink reference signal and / or a target link quality.

[0014] In a fifth aspect, an embodiment of the present application provides a device for determining a data set, applied to a terminal, the device comprising:

[0015] An acquisition unit is used to acquire a data set, the data set including at least one input instance and at least one label instance; the input instance is obtained based on measuring a downlink reference signal in a first downlink reference signal set or a second downlink reference signal set, and the label instance is obtained based on measuring a downlink reference signal in a third downlink reference signal set; wherein the data set is used to adjust a second model on the terminal side, and the second model is used to predict a target downlink reference signal and / or a target link quality.

[0016] In a sixth aspect, an embodiment of the present application provides a device for determining a data set, applied to a network device, the device comprising:

[0017] A receiving unit is used to receive a data set sent by a terminal, the data set including at least one input instance and at least one label instance; the input instance is obtained based on the terminal measuring the downlink reference signal in the first downlink reference signal set or the second downlink reference signal set, and the label instance is obtained based on the terminal measuring the downlink reference signal in the third downlink reference signal set; wherein the data set is used to train a first model on the network device side, and the first model is used to predict a target downlink reference signal and / or a target link quality.

[0018] In the seventh aspect, an embodiment of the present application provides a communication device, which includes a memory and a processor; wherein the memory is used to store computer-executable instructions; the processor is connected to the memory and is used to implement the method described in any of the above aspects by executing the computer-executable instructions.

[0019] In an eighth aspect, an embodiment of the present application provides a chip, which includes a processor configured to call and execute a computer program from a memory, so that a device equipped with the chip executes the method described in any one of the above aspects.

[0020] In a ninth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by at least one processor, it implements the method described in any of the above aspects.

[0021] The technical solution of the embodiment of the present application is as follows: on the one hand, for the training of the first model on the network device side, the terminal sends a data set for training the first model to the network device, the data set including at least one input instance and at least one label instance, wherein the input instance is obtained by the terminal by measuring the downlink reference signal in the first downlink reference signal set or the second downlink reference signal set, and the label instance is obtained by the terminal by measuring the downlink reference signal in the third downlink reference signal set, thereby realizing the collection of the data set for model training on the network device side. On the other hand, for the training of the second model on the terminal side, the terminal can obtain a small amount of data set for adjusting the second model on the terminal side, the data set including at least one input instance and at least one label instance, wherein the input instance is obtained by the terminal by measuring the downlink reference signal in the first downlink reference signal set or the second downlink reference signal set, and the label instance is obtained by the terminal by measuring the downlink reference signal in the third downlink reference signal set, thereby realizing the collection of the data set for model training on the terminal side. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0023] FIG1 is a schematic diagram of an application scenario of an embodiment of the present application;

[0024] Figure 2 is a schematic diagram of an example of downlink beam management in an NR system;

[0025] Figure 3 is a schematic diagram of an example of neuron structure;

[0026] FIG4 is a schematic diagram of an example of a neural network;

[0027] FIG5 is a schematic diagram of an example of a neural network model applicable to an embodiment of the present application;

[0028] FIG6 is a schematic diagram of another example of a neural network model applicable to an embodiment of the present application;

[0029] FIG7 is a flow chart of a method for determining a data set provided in an embodiment of the present application;

[0030] FIG8 is a format of a MAC layer container provided in an embodiment of the present application;

[0031] FIG9 is a schematic diagram of a data set collection process according to Scenario 1 provided in an embodiment of the present application;

[0032] FIG10 is a schematic diagram of a data set collection process according to scenario 2 provided in an embodiment of the present application;

[0033] FIG11 is a flow chart of another method for determining a data set provided in an embodiment of the present application;

[0034] FIG12 is a first schematic diagram of the structure of a device for determining a data set provided in an embodiment of the present application;

[0035] FIG13 is a second schematic diagram of the structure of the apparatus for determining a data set provided in an embodiment of the present application;

[0036] FIG14 is a third schematic diagram of the structure of the apparatus for determining a data set provided in an embodiment of the present application;

[0037] FIG15 is a schematic structural diagram of a communication device provided in an embodiment of the present application;

[0038] FIG16 is a schematic structural diagram of a chip according to an embodiment of the present application;

[0039] Figure 17 is a schematic block diagram of a communication system provided by an embodiment of the present application. DETAILED DESCRIPTION

[0040] The following will describe the technical solutions in the embodiments of this application in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0041] FIG1 is a schematic diagram of an application scenario of an embodiment of the present application.

[0042] As shown in Figure 1, a communication system 100 may include a terminal 110 and a network device 120. The network device 120 may communicate with the terminal 110 via an air interface. The terminal 110 and the network device 120 support multi-service transmission.

[0043] It should be understood that the embodiments of the present application are only illustrative of the communication system 100, but the embodiments of the present application are not limited thereto. That is, the technical solutions of the embodiments of the present application can be applied to various communication systems, such as: Long Term Evolution (LTE) system, LTE Time Division Duplex (TDD), Universal Mobile Telecommunication System (UMTS), Internet of Things (IoT) system, Narrow Band Internet of Things (NB-IoT) system, enhanced Machine-Type Communications (eMTC) system, 5G communication system (also known as New Radio (NR) communication system), or future communication systems.

[0044] 1 , the network device 120 may be an access network device that communicates with the terminal 110. The access network device may provide communication coverage for a specific geographical area and may communicate with the terminal 110 (eg, UE) located within the coverage area.

[0045] The network device 120 may be an evolved Node B (eNB or eNodeB) in a Long Term Evolution (LTE) system, or a Next Generation Radio Access Network (NG RAN) device, or a base station (gNB) in an NR system, or a wireless controller in a Cloud Radio Access Network (CRAN), or the network device 120 may be a relay station, an access point, an in-vehicle device, a wearable device, a hub, a switch, a bridge, a router, or a network device in a future evolved Public Land Mobile Network (PLMN), etc.

[0046] The terminal 110 may be any terminal, including but not limited to a terminal connected to the network device 120 or other terminals by wired or wireless means.

[0047] For example, the terminal 110 may refer to an access terminal, user equipment (UE), a subscriber unit, a subscriber station, a mobile station, a mobile station, a remote station, a remote terminal, a mobile device, a user terminal, a terminal, a wireless communication device, a user agent, or a user device. An access terminal may be a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, an IoT device, a satellite handheld terminal, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA), a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, an in-vehicle device, a wearable device, a terminal in a 5G network or a terminal in a future evolution network, etc.

[0048] The terminal 110 may be used for device-to-device (D2D) communication.

[0049] The wireless communication system 100 may further include a core network device 130 for communicating with the base station. The core network device 130 may be a 5G core network (5G Core, 5GC) device, such as an Access and Mobility Management Function (AMF), an Authentication Server Function (AUSF), a User Plane Function (UPF), or a Session Management Function (SMF). Optionally, the core network device 130 may also be an Evolved Packet Core (EPC) device of an LTE network, such as a Session Management Function + Core Packet Gateway (SMF+PGW-C) device. It should be understood that SMF+PGW-C can simultaneously implement the functions that can be implemented by SMF and PGW-C. During the network evolution process, the above-mentioned core network device may also be called other names, or a new network entity may be formed by dividing the functions of the core network, which is not limited in the embodiments of the present application.

[0050] The functional units in the communication system 100 may also establish connections and implement communication via next generation (NG) network interfaces.

[0051] For example, the terminal establishes an air interface connection with the access network device through the NR interface for transmitting user plane data and control plane signaling; the terminal can establish a control plane signaling connection with the AMF through the NG interface 1 (referred to as N1); the access network device, such as the next generation wireless access base station (gNB), can establish a user plane data connection with the UPF through the NG interface 3 (referred to as N3); the access network device can establish a control plane signaling connection with the AMF through the NG interface 2 (referred to as N2); the UPF can establish a control plane signaling connection with the SMF through the NG interface 4 (referred to as N4); the UPF can exchange user plane data with the data network through the NG interface 6 (referred to as N6); the AMF can establish a control plane signaling connection with the SMF through the NG interface 11 (referred to as N11); the SMF can establish a control plane signaling connection with the PCF through the NG interface 7 (referred to as N7).

[0052] Figure 1 exemplarily shows a base station, a core network device and two terminals. Optionally, the wireless communication system 100 may include multiple base station devices and each base station may include other numbers of terminals within its coverage area, which is not limited in the embodiments of the present application.

[0053] It should be noted that Figure 1 is merely an example of a system applicable to this application. Of course, the methods described in the embodiments of this application can also be applied to other systems. Furthermore, the terms "system" and "network" are often used interchangeably herein. The term "and / or" herein simply describes an association relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " generally indicates that the associated objects are in an "or" relationship. It should also be understood that the "indication" mentioned in the embodiments of this application can be a direct indication, an indirect indication, or an indication of an association relationship. For example, "A indicates B" can mean that A directly indicates B, for example, B can obtain information through A; it can also mean that A indirectly indicates B, for example, A indicates C, and B can obtain information through C; or it can mean that A and B have an association relationship. It should also be understood that the "correspondence" mentioned in the embodiments of this application can mean that there is a direct or indirect correspondence between two objects, or that there is an association relationship between the two objects, or a relationship between an indicator and the indicated, a configuration and the configured, and so on. It should also be understood that the “predefined” or “predefined rules” mentioned in the embodiments of the present application can be implemented by pre-saving corresponding codes, tables or other methods that can be used to indicate relevant information in a device (for example, including a terminal and a network device), and the present application does not limit its specific implementation method. For example, predefined can refer to a definition in a protocol. It should also be understood that in the embodiments of the present application, the “protocol” may refer to a standard protocol in the field of communications, for example, it may include an LTE protocol, an NR protocol, and related protocols used in future communication systems, and the present application does not limit this.

[0054] To facilitate understanding of the technical solutions of the embodiments of the present application, the relevant technologies of the embodiments of the present application are described below. The following relevant technologies can be arbitrarily combined with the technical solutions of the embodiments of the present application as optional solutions, and they all fall within the protection scope of the embodiments of the present application.

[0055] 1. NR beam management

[0056] The first version of the NR system (Rel.15) introduced millimeter wave frequency band communications, which also introduced the beam management mechanism. Beam management can be divided into uplink and downlink beam management. Among them, downlink beam management includes: downlink beam sweeping (Beam Sweeping), UE beam measurement and reporting (Measurement & Reporting), downlink beam indication (Beam Indication) of the network (NW), etc.; uplink beam management includes: uplink beam sweeping (UE sends beam sweeping SRS), NW beam measurement, and uplink beam indication and configuration for UE.

[0057] Figure 2 is a schematic diagram of an example of downlink beam management in the NR system. Taking the downlink beam scanning process as an example, it can be divided into three processes, namely P1, P2 and P3 processes. Among them, the P1 process refers to the NW scanning different transmit beams and the UE scanning different receive beams; the P2 process refers to the NW scanning different transmit beams and the UE using the same receive beam; the P3 process refers to the NW using the same transmit beam and the UE scanning different receive beams. Generally, the P1 process can be executed before the P2 and P3 processes. For example, in the example of Figure 2, the execution order can be P1, P2 and P3. Generally, the NW completes the above beam scanning process by sending a downlink reference signal SSB and / or a channel state information-reference signal (CSI-RS).

[0058] The beam reporting mechanism in NR Rel.15 includes: the UE measures multiple transmit beams (P2 process) or transmit-receive beam pairs (P1 process) and reports the L transmit beams with the highest layer 1 reference signal receiving power (L1-Reference Signal Receiving Power, RSRP, L1-RSRP) and their performance to the NW in the form of channel state information (CSI). After decoding the beam information reported by the UE, the NW can consider the downlink transmission channel and signal, and carry the transmission configuration indicator (TCI) state (TCI State) (including the SSB or CSI-RS resource index as a reference for the UE) through medium access control (MAC) and / or downlink control information (DCI) signaling to indicate the beam information to the UE. The UE uses the receive beam corresponding to the transmit beam of the indicated SSB or CSI-RS for downlink reception.

[0059] Accordingly, NR defines three uplink beam scanning processes, namely U1, U2, and U3. Among them, the U1 process includes: the UE scans different transmit beams, and the NW scans different receive beams; the U2 process includes: the UE uses the same transmit beam, and the NW scans different receive beams; the U3 process includes: the UE scans different transmit beams, and the NW uses the same receive beam. For the uplink beam scanning process, since the NW measures the beam from the UE, the UE beam report is not required. The NW can select the appropriate uplink beam from the measured uplink beam or configure it to the UE for uplink transmission. At the same time, the NW can prepare the corresponding receive beam.

[0060] 2. Neural Network (NN)

[0061] 1) Neuronal structure

[0062] A neural network is a computational model consisting of multiple interconnected neuron nodes. The connections between the nodes represent the weighted values ​​from the input signal to the output signal, which are called weights. Each node performs a weighted summation of different input signals and outputs them through a specific activation function. The neuron structure is shown in Figure 3, where a1, a2, ..., a n and 1 are the inputs of neurons, w1, w2, ..., w n and b represent weights, Sum represents the summation function, f represents the activation function, and t is the output result.

[0063] 2) Deep Neural Networks

[0064] Figure 4 shows a simple neural network. The neural network includes an input layer, a hidden layer, and an output layer. By connecting multiple neurons in different ways, combined with different weights and activation functions, different outputs can be generated, thereby fitting the mapping relationship from input to output. For example, each upper-level node can be connected to all of its lower-level nodes to form a fully connected model. This fully connected model can also be called a deep neural network (DNN). This NN model can be used for spatial-domain DL beam prediction or temporal DL beam prediction.

[0065] 3) Beam management based on artificial intelligence (AI) / machine learning (ML)

[0066] AI / ML-based beam management is the main use case of the Release 18 AI project, and two use cases are defined: spatial-domain DL beam prediction and temporal DL beam prediction.

[0067] Definition of spatial-domain beam prediction (BM-Case 1): Downlink beams in Set A are spatially predicted by measuring beams (pairs) in Set B. Set B is either a subset of Set A or different sets of beams (pairs). Set B can be considered a subset of Set A, while Set A can be considered the full set of beams (pairs).

[0068] Definition of time-domain beam prediction (BM-Case2): The optimal beam (or beam pair) in Set A is predicted P times in the future by measuring the beam (or beam pair) in Set B N times. Set B can be a subset of Set A or the same as Set A.

[0069] Figure 5 is a schematic diagram of an example of a neural network model applicable to an embodiment of the present application. The model is an optimal beam (pair) prediction model, and it can be considered that the model solves a multi-classification problem. The model can be used to fit the relationship between the measurement results of Set B (such as the L1-RSRP of the reference signal / beam (pair) in Set B) and the optimal L beams (pairs) in Set A. Among them, the measurement results of Set B can be used as the input of the model, and the output can be the optimal K beam (pair) indexes selected from the entire set (Set A), that is, the K beams (pairs) with the highest L1-RSRP in Set A. In the example of Figure 5, the number of beams (pairs) in Set B is T, the number of beams (pairs) in Set A is M, K=1, and beam (pair) #2 is the beam (pair) with the highest L1-RSRP, that is, the optimal beam (pair). The label used by the model is the optimal (that is, the highest L1-RSRP) beam (pair) index measured in the entire set.

[0070] Figure 6 is another schematic diagram of a neural network model applicable to an embodiment of the present application, which is an optimal beam quality prediction model and can be understood as a linear regression problem. The input and output relationship of the model is: the relationship from the partial subset input L1-RSRP to the optimal K beam (pair) L1-RSRP. The input part of the model is the same as the input of the model shown in Figure 5, except that the output of the model is K (K≥1) optimal L1-RSRPs and the K beam (pair) indices corresponding to the K optimal L1-RSRPs. In the example of Figure 6, the number of beams (pairs) in Set B is T. The labels used by the model are the optimal K L1-RSRPs measured in the entire set, and the corresponding K beam (pair) indices.

[0071] It should be understood that the output of the model can also be understood as an inference or prediction result, that is, inference and prediction can represent the same meaning and can be replaced with each other.

[0072] To facilitate understanding of the technical solutions of the embodiments of the present application, the technical solutions of the present application are described in detail below through specific embodiments. The above related technologies can be combined arbitrarily with the technical solutions of the embodiments of the present application as optional solutions, and all of them fall within the scope of protection of the embodiments of the present application. The embodiments of the present application include at least part of the following contents.

[0073] It should be noted that the "beam (pair)" in the embodiments of the present application can refer to a beam, including a transmit beam or a receive beam, or a beam pair, such as a pair of transmit beams and a receive beam. The meaning of the "beam (pair)" in the embodiments of the present application applies to downlink transmission. The "beam (pair)" in the embodiments of the present application can also be referred to as a spatial filter, that is, the terms "beam (pair)" and "spatial filter" are interchangeable.

[0074] It should be noted that the "downlink reference signal resource index" in the embodiments of the present application may also be referred to as a beam (pair) index, a downlink reference signal index, or a spatial filter index. That is, the downlink reference signal resource index, beam (pair) index, downlink reference signal index, and spatial filter index may be interchangeable. For example, the downlink reference signal resource index may be a CSI-RS resource indicator (CRI) or an SSB resource indicator (SSBRI).

[0075] It should be noted that the downlink reference signal in the embodiment of the present application may include: Channel State Information-Reference Signal (CSI-RS) and / or SSB. The measurement result obtained by measuring the downlink reference signal characterizes the link quality. In some embodiments, the link quality may include at least one of the following: Reference Signal Receiving Power (RSRP) (such as L1-RSRP), Signal to Interference plus Noise Ratio (SINR) (such as L1-SINR), Received Signal Strength Indicator (RSSI) (such as L1-RSSI), Reference Signal Receiving Quality (RSRQ) (such as L1-RSRQ).

[0076] First, a first downlink reference signal set, a second downlink reference signal set, and a third downlink reference signal set are defined.

[0077] The first downlink reference signal set is Set B, the second downlink reference signal set is Set C, and the third downlink reference signal set is Set A. A "downlink reference signal set" can be understood as a set of downlink reference signals, a set of downlink reference signal resources, or a set of beams (pairs). Therefore, a "downlink reference signal set" can also be referred to as a "downlink reference signal resource set" or a "beam (pair) set."

[0078] In some scenarios, Set B can be a subset of Set A. In this case, Set B can be understood as a partial subset of beams (pairs), and Set A can be understood as the full set of beams (pairs). In other scenarios, Set B and Set A can also be two different sets of beams (pairs). For example, Set B can be used as a measurement set, and Set A can be used as a prediction set. In an embodiment of the present application, a neural network model (referred to as the model) can predict the beams in Set A by measuring the beams in Set B.

[0079] In some scenarios, Set B is the measurement set and also the model input set. In other scenarios, Set C is the measurement set, and the optimal M beams (pairs) are selected from Set C as Set B, which is the model input set. In other scenarios, Set B is a subset of Set C.

[0080] Next, define the first model and the second model.

[0081] The first model is a model on the network device side (also referred to as a model on the NW side), or in other words, the first model is a model deployed on the network device side. The first model is used to predict the downlink beam (pair), specifically, the first model is used to predict the target downlink reference signal and / or target link quality, the target downlink reference signal includes the K downlink reference signals corresponding to the optimal K beams (pairs) in Set A, and the target link quality includes the K link qualities corresponding to the optimal K beams (pairs) in Set A. The first model can realize beam prediction in the spatial domain or beam prediction in the time domain. The input of the first model is: the measurement results of Set B or part of the measurement results of Set C (such as the measurement results of the optimal M beams (pairs) in Set C); the output of the first model is at least one of the indices of the optimal K beams (pairs) in Set A, the optimal K L1-RSRPs in Set A, and the indices of the K beams (pairs) corresponding to the optimal K L1-RSRPs.

[0082] The second model is a model on the terminal side (also called a model on the UE side), or in other words, the second model is a model deployed on the terminal side. The second model is used to predict the downlink beam (pair), specifically, the second model is used to predict the target downlink reference signal and / or target link quality, the target downlink reference signal includes the K downlink reference signals corresponding to the optimal K beams (pairs) in Set A, and the target link quality includes the K link qualities corresponding to the optimal K beams (pairs) in Set A. The second model can realize beam prediction in the spatial domain or beam prediction in the time domain. The input of the second model is: the measurement results of Set B or part of the measurement results of Set C (such as the measurement results of the optimal M beams (pairs) in Set C); the output of the second model is at least one of the indexes of the optimal K beams (pairs) in Set A, the optimal K L1-RSRPs in Set A, and the indexes of the K beams (pairs) corresponding to the optimal K L1-RSRPs.

[0083] It should be noted that the second model can be an initial model (also called a meta-model). Compared to the first model, this initial model does not need to be trained from scratch. This initial model can be a pre-trained model. Therefore, compared to the first model, the second model only requires a small amount of data to adjust the model, that is, fine-tuning the second model with a small amount of data.

[0084] For example, the first model and the second model can be the neural network model shown in Figure 5 or Figure 6 above, or can be a modified model based on the neural network model shown in Figure 5 or Figure 6. This application does not limit the specific network structure of the first model and the second model. The neural network model can also be called an AI model or an ML model.

[0085] For the first model on the network device side, the terminal needs to report a large amount of data (called a data set) to the network device. The terminal transmits the data set to the network device via an air interface.

[0086] FIG7 is a flow chart of a method for determining a data set provided in an embodiment of the present application. As shown in FIG7 , the method includes the following steps:

[0087] Step 701: The terminal sends a data set to the network device, and the network device receives the data set sent by the terminal, where the data set includes at least one input instance and at least one label instance; the input instance is obtained based on the terminal measuring the downlink reference signal in the first downlink reference signal set or the second downlink reference signal set, and the label instance is obtained based on the terminal measuring the downlink reference signal in the third downlink reference signal set; wherein the data set is used to train a first model on the network device side, and the first model is used to predict the target downlink reference signal and / or target link quality.

[0088] For ease of description, in the examples described below, the first downlink reference signal set is referred to as Set B, the second downlink reference signal set is referred to as Set A, and the third downlink reference signal set is referred to as Set C. For more information about Set A, Set B, and Set C, refer to the aforementioned description. The downlink reference signal (or downlink reference signal resource) may include, for example, a CSI-RS (or a CSI-RS resource) and / or an SSB (or an SSB resource).

[0089] Dataset collection

[0090] 1) Collection of input instances

[0091] In this embodiment of the present application, the terminal measures the downlink reference signal in the first downlink reference signal set or the second downlink reference signal set to obtain at least one input instance. In some embodiments, "at least one input instance" can be replaced by multiple input instances. In other embodiments, "at least one input instance" can be replaced by one input instance.

[0092] In some implementations, the terminal performs one measurement on the downlink reference signals in the first downlink reference signal set or the second downlink reference signal set to obtain one input instance. In other implementations, the terminal performs N (N≥1) measurements on the downlink reference signals in the first downlink reference signal set or the second downlink reference signal set to obtain N input instances.

[0093] In the embodiments of this application, "input instance" may also be referred to as input, specifically referring to the input of a model. For the first model, the input is the measurement results of Set B or a portion of the measurement results of Set C (e.g., the measurement results of the optimal M beams (pairs) in Set C).

[0094] Illustratively, in a measurement of Set B, the terminal measures each downlink reference signal in Set B to obtain the link quality corresponding to each downlink reference signal. For example, the link quality may include at least one of the following: RSRP (e.g., L1-RSRP), SINR (e.g., L1-SINR), RSSI (e.g., L1-RSSI), and RSRQ (e.g., L1-RSRQ). As input to the first model, the measurement result of Set B refers to the link quality corresponding to each downlink reference signal in Set B measured by the terminal.

[0095] Illustratively, in a measurement of Set C, the terminal measures each downlink reference signal in Set C to obtain the link quality corresponding to each downlink reference signal. For example, the link quality may include at least one of the following: RSRP (e.g., L1-RSRP), SINR (e.g., L1-SINR), RSSI (e.g., L1-RSSI), and RSRQ (e.g., L1-RSRQ). As input to the first model, the partial measurement results of Set C refer to the link quality corresponding to some downlink reference signals in Set C measured by the terminal, such as the link quality corresponding to the optimal M downlink reference signals in Set C.

[0096] 2) Enter the content of the instance

[0097] In the embodiments of this application, the content of the input instance is clarified. In some embodiments, the content of the input instance can be the following two:

[0098] The first type: the input instance includes M link quality measurement results, where M is a positive integer.

[0099] In some implementations, the M link quality measurement results are obtained based on the terminal measuring the first downlink reference signal set.

[0100] Exemplarily, the M link quality measurement results refer to M link qualities obtained by the terminal by measuring the M downlink reference signals in Set B.

[0101] In some implementations, the M link quality measurement results are input into the first model in a fixed order of Set B, that is, the fixed order represents implicit downlink reference signal resource index information.

[0102] Exemplarily, the first model has M input nodes, and the M link quality measurement results need to be input into the M input nodes in a fixed order in Set B, thereby ensuring that the correspondence between the M link quality measurement results and the M downlink reference signal resource indexes is matched. In this way, the input instance can only include M link quality measurement results (e.g., M L1-RSRPs). By inputting the M link quality measurement results into the first model in a fixed order in Set B, the first model can correctly identify the downlink reference signal resource index corresponding to each link quality measurement result.

[0103] The second type: the input instance includes M link quality measurement results and M downlink reference signal resource indexes, the M link quality measurement results and the M downlink reference signal resource indexes have a corresponding relationship, and M is a positive integer.

[0104] In some embodiments, the M link quality measurement results are obtained based on the terminal measuring the first downlink reference signal set; or, the M link quality measurement results are the best M link quality measurement results among multiple link quality measurement results, and the multiple link quality measurement results are obtained based on the terminal measuring the second downlink reference signal set.

[0105] Exemplarily, the M link quality measurement results refer to the M link qualities obtained by the terminal measuring the M downlink reference signals in Set B, or the M link quality measurement results refer to the optimal M link qualities obtained by the terminal measuring multiple downlink reference signals in Set C. Each link quality measurement result of the M link quality measurement results refers to a link quality measurement result corresponding to a downlink reference signal (or a downlink reference signal resource). Therefore, the M link quality measurement results correspond to the M downlink reference signal resource indexes, and the correspondence is a one-to-one relationship.

[0106] Since the input instance includes M downlink reference signal resource indexes, it can be understood that the input instance provides explicit downlink reference signal resource index information.

[0107] Exemplarily, the first model has M input nodes, and M link quality measurement results (such as M L1-RSRPs) and corresponding M downlink reference signal resource indices (such as M CRIs or SSBRIs) can be input into the M input nodes in any order. Regardless of the order, the correspondence between the M link quality measurement results and the M downlink reference signal resource indices is always matched. In this way, the input instance includes M link quality measurement results (such as M L1-RSRPs) and corresponding M downlink reference signal resource indices (such as M CRIs or SSBRIs), which enables the first model to correctly identify the downlink reference signal resource index corresponding to each link quality measurement result.

[0108] The content of the first input example above can be referred to as input 1 (input 1). The content of the second input example above can be referred to as input 2 (input 2).

[0109] Input 1 contains M link quality measurement results (eg, M L1-RSRPs). In this case, Set B is a fixed measurement set configured and activated by the network device.

[0110] Input 2 contains M link quality measurements (e.g., M L1-RSRPs) and corresponding M downlink reference signal resource indices (e.g., M CRIs or SSBRIs). In this case, Set B is a variable measurement set that follows a pre-defined pattern, or Set B is the optimal M downlink reference signals measured in Set C.

[0111] Here, Set B is a variable measurement set according to a pre-set mode. It can be understood that measurement sets with multiple modes are pre-set. Measurement sets of different modes contain different downlink reference signals. Set B can be flexibly changed in measurement sets of multiple modes.

[0112] Here, Set B is the optimal M downlink reference signals measured in Set C. It can be understood that the optimal M downlink reference signals obtained by measuring multiple downlink reference signals in Set C form Set B, which serves as the input set of the model.

[0113] 3) Collection of label instances

[0114] In this embodiment of the present application, the terminal measures the downlink reference signal in the third downlink reference signal set to obtain at least one label instance. In some embodiments, "at least one label instance" can be replaced by a plurality of label instances. In other embodiments, "at least one label instance" can be replaced by a single label instance.

[0115] In some implementations, the terminal performs one measurement on the downlink reference signal in the third downlink reference signal set to obtain one label instance. In other implementations, the terminal performs P (P ≥ 1) measurements on the downlink reference signal in the third downlink reference signal set to obtain P label instances.

[0116] In the embodiments of the present application, a "label instance" may also be referred to as a label, specifically referring to a label corresponding to a model input. For the first model, the label corresponding to the model input includes at least one of the following: the optimal downlink reference signal resource index in Set A, the K optimal link quality measurement results (K ≥ 1) in Set A, and the K downlink reference signal resource indices corresponding to the K optimal link quality measurement results (K ≥ 1) in Set A.

[0117] Illustratively, in a measurement of Set A, the terminal measures each downlink reference signal in Set A, selects the optimal downlink reference signal resource index in Set A as the label of the first model, and / or selects the optimal K (K≥1) link quality measurement results and / or the corresponding K downlink reference signal resource indexes in Set A as the label of the first model. For example, the link quality may include at least one of the following: RSRP (such as L1-RSRP), SINR (such as L1-SINR), RSSI (such as L1-RSSI), and RSRQ (such as L1-RSRQ).

[0118] 4) Content of the tag instance

[0119] In the embodiments of this application, the content of the tag instance is clarified. In some embodiments, the content of the tag instance can be the following two:

[0120] Type 1: the label instance includes a first label; the first label includes a downlink reference signal resource index that meets a first condition, and the downlink reference signal resource index that meets the first condition is determined based on the terminal measuring the downlink reference signal in the third downlink reference signal set.

[0121] In some implementations, the downlink reference signal resource index that meets the first condition is a downlink reference signal resource index corresponding to an optimal link quality measurement result obtained by the terminal measuring downlink reference signals in the third downlink reference signal set.

[0122] Exemplarily, the terminal measures each downlink reference signal in Set A, and selects the best downlink reference signal resource index in Set A (ie, the downlink reference signal resource index corresponding to the best link quality measurement result) as the first label.

[0123] The second type: the label instance includes a first label and a second label; the first label includes a downlink reference signal resource index that meets the first condition, and the downlink reference signal resource index that meets the first condition is determined based on the terminal measuring the downlink reference signal in the third downlink reference signal set; the second label includes K link quality measurement results and / or K downlink reference signal resource indexes that meet the second condition, the K link quality measurement results and the K downlink reference signal resource indexes have a corresponding relationship, and the K link quality measurement results that meet the second condition are determined based on the terminal measuring the downlink reference signal in the third downlink reference signal set, and K is a positive integer.

[0124] In some implementations, the downlink reference signal resource index that meets the first condition is a downlink reference signal resource index corresponding to an optimal link quality measurement result obtained by the terminal measuring downlink reference signals in the third downlink reference signal set.

[0125] In some implementations, the K link quality measurement results that meet the second condition refer to: the best first K link quality measurement results obtained by the terminal measuring the downlink reference signals in the third downlink reference signal set.

[0126] Exemplarily, the terminal measures each downlink reference signal in Set A, selects the optimal downlink reference signal resource index in Set A (i.e., the downlink reference signal resource index corresponding to the optimal link quality measurement result) as the first label, and selects the optimal K (K≥1) link quality measurement results and / or corresponding K downlink reference signal resource indexes in Set A as the second label.

[0127] The first label can be referred to as label 1 (label1) and the second label can be referred to as label 2 (label2).

[0128] In label 1, the optimal downlink reference signal resource index in Set A (ie, the optimal Top-1 downlink reference signal resource index in Set A) can be predicted using a multi-classification model.

[0129] In label 2, the best K (K ≥ 1) link quality measurement results and / or the corresponding K downlink reference signal resource indices in Set A (i.e., the best Top-K (K ≥ 1) link quality measurement results and / or the corresponding K downlink reference signal resource indices in Set A) can be predicted using linear regression.

[0130] A complete sample needs to include both label instances and input instances. In some implementations, to reduce data collection overhead, a sample can be one of the following combinations of label instances and input instances:

[0131] Combination A of label instance and input instance: label 1 + input 1

[0132] Combination B of label instance and input instance: label 1 + input 2

[0133] Combination C of label instance and input instance: label1+label2+input1

[0134] The combination of label instance and input instance D: label1+label2+input2

[0135] It should be noted that label 1 is used for downlink reference signal resource index prediction and must be included in the sample. Label 2 is used for link quality (such as L1-RSRP) prediction and can be optionally included in the sample.

[0136] Dataset reporting format

[0137] In some implementations, for the M link quality measurement results in the input instance, the M link quality measurement results have the first format or the second format in the input instance:

[0138] The first format includes: M link quality measurement results;

[0139] The second format includes: a first link quality measurement result and M-1 differential values, where the first link quality measurement result is one of the M link quality measurement results, and the M-1 differential values ​​are differential values ​​of the M-1 link quality measurement results other than the first link quality measurement result among the M link quality measurement results relative to the first link quality measurement result.

[0140] In the first format described above, the M link quality measurement results are not differentially reported.

[0141] In the second format, the M link quality measurement results are differentially reported. In some implementations, the first link quality measurement result may be the best link quality measurement result among the M link quality measurement results, or any link quality measurement result among the M link quality measurement results.

[0142] In some implementations, the second format may further include: a downlink reference signal resource index corresponding to the first link quality measurement result.

[0143] In some embodiments, for the K link quality measurement results in the tag instance (for the case where the tag instance includes K link quality measurement results), the K link quality measurement results have the third format or the fourth format in the tag instance;

[0144] The third format includes: K link quality measurement results;

[0145] The fourth format includes: a second link quality measurement result and K-1 differential values, where the second link quality measurement result is one of the K link quality measurement results, and the K-1 differential values ​​are differential values ​​of the K-1 link quality measurement results other than the second link quality measurement result among the K link quality measurement results relative to the second link quality measurement result.

[0146] In the third format above, the K link quality measurement results are not differentially reported.

[0147] In the fourth format, the K link quality measurement results are differentially reported. In some implementations, the second link quality measurement result may be the best link quality measurement result among the K link quality measurement results, or any link quality measurement result among the K link quality measurement results.

[0148] In an embodiment of the present application, a terminal sends a data set to a network device via an air interface. In some implementations, the data set is carried in at least one of the following signaling: Radio Resource Control (RRC) signaling, Uplink Control Information (UCI), or a Media Access Control (MAC) layer container.

[0149] 1) Data set carried in RRC signaling

[0150] The RRC signaling of the control plane belongs to high-layer signaling, which is characterized by a large carrying capacity. Therefore, a large amount of data sets can be transmitted through RRC signaling.

[0151] 2) The dataset is hosted in UCI

[0152] The UCI of the physical layer is characterized by low latency. Therefore, low-latency transmission can be achieved by carrying data sets through UCI.

[0153] As an example, the UCI reporting formats used by several data sets are given below. It should be noted that the downlink reference signal resource index (CRI or SSBR) can also be replaced by a downlink reference signal index or a beam (pair) index or a spatial filter index.

[0154] Table 1 below shows a UCI reporting format for sample A. Sample A includes label 1 and input 1. The M RSRPs of input 1 are given in a fixed order of Set B, and no differential reporting is performed.

[0155] Table 1: UCI reporting format 1 for sample A

[0156] Table 2 below shows an alternative UCI reporting format for sample A. Sample A includes label 1 and input 1. The M RSRPs for input 1 are presented in a fixed order according to Set B and differentially reported. In Table 2, RSRP#1, corresponding to the optimal CRI or SSBRI in Set B, can be placed at the front. The RSRPs corresponding to other CRIs or SSBRIs in Set B (RSRP#2 to RSRP#M) are differentially reported with reference to RSRP#1.

[0157] Table 2: UCI reporting format 2 for sample A

[0158] Table 3 below shows a UCI reporting format for Sample B. Sample B includes Label 1 and Input 2, which contains M RSRPs and the corresponding CRI or SSBRI from Set B or Set C. The M RSRPs in Input 2 are differentially reported. Because this format already contains the CRI or SSBRI, there is no need to additionally report the optimal CRI or SSBRI from Set B.

[0159] Table 3: UCI reporting format for Sample B

[0160] Table 4 below shows a UCI reporting format for sample C. Compared to sample A, sample C adds label 2. That is, sample C includes label 1, label 2, and input 1. The M RSRPs for input 1 are given in the fixed order of Set B, without differential reporting. The K RSRPs for label 2 are given in the fixed order of Set A, with differential reporting.

[0161] Table 4: UCI reporting format 1 for sample C

[0162] Table 5 below shows another UCI reporting format for sample C. In this format, the M RSRPs of input 1 are given in a fixed order according to Set B for differential reporting; the K RSRPs of label 2 are given in a fixed order according to Set A for differential reporting.

[0163] Table 5: UCI reporting format 2 for sample C

[0164] Table 6 below shows a UCI reporting format for Sample D, which includes Label 1, Label 2, and Input 2. In this format, the M RSRPs of Input 2 are differentially reported, and the K RSRPs of Label 2 are differentially reported in a fixed order according to Set A.

[0165] Table 6: UCI reporting format for Sample D

[0166] It should be noted that the above Tables 1 to 6 are only exemplary descriptions. The technical solutions of the embodiments of the present application are not limited to the forms of the above Tables 1 to 6. According to the ideas of the technical solutions of the embodiments of the present application, there may also be other types of UCI reporting formats.

[0167] 3) Datasets are carried in MAC layer containers

[0168] MAC layer containers have a fixed format, with 8 bits as the basic unit (i.e., byte), and variable byte length. MAC layer containers are characterized by low latency, medium carrying capacity, and error control mechanisms (supporting retransmission after MAC CE errors).

[0169] As an example, Figure 8 shows the format of a MAC layer container. It should be noted that the downlink reference signal resource index (CRI or SSBR) can also be replaced by a downlink reference signal index, a beam (pair) index, or a spatial filter index. In Figure 8, K samples are included. In some embodiments, the maximum value of K can be designed to be 2 to the power of 8, that is, 256. The optimal CRI or SSBRI in Set A occupies 6 bits, and 6 bits can support a maximum of 64 CSI-RS resources or SSB resources in Set A.

[0170] It should be noted that in the example of Figure 8, the sample consists of label 1 and input 1. Other combinations of labels and inputs can also be carried by MAC layer containers. That is, the technical solution of the embodiment of the present application is not limited to the form shown in Figure 8. According to the idea of ​​the technical solution of the embodiment of the present application, there can also be other types of forms.

[0171] Configuration of downlink reference signal sets

[0172] The network device may configure the first downlink reference signal set or the second downlink reference signal set, and / or the third downlink reference signal set for the terminal. The network device sends first configuration information and / or second configuration information to the terminal; the terminal receives the first configuration information and / or second configuration information sent by the network device; the first configuration information is used to configure at least one first downlink reference signal set or at least one second downlink reference signal set; and the second configuration information is used to configure at least one third downlink reference signal set.

[0173] As an implementation manner, the first configuration information and / or the second configuration information is carried in RRC signaling.

[0174] In some implementations, the first configuration information is used to configure a first downlink reference signal set or a second downlink reference signal set, and the first downlink reference signal set or the second downlink reference signal set is used to obtain an input instance. In other implementations, the first configuration information is used to configure multiple first downlink reference signal sets or multiple second downlink reference signal sets, the network device sends first indication information to the terminal, and the terminal receives the first indication information sent by the network device, the first indication information is used to indicate activation of one first downlink reference signal set among the multiple first downlink reference signal sets or activation of one second downlink reference signal set among the multiple second downlink reference signal sets, and the first downlink reference signal set or the second downlink reference signal set is used to obtain the input instance.

[0175] As an implementation method, the first indication information is carried in MAC CE or DCI.

[0176] Exemplarily, the network device uses RRC signaling to configure J (J ≥ 1) Set B or Set C for the terminal, where Set Bj or Set Cj is the Set B or Set C identified as j in the J Set B or Set C, and 1 ≤ j ≤ J. If the network device is configured with only one Set B or Set C (i.e., J = 1), then there is no need to activate the Set B or Set C. The Set B or Set C is activated by default and can be used to obtain an input instance. If the network device is configured with multiple Set B or Set C (i.e., J > 1), then the network device needs to use MAC CE or DCI to activate one of the multiple configurations of Set B or Set C based on the actual deployment and antenna configuration. The activated Set B or Set C can be used to obtain an input instance.

[0177] In some implementations, the second configuration information is used to configure a single third downlink reference signal set, and the single third downlink reference signal set is used to obtain the label instance. In other implementations, the second configuration information is used to configure multiple third downlink reference signal sets, and the method further includes: the network device sends second indication information to the terminal, and the terminal receives the second indication information sent by the network device, the second indication information is used to indicate activation of one third downlink reference signal set among the multiple third downlink reference signal sets, and the single third downlink reference signal set is used to obtain the label instance.

[0178] As an implementation method, the second indication information is carried in MAC CE or DCI.

[0179] Exemplarily, the network device uses RRC signaling to configure S (S ≥ 1) Set As for the terminal, where Set As is the Set A identified as s in the S Set As, where 1 ≤ s ≤ S. If the network device is configured with only one Set A (i.e., S = 1), then the Set A is not activated, and the Set A is activated by default and can be used to obtain a label instance. If the network device is configured with multiple Set As (i.e., S > 1), then the network device needs to use MAC CE or DCI to activate one of the multiple configurations of Set A based on the actual deployment and antenna configuration. The activated Set A can be used to obtain a label instance.

[0180] In some implementations, the first configuration information and / or the second configuration information are related to capability information of the terminal. Based on the capability information of the terminal, the network device configures at least one first downlink reference signal set or at least one second downlink reference signal set for the terminal, and / or configures at least one third downlink reference signal set for the terminal.

[0181] Illustratively, the network device configures J (J≥1) Set Bs or Set Cs for the terminal according to the capability information reported by the terminal. Optionally, when J>1, one Set B or Set C among the J Set Bs or Set Cs is further activated.

[0182] Exemplarily, the network device configures S (S≥1) Set A for the terminal according to the capability information reported by the terminal. Optionally, when S>1, one Set A among the S Set A is further activated.

[0183] Terminal capability reporting

[0184] Different terminals have different capabilities for data collection, so the terminal needs to report its ability to collect data for network devices.

[0185] In some implementations, the terminal sends the terminal's capability information to the network device, and the network device receives the terminal's capability information sent by the terminal. The terminal's capability information includes at least one of the following:

[0186] First information, the first information is used to indicate whether the terminal supports collection of a data set for prediction of a target downlink reference signal and / or a target link quality;

[0187] Second information, where the second information is used to indicate a maximum number of first downlink reference signal sets or second downlink reference signal sets supported by the terminal on all downlink carrier components (CCs) or bandwidth parts (BWPs);

[0188] The third information is used to indicate the maximum number of first downlink reference signal sets or second downlink reference signal sets supported by the terminal on each downlink CC or BWP;

[0189] Fourth information, the fourth information is used to indicate a maximum number of downlink reference signal resources supported by the terminal in all first downlink reference signal sets or second downlink reference signal sets;

[0190] Fifth information, the fifth information is used to indicate a maximum number of downlink reference signal resources supported by the terminal in each first downlink reference signal set or second downlink reference signal set;

[0191] Sixth information, the sixth information is used to indicate the maximum number of third downlink reference signal sets supported by the terminal on all downlink CCs or BWPs;

[0192] The seventh information is used to indicate the maximum number of third downlink reference signal sets supported by the terminal on each downlink CC or BWP;

[0193] The eighth information is used to indicate the maximum number of downlink reference signal resources supported by the terminal in all third downlink reference signal sets;

[0194] Ninth information: The ninth information is used to indicate the maximum number of downlink reference signal resources supported by the terminal in each third downlink reference signal set.

[0195] The above-mentioned first information indication capability can be recorded as capability 1. Similarly, the second information indication capability can be recorded as capability 2, the third information indication capability can be recorded as capability 3, the fourth information indication capability can be recorded as capability 4, the fifth information indication capability can be recorded as capability 5, the sixth information indication capability can be recorded as capability 6, the seventh information indication capability can be recorded as capability 7, the eighth information indication capability can be recorded as capability 8, and the ninth information indication capability can be recorded as capability 9.

[0196] Exemplarily, capability 2 may be understood as the maximum number of measurement sets of Set B or Set C supported by the terminal on all downlink CCs or BWPs, for obtaining input instances.

[0197] Exemplarily, capability 3 may be understood as the maximum number of measurement sets of Set B or Set C supported by the terminal on a downlink CC or BWP, for obtaining input instances.

[0198] Exemplarily, capability 4 may be understood as the maximum number of SSB resources and / or CSI-RS resources that the terminal supports for measurement in all measurement sets Set B or Set C.

[0199] Exemplarily, capability 5 can be understood as the maximum number of SSB resources and / or CSI-RS resources that the terminal supports measurement in a measurement set Set B or Set C.

[0200] For example, capability 6 may be understood as the maximum number of measurement sets of Set A supported by the terminal on all downlink CCs or BWPs, for obtaining label instances.

[0201] Exemplarily, capability 7 may be understood as the maximum number of measurement sets of Set A supported by the terminal on a downlink CC or BWP, for obtaining label instances.

[0202] Exemplarily, capability 8 may be understood as the maximum number of SSB resources and / or CSI-RS resources that the terminal supports for measurement in all measurement sets Set A.

[0203] Exemplarily, capability 9 can be understood as the maximum number of SSB resources and / or CSI-RS resources that the terminal supports measurement in a measurement set Set A.

[0204] In some embodiments, the first information is used to indicate whether the terminal supports the collection of a data set. As one implementation, the first information is used to indicate whether the terminal supports the collection of a data set for prediction of a target downlink reference signal and / or target link quality. As another implementation, the first information is used to indicate whether the terminal supports the collection of a data set for prediction of downlink beams (pairs) in the spatial domain (BM-Case1) and / or the time domain (BM-Case2).

[0205] In some embodiments, the capability information of the terminal includes the first information. Optionally, when the first information indicates that the terminal supports the collection of data sets, the capability information of the terminal may include at least one of the second to ninth information.

[0206] In some implementations, the capability information of the terminal includes at least one of the first to ninth information described above.

[0207] Sample composition

[0208] 1) Sample composition method 1

[0209] In some embodiments, an input instance and a label instance in the dataset constitute a sample for training the first model.

[0210] Exemplarily, for the prediction of downlink beams (pairs) in the spatial domain (BM-Case1), the sample of the first model includes an input instance and a label instance.

[0211] 2) Sample composition method 2

[0212] In some embodiments, N input instances and P label instances in a data set constitute a sample for training the first model; the N input instances and P label instances are obtained by the terminal in a time-sharing manner, and N and P are positive integers.

[0213] Exemplarily, for the prediction of downlink beam (pair) in the time domain (BM-Case2), the samples of the first model include N input instances and P label instances, where N≥1 and P≥1.

[0214] In some embodiments, the N input instances have the fifth format or the sixth format in the sample;

[0215] The fifth format includes: N input instances; the ordering of the N input instances in the sample is related to the temporal order of the N input instances;

[0216] The sixth format includes: N input instances and N instance serial numbers; the N input instances and the N instance serial numbers have a corresponding relationship, and the N instance serial numbers are related to the time domain sequence of the N input instances.

[0217] In some embodiments, the P label instances have the seventh format or the eighth format in the sample;

[0218] The seventh format includes: P label instances; the order of the P label instances in the sample is related to the temporal order of the P label instances;

[0219] The eighth format includes: P label instances and P instance serial numbers; the P label instances and the P instance serial numbers have a corresponding relationship, and the P instance serial numbers are related to the time domain order of the P label instances.

[0220] For downlink beam (pair) prediction in the time domain (BM-Case 2), the temporal relationship between samples needs to be reflected during the sample collection process. In some embodiments, for N input instances and P label instances in the sample, the temporal relationship during the collection process can be implemented in the following two ways:

[0221] Solution 1) The N input instances are obtained by the terminal performing N measurements on the downlink reference signals in the first downlink reference signal set or the second downlink reference signal set within the first time period, and the P label instances are obtained by the terminal performing P measurements on the downlink reference signals in the third downlink reference signal set within the second time period.

[0222] For this solution, the terminal performs N measurements on the downlink reference signals in the first downlink reference signal set or the second downlink reference signal set within a first time period to obtain N input instances corresponding to the first time period; the terminal performs P measurements on the downlink reference signals in the third downlink reference signal set within a second time period to obtain P label instances corresponding to the second time period.

[0223] In some implementations, the terminal sends a data set to the network device according to a first reporting unit, where the first reporting unit includes N input instances corresponding to the first time period and P label instances corresponding to the second time period.

[0224] As an example, the following illustrates the dataset collection process using Scenario 1 shown in Figure 9. During time T1, the terminal performs N measurements on Set B or Set C, obtaining N input instances as model input. During time T2, the terminal performs P measurements on Set A, obtaining P labeled instances as model labels.

[0225] During the measurement of time T1, each time the terminal measures Set B or Set C configured on the network device, an input instance is obtained. The input instances obtained from N measurements can be identified as follows:

[0226] {Input instance 1, input instance 2, ... , input instance N}.

[0227] For each input instance, the collection of input instances and the content of input instances can refer to the aforementioned related solutions, and the aforementioned solutions related to "collection of input instances" and "content of input instances" can be combined here for implementation.

[0228] During the measurement process of time T2, each time the terminal measures Set A configured on the network device, a label instance is obtained. The label instances obtained through P measurements can be identified as follows:

[0229] {label instance 1, label instance 2, ... , label instance P}.

[0230] For each tag instance, the collection of tag instances and the content of tag instances can refer to the aforementioned related solutions, and the aforementioned solutions related to "collection of tag instances" and "content of tag instances" can be combined here for implementation.

[0231] Among the N input instances, the temporal order of the input instances (which can also be understood as the measurement time) can be implicitly expressed by the order in which the input instances appear in the sample, or it can be explicitly expressed by adding an instance number to each input instance. Similarly, among the P label instances, the temporal order of the label instances can be implicitly expressed by the order in which the label instances appear in the sample, or it can be explicitly expressed by adding an instance number to each label instance. As an implementation method, in a sample, the input instance is located at the front position and the label instance is located at the back position. As another implementation method, in a sample, the input instance is located at the back position and the label instance is located at the front position.

[0232] It should be noted that there is no need to mark the order of samples. This is because during the training process, the model needs to perform parallel calculations on the samples and does not consider the order of the samples.

[0233] For example, taking the reporting of a data set via UCI as an example, several sample formats are given below. It should be noted that the downlink reference signal resource index (CRI or SSBR) can also be replaced by a downlink reference signal index or a beam (pair) index or a spatial filter index.

[0234] Table 7 below shows a UCI reporting format for scenario 1. In this format, the temporal order of input instances is implicitly expressed by the order in which they appear in the sample, and the temporal order of label instances is implicitly expressed by the order in which they appear in the sample. This example uses the sample consisting of input 1 and label 1 as an example.

[0235] Table 7: UCI reporting format 1 for scenario 1

[0236] Table 8 below shows another UCI reporting format for scenario 1. In this format, the temporal order of input instances is explicitly expressed using instance numbers, and the temporal order of label instances is explicitly expressed using instance numbers. This example uses a sample consisting of input 1 and label 1 as an example.

[0237] Table 8: UCI reporting format 2 for scenario 1

[0238] Solution 2) The N input instances are obtained by the terminal performing N measurements on the downlink reference signals in the first downlink reference signal set or the second downlink reference signal set within N periods, and the P label instances are obtained by the terminal performing P measurements on the downlink reference signals in the third downlink reference signal set within the last period of the N periods.

[0239] For this solution, the terminal performs one measurement on the downlink reference signal in the first downlink reference signal set or the second downlink reference signal set and P measurements on the downlink reference signal in the third downlink reference signal set in each of N periods, and obtains one input instance and P label instances corresponding to each period.

[0240] In some embodiments, the terminal sends a data set to the network device according to a second reporting unit; the second reporting unit includes one input instance and P label instances corresponding to each cycle; or, the second reporting unit includes N input instances corresponding to N cycles and P label instances corresponding to the last cycle of the N cycles; or, the second reporting unit includes all input instances and all label instances corresponding to the N cycles.

[0241] As an example, the process of collecting a data set is described below with reference to Scenario 2 shown in FIG10 .

[0242] In each T1 measurement cycle, the terminal performs one measurement on Set B or Set C and P measurements on Set A, obtaining one input instance and P label instances. In N T1 measurement cycles, a total of N input instances and N × P label instances are obtained. These N input instances and P label instances constitute a sample, and the P label instances that make up the sample can be the P label instances obtained in the last T1 measurement cycle.

[0243] For each input instance, the collection of input instances and the content of input instances can refer to the aforementioned related solutions, and the aforementioned solutions related to "collection of input instances" and "content of input instances" can be combined here for implementation.

[0244] For each tag instance, the collection of tag instances and the content of tag instances can refer to the aforementioned related solutions, and the aforementioned solutions related to "collection of tag instances" and "content of tag instances" can be combined here for implementation.

[0245] The terminal can report the measured input instances and label instances in the following three ways:

[0246] Method 1: The terminal's reporting unit includes one input instance and P tag instances. In this case, the terminal can report once after obtaining one input instance and P tag instances during each T1 measurement period.

[0247] The temporal order (also understood as measurement time) of the P tag instances in a reporting unit can be implicitly expressed by the order in which the tag instances appear in the reporting unit, or can be explicitly expressed by adding an instance number to each tag instance.

[0248] For example, for the implicit mode, the order of each instance of the reporting unit is as follows:

[0249] Input instance n

[0250] Label instance n1

[0251] Tag instance n2

[0252] …

[0253] Tag instance nP.

[0254] For example, in the explicit mode, the order of each instance of the reporting unit is as follows:

[0255] Enter instance ID#n

[0256] Tag instance ID#n3

[0257] Tag instance ID#n1

[0258] …

[0259] Tag instance ID#nP.

[0260] Method 2: The terminal's reporting unit includes N input instances and P label instances after the Nth input instance. In this case, the terminal can report once after measuring N input instances and N×P label instances within N T1 measurement periods.

[0261] The temporal order (also known as measurement time) of the N input instances in a reporting unit can be implicitly expressed by the order in which the input instances appear in the reporting unit, or explicitly expressed by adding an instance number to each input instance. Similarly, the temporal order (also known as measurement time) of the P label instances can be implicitly expressed by the order in which the label instances appear in the reporting unit, or explicitly expressed by adding an instance number to each label instance.

[0262] For example, for the implicit mode, the order of each instance of the reporting unit is as follows:

[0263] Input Example 1

[0264] Input Example 2

[0265] …

[0266] Input instance N

[0267] Label instance N1

[0268] Tag instance N2

[0269] …

[0270] Label instance NP.

[0271] For example, in the explicit mode, the order of each instance of the reporting unit is as follows:

[0272] Enter instance ID#2

[0273] Enter instance ID#N

[0274] …

[0275] Enter instance ID#1

[0276] Tag instance ID#N3

[0277] Tag instance ID#N1

[0278] …

[0279] Tag instance ID#NP.

[0280] Method 3: The terminal's reporting unit includes N input instances and N×P label instances. In this case, the terminal can make a report once it obtains N input instances and N×P label instances through measurement within N T1 measurement periods.

[0281] The temporal order (also known as measurement time) of the N input instances in a reporting unit can be implicitly expressed by the order in which the input instances appear in the reporting unit, or explicitly expressed by adding an instance sequence number to each input instance. Similarly, the temporal order (also known as measurement time) of the N×P label instances can be implicitly expressed by the order in which the label instances appear in the reporting unit, or explicitly expressed by adding an instance sequence number to each label instance.

[0282] For example, for the implicit mode, the order of each instance of the reporting unit is as follows:

[0283] Input Example 1

[0284] Tag Example 11

[0285] Tag Example 12

[0286] …

[0287] Tag instance 1P

[0288] Input Example 2

[0289] Tag Example 21

[0290] Tag Example 22

[0291] …

[0292] Tag Instance 2P

[0293] …

[0294] Input instance N

[0295] Label instance N1

[0296] Tag instance N2

[0297] …

[0298] Label instance NP.

[0299] The technical solution of the embodiment of the present application proposes a method for collecting a data set, clarifies the content of the data set, and the reporting format, in order to train a model with a downlink beam (pair) prediction function. For model training on the network device side, the network device configures and / or activates the downlink reference signal set required for measurement based on the terminal capability information reported by the terminal. The terminal measures the downlink reference signal set, collects a data set, and then reports the data set to the network device. The data set contains a combination of input instances and label instances, providing training material for model training.

[0300] FIG11 is a flow chart of another method for determining a data set provided in an embodiment of the present application. As shown in FIG11 , the method includes the following steps:

[0301] Step 1101: The terminal obtains a data set, where the data set includes at least one input instance and at least one label instance; the input instance is obtained based on measuring a downlink reference signal in a first downlink reference signal set or a second downlink reference signal set, and the label instance is obtained based on measuring a downlink reference signal in a third downlink reference signal set; wherein the data set is used to adjust a second model on the terminal side, and the second model is used to predict a target downlink reference signal and / or a target link quality.

[0302] For the second model on the terminal side, on the one hand, it is difficult for a single terminal to collect a large amount of data to complete the initial training of the model. On the other hand, if the model on the terminal side is trained from scratch, the network device needs to transmit a large amount of data to the terminal, which is a considerable overhead for the downlink data channel. To this end, as an implementation method, the network device sends the trained model to the terminal, and the terminal can obtain the trained model. The network device can obtain the data set and complete the model training through the scheme related to Figure 7. As another implementation method, the terminal has an initial model (also called a meta-model) for beam (pair) prediction. The model can be understood as a factory setting or preset model. In order to improve the model's adaptability to the terminal's environment and the network device's beam configuration, that is, generalization performance, the terminal can use a small amount of data to complete fine-tuning of the model. It can be understood that the data set obtained by the terminal contains a small number of input instances and a small number of label instances.

[0303] In this embodiment of the present application, the terminal can obtain the data set in the following ways:

[0304] Method 1: The terminal may obtain a data set in the following manner: the terminal receives a data set sent by a network device, where the data set is all or part of a data set reported by at least one terminal to the network device, and the at least one terminal may include or exclude the terminal.

[0305] In some implementations, the data set is carried in at least one of the following signaling: RRC signaling, MAC layer container.

[0306] Exemplarily, the network device collects data sets from other terminals, and sends all or part of the collected data sets to the terminal via RRC signaling or MAC layer container, so that the terminal can obtain the data sets without performing measurements.

[0307] Method 2: The terminal may obtain the data set in the following manner: the terminal measures the downlink reference signal in the first downlink reference signal set or the second downlink reference signal set to obtain an input instance; and the terminal measures the downlink reference signal in the third downlink reference signal set to obtain a label instance.

[0308] Here, the terminal may obtain the data set by measuring the downlink reference signal set.

[0309] In some implementations, the network device may configure a downlink reference signal set for the terminal. In some implementations, the network device may configure a downlink reference signal set for the terminal based on terminal capability information reported by the terminal.

[0310] It should be noted that the above-mentioned understanding of the input instance and the label instance can refer to the relevant description of Figure 7.

[0311] It should be noted that the above-mentioned understanding about the network device being able to configure a downlink reference signal set for the terminal can refer to the relevant description in FIG. 7 .

[0312] It should be noted that the above-mentioned understanding of the terminal capability information reported by the terminal can refer to the relevant description of Figure 7.

[0313] According to the technical solution of the embodiment of the present application, for model training on the terminal side, the network device can send a small amount of data to the terminal, or the terminal can obtain data by measuring on its own, thereby achieving fine-tuning of the model.

[0314] The preferred embodiments of the present application are described in detail above in conjunction with the accompanying drawings. However, the present application is not limited to the specific details in the above embodiments. Within the technical concept of the present application, the technical solution of the present application can be subjected to a variety of simple modifications, and these simple modifications all fall within the scope of protection of the present application. For example, the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present application will no longer describe the various possible combinations separately. For another example, the various different embodiments of the present application can also be arbitrarily combined, as long as they do not violate the idea of ​​the present application, they should also be regarded as the contents disclosed in the present application. For another example, under the premise of no conflict, the various embodiments and / or the technical features in each embodiment described in the present application can be arbitrarily combined with the prior art, and the technical solution obtained after the combination should also fall within the scope of protection of the present application.

[0315] It should also be understood that in the various method embodiments of the present application, the sequence numbers of the above-mentioned processes do not imply a precedence in their execution order. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. Furthermore, in the embodiments of the present application, the terms "downlink," "uplink," and "sidelink" are used to indicate the direction of signal or data transmission. "Downlink" indicates a signal or data transmission direction of a first direction, from a site to a user equipment in a cell; "uplink" indicates a signal or data transmission direction of a second direction, from a user equipment in a cell to a site; and "sidelink" indicates a signal or data transmission direction of a third direction, from user equipment 1 to user equipment 2. For example, "downlink signal" indicates that the signal is transmitted in the first direction. Furthermore, in the embodiments of the present application, the term "and / or" merely describes an association relationship between associated objects, indicating that three possible relationships exist. Specifically, "A and / or B" can represent three situations: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " as used herein generally indicates that the associated objects are in an "or" relationship.

[0316] FIG12 is a schematic diagram of the first structure of a device for determining a data set provided in an embodiment of the present application, which is applied to a terminal. As shown in FIG12 , the device includes:

[0317] A sending unit 1201 is configured to send a data set to a network device, where the data set includes at least one input instance and at least one label instance; the input instance is obtained based on a terminal measuring a downlink reference signal in a first downlink reference signal set or a second downlink reference signal set, and the label instance is obtained based on a terminal measuring a downlink reference signal in a third downlink reference signal set; wherein the data set is used to train a first model on the network device side, and the first model is used to predict a target downlink reference signal and / or a target link quality.

[0318] In some embodiments, the apparatus further includes: a receiving unit 1202 for receiving first configuration information and / or second configuration information sent by a network device; the first configuration information is used to configure at least one first downlink reference signal set or at least one second downlink reference signal set; and the second configuration information is used to configure at least one third downlink reference signal set.

[0319] In some embodiments, the first configuration information is used to configure a first downlink reference signal set or a second downlink reference signal set, and the first downlink reference signal set or the second downlink reference signal set is used to obtain an input instance; or, the first configuration information is used to configure multiple first downlink reference signal sets or multiple second downlink reference signal sets, the receiving unit 1202 is used to receive first indication information sent by a network device, the first indication information is used to indicate activation of a first downlink reference signal set in multiple first downlink reference signal sets or activation of a second downlink reference signal set in multiple second downlink reference signal sets, and the first downlink reference signal set or the second downlink reference signal set is used to obtain the input instance.

[0320] In some embodiments, the second configuration information is used to configure a third downlink reference signal set, and the third downlink reference signal set is used to obtain a label instance; or, the second configuration information is used to configure multiple third downlink reference signal sets, and the receiving unit 1202 is used to receive second indication information sent by the network device, the second indication information is used to indicate activation of a third downlink reference signal set among multiple third downlink reference signal sets, and the third downlink reference signal set is used to obtain a label instance.

[0321] In some implementations, the first configuration information and / or the second configuration information is related to capability information of the terminal.

[0322] In some implementations, the sending unit 1201 is configured to send capability information of the terminal to the network device, where the capability information of the terminal includes at least one of the following:

[0323] First information, the first information is used to indicate whether the terminal supports collection of a data set for prediction of a target downlink reference signal and / or a target link quality;

[0324] Second information, where the second information is used to indicate a maximum number of first downlink reference signal sets or second downlink reference signal sets supported by the terminal on all downlink CCs or BWPs;

[0325] The third information is used to indicate the maximum number of first downlink reference signal sets or second downlink reference signal sets supported by the terminal on each downlink CC or BWP;

[0326] Fourth information, the fourth information is used to indicate a maximum number of downlink reference signal resources supported by the terminal in all first downlink reference signal sets or second downlink reference signal sets;

[0327] Fifth information, the fifth information is used to indicate a maximum number of downlink reference signal resources supported by the terminal in each first downlink reference signal set or second downlink reference signal set;

[0328] Sixth information, the sixth information is used to indicate the maximum number of third downlink reference signal sets supported by the terminal on all downlink CCs or BWPs;

[0329] The seventh information is used to indicate the maximum number of third downlink reference signal sets supported by the terminal on each downlink CC or BWP;

[0330] The eighth information is used to indicate the maximum number of downlink reference signal resources supported by the terminal in all third downlink reference signal sets;

[0331] Ninth information: The ninth information is used to indicate the maximum number of downlink reference signal resources supported by the terminal in each third downlink reference signal set.

[0332] In some embodiments, the input instance includes M link quality measurement results; or, the input instance includes M link quality measurement results and M downlink reference signal resource indexes, and the M link quality measurement results and the M downlink reference signal resource indexes have a corresponding relationship; M is a positive integer.

[0333] In some embodiments, the M link quality measurement results are obtained based on the terminal measuring the first downlink reference signal set; or, the M link quality measurement results are the best M link quality measurement results among multiple link quality measurement results, and the multiple link quality measurement results are obtained based on the terminal measuring the second downlink reference signal set, and M is a positive integer.

[0334] In some embodiments, the M link quality measurement results have a first format or a second format in an input instance: the first format includes: M link quality measurement results; the second format includes: a first link quality measurement result and M-1 differential values, where the first link quality measurement result is one of the M link quality measurement results, and the M-1 differential values ​​are differential values ​​of the M-1 link quality measurement results other than the first link quality measurement result in the M link quality measurement results relative to the first link quality measurement result.

[0335] In some embodiments, the label instance includes a first label; or the label instance includes a first label and a second label; the first label includes a downlink reference signal resource index that satisfies a first condition, and the downlink reference signal resource index that satisfies the first condition is determined based on the terminal measuring the downlink reference signal in the third downlink reference signal set; the second label includes K link quality measurement results and / or K downlink reference signal resource indexes that meet the second condition, the K link quality measurement results and the K downlink reference signal resource indexes have a corresponding relationship, and the K link quality measurement results that meet the second condition are determined based on the terminal measuring the downlink reference signal in the third downlink reference signal set, and K is a positive integer.

[0336] In some implementations, the downlink reference signal resource index that meets the first condition is a downlink reference signal resource index corresponding to an optimal link quality measurement result obtained by the terminal measuring downlink reference signals in the third downlink reference signal set.

[0337] In some implementations, the K link quality measurement results that meet the second condition refer to: the best first K link quality measurement results obtained by the terminal measuring the downlink reference signals in the third downlink reference signal set.

[0338] In some embodiments, the K link quality measurement results have a third format or a fourth format in the tag instance; the third format includes: K link quality measurement results; the fourth format includes: a second link quality measurement result and K-1 differential values, where the second link quality measurement result is one of the K link quality measurement results, and the K-1 differential values ​​are differential values ​​of the K-1 link quality measurement results other than the second link quality measurement result in the K link quality measurement results relative to the second link quality measurement result.

[0339] In some implementations, the data set is carried in at least one of the following signaling: RRC signaling, UCI, and MAC layer container.

[0340] In some embodiments, an input instance and a label instance in the dataset constitute a sample for training the first model.

[0341] In some embodiments, N input instances and P label instances in a data set constitute a sample for training the first model; the N input instances and P label instances are obtained by the terminal in a time-sharing manner, and N and P are positive integers.

[0342] In some embodiments, the N input instances have a fifth format or a sixth format in the sample; the fifth format includes: N input instances; the order of the N input instances in the sample is related to the time domain order of the N input instances; the sixth format includes: N input instances and N instance serial numbers; the N input instances have a corresponding relationship with the N instance serial numbers, and the N instance serial numbers are related to the time domain order of the N input instances.

[0343] In some embodiments, the P label instances have a seventh format or an eighth format in the sample; the seventh format includes: P label instances; the order of the P label instances in the sample is related to the time domain order of the P label instances; the eighth format includes: P label instances and P instance serial numbers; the P label instances have a corresponding relationship with the P instance serial numbers, and the P instance serial numbers are related to the time domain order of the P label instances.

[0344] In some embodiments, the N input instances are obtained by the terminal performing N measurements on the downlink reference signals in the first downlink reference signal set or the second downlink reference signal set within a first time period, and the P label instances are obtained by the terminal performing P measurements on the downlink reference signals in the third downlink reference signal set within a second time period.

[0345] In some embodiments, the device further includes: a measurement unit, configured to perform N measurements on the downlink reference signal in the first downlink reference signal set or the second downlink reference signal set within a first time period to obtain N input instances corresponding to the first time period; and perform P measurements on the downlink reference signal in the third downlink reference signal set within a second time period to obtain P label instances corresponding to the second time period.

[0346] In some implementations, the sending unit 1201 is configured to send a data set to a network device according to a first reporting unit, where the first reporting unit includes N input instances corresponding to a first time period and P label instances corresponding to a second time period.

[0347] In some embodiments, the N input instances are obtained by the terminal performing N measurements on the downlink reference signal in the first downlink reference signal set or the second downlink reference signal set within N periods, and the P label instances are obtained by the terminal performing P measurements on the downlink reference signal in the third downlink reference signal set within the last period of the N periods.

[0348] In some embodiments, the measurement unit is configured to perform one measurement on the downlink reference signal in the first downlink reference signal set or the second downlink reference signal set and P measurements on the downlink reference signal in the third downlink reference signal set in each of N periods, to obtain one input instance and P label instances corresponding to each period.

[0349] In some embodiments, the sending unit 1201 is used to send a data set to the network device according to a second reporting unit; the second reporting unit includes one input instance and P label instances corresponding to each cycle; or, the second reporting unit includes N input instances corresponding to N cycles and P label instances corresponding to the last cycle of the N cycles; or, the second reporting unit includes all input instances and all label instances corresponding to the N cycles.

[0350] In some embodiments, the downlink reference signal includes: a channel state information-reference signal CSI-RS and / or a synchronization signal block SSB.

[0351] In some implementations, the link quality includes at least one of the following: RSRP, SINR, and RSRQ.

[0352] Those skilled in the art should understand that the description of the apparatus for determining a data set in the embodiment of the present application can be understood with reference to the description of the method for determining a data set in the embodiment of the present application.

[0353] FIG13 is a second schematic diagram of the structure of the apparatus for determining a data set provided in an embodiment of the present application, which is applied to a terminal. As shown in FIG13 , the apparatus includes:

[0354] An acquisition unit 1301 is configured to acquire a data set, the data set comprising at least one input instance and at least one label instance; the input instance is obtained based on measuring a downlink reference signal in a first downlink reference signal set or a second downlink reference signal set, and the label instance is obtained based on measuring a downlink reference signal in a third downlink reference signal set; wherein the data set is used to adjust a second model on the terminal side, and the second model is used to predict a target downlink reference signal and / or a target link quality.

[0355] In some implementations, the acquiring unit 1301 is configured to receive a data set sent by a network device, where the data set is all or part of a data set reported by at least one terminal to the network device, where the at least one terminal includes a terminal or does not include a terminal.

[0356] In some implementations, the data set is carried in at least one of the following signaling: RRC signaling, MAC layer container.

[0357] In some implementations, the acquiring unit 1301 is configured to measure a downlink reference signal in the first downlink reference signal set or the second downlink reference signal set to obtain an input instance; and measure a downlink reference signal in the third downlink reference signal set to obtain a label instance.

[0358] In some embodiments, the apparatus further includes: a receiving unit 1302 for receiving first configuration information and / or second configuration information sent by a network device; the first configuration information is used to configure at least one first downlink reference signal set or at least one second downlink reference signal set; and the second configuration information is used to configure at least one third downlink reference signal set.

[0359] In some embodiments, the first configuration information is used to configure a first downlink reference signal set or a second downlink reference signal set, and the first downlink reference signal set or the second downlink reference signal set is used to obtain an input instance; or, the first configuration information is used to configure multiple first downlink reference signal sets or multiple second downlink reference signal sets, the receiving unit 1302 is used to receive first indication information sent by a network device, the first indication information is used to indicate activation of a first downlink reference signal set in multiple first downlink reference signal sets or activation of a second downlink reference signal set in multiple second downlink reference signal sets, and the first downlink reference signal set or the second downlink reference signal set is used to obtain the input instance.

[0360] In some embodiments, the second configuration information is used to configure a third downlink reference signal set, and the third downlink reference signal set is used to obtain a label instance; or, the second configuration information is used to configure multiple third downlink reference signal sets, and the receiving unit 1302 is used to receive second indication information sent by the network device, the second indication information is used to indicate activation of a third downlink reference signal set among multiple third downlink reference signal sets, and the third downlink reference signal set is used to obtain a label instance.

[0361] In some implementations, the first configuration information and / or the second configuration information is related to capability information of the terminal.

[0362] In some implementations, the apparatus further includes: a sending unit 1303, configured to send capability information of the terminal to the network device, where the capability information of the terminal includes at least one of the following:

[0363] First information, the first information is used to indicate whether the terminal supports collection of a data set for prediction of a target downlink reference signal and / or a target link quality;

[0364] Second information, where the second information is used to indicate a maximum number of first downlink reference signal sets or second downlink reference signal sets supported by the terminal on all downlink CCs or BWPs;

[0365] The third information is used to indicate the maximum number of first downlink reference signal sets or second downlink reference signal sets supported by the terminal on each downlink CC or BWP;

[0366] Fourth information, the fourth information is used to indicate a maximum number of downlink reference signal resources supported by the terminal in all first downlink reference signal sets or second downlink reference signal sets;

[0367] Fifth information, the fifth information is used to indicate a maximum number of downlink reference signal resources supported by the terminal in each first downlink reference signal set or second downlink reference signal set;

[0368] Sixth information, the sixth information is used to indicate the maximum number of third downlink reference signal sets supported by the terminal on all downlink CCs or BWPs;

[0369] The seventh information is used to indicate the maximum number of third downlink reference signal sets supported by the terminal on each downlink CC or BWP;

[0370] The eighth information is used to indicate the maximum number of downlink reference signal resources supported by the terminal in all third downlink reference signal sets;

[0371] Ninth information: The ninth information is used to indicate the maximum number of downlink reference signal resources supported by the terminal in each third downlink reference signal set.

[0372] In some implementations, the downlink reference signal includes: CSI-RS and / or SSB.

[0373] In some implementations, the link quality includes at least one of the following: RSRP, SINR, RSSI, and RSRQ.

[0374] Those skilled in the art should understand that the relevant description of determining the above-mentioned data set in the embodiment of the present application can be understood with reference to the relevant description of the method for determining the data set in the embodiment of the present application.

[0375] FIG14 is a third structural diagram of the apparatus for determining a data set according to an embodiment of the present application, which is applied to a network device. As shown in FIG14 , the apparatus includes:

[0376] Receiving unit 1401 is used to receive a data set sent by a terminal, where the data set includes at least one input instance and at least one label instance; the input instance is obtained based on the terminal measuring the downlink reference signal in the first downlink reference signal set or the second downlink reference signal set, and the label instance is obtained based on the terminal measuring the downlink reference signal in the third downlink reference signal set; wherein the data set is used to train a first model on the network device side, and the first model is used to predict the target downlink reference signal and / or target link quality.

[0377] In some embodiments, the device also includes: a sending unit 1402, used to send first configuration information and / or second configuration information to the terminal; the first configuration information is used to configure at least one first downlink reference signal set or at least one second downlink reference signal set; the second configuration information is used to configure at least one third downlink reference signal set.

[0378] In some embodiments, the first configuration information is used to configure a first downlink reference signal set or a second downlink reference signal set, and the first downlink reference signal set or the second downlink reference signal set is used to obtain an input instance; or, the first configuration information is used to configure multiple first downlink reference signal sets or multiple second downlink reference signal sets, the sending unit 1402 is used to send first indication information to the terminal, the first indication information is used to indicate activation of a first downlink reference signal set in multiple first downlink reference signal sets or activation of a second downlink reference signal set in multiple second downlink reference signal sets, and the first downlink reference signal set or the second downlink reference signal set is used to obtain the input instance.

[0379] In some embodiments, the second configuration information is used to configure a third downlink reference signal set, and the third downlink reference signal set is used to obtain a label instance; or, the second configuration information is used to configure multiple third downlink reference signal sets, and the sending unit 1402 is used to send second indication information to the terminal, and the second indication information is used to indicate activation of a third downlink reference signal set among multiple third downlink reference signal sets, and the third downlink reference signal set is used to obtain a label instance.

[0380] In some implementations, the first configuration information and / or the second configuration information is related to capability information of the terminal.

[0381] In some implementations, the receiving unit 1401 is configured to receive terminal capability information sent by a terminal, where the terminal capability information includes at least one of the following:

[0382] First information, the first information is used to indicate whether the terminal supports collection of a data set for prediction of a target downlink reference signal and / or a target link quality;

[0383] Second information, where the second information is used to indicate a maximum number of first downlink reference signal sets or second downlink reference signal sets supported by the terminal on all downlink CCs or BWPs;

[0384] The third information is used to indicate the maximum number of first downlink reference signal sets or second downlink reference signal sets supported by the terminal on each downlink CC or BWP;

[0385] Fourth information, the fourth information is used to indicate a maximum number of downlink reference signal resources supported by the terminal in all first downlink reference signal sets or second downlink reference signal sets;

[0386] Fifth information, the fifth information is used to indicate a maximum number of downlink reference signal resources supported by the terminal in each first downlink reference signal set or second downlink reference signal set;

[0387] Sixth information, the sixth information is used to indicate the maximum number of third downlink reference signal sets supported by the terminal on all downlink CCs or BWPs;

[0388] The seventh information is used to indicate the maximum number of third downlink reference signal sets supported by the terminal on each downlink CC or BWP;

[0389] The eighth information is used to indicate the maximum number of downlink reference signal resources supported by the terminal in all third downlink reference signal sets;

[0390] Ninth information: The ninth information is used to indicate the maximum number of downlink reference signal resources supported by the terminal in each third downlink reference signal set.

[0391] In some embodiments, the input instance includes M link quality measurement results; or, the input instance includes M link quality measurement results and M downlink reference signal resource indexes, and the M link quality measurement results and the M downlink reference signal resource indexes have a corresponding relationship; M is a positive integer.

[0392] In some embodiments, the M link quality measurement results are obtained based on the terminal measuring the first downlink reference signal set; or, the M link quality measurement results are the best M link quality measurement results among multiple link quality measurement results, and the multiple link quality measurement results are obtained based on the terminal measuring the second downlink reference signal set, and M is a positive integer.

[0393] In some embodiments, the M link quality measurement results have a first format or a second format in an input instance: the first format includes: M link quality measurement results; the second format includes: a first link quality measurement result and M-1 differential values, where the first link quality measurement result is one of the M link quality measurement results, and the M-1 differential values ​​are differential values ​​of the M-1 link quality measurement results other than the first link quality measurement result in the M link quality measurement results relative to the first link quality measurement result.

[0394] In some embodiments, the label instance includes a first label; or the label instance includes a first label and a second label; the first label includes a downlink reference signal resource index that satisfies a first condition, and the downlink reference signal resource index that satisfies the first condition is determined based on the terminal measuring the downlink reference signal in the third downlink reference signal set; the second label includes K link quality measurement results and / or K downlink reference signal resource indexes that meet the second condition, the K link quality measurement results and the K downlink reference signal resource indexes have a corresponding relationship, and the K link quality measurement results that meet the second condition are determined based on the terminal measuring the downlink reference signal in the third downlink reference signal set, and K is a positive integer.

[0395] In some implementations, the downlink reference signal resource index that meets the first condition is a downlink reference signal resource index corresponding to an optimal link quality measurement result obtained by the terminal measuring downlink reference signals in the third downlink reference signal set.

[0396] In some implementations, the K link quality measurement results that meet the second condition refer to: the best first K link quality measurement results obtained by the terminal measuring the downlink reference signals in the third downlink reference signal set.

[0397] In some embodiments, the K link quality measurement results have a third format or a fourth format in the tag instance; the third format includes: K link quality measurement results; the fourth format includes: a second link quality measurement result and K-1 differential values, where the second link quality measurement result is one of the K link quality measurement results, and the K-1 differential values ​​are differential values ​​of the K-1 link quality measurement results other than the second link quality measurement result in the K link quality measurement results relative to the second link quality measurement result.

[0398] In some implementations, the data set is carried in at least one of the following signaling: RRC signaling, UCI, and MAC layer container.

[0399] In some embodiments, an input instance and a label instance in a dataset constitute a sample for training the first model.

[0400] In some embodiments, N input instances and P label instances in a data set constitute a sample for training the first model; the N input instances and P label instances are obtained by the terminal in a time-sharing manner, and N and P are positive integers.

[0401] In some embodiments, the N input instances have a fifth format or a sixth format in the sample; the fifth format includes: N input instances; the order of the N input instances in the sample is related to the time domain order of the N input instances; the sixth format includes: N input instances and N instance serial numbers; the N input instances have a corresponding relationship with the N instance serial numbers, and the N instance serial numbers are related to the time domain order of the N input instances.

[0402] In some embodiments, the P label instances have a seventh format or an eighth format in the sample; the seventh format includes: P label instances; the order of the P label instances in the sample is related to the time domain order of the P label instances; the eighth format includes: P label instances and P instance serial numbers; the P label instances have a corresponding relationship with the P instance serial numbers, and the P instance serial numbers are related to the time domain order of the P label instances.

[0403] In some implementations, the downlink reference signal includes: CSI-RS and / or SSB.

[0404] In some implementations, the link quality includes at least one of the following: RSRP, SINR, RSSI, and RSRQ.

[0405] Those skilled in the art should understand that the relevant description of determining the above-mentioned data set in the embodiment of the present application can be understood with reference to the relevant description of the method for determining the data set in the embodiment of the present application.

[0406] Figure 15 is a schematic diagram of a communication device 1500 provided in an embodiment of the present application. The communication device can be a terminal or a network device. The communication device 1500 shown in Figure 15 includes a processor 1510, which can call and execute a computer program from a memory to implement the method in the embodiment of the present application.

[0407] Optionally, as shown in FIG15 , the communication device 1500 may further include a memory 1520. The processor 1510 may call and execute a computer program from the memory 1520 to implement the method in the embodiment of the present application.

[0408] The memory 1520 may be a separate device independent of the processor 1510 , or may be integrated into the processor 1510 .

[0409] Optionally, as shown in FIG15 , the communication device 1500 may further include a transceiver 1530 , and the processor 1510 may control the transceiver 1530 to communicate with other devices, specifically, to send information or data to other devices, or to receive information or data sent by other devices.

[0410] The transceiver 1530 may include a transmitter and a receiver. The transceiver 1530 may further include an antenna, and the number of antennas may be one or more.

[0411] Optionally, the communication device 1500 may specifically be a network device in an embodiment of the present application, and the communication device 1500 may implement the corresponding processes implemented by the network device in each method in the embodiment of the present application. For the sake of brevity, they will not be repeated here.

[0412] Optionally, the communication device 1500 may specifically be a mobile terminal / terminal of an embodiment of the present application, and the communication device 1500 may implement the corresponding processes implemented by the mobile terminal / terminal in each method of the embodiment of the present application. For the sake of brevity, they will not be repeated here.

[0413] Figure 16 is a schematic structural diagram of a chip according to an embodiment of the present application. The chip 1600 shown in Figure 16 includes a processor 1610, which can call and run a computer program from a memory to implement the method according to the embodiment of the present application.

[0414] Optionally, as shown in FIG16 , the chip 1600 may further include a memory 1620. The processor 1610 may call and execute a computer program from the memory 1620 to implement the method in the embodiment of the present application.

[0415] The memory 1620 may be a separate device independent of the processor 1610 , or may be integrated into the processor 1610 .

[0416] Optionally, the chip 1600 may further include an input interface 1630. The processor 1610 may control the input interface 1630 to communicate with other devices or chips, and specifically, may obtain information or data sent by other devices or chips.

[0417] Optionally, the chip 1600 may further include an output interface 1640. The processor 1610 may control the output interface 1640 to communicate with other devices or chips, and specifically, may output information or data to other devices or chips.

[0418] Optionally, the chip can be applied to the network device in the embodiments of the present application, and the chip can implement the corresponding processes implemented by the network device in each method of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0419] Optionally, the chip can be applied to the mobile terminal / terminal in the embodiments of the present application, and the chip can implement the corresponding processes implemented by the mobile terminal / terminal in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0420] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0421] FIG17 is a schematic block diagram of a communication system 1700 provided in an embodiment of the present application. As shown in FIG17 , the communication system 1700 includes a terminal 1710 and a network device 1720 .

[0422] Among them, the terminal 1710 can be used to implement the corresponding functions implemented by the terminal in the above method, and the network device 1720 can be used to implement the corresponding functions implemented by the network device in the above method. For the sake of brevity, they are not repeated here.

[0423] It should be understood that the processor of the embodiments of the present application may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method embodiment can be completed by hardware integrated logic circuits in the processor or software instructions. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The various methods, steps, and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of the present application can be directly embodied as being executed by a hardware decoding processor, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium mature in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0424] It is understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0425] It should be understood that the above-mentioned memories are exemplary but not restrictive. For example, the memories in the embodiments of the present application may also be static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM RAM (DR RAM), etc. In other words, the memories in the embodiments of the present application are intended to include, but are not limited to, these and any other suitable types of memories.

[0426] An embodiment of the present application also provides a computer-readable storage medium for storing a computer program.

[0427] Optionally, the computer-readable storage medium can be applied to the network device in the embodiments of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the network device in the various methods of the embodiments of the present application. For the sake of brevity, they are not repeated here.

[0428] Optionally, the computer-readable storage medium can be applied to the mobile terminal / terminal in the embodiments of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the mobile terminal / terminal in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0429] An embodiment of the present application also provides a computer program product, including computer program instructions.

[0430] Optionally, the computer program product can be applied to the network device in the embodiments of the present application, and the computer program instructions enable the computer to execute the corresponding processes implemented by the network device in the various methods of the embodiments of the present application. For the sake of brevity, they are not repeated here.

[0431] Optionally, the computer program product can be applied to the mobile terminal / terminal in the embodiments of the present application, and the computer program instructions enable the computer to execute the corresponding processes implemented by the mobile terminal / terminal in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0432] The embodiment of the present application also provides a computer program.

[0433] Optionally, the computer program can be applied to the network device in the embodiments of the present application. When the computer program runs on a computer, the computer executes the corresponding processes implemented by the network device in the various methods of the embodiments of the present application. For the sake of brevity, they are not described here.

[0434] Optionally, the computer program can be applied to the mobile terminal / terminal in the embodiments of the present application. When the computer program runs on the computer, the computer executes the corresponding processes implemented by the mobile terminal / terminal in the various methods of the embodiments of the present application. For the sake of brevity, they are not repeated here.

[0435] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0436] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0437] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0438] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0439] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0440] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0441] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A method for determining a data set, the method comprising: The terminal sends a data set to the network device, wherein the data set includes at least one input instance and at least one label instance; The input instance is obtained based on the terminal measuring a downlink reference signal in a first downlink reference signal set or a second downlink reference signal set, and the label instance is obtained based on the terminal measuring a downlink reference signal in a third downlink reference signal set; wherein the data set is used to train a first model on the network device side, and the first model is used to predict a target downlink reference signal and / or a target link quality.

2. The method according to claim 1, wherein: The method further comprises: The terminal receives the first configuration information and / or the second configuration information sent by the network device; The first configuration information is used to configure at least one first downlink reference signal set or at least one second downlink reference signal set; and the second configuration information is used to configure at least one third downlink reference signal set.

3. The method according to claim 2, wherein: The first configuration information is used to configure a first downlink reference signal set or a second downlink reference signal set, and the first downlink reference signal set or the second downlink reference signal set is used to obtain the input instance; or, The first configuration information is used to configure a situation where multiple first downlink reference signal sets or multiple second downlink reference signal sets, and the method also includes: the terminal receives first indication information sent by the network device, the first indication information is used to indicate activation of a first downlink reference signal set in the multiple first downlink reference signal sets or activation of a second downlink reference signal set in the multiple second downlink reference signal sets, and the one first downlink reference signal set or the one second downlink reference signal set is used to obtain the input instance.

4. The method according to claim 2, wherein: A case where the second configuration information is used to configure a third downlink reference signal set, and the third downlink reference signal set is used to obtain the label instance; or, The second configuration information is used to configure multiple third downlink reference signal sets. The method also includes: the terminal receives second indication information sent by the network device, the second indication information is used to indicate the activation of a third downlink reference signal set among the multiple third downlink reference signal sets, and the third downlink reference signal set is used to obtain the label instance.

5. The method according to any one of claims 2 to 4, wherein: The first configuration information and / or the second configuration information are related to capability information of the terminal.

6. The method according to any one of claims 2 to 5, wherein: The method further comprises: The terminal sends the capability information of the terminal to the network device, where the capability information of the terminal includes at least one of the following: first information, where the first information is used to indicate whether the terminal supports collection of a data set for prediction of the target downlink reference signal and / or target link quality; second information, where the second information is used to indicate a maximum number of the first downlink reference signal sets or the second downlink reference signal sets supported by the terminal on all downlink component carriers CC or bandwidth parts BWP; third information, where the third information is used to indicate the maximum number of the first downlink reference signal sets or the second downlink reference signal sets supported by the terminal on each downlink CC or BWP; Fourth information, where the fourth information is used to indicate a maximum number of downlink reference signal resources supported by the terminal in all first downlink reference signal sets or second downlink reference signal sets; fifth information, where the fifth information is used to indicate a maximum number of downlink reference signal resources supported by the terminal in each first downlink reference signal set or second downlink reference signal set; Sixth information, where the sixth information is used to indicate a maximum number of third downlink reference signal sets supported by the terminal on all downlink CCs or BWPs; seventh information, where the seventh information is used to indicate a maximum number of third downlink reference signal sets supported by the terminal on each downlink CC or BWP; eighth information, where the eighth information is used to indicate the maximum number of downlink reference signal resources supported by the terminal in all third downlink reference signal sets; Ninth information, the ninth information is used to indicate the maximum number of downlink reference signal resources supported by the terminal in each third downlink reference signal set.

7. The method according to any one of claims 1 to 6, wherein: The input instance includes M link quality measurement results; or, The input instance includes M link quality measurement results and M downlink reference signal resource indexes, and the M link quality measurement results and the M downlink reference signal resource indexes have a corresponding relationship; M is a positive integer.

8. The method according to claim 7, wherein: The M link quality measurement results are obtained based on the terminal measuring the first downlink reference signal set; or, The M link quality measurement results are optimal M link quality measurement results among multiple link quality measurement results, the multiple link quality measurement results are obtained based on the terminal measuring the second downlink reference signal set, and M is a positive integer.

9. The method according to claim 7 or 8, wherein: The M link quality measurement results have a first format or a second format in the input instance: The first format includes: the M link quality measurement results; The second format includes: a first link quality measurement result and M-1 differential values, wherein the first link quality measurement result is one of the M link quality measurement results, and the M-1 differential values ​​are differential values ​​of M-1 link quality measurement results other than the first link quality measurement result among the M link quality measurement results relative to the first link quality measurement result.

10. The method according to any one of claims 1 to 9, wherein: The tag instance includes a first tag; or, The label instance includes a first label and a second label; The first tag includes a downlink reference signal resource index that satisfies a first condition, and the downlink reference signal resource index that satisfies the first condition is determined based on the terminal measuring a downlink reference signal in the third downlink reference signal set; The second label includes K link quality measurement results and / or K downlink reference signal resource indexes that meet the second condition, the K link quality measurement results and the K downlink reference signal resource indexes have a corresponding relationship, the K link quality measurement results that meet the second condition are determined based on the terminal measuring the downlink reference signal in the third downlink reference signal set, and K is a positive integer.

11. The method according to claim 10, wherein: The downlink reference signal resource index that meets the first condition refers to: A downlink reference signal resource index corresponding to an optimal link quality measurement result obtained by measuring, by the terminal, a downlink reference signal in the third downlink reference signal set.

12. The method according to claim 10 or 11, wherein: The K link quality measurement results satisfying the second condition are: The terminal measures the downlink reference signals in the third downlink reference signal set and obtains the best first K link quality measurement results.

13. The method according to any one of claims 10 to 12, wherein: The K link quality measurement results have a third format or a fourth format in the tag instance; The third format includes: the K link quality measurement results; The fourth format includes: a second link quality measurement result and K-1 differential values, wherein the second link quality measurement result is one of the K link quality measurement results, and the K-1 differential values ​​are differential values ​​of K-1 link quality measurement results among the K link quality measurement results other than the second link quality measurement result relative to the second link quality measurement result.

14. The method according to any one of claims 1 to 13, wherein: The data set is carried in at least one of the following signalings: radio resource control RRC signaling, uplink control information UCI, and media access control MAC layer container.

15. The method according to any one of claims 1 to 14, wherein: An input instance and a label instance in the data set constitute a sample for training the first model.

16. The method according to any one of claims 1 to 14, wherein: The N input instances and P label instances in the data set constitute a sample for training the first model; the N input instances and the P label instances are obtained by the terminal in a time-sharing manner, and N and P are positive integers.

17. The method according to claim 16, wherein: The N input instances have the fifth format or the sixth format in the sample; The fifth format includes: the N input instances; the order of the N input instances in the sample is related to the time domain order of the N input instances; The sixth format includes: the N input instances and N instance serial numbers; the N input instances have a corresponding relationship with the N instance serial numbers, and the N instance serial numbers are related to the time domain sequence of the N input instances.

18. The method according to claim 16 or 17, wherein: The P label instances have the seventh format or the eighth format in the sample; The seventh format includes: the P tag instances; the order of the P tag instances in the sample is related to the time domain order of the P tag instances; The eighth format includes: the P label instances and P instance serial numbers; the P label instances have a corresponding relationship with the P instance serial numbers, and the P instance serial numbers are related to the time domain order of the P label instances.

19. The method according to any one of claims 16 to 18, wherein: The N input instances are obtained by the terminal performing N measurements on the downlink reference signals in the first downlink reference signal set or the second downlink reference signal set within a first time period, and the P label instances are obtained by the terminal performing P measurements on the downlink reference signals in the third downlink reference signal set within a second time period.

20. The method according to claim 19, wherein: The method further comprises: The terminal performs N measurements on the downlink reference signal in the first downlink reference signal set or the second downlink reference signal set within a first time period to obtain N input instances corresponding to the first time period; The terminal performs P measurements on the downlink reference signal in the third downlink reference signal set within a second time period to obtain P label instances corresponding to the second time period.

21. The method according to claim 19 or 20, wherein: The terminal sends a data set to a network device, including: The terminal sends a data set to a network device according to a first reporting unit, where the first reporting unit includes N input instances corresponding to the first time period and P label instances corresponding to the second time period.

22. The method according to any one of claims 16 to 18, wherein: The N input instances are obtained by the terminal performing N measurements on the downlink reference signals in the first downlink reference signal set or the second downlink reference signal set within N cycles, and the P label instances are obtained by the terminal performing P measurements on the downlink reference signals in the third downlink reference signal set within the last cycle of the N cycles.

23. The method according to claim 22, wherein: The method further comprises: The terminal performs one measurement on a downlink reference signal in the first downlink reference signal set or the second downlink reference signal set and performs P measurements on a downlink reference signal in the third downlink reference signal set in each of the N periods, to obtain an input instance and P label instances corresponding to each period.

24. The method according to claim 22 or 23, wherein: The terminal sends a data set to a network device, including: The terminal sends the data set to the network device according to the second reporting unit; The second reporting unit includes one input instance and P label instances corresponding to each cycle; or, The second reporting unit includes N input instances corresponding to the N cycles and P label instances corresponding to the last cycle of the N cycles; or, The second reporting unit includes all input instances and all label instances corresponding to the N cycles.

25. The method according to any one of claims 1 to 24, wherein: The downlink reference signal includes: a channel state information-reference signal CSI-RS and / or a synchronization signal block SSB.

26. The method according to any one of claims 1 to 25, wherein: The link quality includes at least one of the following: reference signal received power RSRP, signal to interference plus noise ratio SINR, received signal strength indication RSSI and reference signal received quality RSRQ.

27. A method for determining a data set, the method comprising: The terminal obtains a data set, which includes at least one input instance and at least one label instance; the input instance is obtained based on measuring a downlink reference signal in a first downlink reference signal set or a second downlink reference signal set, and the label instance is obtained based on measuring a downlink reference signal in a third downlink reference signal set; wherein the data set is used to adjust a second model on the terminal side, and the second model is used to predict a target downlink reference signal and / or a target link quality.

28. The method according to claim 27, wherein: The terminal obtains a data set, including: The terminal receives a data set sent by a network device, where the data set is all or part of a data set reported by at least one terminal to the network device, and the at least one terminal includes the terminal or does not include the terminal.

29. The method according to claim 28, wherein: The data set is carried in at least one of the following signalings: RRC signaling, MAC layer container.

30. The method of claim 27, wherein: The terminal obtains a data set, including: The terminal measures a downlink reference signal in the first downlink reference signal set or the second downlink reference signal set to obtain the input instance; and, The terminal measures a downlink reference signal in a third downlink reference signal set to obtain the label instance.

31. The method according to claim 30, wherein: The method further comprises: The terminal receives the first configuration information and / or the second configuration information sent by the network device; The first configuration information is used to configure at least one first downlink reference signal set or at least one second downlink reference signal set; and the second configuration information is used to configure at least one third downlink reference signal set.

32. The method according to claim 31, wherein: The first configuration information is used to configure a first downlink reference signal set or a second downlink reference signal set, and the first downlink reference signal set or the second downlink reference signal set is used to obtain the input instance; or, The first configuration information is used to configure a situation where multiple first downlink reference signal sets or multiple second downlink reference signal sets, and the method also includes: the terminal receives first indication information sent by the network device, the first indication information is used to indicate activation of a first downlink reference signal set in the multiple first downlink reference signal sets or activation of a second downlink reference signal set in the multiple second downlink reference signal sets, and the one first downlink reference signal set or the one second downlink reference signal set is used to obtain the input instance.

33. The method according to claim 31, wherein: A case where the second configuration information is used to configure a third downlink reference signal set, and the third downlink reference signal set is used to obtain the label instance; or, The second configuration information is used to configure multiple third downlink reference signal sets. The method also includes: the terminal receives second indication information sent by the network device, the second indication information is used to indicate the activation of a third downlink reference signal set among the multiple third downlink reference signal sets, and the third downlink reference signal set is used to obtain the label instance.

34. A method according to any one of claims 31 to 33, wherein: The first configuration information and / or the second configuration information are related to capability information of the terminal.

35. The method according to any one of claims 31 to 34, wherein: The method further comprises: The terminal sends the capability information of the terminal to the network device, where the capability information of the terminal includes at least one of the following: first information, where the first information is used to indicate whether the terminal supports collection of a data set for prediction of the target downlink reference signal and / or target link quality; second information, where the second information is used to indicate a maximum number of the first downlink reference signal sets or second downlink reference signal sets supported by the terminal on all downlink CCs or BWPs; third information, where the third information is used to indicate the maximum number of the first downlink reference signal sets or the second downlink reference signal sets supported by the terminal on each downlink CC or BWP; Fourth information, where the fourth information is used to indicate a maximum number of downlink reference signal resources supported by the terminal in all first downlink reference signal sets or second downlink reference signal sets; fifth information, where the fifth information is used to indicate a maximum number of downlink reference signal resources supported by the terminal in each first downlink reference signal set or second downlink reference signal set; Sixth information, where the sixth information is used to indicate a maximum number of third downlink reference signal sets supported by the terminal on all downlink CCs or BWPs; seventh information, where the seventh information is used to indicate a maximum number of third downlink reference signal sets supported by the terminal on each downlink CC or BWP; eighth information, where the eighth information is used to indicate the maximum number of downlink reference signal resources supported by the terminal in all third downlink reference signal sets; Ninth information, the ninth information is used to indicate the maximum number of downlink reference signal resources supported by the terminal in each third downlink reference signal set.

36. A method according to any one of claims 27 to 35, wherein: The downlink reference signal includes: CSI-RS and / or SSB.

37. The method according to any one of claims 27 to 36, wherein: The link quality includes at least one of the following: RSRP, SINR, RSSI and RSRQ.

38. A method for determining a data set, the method comprising: The network device receives a data set sent by a terminal, wherein the data set includes at least one input instance and at least one label instance; The input instance is obtained based on the terminal measuring a downlink reference signal in a first downlink reference signal set or a second downlink reference signal set, and the label instance is obtained based on the terminal measuring a downlink reference signal in a third downlink reference signal set; wherein the data set is used to train a first model on the network device side, and the first model is used to predict a target downlink reference signal. Reference signal and / or target link quality.

39. The method of claim 38, wherein: The method further comprises: The network device sends first configuration information and / or second configuration information to the terminal; The first configuration information is used to configure at least one first downlink reference signal set or at least one second downlink reference signal set; and the second configuration information is used to configure at least one third downlink reference signal set.

40. The method of claim 39, wherein: The first configuration information is used to configure a first downlink reference signal set or a second downlink reference signal set, and the first downlink reference signal set or the second downlink reference signal set is used to obtain the input instance; or, The first configuration information is used to configure a plurality of first downlink reference signal sets or a plurality of second downlink reference signal sets, and the method further includes: the network device sends first indication information to the terminal, the first indication information is used to indicate activation of a first downlink reference signal set in the plurality of first downlink reference signal sets or activation of a second downlink reference signal set in the plurality of second downlink reference signal sets, and the first downlink reference signal set or the second downlink reference signal set is used to obtain the input instance.

41. The method of claim 39, wherein: A case where the second configuration information is used to configure a third downlink reference signal set, and the third downlink reference signal set is used to obtain the label instance; or, The second configuration information is used to configure multiple third downlink reference signal sets. The method also includes: the network device sends second indication information to the terminal, the second indication information is used to indicate the activation of a third downlink reference signal set among the multiple third downlink reference signal sets, and the third downlink reference signal set is used to obtain the label instance.

42. The method according to any one of claims 39 to 41, wherein: The first configuration information and / or the second configuration information are related to capability information of the terminal.

43. A method according to any one of claims 39 to 42, wherein: The method further comprises: The network device receives capability information of the terminal sent by the terminal, where the capability information of the terminal includes at least one of the following: first information, where the first information is used to indicate whether the terminal supports collection of a data set for prediction of the target downlink reference signal and / or target link quality; second information, where the second information is used to indicate a maximum number of the first downlink reference signal sets or second downlink reference signal sets supported by the terminal on all downlink CCs or BWPs; third information, where the third information is used to indicate the maximum number of the first downlink reference signal sets or the second downlink reference signal sets supported by the terminal on each downlink CC or BWP; Fourth information, where the fourth information is used to indicate a maximum number of downlink reference signal resources supported by the terminal in all first downlink reference signal sets or second downlink reference signal sets; fifth information, where the fifth information is used to indicate a maximum number of downlink reference signal resources supported by the terminal in each first downlink reference signal set or second downlink reference signal set; Sixth information, where the sixth information is used to indicate a maximum number of third downlink reference signal sets supported by the terminal on all downlink CCs or BWPs; seventh information, where the seventh information is used to indicate a maximum number of third downlink reference signal sets supported by the terminal on each downlink CC or BWP; eighth information, where the eighth information is used to indicate the maximum number of downlink reference signal resources supported by the terminal in all third downlink reference signal sets; Ninth information, the ninth information is used to indicate the maximum number of downlink reference signal resources supported by the terminal in each third downlink reference signal set.

44. A method according to any one of claims 38 to 43, wherein: The input instance includes M link quality measurement results; or, The input instance includes M link quality measurement results and M downlink reference signal resource indexes, and the M link quality measurement results and the M downlink reference signal resource indexes have a corresponding relationship; M is a positive integer.

45. The method of claim 44, wherein: The M link quality measurement results are obtained based on the terminal measuring the first downlink reference signal set; or, The M link quality measurement results are optimal M link quality measurement results among multiple link quality measurement results, the multiple link quality measurement results are obtained based on the terminal measuring the second downlink reference signal set, and M is a positive integer.

46. ​​The method according to claim 44 or 45, wherein: The M link quality measurement results have a first format or a second format in the input instance: The first format includes: the M link quality measurement results; The second format includes: a first link quality measurement result and M-1 differential values, wherein the first link quality measurement result is one of the M link quality measurement results, and the M-1 differential values ​​are differential values ​​of M-1 link quality measurement results other than the first link quality measurement result among the M link quality measurement results relative to the first link quality measurement result.

47. A method according to any one of claims 38 to 46, wherein: The tag instance includes a first tag; or, The label instance includes a first label and a second label; The first tag includes a downlink reference signal resource index that satisfies a first condition, and the downlink reference signal resource index that satisfies the first condition is determined based on the terminal measuring a downlink reference signal in the third downlink reference signal set; The second label includes K link quality measurement results and / or K downlink reference signal resource indexes that meet the second condition, the K link quality measurement results and the K downlink reference signal resource indexes have a corresponding relationship, the K link quality measurement results that meet the second condition are determined based on the terminal measuring the downlink reference signal in the third downlink reference signal set, and K is a positive integer.

48. The method of claim 47, wherein: The downlink reference signal resource index that meets the first condition refers to: A downlink reference signal resource index corresponding to an optimal link quality measurement result obtained by measuring, by the terminal, a downlink reference signal in the third downlink reference signal set.

49. The method according to claim 47 or 48, wherein: The K link quality measurement results satisfying the second condition are: The terminal measures the downlink reference signals in the third downlink reference signal set and obtains the best first K link quality measurement results.

50. The method according to any one of claims 47 to 49, wherein: The K link quality measurement results have a third format or a fourth format in the tag instance; The third format includes: the K link quality measurement results; The fourth format includes: a second link quality measurement result and K-1 differential values, wherein the second link quality measurement result is one of the K link quality measurement results, and the K-1 differential values ​​are differential values ​​of K-1 link quality measurement results among the K link quality measurement results other than the second link quality measurement result relative to the second link quality measurement result.

51. A method according to any one of claims 38 to 50, wherein: The data set is carried in at least one of the following signaling: RRC signaling, UCI, and MAC layer container.

52. The method according to any one of claims 38 to 51, wherein: An input instance and a label instance in the data set constitute a sample for training the first model.

53. The method according to any one of claims 38 to 51, wherein: The N input instances and P label instances in the data set constitute a sample for training the first model; the N input instances and the P label instances are obtained by the terminal in a time-sharing manner, and N and P are positive integers.

54. The method of claim 53, wherein: The N input instances have the fifth format or the sixth format in the sample; The fifth format includes: the N input instances; the order of the N input instances in the sample is related to the time domain order of the N input instances; The sixth format includes: the N input instances and N instance serial numbers; the N input instances have a corresponding relationship with the N instance serial numbers, and the N instance serial numbers are related to the time domain sequence of the N input instances.

55. The method of claim 53 or 54, wherein: The P label instances have the seventh format or the eighth format in the sample; The seventh format includes: the P tag instances; the order of the P tag instances in the sample is related to the time domain order of the P tag instances; The eighth format includes: the P label instances and P instance serial numbers; the P label instances have a corresponding relationship with the P instance serial numbers, and the P instance serial numbers are related to the time domain order of the P label instances.

56. The method according to any one of claims 38 to 55, wherein: The downlink reference signal includes: CSI-RS and / or SSB.

57. A method according to any one of claims 38 to 56, wherein: The link quality includes at least one of the following: RSRP, SINR, RSSI and RSRQ.

58. A device for determining a data set, applied to a terminal, the device comprising: A sending unit, configured to send a data set to a network device, wherein the data set includes at least one input instance and at least one label instance; The input instance is obtained based on the terminal measuring a downlink reference signal in a first downlink reference signal set or a second downlink reference signal set, and the label instance is obtained based on the terminal measuring a downlink reference signal in a third downlink reference signal set; wherein the data set is used to train a first model on the network device side, and the first model is used to predict a target downlink reference signal and / or a target link quality.

59. A device for determining a data set, applied to a terminal, the device comprising: An acquisition unit, configured to acquire a data set, wherein the data set includes at least one input instance and at least one label instance; The input instance is obtained based on measuring a downlink reference signal in a first downlink reference signal set or a second downlink reference signal set, and the label instance is obtained based on measuring a downlink reference signal in a third downlink reference signal set; wherein the data set is used to adjust a second model on the terminal side, and the second model is used to predict a target downlink reference signal and / or a target link quality.

60. A device for determining a data set, applied to a network device, the device comprising: A receiving unit, configured to receive a data set sent by a terminal, wherein the data set includes at least one input instance and at least one label instance; The input instance is obtained based on the terminal measuring a downlink reference signal in a first downlink reference signal set or a second downlink reference signal set, and the label instance is obtained based on the terminal measuring a downlink reference signal in a third downlink reference signal set; wherein the data set is used to train a first model on the network device side, and the first model is used to predict a target downlink reference signal and / or a target link quality.

61. A communication device, the communication device comprising: A memory for storing computer executable instructions; A processor, connected to the memory, for implementing the method of any one of claims 1 to 26, or implementing the method of any one of claims 27 to 37, by executing the computer executable instructions; implementing the method of any one of claims 38 to 57.

62. A chip, comprising: A processor, used to call and run a computer program from a memory so that a device equipped with the chip executes a method as described in any one of claims 1 to 26, or implements a method as described in any one of claims 27 to 37; implements a method as described in any one of claims 38 to 57.

63. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by at least one processor, implements the method as described in any one of claims 1 to 26, or implements the method as described in any one of claims 27 to 37; implements the method as described in any one of claims 38 to 57.

Images