Beam management method and device, user equipment and network equipment

Abstract

The embodiment of the invention provides a beam management method and device, user equipment and network equipment. The method comprises the following steps: the UE measures link quality of a plurality of beams or beam pairs; the UE determines at least one receiving beam index and / or uplink feedback information based on the link quality of the plurality of beams or beam pairs by using a first sub-model; wherein the uplink feedback information is used for determining at least one transmitting beam index and / or at least one link quality corresponding to the at least one transmitting beam index. The network equipment receives uplink feedback information sent by the UE; and the network device determines at least one transmitting beam index and / or at least one link quality corresponding to the at least one transmitting beam index based on the uplink feedback information by using a second sub-model.

Description

Beam management method and device, user equipment, and network equipment Technical Field

[0001] The embodiments of the present application relate to the field of mobile communication technology, and specifically to a beam management method and device, user equipment, and network equipment. Background Art

[0002] Analog beamforming technology can enhance network coverage while reducing device implementation complexity. Analog beamforming can be used not only on network equipment but also on user equipment (UE).

[0003] To achieve good communication quality, it is necessary to determine the optimal transmit beam and / or receive beam. In the current downlink beam management process, to determine the optimal transmit beam and / or receive beam, the UE needs to traverse all transmit beams and / or all receive beams to perform the corresponding link quality measurements, which incurs significant overhead and latency.

[0004] Summary of the Invention

[0005] Embodiments of the present application provide a beam management method and apparatus, user equipment, network equipment, a chip, a computer-readable storage medium, a computer program product, and a computer program.

[0006] The beam management method provided in the embodiment of the present application includes:

[0007] The UE measures the link quality of multiple beams or beam pairs;

[0008] The UE uses the first sub-model to determine at least one receive beam index and / or uplink feedback information based on the link quality of the multiple beams or beam pairs; wherein the uplink feedback information is used to determine at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index.

[0009] The beam management method provided in the embodiment of the present application includes:

[0010] The network device receives uplink feedback information sent by the UE;

[0011] The network device determines at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index based on the uplink feedback information using the second sub-model.

[0012] The beam management device provided in an embodiment of the present application is applied to a UE, and the device includes:

[0013] a measurement unit, configured to measure link qualities of a plurality of beams or beam pairs;

[0014] A processing unit, used to determine at least one receive beam index and / or uplink feedback information based on the link quality of the multiple beams or beam pairs using a first sub-model; wherein the uplink feedback information is used to determine at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index.

[0015] The beam management device provided in an embodiment of the present application is applied to a network device, and the device includes:

[0016] A receiving unit, configured to receive uplink feedback information sent by the UE;

[0017] A processing unit is configured to determine, using a second sub-model based on the uplink feedback information, at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index.

[0018] The user equipment provided in the embodiment of the present application includes a processor and a memory. The memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory to perform the above-mentioned beam management method.

[0019] The network device provided in an embodiment of the present application includes a processor and a memory. The memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory to perform the above-mentioned beam management method.

[0020] The chip provided in the embodiment of the present application is used to implement the above-mentioned beam management method.

[0021] Specifically, the chip includes: a processor, configured to call and run a computer program from a memory, so that a device equipped with the chip executes the above-mentioned beam management method.

[0022] The computer-readable storage medium provided in an embodiment of the present application is used to store a computer program, which enables a computer to execute the above-mentioned beam management method.

[0023] The computer program product provided in an embodiment of the present application includes computer program instructions, which enable a computer to execute the above-mentioned beam management method.

[0024] The computer program provided in the embodiment of the present application, when executed on a computer, enables the computer to execute the above-mentioned beam management method.

[0025] The technical solution of the embodiment of the present application deploys a first sub-model on the UE side, so that the UE can predict (i.e., infer) at least one optimal receiving beam index and / or uplink feedback information based on the measured link quality of a small number of beams or beam pairs through the first sub-model. On the other hand, a second sub-model is deployed on the network device side, so that the network device can predict at least one optimal transmitting beam index and / or corresponding at least one link quality based on the uplink feedback information of the UE through the second sub-model. By adopting the technical solution of the embodiment of the present application, the optimal transmitting beam and / or the optimal receiving beam can be quickly determined, and the resource overhead and delay in the beam scanning process can be significantly reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] 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:

[0027] FIG1 is a schematic diagram of an application scenario of an embodiment of the present application;

[0028] FIG2 is a schematic diagram of a neuron structure;

[0029] FIG3 is a schematic diagram of a deep neural network;

[0030] FIG4 is a schematic diagram of a convolutional neural network;

[0031] FIG5 is a schematic diagram of an LSTM unit structure;

[0032] FIG6 is a schematic diagram of a downlink beam scanning process;

[0033] FIG7 is a flow chart of a beam management method provided in an embodiment of the present application;

[0034] FIG8 is a schematic diagram of a model provided in an embodiment of the present application that is deployed separately on the UE side and the NW side;

[0035] FIG9-1 is a structural diagram of a UE-side sub-model according to an embodiment of the present application;

[0036] Figure 9-2 is a second structural diagram of the UE-side sub-model provided in an embodiment of the present application;

[0037] FIG9-3 is a third structural diagram of a UE-side sub-model provided in an embodiment of the present application;

[0038] FIG9-4 is a fourth structural diagram of a UE-side sub-model provided in an embodiment of the present application;

[0039] FIG10 is a schematic diagram of a model training process provided in an embodiment of the present application;

[0040] FIG11 is a schematic diagram of a beam indication process provided in an embodiment of the present application;

[0041] FIG12 is a schematic diagram of the structure of a beam management device according to an embodiment of the present application;

[0042] FIG13 is a second schematic diagram of the structure of a beam management device provided in an embodiment of the present application;

[0043] FIG14 is a schematic structural diagram of a communication device provided in an embodiment of the present application;

[0044] FIG15 is a schematic structural diagram of a chip according to an embodiment of the present application;

[0045] FIG16 is a schematic block diagram of a communication system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] 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.

[0047] Figure 1 is a schematic diagram of an application scenario of an embodiment of the present application. As shown in Figure 1, a communication system may include user equipment 110 and network equipment 120. Network equipment 120 may communicate with user equipment 110 via an air interface.

[0048] It should be understood that although the embodiments of the present application are illustratively described with reference to the communication system shown in FIG1 , the embodiments of the present application are not limited thereto. In other words, the technical solutions of the embodiments of the present application can be applied to various communication systems, such as a 5G communication system (also known as a New Radio (NR) communication system), or future communication systems.

[0049] In the communication system shown in FIG1 , network device 120 may be an access network device that communicates with user equipment 110. The access network device may provide communication coverage for a specific geographic area and may communicate with user equipment 110 located within the coverage area. As an example, the access network device may be a base station (gNB) in an NR system.

[0050] In the communication system shown in FIG1 , the user equipment 110 may be any user equipment, such as a mobile phone, a handheld device with wireless communication function, a vehicle-mounted device, a wearable device, etc.

[0051] The wireless communication system shown in FIG1 may further include a core network device 130 for communicating with the access network device. During the network evolution process, the name of the core network device 130 may be different.

[0052] It should be noted that the terms "system" and "network" are often used interchangeably in this article. The term "and / or" in this article is merely a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the objects associated before and after 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, which can mean that A directly indicates B, for example, B can be obtained through A; it can also mean that A indirectly indicates B, for example, A indicates C, and B can be obtained through C; it can also mean that there is an association relationship between A and B. 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 the two, or it can mean that there is an association relationship between the two, or it can mean a relationship between indication and indication, configuration and configuration, etc. 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 user equipment and network equipment). 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 NR protocol and related protocols used in future communication systems, and the present application does not limit this.

[0053] 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.

[0054] Neural Networks and Machine Learning

[0055] A neural network is a computational model consisting of multiple interconnected neurons (referred to as nodes for short), where the connections between nodes represent weighted values, called weights. A node can be connected to multiple upper-level nodes, with the outputs of multiple upper-level nodes serving as the inputs of the node. The node uses weights to perform a weighted summation of the inputs and processes the result through a specific activation function. As an example: the neuron structure is shown in Figure 2. The neuron uses n weights (denoted as w1, w2..., wn) to perform a weighted summation of n inputs (denoted as a1, a2,..., an), and processes the result through an activation function (denoted as f), ultimately outputting t; optionally, after performing the weighted summation, the neuron can also add a bias b to the result.

[0056] There are many types of neural networks, and deep neural networks (DNNs) are a common type. In addition to deep neural networks, other types of neural networks include convolutional neural networks (CNNs) and recurrent neural networks (RNNs).

[0057] Figure 3 illustrates a typical deep neural network. As shown in Figure 3, the basic structure of a deep neural network consists of an input layer, hidden layers, and an output layer. The input layer's inputs serve as the deep neural network's inputs, while the output layer's outputs serve as the deep neural network's outputs. By designing different node connections, weights, and activation functions, a deep neural network can produce different outputs, thereby fitting the mapping from input to output. The deep neural network shown in Figure 3 is a fully connected model, meaning that each node is connected to all nodes in the next level below it.

[0058] Figure 4 shows a typical convolutional neural network. As shown in Figure 4, the basic structure of a convolutional neural network includes an input layer, multiple convolutional layers, multiple pooling layers, a fully connected layer, and an output layer. The input of the input layer is the input of the convolutional neural network, and the output of the output layer is the output of the convolutional neural network. For a convolutional neural network, each neuron in the convolution kernel of the convolution layer is locally connected to its input. By introducing a pooling layer to extract the local maximum or average features of a certain layer, this effectively reduces the network parameters and exploits local features, enabling the convolutional neural network to converge quickly and achieve excellent performance.

[0059] Recurrent neural networks (RNNs) are neural networks that model sequential data and have achieved remarkable success in natural language processing applications such as machine translation and speech recognition. Specifically, RNNs memorize information from past moments and use it in calculating current outputs. Specifically, nodes in hidden layers are no longer disconnected but connected, and the input to a hidden layer includes not only the input from the previous layer but also the output from the previous hidden layer. RNNs typically incorporate a long-short-term memory (LSTM) artificial neural network (ANN) cell structure. Figure 5 illustrates a typical LSTM cell structure. As shown in Figure 5, the LSTM cell structure includes a tanh node and a σ node. Unlike convolutional neural networks, which only consider the most recent state, the LSTM cell state determines which states should be retained and which should be forgotten, addressing the long-term memory limitations of traditional CNNs.

[0060] It should be noted that the neural network in the technical solution of the embodiment of the present application can be a deep neural network, but is not limited to this. The deep neural network can also be replaced by a convolutional neural network or a recurrent neural network as the neural network in the technical solution of the embodiment of the present application.

[0061] Beam management

[0062] Beam management mechanisms are divided into uplink and downlink beam management. Downlink beam management involves downlink beam scanning, the UE reporting the optimal beam, and the network device indicating the downlink beam. It should be noted that for network devices, the downlink beam refers to the network device's transmit beam; for UEs, the downlink beam refers to the UE's receive beam.

[0063] As shown in Figure 6, the downlink beam scanning process has three processes, namely the P1 process, the P2 process, and the P3 process. For the P1 process, the network device traverses all transmit beams and the UE traverses all receive beams, so that all beam pair combinations can be traversed, and the UE measures the corresponding link quality for each beam pair. For the P2 process, the UE's receive beam is specific, and the network device traverses all transmit beams, so that all transmit beams can be traversed for a specific receive beam, and the UE measures the corresponding link quality for each transmit beam. For the P3 process, the network device's transmit beam is specific, and the UE traverses all receive beams, so that all receive beams can be traversed for a specific transmit beam, and the UE measures the corresponding link quality for each receive beam.

[0064] After the UE performs link quality measurement, it selects the K beams or beam pairs with the highest link quality (i.e., the K optimal beams or beam pairs) by comparing the link quality of all measured beams or beam pairs, and then reports them to the network device through uplink control information, where K is a positive integer.

[0065] After obtaining the K optimal beams or beam pairs, the network device selects a transmit beam for downlink transmission from the K optimal beams or beam pairs and indicates the transmit beam using the Transmission Configuration Indication (TCI) status carried in the Media Access Control-Control Element (MAC CE) or Downlink Control Information (DCI). The UE uses the receive beam corresponding to the transmit beam to receive downlink transmissions.

[0066] The beam scanning process described above requires traversing all transmit beams and / or receive beams, which results in significant overhead and latency. For example, assuming the network device has 64 transmit beams and the UE has 4 receive beams, the UE needs to measure 256 (i.e., 64×4) beam pairs for the P1 process, resulting in 256 downlink resource overheads. From a time perspective, the measurement of each beam pair takes a certain amount of time to complete, and the measurement of 256 beam pairs will take a significant amount of time. Therefore, the current beam scanning process results in significant overhead and latency.

[0067] The technical solution of the embodiment of the present application uses artificial intelligence / machine learning (AI / ML) technology to train a neural network by constructing a data set and deploying it on the UE side and / or the network device side, so that the UE can predict (i.e., infer) at least one optimal receiving beam index and / or uplink feedback information through the neural network when the measurement results of some beams or beam pairs are used as inputs to the neural network. The network device predicts at least one optimal transmitting beam index and / or corresponding at least one link quality based on the uplink feedback information reported by the UE through the neural network. In this way, the network device and the UE can quickly match the optimal transmitting beam and / or the optimal receiving beam, and significantly reduce the resource overhead and delay in the beam scanning process.

[0068] 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.

[0069] It should be noted that the beam described in the embodiments of the present application refers to a transmit beam or a receive beam, and a beam pair refers to a pair of a transmit beam and a receive beam.

[0070] It should be noted that the “model” described in the embodiments of the present application may also be referred to as a “neural network”, which may be, but is not limited to, a deep neural network, a convolutional neural network, and a recurrent neural network.

[0071] FIG7 is a flow chart of a beam management method provided in an embodiment of the present application. As shown in FIG7 , the beam management method includes the following steps:

[0072] Step 701: The UE measures the link quality of multiple beams or beam pairs.

[0073] Step 702: The UE uses the first sub-model to determine at least one receive beam index and / or uplink feedback information based on the link quality of the multiple beams or beam pairs; wherein the uplink feedback information is used to determine at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index.

[0074] In some optional implementations, the multiple beams measured by the UE are a subset of all beams. Here, all beams refer to all transmit beams or all receive beams, that is, all transmit beams of the network device or all receive beams of the UE.

[0075] In some optional implementations, the beam pairs measured by the UE are a subset of all beam pairs. Here, all beam pairs refer to all transmit beam and receive beam pairs, that is, transmit beam and receive beam pairs formed by all possible combinations of all receive beams of the UE and all transmit beams of the network device.

[0076] In the embodiment of the present application, the link quality of a beam refers to the link quality corresponding to the transmit beam. The link quality of a beam pair refers to the link quality corresponding to the transmit beam and the receive beam pair.

[0077] In the embodiment of the present application, the measurement of a beam or a beam pair refers to the measurement of a reference signal associated with the beam or the beam pair.

[0078] In some optional implementations, the reference signal includes at least one of the following: a synchronization signal block (Synchronization Signal and PBCH Block, SSB), a channel state information-reference signal (Channel State Information-Reference Signal, CSI-RS).

[0079] In some optional embodiments, the link quality includes at least one of the following: Layer 1-Reference Signal Receiving Power (L1-RSRP), Layer 1-Signal to Interference plus Noise Ratio (L1-SINR), and Layer 1-Reference Signal Receiving Quality (L1-RSRQ).

[0080] In an embodiment of the present application, after measuring the link quality of multiple beams or beam pairs, the UE uses a first submodel to determine at least one receive beam index and / or uplink feedback information based on the link quality of the multiple beams or beam pairs. Here, the multiple beams are multiple receive beams or multiple transmit beams, and the multiple beam pairs are multiple transmit beams and receive beam pairs.

[0081] As an implementation method, after measuring the link quality of multiple beams (here referring to multiple transmit beams), the UE uses the first sub-model to determine uplink feedback information based on the link quality of the multiple beams.

[0082] As another implementation method, after measuring the link quality of multiple beams (here referring to multiple receiving beams), the UE uses the first sub-model to determine at least one receiving beam index based on the link quality of the multiple beams.

[0083] As another implementation method, after the UE measures the link quality of multiple beam pairs (here referring to multiple pairs of transmit beams and receive beams), it uses the first sub-model to determine at least one receive beam index and uplink feedback information based on the link quality of the multiple beam pairs.

[0084] In some optional embodiments, when the UE determines the uplink feedback information, the UE sends the uplink feedback information to the network device. Accordingly, the network device receives the uplink feedback information sent by the UE, and the network device uses the second sub-model to determine at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index based on the uplink feedback information.

[0085] Here, the uplink feedback information is used by the network device to determine at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index using the second sub-model.

[0086] In the above solution, the at least one receive beam index is the index of the K receive beams with the best link quality among the multiple beams or beam pairs, where K is a positive integer. Here, the at least one receive beam index can be recorded as K optimal receive beam indices.

[0087] In the above solution, the at least one transmit beam index is the index of the K transmit beams with the best link quality among the multiple beams or beam pairs, where K is a positive integer. Here, the at least one transmit beam index can be recorded as K optimal transmit beam indices.

[0088] Here, the value of K may be configured by a network device or may be predefined or determined based on the UE's own implementation.

[0089] In the embodiments of the present application, since link quality refers to the link quality of a beam or beam pair, there is a corresponding relationship between the transmit beam and the link quality, or between the receive beam and the link quality, or between the transmit beam and receive beam pair and the link quality. Based on this, in the above solution, there is a corresponding relationship between any two of the at least one receive beam index, the at least one transmit beam index, and the at least one link quality.

[0090] In some optional implementations, the correspondence is determined by the sorting of the at least one receive beam index, the sorting of the at least one transmit beam index, and the sorting of the at least one link quality.

[0091] As an example, the first submodel on the UE side outputs K optimal receive beam indices, and the second submodel on the network device side outputs K optimal transmit beam indices and K link qualities. The K optimal receive beam indices are denoted as {n1, n2, ... nK}, the K optimal transmit beam indices are denoted as {m1, m2, ... mK}, and the K link qualities are denoted as {r1, r2, ... rK}. Optionally, the link qualities can be sorted from high to low, i.e., {r1>r2> ...>rK}. The K optimal receive beam indices, the K optimal transmit beam indices, and the K link qualities have a one-to-one correspondence. This correspondence is implicitly indicated by the order of the K optimal receive beam indices, the order of the K optimal transmit beam indices, and the order of the K link qualities. That is, the optimal receive beam indices, the optimal transmit beam indices, and the link qualities sorted at the same position correspond to each other. In specific implementation, the K optimal receive beam indices output by the first sub-model on the UE side are sorted according to the order of the corresponding link qualities (e.g., from high to low); and the K optimal transmit beam indices output by the second sub-model on the network device side are sorted according to the order of the corresponding link qualities (e.g., from high to low). In this way, the K optimal receive beam indices, the K optimal transmit beam indices, and the K link qualities can be represented as a set of triples according to the corresponding relationship, denoted as {(m1,n1,r1),(m2,n2,r2),…,(mK,nK,rK)}. It should be noted that there is no mutual exclusion relationship between the elements in {n1,n2,…nK}, that is, the optimal receiving beam indices in the K optimal receiving beam indices can be the same or different; similarly, there is no mutual exclusion relationship between the elements in {m1,m2,…mK}, that is, the optimal transmitting beam indices in the K optimal transmitting beam indices can be the same or different; and there is a mutual exclusion relationship between the elements in {(m1,n1),(m2,n2),…,(mK,nK)}, that is, the optimal beam pairs in the K optimal beam pairs are different.

[0092] In the embodiment of the present application, the input of the first sub-model on the UE side is the link quality of multiple beams or beam pairs measured by the UE. The output of the first sub-model on the UE side includes at least one receive beam index and / or uplink feedback information.

[0093] In the case where the output of the first sub-model on the UE side is at least one receiving beam index, the first sub-model has an output layer, and the UE outputs at least one receiving beam index through the output layer.

[0094] In the case where the output of the first sub-model on the UE side is uplink feedback information, the first sub-model has an output layer, the UE outputs the first information through the output layer, and the first information is used to determine the uplink feedback information. Furthermore, the first sub-model also includes a quantization layer, and the UE quantizes the first information through the quantization layer to obtain the uplink feedback information.

[0095] For the case where the output of the first sub-model on the UE side includes at least one receiving beam index and uplink feedback information, the output layer of the first sub-model can be implemented as follows.

[0096] Mode A: The first sub-model includes an output layer; the UE processes the link quality of the multiple beams or beam pairs using the first sub-model, outputs at least one receive beam index through a first portion of nodes in the output layer, and outputs first information through a second portion of nodes in the output layer, where the first information is used to determine uplink feedback information. Furthermore, the first sub-model also includes a quantization layer, where the UE quantizes the first information through the quantization layer to obtain the uplink feedback information.

[0097] Mode B: The first submodel includes a first output layer and a second output layer; the UE processes the link quality of the multiple beams or beam pairs using the first submodel, outputs at least one receive beam index through the first output layer, and outputs first information through the second output layer, where the first information is used to determine uplink feedback information. Furthermore, the first submodel also includes a quantization layer, and the UE quantizes the first information through the quantization layer to obtain the uplink feedback information.

[0098] Here, in the first sub-model, the first output layer is adjacent to the second output layer, or the first output layer is separated from the second output layer by at least one neural network layer.

[0099] Here, in the first sub-model, the first output layer is located before the second output layer, or the first output layer is located after the second output layer.

[0100] Option B-1) In some optional embodiments, the first output layer is located before the second output layer, and the second output layer is the last neural network layer of the first sub-model; the first output layer is the middle neural network layer of the first sub-model.

[0101] Here, the first output layer can be all or part of a middle neural network layer of the first sub-model. Taking a deep neural network as an example, the middle neural network layer of the first sub-model refers to a hidden layer.

[0102] Option B-2) In some optional embodiments, when the first output layer is located after the second output layer, the first output layer is the last neural network layer of the first sub-model; and the second output layer is the middle neural network layer of the first sub-model.

[0103] Here, the second output layer can be all or part of a middle neural network layer of the first sub-model. Taking a deep neural network as an example, the middle neural network layer of the first sub-model refers to a hidden layer.

[0104] Option B-3) In some optional embodiments, when the first output layer is located after the second output layer, the first output layer is the last neural network layer of the first sub-model; and the second output layer is the first neural network layer of the first sub-model.

[0105] Here, the second output layer may be the entire first neural network layer. Taking a deep neural network as an example, the first neural network layer of the first sub-model refers to an input layer, and the input of the first neural network layer is the link quality of the multiple beams or beam pairs.

[0106] In the above solution, there is an association between the first output layer and the second output layer, or there is no association between the first output layer and the second output layer.

[0107] Here, the meaning of the association between the first output layer and the second output layer is that there is an association between the nodes of the first output layer and the nodes of the second output layer. Here, the association between the nodes of the first output layer and the nodes of the second output layer means that the output of the nodes of the first output layer is associated with the input of the nodes of the second output layer (this is for the case where the first output layer is located before the second output layer), or the output of the nodes of the second output layer is associated with the input of the nodes of the first output layer (this is for the case where the second output layer is located before the first output layer). It should be pointed out that this association can be a direct association (corresponding to the case where the two layers are adjacent) or an indirect association (corresponding to the case where the two layers are not adjacent). In addition, all or part of the nodes of the first output layer can be associated with all or part of the nodes of the second output layer.

[0108] In the embodiment of the present application, the uplink feedback information obtained by the UE after quantization by the quantization layer may be a bit sequence.

[0109] In some optional embodiments, for the above-mentioned scheme A, scheme B-1) and scheme B-2), the uplink feedback information is a first bit sequence, and the first bit sequence is used to determine at least one transmit beam index and at least one link quality corresponding to the at least one transmit beam index.

[0110] In some optional embodiments, for the above-mentioned solution B-3), the uplink feedback information is a second bit sequence, and the second bit sequence is used to indicate the link quality of the multiple beams or beam pairs, and the link quality of the multiple beams or beam pairs is used to determine at least one transmit beam index and at least one link quality corresponding to the at least one transmit beam index.

[0111] In some optional embodiments, when the at least one receiving beam index is one receiving beam index, the UE uses the receiving beam indicated by the one receiving beam index to receive downlink transmission.

[0112] In some optional implementations, when the at least one transmit beam index is one transmit beam index, the network device uses the transmit beam indicated by the one transmit beam index to send downlink transmission.

[0113] In some optional embodiments, when the at least one receiving beam index is a plurality of receiving beam indexes, the UE determines a first receiving beam index from the plurality of receiving beam indexes, and uses the receiving beam indicated by the first receiving beam index to receive downlink transmission.

[0114] Here, the UE receives first indication information sent by a network device, where the first indication information is used to indicate a first transmit beam index used by the network device; and the UE determines a first receive beam index corresponding to the first transmit beam index based on a correspondence between the at least one receive beam index and the at least one transmit beam index. Optionally, the first indication information is carried in a MAC CE or DCI.

[0115] In some optional embodiments, when the at least one transmit beam index is a plurality of transmit beam indexes, the network device selects a first transmit beam index from the plurality of transmit beam indexes, and uses the transmit beam indicated by the first transmit beam index to send downlink transmission.

[0116] Here, the network device sends first indication information to the UE, where the first indication information is used to indicate a first transmit beam index used by the network device. Optionally, the first indication information is carried in a MAC CE or a DCI.

[0117] The first sub-model in the above solution is pre-trained, and the first sub-model can be trained by the network device, and the trained first sub-model is sent to the UE.

[0118] Similarly, the second sub-model in the above solution is also pre-trained, and the second sub-model can be trained through the network device.

[0119] Of course, the first sub-model and / or the second sub-model are not limited to being trained by the network device, and the first sub-model and / or the second sub-model can also be trained by other devices with processing capabilities.

[0120] In some optional embodiments, the first sub-model and / or the second sub-model are trained by a network device, including: the UE reports a data set to the network device; the network device receives the data set reported by the UE, and the network device uses the data set to train the first sub-model and / or the second sub-model, where the first sub-model is used by the UE to determine at least one receive beam index and / or uplink feedback information based on the link quality of multiple beams or beam pairs, and the second sub-model is used by the network device to determine at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index based on the uplink feedback information.

[0121] Here, the data set reported by the UE to the network device is used by the network device to train the first sub-model and / or the second sub-model.

[0122] As an implementation method, when the data set is used by a network device to train the first sub-model or the first sub-model plus the second sub-model (or the network uses the data set to train the first sub-model or the first sub-model plus the second sub-model), the network device sends the trained first sub-model to the UE; the UE receives the trained first sub-model sent by the network device.

[0123] In some optional embodiments, the data set includes:

[0124] Input data of the first sub-model, the input data being link quality of at least some beams or beam pairs in the first set of beams or beam pairs;

[0125] a label of the output data of the first sub-model, the label being the indexes of K receiving beams with the best link quality in the first beam or beam pair set;

[0126] The label of the output data of the second sub-model is the index of the K transmitting beams with the best link quality in the first beam or beam pair set and the corresponding K link qualities, where K is a positive integer.

[0127] Here, the first beam or beam pair set refers to the first beam set or the first beam pair set. Here, the first beam set includes all transmit beams or all receive beams, i.e., all transmit beams of a network device or all receive beams of a UE. The first beam pair set includes all transmit beams and receive beam pairs, i.e., transmit beams and receive beam pairs formed by all possible combinations of all receive beams of a UE and all transmit beams of a network device.

[0128] In an embodiment of the present application, the first sub-model and the second sub-model are trained jointly; or, the first sub-model and the second sub-model are trained independently.

[0129] Here, in the process of jointly training the first sub-model and the second sub-model, the first sub-model and the second sub-model share a loss function, which can be understood as the joint loss function of the first sub-model and the second sub-model, and the weights of the two sub-models are jointly optimized through the joint loss function. Specifically, the loss value between the output data and the label of the entire model is calculated by the joint loss function, and it is judged whether the loss value meets the preset conditions. If not, the first sub-model and the second sub-model are updated according to the loss value (including but not limited to updating the weights of the first sub-model and the second sub-model), and the above steps are continuously iterated until the loss value meets the preset conditions, and the joint training of the first sub-model and the second sub-model is completed. Here, the output data of the entire model includes the indexes of K receiving beams, the indexes of K transmitting beams, and the corresponding K link qualities.

[0130] Here, during the training of the first sub-model and the independent training of the second sub-model, the first sub-model and the second sub-model have independent loss functions, and the weights of the two sub-models are optimized respectively by the two loss functions. Specifically, the loss value between the output data of the first sub-model and the label of the output data of the first sub-model is calculated by the first loss function, and it is judged whether the loss value meets the preset conditions. If not, the first sub-model is updated according to the loss value (including but not limited to updating the weight of the first sub-model), and the above steps are continuously iterated until the loss value meets the preset conditions, and the training of the first sub-model is completed. Similarly, the loss value between the output data of the second sub-model and the label of the output data of the second sub-model is calculated by the second loss function, and it is judged whether the loss value meets the preset conditions. If not, the second sub-model is updated according to the loss value (including but not limited to updating the weight of the second sub-model), and the above steps are continuously iterated until the loss value meets the preset conditions, and the training of the second sub-model is completed. Here, the output data of the first sub-model includes the indexes of K receive beams, and the output data of the second sub-model includes the indexes of K transmit beams and the corresponding K link qualities.

[0131] In some optional implementations, for the above-mentioned solutions A, B-1), and B-2), since the input of the second sub-model (i.e., uplink feedback information) is obtained by the first sub-model after processing the link quality of multiple beams or beam pairs (here, processing other than the quantization layer), the second sub-model and the first sub-model need to be trained in pairs, that is, the first sub-model and the second sub-model need to be trained jointly. In addition, in this case, the first sub-model and the second sub-model also need to be deployed in pairs.

[0132] In some optional implementations, for the above solution B-3), since the input of the second sub-model (i.e., uplink feedback information) does not need to be obtained by the first sub-model after processing the link quality of multiple beams or beam pairs (here, processing other than the quantization layer), the second sub-model and the first sub-model do not need to be paired for training, and the first sub-model and the second sub-model can be trained independently. In addition, in this case, the first sub-model and the second sub-model can also be deployed independently.

[0133] The technical solution of the embodiment of the present application realizes AI / ML-based beam management, and the trained neural network is deployed on the network device and / or UE side respectively. The beam management process includes at least one of the following: UE measurement process, UE reasoning (ie, prediction) process, UE reporting and network device reasoning (ie, prediction) process. For the UE measurement process, the UE measures all beams or some beams or beam pairs in the beam pair; for the UE reasoning (ie, prediction) process, the UE predicts the optimal receiving beam index and / or uplink feedback information based on the measurement results through the local neural network (ie, the first sub-model); for the UE reporting and network device reasoning (ie, prediction) process, the UE reports the uplink feedback information to the network device, and the network device predicts the optimal transmit beam and the corresponding link quality based on the uplink feedback information through the local neural network (ie, the second sub-model). Through this solution, the network device and the UE can select a beam or beam pair with better channel quality, and reduce a large amount of overhead and delay generated during the beam scanning process.

[0134] The following is an explanation of all or part of the above technical solutions in the embodiments of the present application with reference to specific examples.

[0135] It should be noted that the following examples are described using a "beam pair" as an example, where a beam pair refers to a pair of a transmit beam and a receive beam.

[0136] It should be noted that the following examples are described using the link quality "L1-RSRP" as an example, but are certainly not limited thereto. The link quality may also be L1-SINR or L1-RSRQ.

[0137] It should be noted that the network (NW) in the following examples corresponds to the “network device” in the above solution.

[0138] Example 1

[0139] In this example, the model is deployed separately on the UE side and the NW side. As shown in Figure 8, the model (the part contained in the dotted box) includes a UE-side sub-model (i.e., the first sub-model) and an NW-side sub-model (i.e., the second sub-model), wherein the input of the UE-side sub-model is the L1-RSRP of the beam pair measured by the UE, and the measured beam pair is a subset of all beam pairs. Here, all beam pairs refer to all transmit beams and receive beam pairs, that is, all possible combinations of all receive beams of the UE and all transmit beams of the NW to form transmit beams and receive beam pairs. The UE-side sub-model outputs K optimal receive beam indices, and the UE side outputs uplink feedback information, which can be reported to the NW side via MAC CE (corresponding to PUSCH) or DCI (corresponding to PDCCH). The input of the NW-side sub-model is uplink feedback information, and the output is K optimal transmit beam indices and the L1-RSRP corresponding to the K optimal transmit beam indices, where K is a positive integer.

[0140] Case 1) K = 1

[0141] The UE-side sub-model outputs one optimal receive beam index, and the NW-side sub-model outputs one optimal transmit beam index and the L1-RSRP corresponding to the optimal transmit beam index.

[0142] In this case, the UE selects the optimal receive beam output by the UE-side sub-model for downlink reception, and the NW selects the optimal transmit beam output by the NW-side sub-model for downlink transmission. Here, the optimal receive beam indicated by the optimal receive beam index output by the UE-side sub-model and the optimal transmit beam indicated by the optimal transmit beam index output by the NW-side sub-model automatically match the optimal beam pair, eliminating the need for the UE and NW to specify the beams to use. The link quality of this optimal beam pair is determined by the L1-RSRP output by the NW-side sub-model.

[0143] Case 2) K > 1

[0144] The UE-side sub-model outputs K (K>1) optimal receive beam indices, and the NW-side sub-model outputs K optimal transmit beam indices and the L1-RSRP corresponding to the K optimal transmit beam indices.

[0145] In this case, the K optimal receive beam indices (represented by n here) output by the UE-side sub-model and the K optimal transmit beam indices (represented by m here) and L1-RSRP (represented by the symbol r here) output by the NW-side sub-model contain an implicit one-to-one pairing relationship, that is, the K beam pairs and the corresponding L1-RSRP can be represented as a triple set {(m1,n1,r1),(m2,n2,r2),…,(mK,nK,rK)}, where {m1,m2,…mK} ​​represent the K optimal transmit beam indices, {n1,n2,…nK} represent the K optimal receive beam indices, and {r1,r2,…,rK} represent the L1-RSRP corresponding to the K optimal transmit beam indices. Optionally, {r1>r2>…>rK}. The pairing relationship (i.e., the corresponding relationship) between the optimal transmit beam index and the optimal receive beam index is implicitly indicated by the order of the beam index sequences output by the UE-side sub-model and the NW-side sub-model, rather than explicitly indicated by the uplink feedback information, that is, the K optimal receive beam indices output by the UE-side sub-model and the K optimal transmit beam indices output by the NW-side sub-model are arranged in descending order of the corresponding L1-RSRP.

[0146] Example 2

[0147] In this example, the structure of the UE-side sub-model (i.e., the first sub-model) can be implemented in a variety of ways. The following are several structures of the UE-side sub-model. It should be noted that this example uses a deep neural network implementation to illustrate the structure of the UE-side sub-model, but is not limited to this. Other implementation methods such as convolutional neural networks and recurrent neural networks can also implement the structure of the UE-side sub-model. The implementation method of other neural networks can be expanded based on this example and using the same input / output interface.

[0148] As an implementation, as shown in Figure 9-1, the UE-side sub-model includes an input layer, a hidden layer, and an output layer. The first information and the K optimal receive beam indices are output in parallel at the output layer, sharing all hidden layer weights. The input layer is a matrix of the measured L1-RSRPs of M beam pairs arranged in a specified order. The hidden layer consists of multiple fully connected layers. The output layer consists of two parts: the K nodes in the output layer output the K optimal receive beam indices, and the other nodes in the output layer output the first information. After the first information passes through the quantization layer, it outputs uplink feedback information. The uplink feedback information is reported to the NW side and serves as input to the NW-side sub-model. Here, the uplink feedback information is a bit sequence that does not explicitly indicate the K optimal transmit beam indices and the corresponding L1-RSRPs. Instead, it implicitly contains information about the K optimal transmit beam indices and the corresponding L1-RSRPs. This information can be extracted by the matching NW-side sub-model. Therefore, the UE-side sub-model and the NW-side sub-model need to be paired for training and deployment, so that the NW can correctly predict the K optimal transmit beam indices and the L1-RSRP corresponding to the K optimal transmit beam indices from the uplink feedback information.

[0149] As an implementation, as shown in Figure 9-2, the UE-side sub-model includes an input layer, a hidden layer, an uplink feedback information output layer, and an optimal receive beam index output layer. The uplink feedback information output layer and the optimal receive beam index output layer are not on the same layer and only share some hidden layer weights. The input layer is a matrix of the measured L1-RSRPs of M beam pairs arranged in a specified order. The hidden layer consists of multiple fully connected layers. The uplink feedback information output layer is part of a neural network layer of the hidden layer and precedes the optimal receive beam index output layer. The optimal receive beam index output layer outputs K optimal receive beam indices. The uplink feedback information output layer outputs the first system information, which is then quantized and output as uplink feedback information. After being reported to the NW side, the uplink feedback information serves as the input of the NW-side sub-model. Here, the uplink feedback information is a bit sequence that does not explicitly indicate the K optimal transmit beam indices and the L1-RSRPs corresponding to the K optimal transmit beam indices. Instead, it implicitly contains information about the K optimal transmit beam indices and the L1-RSRPs corresponding to the K optimal transmit beam indices, which can be extracted by the matching NW-side sub-model. Therefore, the UE-side sub-model and the NW-side sub-model need to be paired for training and deployment so that the NW can correctly predict the K optimal transmit beam indices and the L1-RSRPs corresponding to the K optimal transmit beam indices from the uplink feedback information. It should be noted that although the uplink feedback information output layer in Figure 9-2 is part of a neural network layer of the hidden layer, this is not limited to this. The uplink feedback information output layer can also be the entire neural network layer of the hidden layer. Although the uplink feedback information output layer in Figure 9-2 is adjacent to the optimal receive beam index output layer, this is not limited to this. The uplink feedback information output layer can also be separated from the optimal receive beam index output layer by one or more neural network layers. In addition, the output of the uplink feedback information output layer can be used or not used (dashed weight connection in Figure 9-2) to obtain partial implicit information of the K optimal receiving beam indices.

[0150] As an implementation, as shown in Figure 9-3, the UE-side submodel includes an input layer, a hidden layer, an optimal receive beam index output layer, and an uplink feedback information output layer. The optimal receive beam index output layer and the uplink feedback information output layer are not on the same layer and only share some hidden layer weights. The input layer is a matrix of the measured L1-RSRPs of M beam pairs arranged in a specified order. The hidden layer consists of multiple fully connected layers. The optimal receive beam index output layer is part of a neural network layer within the hidden layer and precedes the uplink feedback information output layer. The optimal receive beam index output layer outputs K optimal receive beam indices. The uplink feedback information output layer outputs the first system information, which is then quantized and output as uplink feedback information. After being reported to the NW side, the uplink feedback information serves as the input to the NW-side submodel. Here, the uplink feedback information is a bit sequence that does not explicitly indicate the K optimal transmit beam indices and the L1-RSRPs corresponding to the K optimal transmit beam indices. Instead, it implicitly contains information about the K optimal transmit beam indices and the L1-RSRPs corresponding to the K optimal transmit beam indices, which can be extracted by the matching NW-side sub-model. Therefore, the UE-side sub-model and the NW-side sub-model need to be paired for training and deployment so that the NW can correctly predict the K optimal transmit beam indices and the L1-RSRPs corresponding to the K optimal transmit beam indices from the uplink feedback information. It should be noted that although the optimal receive beam index output layer in Figure 9-3 is part of a neural network layer of the hidden layer, this is not limited to this. The optimal receive beam index output layer can also be the entire neural network layer of the hidden layer. Although the optimal receive beam index output layer in Figure 9-3 is adjacent to the uplink feedback information output layer, this is not limited to this. The optimal receive beam index output layer can also be separated from the uplink feedback information output layer by one or more neural network layers. In addition, the output of the optimal receive beam index output layer can be used or not used (dashed weight connection in Figure 9-3) to obtain part of the implicit information of the uplink feedback information.

[0151] As an implementation, as shown in Figure 9-4, the UE-side sub-model includes an input layer, a hidden layer, and an output layer. The input layer is a matrix of the measured L1-RSRPs of M beam pairs arranged in a specified order. The hidden layer consists of multiple fully connected layers. The output layer outputs K optimal receive beam indices. The L1-RSRPs of the M beam pairs in the input layer are used as the first information and then pass through the quantization layer to output uplink feedback information. After being reported to the network-side network (NW), the uplink feedback information serves as the input of the NW-side sub-model. Because the uplink feedback information is directly derived from the L1-RSRPs of the M beam pairs in the input layer through the quantization layer, it does not pass through any hidden layer for implicit feature extraction. Therefore, the uplink feedback information is a bit sequence that explicitly indicates the measured L1-RSRPs of the M beam pairs and is directly reported to the NW. The NW-side sub-model then derives the K optimal transmit beam indices and the L1-RSRPs corresponding to the K optimal transmit beam indices from the measured L1-RSRPs of the beam pairs. Therefore, the UE-side sub-model and the NW-side sub-model do not require a strict pairing relationship and can be trained and deployed independently. It is important to note that the input of the UE-side sub-model is the directly measured L1-RSRP value of the beam pair, which does not have quantization error; while the input of the NW-side sub-model is obtained from uplink feedback information, which is subject to quantization error.

[0152] Example 3

[0153] In this example, the UE-side sub-model and the NW-side sub-model are trained offline. That is, the sub-models deployed on the UE side and the NW side are models that have been adapted to the current cell or communication scenario.

[0154] During offline training, the dataset required by the model includes the following parts:

[0155] Input: L1-RSRP of some beam pairs among all beam pairs. Here, all beam pairs include all possible combinations of all UE receive beams and all network device transmit beams to form transmit beam and receive beam pairs.

[0156] Label: The indices of the K beam pairs with the highest L1-RSRP among all beam pairs (i.e., the K optimal transmit beam indices and the K optimal receive beam indices) and their corresponding L1-RSRP.

[0157] The model output is divided into two parts: the output of the UE-side sub-model and the output of the NW-side sub-model. The UE-side sub-model outputs the K optimal receive beam indices, while the NW-side sub-model outputs the K optimal transmit beam indices and the corresponding L1-RSRPs.

[0158] It should be noted that if K optimal beam pairs are output, the model can infer the case where there are less than K optimal beam pairs.

[0159] Considering that NW has stronger computing power, offline training tasks are completed on the NW side. Specifically, during the offline training process, the NW side saves the complete structure of the UE side sub-model and the NW side sub-model, and after the training is completed, the trained UE side sub-model (specifically including the structure and parameters of the UE side sub-model) is sent to the UE through the downlink channel. Specifically, the training process is shown in Figure 10 and includes the following steps:

[0160] Step 1001: NW performs downlink beam scanning.

[0161] Here, when performing downlink beam scanning, the NW scans all downlink beams (i.e., all transmit beams of the network device), such as the 64 beams used for SSB / CSI-RS. For each downlink beam (i.e., transmit beam), the UE traverses all receive beams to measure the L1-RSRP, ultimately obtaining the L1-RSRP for all beam pairs. Based on the L1-RSRP of all beam pairs, the UE determines the indices of the K beam pairs with the highest L1-RSRP (i.e., the K optimal transmit beam indices and the K optimal receive beam indices) and the corresponding K L1-RSRPs.

[0162] Step 1002: The UE reports a data set.

[0163] Here, the dataset consists of the following two parts: the first part is the input, that is, the input of the UE-side sub-model, specifically the L1-RSRP of some beam pairs among all measured beam pairs; the second part is the label, that is, the label of the UE-side sub-model and the NW-side sub-model, specifically the indexes of the K optimal transmit beams, the K optimal receive beam indices and the corresponding K L1-RSRPs.

[0164] Step 1003: The NW trains the UE-side sub-model and the NW-side sub-model according to the data set reported by the UE.

[0165] Here, for some UE-side sub-model structures (such as those shown in Figures 9-1 to 9-3), uplink feedback information is processed by the UE-side sub-model and used as input to the NW-side sub-model. Therefore, the UE-side sub-model and the NW-side sub-model must be trained jointly to ensure matching between the UE-side sub-model and the NW-side sub-model. For other UE-side sub-model structures (such as that shown in Figure 9-4), the UE-side sub-model and the NW-side sub-model share the input layer. Therefore, the UE-side sub-model and the NW-side sub-model can be trained separately, without the need for strict model matching.

[0166] Step 1004: The NW sends the UE-side sub-model.

[0167] Here, the NW sends the trained UE-side sub-model to the UE via the downlink channel, completing the model's online deployment. The NW and UE can then implement AI / ML-based beam management according to the aforementioned solutions.

[0168] Example 4

[0169] In this example, based on offline trained and online deployed NW-side and UE-side sub-models, online inference is used to implement AI / ML-based beam management. K optimal receive beam indices, K optimal transmit beam indices, and the corresponding L1-RSRPs for these K optimal transmit beam indices are obtained, where K is a positive integer. Furthermore, this example also provides an AI / ML-based beam indication method, which, through necessary signaling support, assists the NW and UE in collaboratively matching the optimal transmit and receive beams.

[0170] Case 1) K = 1

[0171] The UE-side sub-model outputs one optimal receive beam index, and the NW-side sub-model outputs one optimal transmit beam index and the L1-RSRP corresponding to the optimal transmit beam index.

[0172] In this case, the UE selects the optimal receive beam output by the UE-side sub-model for downlink reception, and the NW selects the optimal transmit beam output by the NW-side sub-model for downlink transmission. Here, the optimal receive beam indicated by the optimal receive beam index output by the UE-side sub-model and the optimal transmit beam indicated by the optimal transmit beam index output by the NW-side sub-model automatically match the optimal beam pair, eliminating the need for the UE and NW to specify the beam to use.

[0173] Case 2) K > 1

[0174] The UE-side sub-model outputs K (K>1) optimal receive beam indices, and the NW-side sub-model outputs K optimal transmit beam indices and the L1-RSRP corresponding to the K optimal transmit beam indices.

[0175] In this case, the NW selects an optimal transmit beam index from the K optimal transmit beam indices, and uses the optimal transmit beam indicated by the optimal transmit beam index to send downlink transmission. The NW indicates the optimal transmit beam index used to the UE. The UE can select the optimal receive beam index that matches the optimal transmit beam index used by the NW based on the correspondence between the optimal transmit beam index and the optimal receive beam index, and then use the optimal receive beam indicated by the optimal receive beam index to receive downlink transmission. Here, the NW can indicate the optimal transmit beam index used by the NW through MAC CE or DCI. For the K optimal receive beam indices, if you want to indicate an optimal receive beam index from them, log2K bits are required for indication. For example, when K=4, 2 bits are required for indication. Specifically, the beam indication process is shown in Figure 11, which includes the following steps:

[0176] Step 1101: NW performs downlink beam scanning.

[0177] Here, when performing downlink beam scanning, the NW scans all or part of the downlink beams (i.e., all or part of the network device's transmit beams). The UE measures the L1-RSRP of all or part of the downlink beams (i.e., transmit beams) by traversing all or part of their receive beams, ultimately obtaining the L1-RSRP of some of the beam pairs.

[0178] Step 1102: UE-side sub-model inference.

[0179] Here, the UE infers (i.e., predicts) K optimal receiving beam indices and uplink feedback information based on the L1-RSRP of some beam pairs through the UE-side sub-model.

[0180] Step 1103: The UE reports uplink feedback information.

[0181] Step 1104: NW side sub-model inference.

[0182] Here, the NW uses the NW-side sub-model to infer (i.e., predict) K optimal transmit beam indices and their corresponding K L1-RSRPs based on uplink feedback information. If K > 1, the NW selects one of the K optimal transmit beam indices as the transmit beam index for downlink transmission.

[0183] Step 1105: The NW indicates the optimal transmit beam index to the UE.

[0184] Here, the UE determines the corresponding optimal receiving beam index according to the optimal transmitting beam index indicated by the NW, and uses the receiving beam indicated by the optimal receiving beam index to receive downlink transmission.

[0185] 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.

[0186] 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 the order of execution. 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. In addition, in the embodiments of the present application, the terms "downlink," "uplink," and "sidelink" are used to indicate the transmission direction of signals or data, where "downlink" is used to indicate the first direction of transmission of signals or data from a site to a user equipment in a cell, "uplink" is used to indicate the second direction of transmission of signals or data from a user equipment in a cell to a site, and "sidelink" is used to indicate the third direction of transmission of signals or data from user equipment 1 to user equipment 2. For example, "downlink signal" indicates that the transmission direction of the signal is the first direction. In addition, in the embodiments of the present application, the term "and / or" is merely a description of the association relationship between associated objects, indicating that three relationships can exist. Specifically, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0187] FIG12 is a schematic diagram of the structure of a beam management device provided in an embodiment of the present application, which is applied to a user equipment. As shown in FIG12 , the beam management device includes:

[0188] The measuring unit 1201 is configured to measure the link quality of multiple beams or beam pairs;

[0189] Processing unit 1202 is used to determine at least one receive beam index and / or uplink feedback information based on the link quality of the multiple beams or beam pairs using the first sub-model; wherein the uplink feedback information is used to determine at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index.

[0190] In some optional embodiments, the device also includes: a sending unit 1203, used to send the uplink feedback information to the network device, and the uplink feedback information is used by the network device to determine at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index using a second sub-model.

[0191] In some optional embodiments, the multiple beams are multiple receive beams or multiple transmit beams, and the multiple beam pairs are multiple transmit beam and receive beam pairs.

[0192] In some optional implementations, the at least one receive beam index is an index of K receive beams with the best link quality among the multiple beams or beam pairs, where K is a positive integer.

[0193] In some optional implementations, the at least one transmit beam index is the index of K transmit beams with the best link quality among the multiple beams or beam pairs, where K is a positive integer.

[0194] In some optional implementations, there is a correspondence between any two of the at least one receive beam index, the at least one transmit beam index, and the at least one link quality.

[0195] In some optional implementations, the correspondence is determined by the sorting of the at least one receive beam index, the sorting of the at least one transmit beam index, and the sorting of the at least one link quality.

[0196] In some optional embodiments, the first sub-model includes an output layer; the processing unit 1202 is used to process the link quality of the multiple beams or beam pairs using the first sub-model, output at least one receiving beam index through the first part of the nodes of the output layer, and output first information through the second part of the nodes of the output layer, wherein the first information is used to determine uplink feedback information.

[0197] In some optional embodiments, the first sub-model includes a first output layer and a second output layer; the processing unit 1202 is used to use the first sub-model to process the link quality of the multiple beams or beam pairs, output at least one receiving beam index through the first output layer, and output first information through the second output layer, and the first information is used to determine uplink feedback information.

[0198] In some optional embodiments, in the first sub-model, the first output layer is adjacent to the second output layer, or the first output layer is separated from the second output layer by at least one neural network layer.

[0199] In some optional embodiments, in the first sub-model, the first output layer is located before the second output layer, or the first output layer is located after the second output layer.

[0200] In some optional embodiments, when the first output layer is located before the second output layer, the second output layer is the last neural network layer of the first sub-model; and the first output layer is the middle neural network layer of the first sub-model.

[0201] In some optional embodiments, when the first output layer is located after the second output layer, the first output layer is the last neural network layer of the first sub-model; and the second output layer is the middle neural network layer of the first sub-model.

[0202] In some optional embodiments, when the first output layer is located after the second output layer, the first output layer is the last neural network layer of the first sub-model; and the second output layer is the first neural network layer of the first sub-model.

[0203] In some optional embodiments, the input of the first neural network layer is the link quality of the multiple beams or beam pairs.

[0204] In some optional implementations, there is an association between the first output layer and the second output layer, or there is no association between the first output layer and the second output layer.

[0205] In some optional implementations, the first sub-model further includes a quantization layer, and the processing unit 1202 is configured to perform quantization processing on the first information through the quantization layer to obtain the uplink feedback information.

[0206] In some optional implementations, the uplink feedback information is a first bit sequence, and the first bit sequence is used to determine at least one transmit beam index and at least one link quality corresponding to the at least one transmit beam index.

[0207] In some optional embodiments, the uplink feedback information is a second bit sequence, and the second bit sequence is used to indicate the link quality of the multiple beams or beam pairs, and the link quality of the multiple beams or beam pairs is used to determine at least one transmit beam index and at least one link quality corresponding to the at least one transmit beam index.

[0208] In some optional embodiments, the apparatus further includes: a receiving unit 1204 configured to, when the at least one receiving beam index is one receiving beam index, use the receiving beam indicated by the one receiving beam index to receive downlink transmission.

[0209] In some optional embodiments, the device also includes: a receiving unit 1204, which is used to determine a first receiving beam index from the multiple receiving beam indexes when the at least one receiving beam index is a plurality of receiving beam indexes, and use the receiving beam indicated by the first receiving beam index to receive downlink transmission.

[0210] In some optional embodiments, the receiving unit 1204 is used to receive first indication information sent by a network device, where the first indication information is used to indicate a first transmit beam index adopted by the network device; and based on the correspondence between the at least one receive beam index and the at least one transmit beam index, determine a first receive beam index corresponding to the first transmit beam index.

[0211] In some optional implementations, the first indication information is carried in MAC CE or DCI.

[0212] In some optional implementations, the sending unit 1203 is used to report a data set to a network device, where the data set is used by the network device to train the first sub-model; and the receiving unit 1204 is used to receive the trained first sub-model sent by the network device.

[0213] In some optional embodiments, the data set is also used by the network device to train a second sub-model, and the second sub-model is used by the network device to determine at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index based on the uplink feedback information.

[0214] In some optional embodiments, the first sub-model and the second sub-model are trained jointly; or, the first sub-model and the second sub-model are trained independently.

[0215] In some optional embodiments, the data set includes:

[0216] Input data of the first sub-model, the input data being link quality of at least some beams or beam pairs in the first set of beams or beam pairs;

[0217] a label of the output data of the first sub-model, the label being the indexes of K receiving beams with the best link quality in the first beam or beam pair set;

[0218] The label of the output data of the second sub-model is the index of the K transmitting beams with the best link quality in the first beam or beam pair set and the corresponding K link qualities, where K is a positive integer.

[0219] In some optional implementations, the link quality includes at least one of the following: L1-RSRP, L1-SINR, and L1-RSRQ.

[0220] Those skilled in the art should understand that the relevant description of the above-mentioned beam management device in the embodiment of the present application can be understood with reference to the relevant description of the beam management method in the embodiment of the present application.

[0221] FIG13 is a second schematic diagram of the structure of a beam management device provided in an embodiment of the present application, which is applied to a network device. As shown in FIG13 , the beam management device includes:

[0222] The receiving unit 1301 is configured to receive uplink feedback information sent by the UE;

[0223] The processing unit 1302 is configured to determine at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index based on the uplink feedback information using a second submodel.

[0224] In some optional implementations, the uplink feedback information is a first bit sequence, and the first bit sequence is used to determine at least one transmit beam index and at least one link quality corresponding to the at least one transmit beam index.

[0225] In some optional embodiments, the uplink feedback information is a second bit sequence, which is used to indicate the link quality of multiple beams or beam pairs, and the link quality of the multiple beams or beam pairs is used to determine at least one transmit beam index and at least one link quality corresponding to the at least one transmit beam index.

[0226] In some optional implementations, the apparatus further includes: a sending unit 1303 configured to, when the at least one transmit beam index is one transmit beam index, use the transmit beam indicated by the one transmit beam index to perform downlink transmission.

[0227] In some optional embodiments, the device also includes: a sending unit 1303, which is used to select a first transmission beam index from the multiple transmission beam indexes when the at least one transmission beam index is a plurality of transmission beam indexes, and use the transmission beam indicated by the first transmission beam index to send downlink transmission.

[0228] In some optional implementations, the sending unit 1303 is used to send first indication information to the UE, where the first indication information is used to indicate a first transmit beam index adopted by the network device.

[0229] In some optional implementations, the first indication information is carried in MAC CE or DCI.

[0230] In some optional implementations, the receiving unit 1301 is used to receive a data set reported by the UE; and the processing unit 1302 is used to train the second sub-model using the data set.

[0231] In some optional embodiments, the data set is also used by the network device to train a first sub-model, and the first sub-model is used by the UE to determine at least one receiving beam index and / or uplink feedback information based on the link quality of multiple beams or beam pairs.

[0232] In some optional implementations, the at least one receive beam index is an index of K receive beams with the best link quality among the multiple beams or beam pairs, where K is a positive integer.

[0233] In some optional implementations, the at least one transmit beam index is the index of K transmit beams with the best link quality among the multiple beams or beam pairs, where K is a positive integer.

[0234] In some optional implementations, there is a correspondence between any two of the at least one receive beam index, the at least one transmit beam index, and the at least one link quality.

[0235] In some optional implementations, the correspondence is determined by the sorting of the at least one receive beam index, the sorting of the at least one transmit beam index, and the sorting of the at least one link quality.

[0236] In some optional embodiments, the first sub-model and the second sub-model are trained jointly; or, the first sub-model and the second sub-model are trained independently.

[0237] In some optional implementations, the sending unit 1303 is configured to send the trained first sub-model to the UE.

[0238] In some optional embodiments, the data set includes:

[0239] Input data of the first sub-model, the input data being link quality of at least some beams or beam pairs in the first set of beams or beam pairs;

[0240] a label of the output data of the first sub-model, the label being the indexes of K receiving beams with the best link quality in the first beam or beam pair set;

[0241] The label of the output data of the second sub-model is the index of the K transmitting beams with the best link quality in the first beam or beam pair set and the corresponding K link qualities, where K is a positive integer.

[0242] In some optional implementations, the link quality includes at least one of the following: L1-RSRP, L1-SINR, and L1-RSRQ.

[0243] Those skilled in the art should understand that the relevant description of the above-mentioned beam management device in the embodiment of the present application can be understood with reference to the relevant description of the beam management method in the embodiment of the present application.

[0244] Figure 14 is a schematic diagram of a communication device 1400 provided in an embodiment of the present application. The communication device can be either a user device or a network device. The communication device 1400 shown in Figure 14 includes a processor 1410, which can call and execute a computer program from a memory to implement the method in the embodiment of the present application.

[0245] Optionally, as shown in FIG14 , the communication device 1400 may further include a memory 1420. The processor 1410 may call and execute a computer program from the memory 1420 to implement the method in the embodiment of the present application.

[0246] The memory 1420 may be a separate device independent of the processor 1410 , or may be integrated into the processor 1410 .

[0247] Optionally, as shown in FIG14 , the communication device 1400 may further include a transceiver 1430 , and the processor 1410 may control the transceiver 1430 to communicate with other devices, specifically, to send information or data to other devices, or to receive information or data sent by other devices.

[0248] The transceiver 1430 may include a transmitter and a receiver. The transceiver 1430 may further include an antenna, and the number of antennas may be one or more.

[0249] Optionally, the communication device 1400 may specifically be a network device in an embodiment of the present application, and the communication device 1400 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.

[0250] Optionally, the communication device 1400 may specifically be a user device of an embodiment of the present application, and the communication device 1400 may implement the corresponding processes implemented by the user device in each method of the embodiment of the present application. For the sake of brevity, they will not be repeated here.

[0251] Figure 15 is a schematic structural diagram of a chip according to an embodiment of the present application. The chip 1500 shown in Figure 15 includes a processor 1510, which can call and run a computer program from a memory to implement the method according to the embodiment of the present application.

[0252] Optionally, as shown in FIG15 , the chip 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.

[0253] The memory 1520 may be a separate device independent of the processor 1510 , or may be integrated into the processor 1510 .

[0254] Optionally, the chip 1500 may further include an input interface 1530. The processor 1510 may control the input interface 1530 to communicate with other devices or chips, and specifically, may obtain information or data sent by other devices or chips.

[0255] Optionally, the chip 1500 may further include an output interface 1540. The processor 1510 may control the output interface 1540 to communicate with other devices or chips, and specifically, may output information or data to other devices or chips.

[0256] 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.

[0257] Optionally, the chip can be applied to the user equipment in the embodiments of the present application, and the chip can implement the corresponding processes implemented by the user equipment in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0258] 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.

[0259] FIG16 is a schematic block diagram of a communication system 1600 provided in an embodiment of the present application. As shown in FIG16 , the communication system 1600 includes a user device 1610 and a network device 1620 .

[0260] Among them, the user equipment 1610 can be used to implement the corresponding functions implemented by the user equipment in the above method, and the network equipment 1620 can be used to implement the corresponding functions implemented by the network equipment in the above method. For the sake of brevity, they are not repeated here.

[0261] 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 implemented as a hardware decoding processor, or can be implemented 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.

[0262] 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.

[0263] 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.

[0264] An embodiment of the present application also provides a computer-readable storage medium for storing a computer program.

[0265] 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.

[0266] Optionally, the computer-readable storage medium can be applied to the user equipment in the embodiments of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the user equipment in the various methods of the embodiments of the present application. For the sake of brevity, they are not repeated here.

[0267] An embodiment of the present application also provides a computer program product, including computer program instructions.

[0268] 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.

[0269] Optionally, the computer program product can be applied to the user equipment in the embodiments of the present application, and the computer program instructions enable the computer to execute the corresponding processes implemented by the user equipment in the various methods of the embodiments of the present application. For the sake of brevity, they are not repeated here.

[0270] The embodiment of the present application also provides a computer program.

[0271] 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.

[0272] Optionally, the computer program can be applied to the user equipment in the embodiments of the present application. When the computer program runs on the computer, the computer executes the corresponding processes implemented by the user equipment in the various methods of the embodiments of the present application. For the sake of brevity, they are not repeated here.

[0273] 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.

[0274] 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.

[0275] 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.

[0276] 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.

[0277] 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.

[0278] 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.

[0279] 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 beam management method, the method comprising: The user equipment UE measures the link quality of multiple beams or beam pairs; The UE uses the first sub-model to determine at least one receive beam index and / or uplink feedback information based on the link quality of the multiple beams or beam pairs; wherein the uplink feedback information is used to determine at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index.

2. The method according to claim 1, wherein When the UE determines the uplink feedback information, the method further includes: The UE sends the uplink feedback information to the network device, where the uplink feedback information is used by the network device to determine at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index using a second sub-model.

3. The method according to claim 1 or 2, wherein: The multiple beams are multiple receive beams or multiple transmit beams, and the multiple beam pairs are multiple transmit beam and receive beam pairs.

4. The method according to claim 3, wherein: The at least one receiving beam index is an index of K receiving beams with the best link quality among the multiple beams or beam pairs, where K is a positive integer.

5. The method according to claim 3, wherein The at least one transmit beam index is an index of K transmit beams with the best link quality among the multiple beams or beam pairs, where K is a positive integer.

6. The method according to any one of claims 1 to 5, wherein There is a corresponding relationship between any two of the at least one receive beam index, the at least one transmit beam index, and the at least one link quality.

7. The method according to claim 6, wherein: The corresponding relationship is determined by the sorting of the at least one receive beam index, the sorting of the at least one transmit beam index, and the sorting of the at least one link quality.

8. The method according to any one of claims 1 to 7, wherein The first sub-model includes an output layer; The UE determines, using the first submodel based on the link qualities of the multiple beams or beam pairs, at least one receive beam index and uplink feedback information, including: The UE processes the link quality of the multiple beams or beam pairs using a first sub-model, outputs at least one receiving beam index through a first part of the nodes of the output layer, and outputs first information through a second part of the nodes of the output layer, where the first information is used to determine uplink feedback information.

9. The method according to any one of claims 1 to 7, wherein The first sub-model includes a first output layer and a second output layer; The UE determines, using the first submodel based on the link qualities of the multiple beams or beam pairs, at least one receive beam index and uplink feedback information, including: The UE processes the link quality of the multiple beams or beam pairs using the first sub-model, outputs at least one receiving beam index through the first output layer, and outputs first information through the second output layer, where the first information is used to determine uplink feedback information.

10. The method according to claim 9, wherein: In the first sub-model, the first output layer is adjacent to the second output layer, or the first output layer is separated from the second output layer by at least one neural network layer.

11. The method according to claim 9 or 10, wherein: In the first sub-model, the first output layer is located before the second output layer, or the first output layer is located after the second output layer.

12. The method according to any one of claims 9 to 11, wherein When the first output layer is located before the second output layer, The second output layer is the last neural network layer of the first sub-model; The first output layer is a middle neural network layer of the first sub-model.

13. The method according to any one of claims 9 to 11, wherein When the first output layer is located after the second output layer, The first output layer is the last neural network layer of the first sub-model; The second output layer is a middle neural network layer of the first sub-model.

14. The method according to any one of claims 9 to 11, wherein When the first output layer is located after the second output layer, The first output layer is the last neural network layer of the first sub-model; The second output layer is the first neural network layer of the first sub-model.

15. The method according to claim 14, wherein The input of the first neural network layer is the link quality of the multiple beams or beam pairs.

16. The method according to any one of claims 9 to 15, wherein There is an association between the first output layer and the second output layer, or there is no association between the first output layer and the second output layer.

17. The method according to any one of claims 8 to 16, wherein The first sub-model further includes a quantization layer, and the method further includes: The UE performs quantization processing on the first information through the quantization layer to obtain the uplink feedback information.

18. The method according to claim 17, wherein The uplink feedback information is a first bit sequence, and the first bit sequence is used to determine at least one transmit beam index and at least one link quality corresponding to the at least one transmit beam index.

19. The method according to claim 17, wherein The uplink feedback information is a second bit sequence, which is used to indicate the link quality of the multiple beams or beam pairs, and the link quality of the multiple beams or beam pairs is used to determine at least one transmit beam index and at least one link quality corresponding to the at least one transmit beam index.

20. The method according to any one of claims 1 to 19, wherein The method further comprises: In the case where the at least one receiving beam index is one receiving beam index, the UE uses the receiving beam indicated by the one receiving beam index to receive downlink transmission.

21. The method according to any one of claims 1 to 19, wherein The method further comprises: In the case where the at least one receiving beam index is multiple receiving beam indexes, the UE determines a first receiving beam index from the multiple receiving beam indexes, and uses the receiving beam indicated by the first receiving beam index to receive downlink transmission.

22. The method according to claim 21, wherein The UE determines, from the multiple receive beam indexes, a first receive beam index, including: The UE receives first indication information sent by a network device, where the first indication information is used to indicate a first transmit beam index used by the network device; The UE determines a first receive beam index corresponding to the first transmit beam index based on a correspondence between the at least one receive beam index and the at least one transmit beam index.

23. The method according to claim 22, wherein The first indication information is carried in a media access control element MAC CE or downlink control information DCI.

24. The method according to any one of claims 1 to 23, wherein Before the UE determines, using the first submodel, at least one receive beam index and / or uplink feedback information based on the link qualities of the multiple beams or beam pairs, the method further includes: The UE reports a data set to the network device, where the data set is used by the network device to train the first sub-model; The UE receives the trained first sub-model sent by the network device.

25. The method according to claim 24, wherein The data set is also used by the network device to train a second sub-model, and the second sub-model is used by the network device to determine at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index based on the uplink feedback information.

26. The method according to claim 25, wherein The first sub-model and the second sub-model are trained jointly; or, The first sub-model and the second sub-model are trained independently.

27. The method according to any one of claims 24 to 26, wherein The dataset includes: Input data of the first sub-model, the input data being link quality of at least some beams or beam pairs in the first set of beams or beam pairs; a label of the output data of the first sub-model, the label being the indexes of K receiving beams with the best link quality in the first beam or beam pair set; The label of the output data of the second sub-model is the index of the K transmitting beams with the best link quality in the first beam or beam pair set and the corresponding K link qualities, where K is a positive integer.

28. The method according to any one of claims 1 to 27, wherein The link quality includes at least one of the following: layer 1-reference signal received power L1-RSRP, layer 1-signal to interference plus noise ratio L1-SINR, layer 1-reference signal received quality L1-RSRQ.

29. A beam management method, the method comprising: The network device receives uplink feedback information sent by the UE; The network device determines at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index based on the uplink feedback information using the second sub-model.

30. The method according to claim 29, wherein The uplink feedback information is a first bit sequence, and the first bit sequence is used to determine at least one transmit beam index and at least one link quality corresponding to the at least one transmit beam index.

31. The method according to claim 29, wherein The uplink feedback information is a second bit sequence, which is used to indicate the link quality of multiple beams or beam pairs. The link quality of the multiple beams or beam pairs is used to determine at least one transmit beam index and at least one link quality corresponding to the at least one transmit beam index.

32. The method according to any one of claims 29 to 31, wherein The method further comprises: In the case where the at least one transmit beam index is one transmit beam index, the network device uses the transmit beam indicated by the one transmit beam index to send downlink transmission.

33. The method according to any one of claims 29 to 31, wherein The method further comprises: In a case where the at least one transmit beam index is a plurality of transmit beam indexes, the network device selects a first transmit beam index from the plurality of transmit beam indexes, and uses the transmit beam indicated by the first transmit beam index to send downlink transmission.

34. The method according to claim 33, wherein The method further comprises: The network device sends first indication information to the UE, where the first indication information is used to indicate a first transmit beam index adopted by the network device.

35. The method according to claim 34, wherein The first indication information is carried in MAC CE or DCI.

36. The method of claim 34, wherein: Before the network device receives the uplink feedback information sent by the UE, the method further includes: The network device receives the data set reported by the UE; The network device trains the second sub-model using the data set.

37. The method according to claim 36, wherein The data set is also used by the network device to train a first sub-model, and the first sub-model is used by the UE to determine at least one receiving beam index and / or uplink feedback information based on the link quality of multiple beams or beam pairs.

38. The method of claim 37, wherein: The at least one receiving beam index is an index of K receiving beams with the best link quality among the multiple beams or beam pairs, where K is a positive integer.

39. The method according to claim 37, wherein The at least one transmit beam index is an index of K transmit beams with the best link quality among the multiple beams or beam pairs, where K is a positive integer.

40. The method according to any one of claims 37 to 39, wherein There is a corresponding relationship between any two of the at least one receive beam index, the at least one transmit beam index, and the at least one link quality.

41. The method according to claim 40, wherein The corresponding relationship is determined by the sorting of the at least one receive beam index, the sorting of the at least one transmit beam index, and the sorting of the at least one link quality.

42. The method according to any one of claims 37 to 41, wherein The first sub-model and the second sub-model are trained jointly; or, The first sub-model and the second sub-model are trained independently.

43. The method according to any one of claims 37 to 42, wherein The method further comprises: The network device sends the trained first sub-model to the UE.

44. The method according to any one of claims 36 to 43, wherein The dataset includes: Input data of the first sub-model, the input data being link quality of at least some beams or beam pairs in the first set of beams or beam pairs; a label of the output data of the first sub-model, the label being the indexes of K receiving beams with the best link quality in the first beam or beam pair set; The label of the output data of the second sub-model is the index of the K transmitting beams with the best link quality in the first beam or beam pair set and the corresponding K link qualities, where K is a positive integer.

45. The method according to any one of claims 29 to 44, wherein The link quality includes at least one of the following: L1-RSRP, L1-SINR, and L1-RSRQ.

46. ​​A beam management device, applied to a UE, the device comprising: a measurement unit, configured to measure link qualities of a plurality of beams or beam pairs; A processing unit, used to determine at least one receive beam index and / or uplink feedback information based on the link quality of the multiple beams or beam pairs using a first sub-model; wherein the uplink feedback information is used to determine at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index.

47. A beam management device, applied to a network device, comprising: A receiving unit, configured to receive uplink feedback information sent by the UE; A processing unit is configured to determine, using a second sub-model based on the uplink feedback information, at least one transmit beam index and / or at least one link quality corresponding to the at least one transmit beam index.

48. A user equipment comprising: A processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory to execute the method according to any one of claims 1 to 28.

49. A network device comprising: A processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory to execute the method according to any one of claims 29 to 45.

50. A chip, comprising: A processor, configured to call and execute a computer program from a memory, so that a device equipped with the chip executes the method according to any one of claims 1 to 28.

51. A chip, comprising: A processor, configured to call and execute a computer program from a memory, so that a device equipped with the chip executes the method according to any one of claims 29 to 45.

52. A computer-readable storage medium for storing a computer program, wherein the computer program causes a computer to execute the method according to any one of claims 1 to 28.

53. A computer-readable storage medium for storing a computer program, wherein the computer program causes a computer to execute the method according to any one of claims 29 to 45.

54. A computer program product comprising computer program instructions for causing a computer to perform the method of any one of claims 1 to 28.

55. A computer program product comprising computer program instructions for causing a computer to perform the method of any one of claims 29 to 45.

56. A computer program causing a computer to execute the method of any one of claims 1 to 28.

57. A computer program causing a computer to perform the method of any one of claims 29 to 45.

Images