Communication method, communication device, electronic device and storage medium

Abstract

The embodiments of the present disclosure provide a communication method, a communication device, an electronic device, and a storage medium, and relate to the field of wireless communication technology. The method includes: the first base station CU sends a first model request message to the OAM node, and receives a first request response message sent by the OAM node; the first base station CU obtains the first model training data according to the first request response message, and sends it to the OAM node, so that the OAM node performs model training according to the first model training data and the second model training data sent by the second base station CU, obtains a trained spectrum sharing model, and sends a first model indication message carrying the parameters of the spectrum sharing model to the first base station CU; the first base station CU obtains the spectrum sharing model according to the first model indication message. The method trains the spectrum sharing model through information interaction between the base station and the operation and maintenance management node, and can use the spectrum sharing model to realize spectrum sharing between base stations.

Description

Technical Field

[0001] The present disclosure relates to the field of wireless communication technology, and in particular, to a communication method, a communication device, an electronic device, and a computer-readable storage medium. Background Art

[0002] Spectrum sharing technology is an effective technology that can improve both spectrum efficiency and energy efficiency. Spectrum sharing can dynamically utilize idle spectrum without affecting the spectrum usage rights of primary users. Existing spectrum sharing technologies focus on planned allocation and passive response, with low management efficiency and poor timeliness, and are difficult to meet the explosive growth of future frequency demand and the rapid changes in the electromagnetic environment.

[0003] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0004] The embodiments of the present disclosure provide a communication method, a communication device, an electronic device and a computer-readable storage medium. By training a spectrum sharing model through information interaction between a base station and an operation and maintenance management node, the spectrum sharing model can be used to realize spectrum sharing between base stations, thereby solving the problems existing in the existing spectrum sharing technology to at least a certain extent.

[0005] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0006] According to one aspect of the present disclosure, a communication method is provided, the method comprising: a first base station centralized unit CU sends a first model request message to an operation maintenance management OAM node, and receives a first request response message sent by the OAM node in response to the first model request message; the first base station CU obtains first model training data according to the first request response message, and sends the first model training data to the OAM node, so that the OAM node performs model training according to the first model training data and the second model training data to obtain a trained spectrum sharing model, and sends a first model indication message carrying parameters of the spectrum sharing model to the first base station CU; wherein the first base station CU and the second base station CU belong to base station CUs within a target spectrum optimization range, and the second model training data is sent by the second base station CU to the OAM node; the first base station CU obtains the spectrum sharing model according to the first model indication message.

[0007] In some embodiments of the present disclosure, the first request response message includes: training request indication information and training data provider list information, and the training request indication information includes: model feature input variable information; wherein, the first base station CU obtains the first model training data according to the first request response message, and sends the first model training data to the OAM node, including: the first base station CU obtains the training feature data of the first base station and the training feature data of the neighboring base stations of the first base station according to the model feature input variable information and the training data provider list information; the first base station CU generates measurement configuration information according to the model feature input variable information, and sends the measurement configuration information to the first terminal device through the first base station distribution unit DU, so that the first terminal device obtains the training feature data of the first terminal device according to the measurement configuration information, and sends it to the first base station CU; wherein the first terminal device is within the coverage of the first base station, or the first terminal device is in the public land mobile network PLMN list managed by the first base station; the first base station CU sends the training feature data of the first base station, the training feature data of the neighboring base stations of the first base station and the training feature data of the first terminal device to the OAM node.

[0008] In some embodiments of the present disclosure, the first base station centralized unit CU sends a first model request message to an operation and maintenance management OAM node, and receives a first request response message sent by the OAM node in response to the first model request message, including: the first base station CU sends the first model request message to the OAM node, so that the OAM node determines the target spectrum optimization range according to the first model request message, sets the first base station as an anchor base station, and sends the first request response message to the first base station CU; the first base station CU receives the first request response message sent by the OAM node.

[0009] In some embodiments of the present disclosure, the first request response message includes: anchor base station indication information; wherein, after the first base station CU receives the first request response message sent by the OAM node, the method further includes: the first base station CU determines that the first base station is the anchor base station according to the anchor base station indication information.

[0010] In some embodiments of the present disclosure, the second model training data is acquired by the second base station CU according to the second request response message sent by the OAM node, and the second request response message is sent by the OAM node to the second base station CU in response to the second model request message sent by the second base station CU; wherein the second model training data includes: training feature data of the second base station, training feature data of neighboring base stations of the second base station, and training feature data of the second terminal device, and the second terminal device is within the coverage range of the second base station, or the second terminal device is in the PLMN list managed by the second base station.

[0011] In some embodiments of the present disclosure, the second model request message includes: identification information of the second base station CU, and the identification information of the second base station CU is used by the OAM node to set the second base station as an auxiliary base station; and, the second request response message includes: auxiliary base station indication information and anchor base station information, and the auxiliary base station indication information is used by the second base station CU to determine that the second base station is the auxiliary base station, and the anchor base station information is used by the second base station CU to determine the anchor base station.

[0012] In some embodiments of the present disclosure, the second base station CU may obtain the spectrum sharing model according to a second model indication message sent by the OAM node and carrying parameters of the spectrum sharing model.

[0013] In some embodiments of the present disclosure, the first model indication message includes: a list of auxiliary base stations; wherein the method further includes: the first base station CU determines the second base station CU based on the list of auxiliary base stations; the first base station CU performs model reasoning through the second base station CU based on the spectrum sharing model to obtain a model reasoning result, so that the first base station and the second base station achieve spectrum sharing communication according to the model reasoning result.

[0014] In some embodiments of the present disclosure, the first base station CU performs model reasoning through the second base station CU based on the spectrum sharing model to obtain a model reasoning result, including: the first base station CU acquires first model reasoning data; wherein the first model reasoning data includes: target feature data of the first base station, target feature data of neighboring base stations of the first base station and target feature data of the first terminal device; the first base station CU sends a data request message to the second base station CU, and receives second model reasoning data sent by the second base station CU; wherein the second model reasoning data includes: target feature data of the second base station, target feature data of neighboring base stations of the second base station and target feature data of the second terminal device; the first base station CU performs model reasoning according to the first model reasoning data and the second model reasoning data based on the spectrum sharing model to obtain a model reasoning result; the first base station CU realizes spectrum sharing communication with the second base station CU according to the model reasoning result.

[0015] In some embodiments of the present disclosure, after the first base station CU sends a data request message to the second base station CU, the method further includes: the first base station CU receiving a data request reply message sent by the second base station CU in response to the data request message.

[0016] In some embodiments of the present disclosure, the data request message includes the address information of the first base station CU, and the data request reply message includes the address information of the second base station CU; wherein, the method also includes: the first base station CU establishes a user plane channel with the second base station CU according to the address information of the first base station CU and the address information of the second base station CU, so as to transmit data through the user plane channel.

[0017] In some embodiments of the present disclosure, the data request reply message includes a data request failure and a request retransmission time; wherein, after the first base station CU receives the data request reply message sent by the second base station CU in response to the data request message, the method further includes: the first base station CU parses the data request reply message, determines that the data request fails, and sends the data request message to the second base station CU according to the request retransmission time.

[0018] In some embodiments of the present disclosure, the spectrum sharing model is an artificial intelligence / machine learning model.

[0019] According to another aspect of the present disclosure, a communication method is provided, the method comprising: an OAM node receives a first model request message sent by a first base station CU, and sends a first request response message to the first base station CU in response to the first model request message, so that the first base station CU obtains first model training data according to the first request response message, and sends the first model training data to the OAM node; the OAM node receives a second model request message sent by a second base station CU, and sends a second request response message to the second base station CU in response to the second model request message, so that the second base station CU obtains second model training data according to the second request response message, and sends the second model training data to the OAM node; the OAM node performs model training according to the first model training data and the second model training data to obtain a trained spectrum sharing model, and sends a model indication message carrying parameters of the spectrum sharing model to the first base station CU and the second base station CU, so that the first base station CU and the second base station CU obtain the spectrum sharing model.

[0020] According to another aspect of the present disclosure, a communication device is provided, which is applied to a first base station CU, and the device includes: a first message transceiver unit, used to send a first model request message to an OAM node, and receive a first request response message sent by the OAM node in response to the first model request message; a data acquisition unit, used to acquire first model training data according to the first request response message, and send the first model training data to the OAM node, so that the OAM node performs model training according to the first model training data and the second model training data, obtains a trained spectrum sharing model, and sends a first model indication message carrying parameters of the spectrum sharing model to the first base station CU; wherein the first base station CU and the second base station CU belong to base station CUs within a target spectrum optimization range, and the second model training data is sent by the second base station CU to the OAM node; the model acquisition unit is used to acquire the spectrum sharing model according to the first model indication message.

[0021] In some embodiments of the present disclosure, the first model indication message includes: a list of auxiliary base stations; wherein the device further includes: a model inference unit, used to determine the second base station CU according to the list of auxiliary base stations; and, based on the spectrum sharing model, perform model inference through the second base station CU to obtain a model inference result, so that the first base station and the second base station can achieve spectrum sharing communication according to the model inference result.

[0022] According to another aspect of the present disclosure, a communication device is provided, which is applied to an OAM node, and the device includes: a second message transceiver unit, used to receive a first model request message sent by a first base station CU, and send a first request response message to the first base station CU in response to the first model request message, so that the first base station CU obtains first model training data according to the first request response message, and sends it to the OAM node; and, receiving a second model request message sent by a second base station CU, and sending a second request response message to the second base station CU in response to the second model request message, so that the second base station CU obtains second model training data according to the second request response message, and sends it to the OAM node; a model training unit, used to perform model training according to the first model training data and the second model training data to obtain a trained spectrum sharing model, and send a model indication message carrying parameters of the spectrum sharing model to the first base station CU and the second base station CU, so that the first base station CU and the second base station CU obtain the spectrum sharing model.

[0023] According to another aspect of the present disclosure, an electronic device is provided, comprising: one or more processors; and a storage device configured to store one or more programs, so that when the one or more programs are executed by the one or more processors, the one or more processors implement the communication method as described in the above embodiments.

[0024] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the communication method as described in the above embodiments is implemented.

[0025] The communication method provided by the embodiment of the present disclosure is that when the first base station wants to conduct spectrum sharing communication, it sends a first model request message to the OAM node through the first base station CU, receives a first request response message returned by the OAM node in response to the request message, and then obtains the first model training data according to the response message and sends it to the OAM node, and then the OAM node performs model training according to the first model training data and the second model training data sent by the second base station CU, wherein the first base station and the second base station are base stations within the target spectrum optimization range, and the OAM node performs model training according to the training feature data collected by the first base station CU and the second base station CU, and deploys or updates the trained spectrum sharing model on the first base station CU. In this way, the spectrum sharing model is trained through information interaction between the base station and the OAM node, and the model training is deployed on the OAM node, so that the spectrum sharing model can be used to realize spectrum sharing between base stations, at least to a certain extent solving the problems existing in the existing spectrum sharing technology.

[0026] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute improper limitations on the present disclosure.

[0028] Figure 1 A schematic diagram of a network architecture of a communication system applicable to an embodiment of the present disclosure is shown;

[0029] Figure 2 A flow chart of a communication method according to an embodiment of the present disclosure is shown;

[0030] Figure 3 A schematic diagram of an intelligent functional architecture of a wireless network based on AI / ML technology is shown;

[0031] Figure 4 A flowchart is shown in which a first base station CU acquires first model training data and sends the data to an OAM node in an embodiment of the present disclosure;

[0032] Figure 5 A flow chart of interaction between the second base station CU and the OAM node in an embodiment of the present disclosure is shown;

[0033] Figure 6 A flowchart of performing model reasoning based on a spectrum sharing model according to an embodiment of the present disclosure is shown;

[0034] Figure 7 A flowchart showing model reasoning performed by the first base station CU and the second base station CU based on a spectrum sharing model according to an embodiment of the present disclosure is shown;

[0035] Figure 8 A flow chart showing a communication method according to another embodiment of the present disclosure is shown;

[0036] Fig. 9 An interaction diagram showing a communication method according to an embodiment of the present disclosure;

[0037] Fig.10 A schematic diagram showing the structure of a communication device according to an embodiment of the present disclosure is shown;

[0038] Fig.11 A schematic structural diagram of a communication device according to another embodiment of the present disclosure is shown;

[0039] Fig.12 A structural block diagram of an electronic device in an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0040] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the disclosure will be more comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0041] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0042] It should be noted that the ordinal numbers such as "first" and "second" mentioned in the embodiments of the present disclosure are used to distinguish multiple objects, and are not used to limit the order, timing, priority or importance of multiple objects, and the descriptions of "first" and "second" do not limit the objects to be different.

[0043] Figure 1 FIG. 1 is a schematic diagram showing a network architecture of a communication system applicable to an embodiment of the present disclosure. Figure 1 As shown, the network architecture may include: an operation administration and maintenance (OAM) node 110, a first base station 120, a second base station 130, a first terminal device 140 and a second terminal device 150.

[0044] The OAM node 110 may be understood as a network management device for managing base stations. Figure 1 In the example, the OAM node establishes connections with the first base station 120 and the second base station 130 respectively, and can be used to manage the first base station 120 and the second base station 130.

[0045] The first base station 120 and the second base station 130 are access network devices. Figure 1 As shown, the first base station 120 may include a first base station centralized unit (CU) 121 and a first base station distributed unit (DU) 122. The second base station 130 may include a second base station CU 131 and a second base station DU 132.

[0046] The first terminal device 140 and the second terminal device 150 can be various electronic devices, which can be deployed on land, including indoors or outdoors, handheld, wearable or vehicle-mounted; they can also be deployed on the water surface; they can also be deployed in the air, etc. Optionally, the client of the application installed in different terminal devices is the same, or the client of the same type of application based on different operating systems. Based on different terminal platforms, the specific form of the client of the application can also be different, for example, the application client can be a mobile client, a computer client, etc.

[0047] Among them, the first terminal device 140 can be a terminal device within the coverage of the first base station 120, that is, the first terminal device 140 can access the network through the first base station 120; the second terminal device 150 can be a terminal device within the coverage of the second base station 130, that is, the second terminal device 150 can access the network through the second base station 130.

[0048] It should be noted that the embodiments of the present disclosure do not limit the number of OAM nodes, the number of access network devices, and the number of terminal devices in the network architecture of the communication system, and Figure 1 The network architecture shown may also include other devices, such as core network devices, relay devices, and backhaul devices, which are not limited in the embodiments of the present disclosure. Figure 1 In the network architecture shown, the functions of the first base station CU and the functions of the second base station CU can be integrated into independent physical devices respectively during specific implementation, or can be integrated into one physical device, which is not limited in the embodiments of the present disclosure.

[0049] Figure 2 A flow chart of a communication method according to an embodiment of the present disclosure is shown. Figure 2 The communication method provided in the embodiment can be performed by Figure 1 The first base station CU in the process is executed. Figure 2 As shown, the communication method specifically includes the following steps S210 to S230.

[0050] Step S210: The first base station CU sends a first model request message to the OAM node, and receives a first request response message sent by the OAM node in response to the first model request message.

[0051] The first base station has spectrum sharing capability, and the first base station includes a first base station CU and a first base station DU. If the first base station wants to perform spectrum sharing, the first base station CU sends a first model request message to the OAM node to request a spectrum sharing model from the OAM node.

[0052] In the disclosed embodiment, the spectrum sharing model is an artificial intelligence (AI) / machine learning (ML) model. The wireless network spectrum sharing scenario based on AI / ML can utilize the processing and orchestration capabilities of artificial intelligence and machine learning technologies for big data, collect various data in the wireless network to predict the spectrum and load status of the base station, optimize the spectrum sharing decision, and dynamically configure a more flexible spectrum sharing strategy for the base station in advance. The spectrum sharing model of the disclosed embodiment is an AI / ML model, which enables the base station to realize network spectrum sharing based on AI / ML technology.

[0053] The first model request message may include information indicating the function of the AI / ML model and the reasoning type of the AI / ML model. Specifically, the first model request message may include the following content.

[0054] (1) Identification information of the first base station CU.

[0055] (2) AI / ML model indication information, used to indicate the enabled AI / ML model or function. The first model request message indicates that the indication information of a related model is included, and the trained model numbers can be enumerated, such as represented by integer values ​​0, 1, 2, etc. Each model can include spectrum sharing indication, network load prediction, and terminal trajectory prediction functions.

[0056] (3) Identification information indicated by the training request, which is used to indicate the information required by the sender of the request message to achieve the expected performance, including but not limited to model feature input variable identification information, training mode identification information, and model performance identification information. The sender of the request message here is the first base station or the first base station CU.

[0057] The model feature input variable identification information is used to obtain the model feature input variable.

[0058] The training mode identification information is used to obtain the model algorithm.

[0059] Model performance identification information may include: model parameter identification information, used to indicate the configuration variables within the model, such as weights, biases, and learning rates; inference output name identification information, used to indicate the name of the inference output of the AI / ML function, including at least one or more of network load analysis, network performance analysis, and terminal device-related analysis; performance indicator identification information, used to evaluate the performance indicators of the AI / ML function, such as accuracy, precision, and other indicators; performance score identification information, used to indicate the performance score of the AI / ML function when reasoning on a specific data set; confidence score identification, used to indicate the numerical value of the reliability / quality of a given decision generated by the AI / ML function.

[0060] Among them, the AI / ML functional body can be understood as a unit or module in the OAM node used to train the AI / ML model.

[0061] (4) Training data source indication information, used to obtain the address information of the training data source required for model training.

[0062] (5) AI / ML functional body context information, which may include the following context information: training context information, which is used to indicate the status and conditions related to training; expected runtime context information, which is used to indicate the context of the AI / ML functional body of the expected application, including the specific time conditions for model training; runtime context information, which is used to indicate the context of the application model.

[0063] (6) Request status information, which is used to indicate the status of the training request, and may include status indications such as “NOT STARTED”, “TRAINING IN PROGRESS”, “SUSPENDED”, “FINISHED”, and “CANCELLED”.

[0064] (7) Cancel request indication information, used to indicate whether the sender of the request message cancels the training request. Setting this attribute to "TRUE" will cancel the corresponding training request. Otherwise, setting this attribute to "FALSE" will not cancel the corresponding training request. The default value of this attribute is set to "FALSE". The sender of the request message here is the first base station or the first base station CU.

[0065] (8) Pause request indication information, used to indicate whether the request message sender pauses the training request. When the request status is not in the "FINISHED" state, it can be paused, that is, the attribute is set to "TRUE", otherwise the attribute value is "FALSE". The request message sender here is the first base station or the first base station CU.

[0066] After receiving the first model request message sent by the first base station CU, the OAM node sends a model reply message, ie, a first request response message, to the first base station CU in response to the first model request message. Specifically, the first request response message may include the following information.

[0067] (1) Anchor base station indication information, used to indicate that the base station receiving the information is an anchor base station.

[0068] (2) AI / ML model indication information, used to indicate the enabled AI / ML model or function.

[0069] (3) Priority indication information, used to indicate the priority of the training process. When training AI / ML models, which training factors need to be prioritized to ensure the effectiveness and efficiency of the training? Training factors may include the quality and quantity of training data, the availability of computing resources, the model structure, and the selection of hyperparameters. Different application scenarios and goals may require different training factors to be prioritized, so the priority of the training process needs to be determined based on the specific situation. For example, the measurement information of the terminal device may be prioritized, or the measurement information of the base station may be prioritized. When training AI / ML models, priorities can be used to schedule the training process.

[0070] (4) Training request indication information, corresponding to the identification information of the training request indication, including but not limited to model feature input variable information, training mode information, and model performance information.

[0071] The model feature input variables may include: terminal device feature input information, base station feature input information, and neighboring base station feature input information.

[0072] Among them, the terminal device characteristic input information includes but is not limited to: the location information of the terminal device, such as the location coordinates of the terminal device, the information of the service cell, the moving speed, etc.; the wireless measurement information of the terminal device, such as the reference signal receiving power, the reference signal receiving quality, etc.

[0073] The base station characteristic input information is the characteristic input information of the first base station, including but not limited to: the current energy efficiency of the base station; the current wireless resource status of the base station, such as the cell physical resource block utilization, the average number of wireless resource control connections, the packet loss rate, etc.

[0074] The neighboring base station characteristic input information is characteristic input information of the neighboring base stations of the first base station, wherein the neighboring base stations of the first base station may be base stations having a communication interface with the first base station, or base stations that can be switched with the first base station. The neighboring base station characteristic input information includes but is not limited to: the current energy efficiency of the neighboring base station; the current wireless resource status of the neighboring base station; the current spectrum status of the neighboring base station, for example, the current spectrum status of the neighboring base station is divided into active, high, low, and inactive according to the spectrum usage of the neighboring base station.

[0075] The training mode information is used to indicate the model algorithm, which may include linear regression, logistic regression, decision tree, support vector machine, random forest and other algorithms.

[0076] Model performance information may include model parameter information, inference output name information, performance indicator information, performance score, and confidence score.

[0077] Among them, the model parameter information is used to indicate the configuration variables within the model, including but not limited to weights, biases, learning rates, etc.

[0078] The inference output name information is used to indicate the name of the inference output performed by the AI / ML function, including one or more of network load analysis, network performance analysis, UE-related analysis, etc.

[0079] Performance indicator information is used to evaluate the performance indicators of AI / ML functions, such as accuracy, precision, and other indicators.

[0080] The performance score is a numerical value in the range of 0-100, which is used to represent the performance score of the AI / ML function when performing reasoning on a specific data set. Depending on the nature of the model, the performance metrics of different types of models may be different. For example, for numeric prediction, the metric may be accuracy; for classification, the metric may be a combination of precision and recall.

[0081] The confidence score is a numerical value in the range of 0-100, which is used to represent the reliability / quality of a given decision generated by the AI / ML function, where the lowest value represents the lowest level of reliability of the decision.

[0082] (5) A list of training data providers, used to indicate entities that have provided or should provide data related to AI / ML model training.

[0083] (6) Training data source indication information, which indicates and distinguishes the source of training data through the address information of the training data source. The data source can be different network functional entities and operators, etc., where different network functional entities include base stations, core networks, etc.

[0084] In step S220, the first base station CU obtains the first model training data according to the first request response message, and sends the first model training data to the OAM node, so that the OAM node performs model training according to the first model training data and the second model training data, obtains the trained spectrum sharing model, and sends the first model indication message carrying the parameters of the spectrum sharing model to the first base station CU.

[0085] Step S230: The first base station CU obtains a spectrum sharing model according to the first model indication message.

[0086] The first base station and the second base station are base stations within the target spectrum optimization range, through which the bandwidth, time and other information of spectrum sharing can be determined. Since the first base station and the second base station are base stations within the target spectrum optimization range, the first base station and the second base station can achieve spectrum sharing communication. It should be noted that the first base station is the base station that first sends a model request message to the OAM node. The first base station can be a spectrum requester or a spectrum provider. If the first base station is a spectrum requester, then the second base station is a spectrum provider. Conversely, if the first base station is a spectrum provider, then the second base station is a spectrum requester.

[0087] Among them, the second base station may include a second base station CU and a second base station DU. In step S220, the OAM node performs model training according to the first model training data and the second model training data to obtain a trained spectrum sharing model. The first model training data is sent by the first base station CU to the OAM node, and may specifically include training feature data collected by the first base station CU; the second model training data is sent by the second base station CU to the OAM node, and may specifically include training feature data collected by the second base station CU.

[0088] The OAM node receives data sets sent from base stations within the target spectrum optimization range, namely, the first model training data sent by the first base station CU and the second model training data sent by the second base station CU, and performs preprocessing and data segmentation on the data sets to form training data sets, verification data sets, and test data sets. The OAM node performs online model training based on the above data sets, and adjusts parameters and optimizes features through the model training, verification, and testing processes to obtain a trained spectrum sharing model.

[0089] In the disclosed embodiment, the feature data can be randomly divided according to the proportion. Specifically, the training data set is used to train the model, which usually accounts for about 60% to 80% of the total data set. When dividing the training data set, it is necessary to ensure the diversity and representativeness of the data to fully train the model and avoid overfitting. The validation data set is used to adjust the hyperparameters of the model and evaluate the performance of the model, which usually accounts for about 10% to 20% of the total data set. When dividing the validation data set, it is necessary to ensure the independence and randomness of the data to fully evaluate the generalization ability of the model. The test data set is used to finally evaluate the performance of the model, which usually accounts for about 10% to 20% of the total data set. When dividing the test data set, it is necessary to ensure the independence and randomness of the data to fully evaluate the generalization ability of the model.

[0090] The following should be noted when dividing the data set: the division of the data set should be random to avoid specific patterns or regularities in the data set; the division of the data set should be fixed to facilitate comparisons on different models or algorithms; the division of the data set should ensure that there is no overlap between the training, validation, and test data sets to avoid the model using the same data during training, validation, and testing.

[0091] After the OAM node obtains the spectrum sharing model through training, it may send a first model indication message carrying the parameters of the spectrum sharing model to the first base station CU. Then in step S230, the first base station CU deploys or updates the model according to the first model indication message to obtain the spectrum sharing model.

[0092] In some embodiments of the present disclosure, the second base station CU may obtain the spectrum sharing model according to a second model indication message carrying parameters of the spectrum sharing model sent by the OAM node.

[0093] After the OAM node obtains the spectrum sharing model through training, it can send a second model indication message carrying the parameters of the spectrum sharing model to the second base station CU, so that the second base station CU deploys or updates the model according to the second model indication message to obtain the spectrum sharing model. In this way, both the first base station and the second base station can obtain the spectrum sharing model trained by the OAM node to perform spectrum sharing communication based on the spectrum sharing model.

[0094] In the embodiment of the present disclosure, the spectrum sharing model may be an AI / ML model, and the first base station and the second base station may realize network spectrum sharing based on AI / ML technology. Figure 3 The figure shows a schematic diagram of the intelligent functional architecture of a wireless network based on AI / ML technology. Figure 3 As shown, the intelligent functional architecture of the wireless network based on AI / ML technology may include a data collection function 310, a model training function 320, a model reasoning function 330 and an execution function 340.

[0095] The data collection function 310 provides training data to the model training function 320, and provides inference data to the model inference function 330. Specifically, the training data and inference data include measurements from terminal devices or different network entities, feedback data from the execution function 340, and output data from the AI / ML model.

[0096] The model training function 320 performs the training, verification and testing functions of the AI / ML model, and can also use the generated model performance indicators as part of the model testing process. The model training function 320 is also responsible for data preprocessing, cleaning, formatting and conversion based on the training data provided by the data collection function 330.

[0097] The model reasoning function 330 provides AI / ML model reasoning output based on the trained model. The model reasoning output includes prediction and decision information for each network entity to judge and execute. In addition, the model reasoning function 330 also has the ability to process reasoning data.

[0098] The execution function 340 receives the output of the model reasoning function 330 and triggers or executes corresponding actions.

[0099] The communication method provided by the embodiment of the present disclosure is that when the first base station wants to conduct spectrum sharing communication, it sends a first model request message to the OAM node through the first base station CU, receives a first request response message returned by the OAM node in response to the request message, and then obtains the first model training data according to the response message and sends it to the OAM node, and then the OAM node performs model training according to the first model training data and the second model training data sent by the second base station CU, wherein the first base station and the second base station are base stations within the target spectrum optimization range, and the OAM node performs model training according to the training feature data collected by the first base station CU and the second base station CU, and deploys or updates the trained spectrum sharing model on the first base station CU. In this way, the spectrum sharing model is trained through information interaction between the base station and the OAM node, and the model training is deployed on the OAM node, so that the spectrum sharing model can be used to realize spectrum sharing between base stations, at least to a certain extent solving the problems existing in the existing spectrum sharing technology.

[0100] To facilitate understanding, each step of the communication method of the embodiment of the present disclosure is described in more detail below.

[0101] In some embodiments of the present disclosure, the first base station CU in step S210 sends a first model request message to the OAM node, and receives a first request response message sent by the OAM node in response to the first model request message, which may further include: the first base station CU sends a first model request message to the OAM node, so that the OAM node determines the target spectrum optimization range according to the first model request message, sets the first base station as an anchor base station, and sends a first request response message to the first base station CU; the first base station CU receives the first request response message sent by the OAM node.

[0102] The target spectrum optimization range may include information such as bandwidth and time of spectrum sharing. After receiving the first model request message, the OAM node may predict based on the trained model that the spectrum utilization rate of a certain base station is low in the future, and the spectrum bandwidth of the base station is the same as the spectrum bandwidth of the first base station, and the spectrum utilization time of the base station does not conflict with the spectrum utilization time of the first base station. Then the base station and the first base station are within the target spectrum optimization range, that is, the base station may be determined as a base station for spectrum sharing with the first base station.

[0103] Furthermore, if the first base station is the base station that first requests spectrum sharing within the target spectrum optimization range, the OAM node may set the first base station as an anchor base station.

[0104] In some embodiments of the present disclosure, the first request response message may include: anchor base station indication information; wherein, after the first base station CU receives the first request response message sent by the OAM node, the communication method also includes: the first base station CU determines that the first base station is an anchor base station based on the anchor base station indication information.

[0105] The anchor base station indication information is used to indicate that the base station receiving the information is an anchor base station. After receiving the first request response message, the first base station CU parses the response message including the anchor base station indication information, and confirms that the first base station is an anchor base station.

[0106] In the disclosed embodiment, the anchor base station can be understood as a base station for model reasoning. The OAM node receives the first model request message sent by the first base station CU, and can confirm the target spectrum optimization range, and then confirm the base station that can perform spectrum sharing communication with the first base station based on the target spectrum optimization range. In addition, the OAM node can set the first base station as the anchor base station, and carry the anchor base station indication information in the first request response message, so that the first base station CU can determine that the first base station is the anchor base station based on the first request response message received, and then perform model reasoning.

[0107] Figure 4 FIG. 1 is a flow chart showing an embodiment of the present disclosure in which a first base station CU obtains first model training data and sends the data to an OAM node. Figure 4 As shown, the first base station CU in step S220 obtains the first model training data according to the first request response message, and sending the first model training data to the OAM node may further include the following steps.

[0108] Step S221: The first base station CU obtains the training feature data of the first base station and the training feature data of neighboring base stations of the first base station according to the model feature input variable information and the training data provider list information.

[0109] In the embodiment of the present disclosure, the first request response message may include training request indication information and training data provider list information, and the training request indication information includes model feature input variable information.

[0110] Among them, the model feature input variable information also includes base station feature input information and neighboring base station feature input information. According to the training data provider list information, the neighboring base stations of the first base station, that is, the base stations having a communication interface with the first base station, can be determined, and according to the model feature input variable information, the training feature data of the first base station and the neighboring base stations of the first base station can be obtained.

[0111] Specifically, the training feature data of the first base station may include: the current energy efficiency of the first base station, the current wireless resource status of the first base station. The training feature data of the neighboring base stations of the first base station may include: the current energy efficiency of the neighboring base stations, the current wireless resource status of the neighboring base stations, and the current spectrum status of the neighboring base stations.

[0112] Step S222: The first base station CU generates measurement configuration information according to the model feature input variable information, and sends the measurement configuration information to the first terminal device through the first base station DU, so that the first terminal device obtains the training feature data of the first terminal device according to the measurement configuration information and sends it to the first base station CU.

[0113] The first terminal device is within the coverage of the first base station, or the first terminal device is in the public land mobile network PLMN list managed by the first base station.

[0114] In the embodiment of the present disclosure, the model feature input variable information also includes terminal device feature input information. The first base station CU can generate measurement configuration information based on the terminal device feature input information in the model feature input variable information, and send the measurement configuration information to the first base station DU, and then the first base station DU sends the measurement configuration information to the first terminal device. In this way, the first terminal device can obtain training feature data based on the measurement configuration information, and send the obtained training feature data to the first base station CU.

[0115] The measurement configuration information includes wireless resource management measurement configuration information and minimization path measurement configuration information. The wireless resource management measurement configuration information may include: support for periodic measurement triggering, specifically triggering period and recording period; terminal device wireless measurement information, specifically one or more of reference signal received power and reference signal received quality. The minimization path measurement configuration information may include: support for periodic measurement triggering, specifically triggering period and recording period; terminal device location information; terminal device moving speed information.

[0116] The training feature data obtained by the first terminal device may include: location information of the first terminal device, moving speed information of the first terminal device, wireless measurement information of the first terminal device, and timestamp information.

[0117] Step S223: The first base station CU sends the training characteristic data of the first base station, the training characteristic data of the neighboring base stations of the first base station, and the training characteristic data of the first terminal device to the OAM node.

[0118] The first model training data may include training feature data of the first base station, training feature data of neighboring base stations of the first base station, and training feature data of the first terminal device. After obtaining the training feature data of the first base station, the training feature data of neighboring base stations of the first base station, and the training feature data of the first terminal device, the first base station CU sends these training feature data, i.e., the first model training data, to the OAM node.

[0119] In the disclosed embodiment, the first model request message may include information indicating the function of the AI / ML model and the reasoning type of the AI / ML model, and the first request response message is a reply message of the OAM node in response to the first model request message. The first base station CU may collect the first model training data according to the information in the first request response message, so that the OAM node performs model training according to the training data.

[0120] It has been explained in the above step S220 that the second model training data is sent by the second base station CU to the OAM node, and the first base station CU and the second base station CU belong to the base station CU within the target spectrum optimization range. In some embodiments of the present disclosure, the second model training data is obtained by the second base station CU according to the second request response message sent by the OAM node, and the second request response message is sent by the OAM node to the second base station CU in response to the second model request message sent by the second base station CU. Among them, the second model training data includes: training feature data of the second base station, training feature data of the neighboring base stations of the second base station, and training feature data of the second terminal device, the second terminal device is within the coverage of the second base station, or the second terminal device is in the PLMN list managed by the second base station.

[0121] It can be seen that the second model training data is obtained through information interaction between the second base station CU and the OAM node. For ease of understanding, a detailed embodiment is provided to illustrate the interaction process between the second base station CU and the OAM node.

[0122] Figure 5 A flow chart of interaction between the second base station CU and the OAM node in an embodiment of the present disclosure is shown. Figure 5 It shows that the second base station CU and the OAM node interact with each other, so that the OAM node obtains the second model training data to perform model training.

[0123] like Figure 5 As shown, the interaction process between the second base station CU and the OAM may include the following steps.

[0124] Step S501: The second base station CU sends a second model request message to the OAM node.

[0125] The second model request message may include information for indicating the AI / ML model function and the inference type of the AI / ML model. Specifically, the second model request message may include the following: identification information of the second base station CU, AI / ML model indication information, identification information of the training request indication, training data source indication information, AI / ML function body context information, request status information, cancel request indication information, and pause request indication information.

[0126] The identification information of the second base station CU is used by the OAM node to set the second base station as an auxiliary base station. The identification information indicated by the training request is used to indicate the information required by the sender of the request message to achieve the expected performance, including but not limited to model feature input variable identification information, training mode identification information, and model performance identification information. The sender of the request message here is the second base station or the second base station CU.

[0127] Step S502: After receiving the second model request message, the OAM node sets the second base station as a secondary base station and sends a second request response message to the second base station CU.

[0128] The second request response message may include the following information: auxiliary base station indication information, anchor base station information, AI / ML model indication information, priority indication information, training request indication information, training data provider list, and training data source indication information.

[0129] The auxiliary base station indication information is used by the second base station CU to determine that the second base station is the auxiliary base station, and the anchor base station information is used by the second base station CU to determine the anchor base station.

[0130] After the OAM node receives the second model request message sent by the second base station CU, it determines that the second base station and the first base station are within the target spectrum optimization range, and then sets the second base station as an auxiliary base station to provide data to the first base station during model reasoning. The OAM node adds the auxiliary base station indication information and the anchor base station information to the second request response message, so that after the second base station CU receives the second request response message, it can determine that the second base station is an auxiliary base station of the first base station (i.e., the anchor base station).

[0131] Furthermore, the training request indication information in the second request response message corresponds to the identification information of the training request indication in the second model request message, including but not limited to model feature input variable information, training mode information, and model performance information.

[0132] The model feature input variables may include: terminal device feature input information, base station feature input information, and neighboring base station feature input information. The terminal device feature input information includes, but is not limited to: location information of the terminal device; wireless measurement information of the terminal device. The base station feature input information is the feature input information of the second base station. The neighboring base station feature input information is the feature input information of the neighboring base stations of the second base station, wherein the neighboring base stations of the second base station may be base stations having a communication interface with the second base station, or base stations that can be switched with the second base station.

[0133] Step S503: The second base station CU obtains second model training data according to the second request response message, and sends the second model training data to the OAM node.

[0134] The second model training data includes training feature data of the second base station, training feature data of neighboring base stations of the second base station, and training feature data of the second terminal device.

[0135] Specifically, the second base station CU can determine the neighboring base stations of the second base station, that is, the base stations having a communication interface with the second base station, according to the training data provider list information in the second request response message, and can obtain the training feature data of the second base station and the neighboring base stations of the second base station according to the model feature input variable information. Among them, the training feature data of the second base station may include: the current energy efficiency of the second base station, the current wireless resource status of the second base station. The training feature data of the neighboring base stations of the second base station may include: the current energy efficiency of the neighboring base stations, the current wireless resource status of the neighboring base stations, and the current spectrum status of the neighboring base stations.

[0136] And, the second base station CU can generate measurement configuration information according to the terminal device feature input information in the model feature input variable information, send the measurement configuration information to the second base station DU, and then the second base station DU sends the measurement configuration information to the second terminal device. In this way, the second terminal device can obtain training feature data according to the measurement configuration information, and send the obtained training feature data to the second base station CU. The second terminal device is within the coverage of the second base station, or the second terminal device is in the PLMN list managed by the second base station.

[0137] Therefore, the second model training data may include the training feature data of the second base station, the training feature data of the neighboring base stations of the second base station, and the training feature data of the second terminal device. After obtaining the training feature data of the second base station, the training feature data of the neighboring base stations of the second base station, and the training feature data of the second terminal device, the second base station CU sends these training feature data, i.e., the second model training data, to the OAM node.

[0138] In the embodiment of the present disclosure, the second model request message may include information indicating the function of the AI / ML model and the reasoning type of the AI / ML model, and the second request response message is a reply message of the OAM node in response to the second model request message. The second base station CU may collect the second model training data according to the information in the second request response message and send it to the OAM node so that the OAM node performs model training according to the training data.

[0139] After the OAM node has trained the spectrum sharing model, it sends the first model indication information carrying the model parameters to the first base station CU, and sends the second model indication information carrying the model parameters to the second base station CU, so that the first base station CU and the second base station CU can deploy or update the model, obtain the spectrum sharing model, perform model reasoning based on the spectrum sharing model, and use the model reasoning result to perform spectrum sharing communication. The above describes the communication method of the embodiment of the present disclosure including the process of model training, and the model reasoning process is described next.

[0140] Figure 6 FIG. 4 is a flowchart of performing model reasoning based on a spectrum sharing model according to an embodiment of the present disclosure. Figure 6 As shown, the model reasoning process can include the following steps.

[0141] Step S610: The first base station CU determines the second base station CU according to the auxiliary base station list in the first model indication message.

[0142] Step S620: The first base station CU performs model reasoning through the second base station CU based on the spectrum sharing model to obtain a model reasoning result, so that the first base station and the second base station can achieve spectrum sharing communication according to the model reasoning result.

[0143] After the OAM node obtains the trained spectrum sharing model, it sends a first model indication message carrying the parameters of the spectrum sharing model to the first base station CU. The first model indication message may include: auxiliary base station list, AI / ML model indication information, priority indication information, training request indication information, training data provider list, and training data source indication information.

[0144] The auxiliary base station list is used to indicate the information list of the auxiliary base stations. The first base station CU can determine which auxiliary base stations are included according to the auxiliary base station list, and collect the data required for model inference from the auxiliary base stations. In addition, the AI / ML model indication information, priority indication information, training request indication information, training data provider list, and training data source indication information have been explained in detail above, and will not be repeated here.

[0145] During the model reasoning process, the first base station CU determines the second base station CU based on the auxiliary base station list, and then the first base station CU and the second base station CU perform model reasoning based on the spectrum sharing model to obtain model reasoning results, and then realize spectrum sharing communication between the first base station and the second base station based on the model reasoning results.

[0146] Figure 7 FIG. 1 shows a flow chart of a first base station CU and a second base station CU performing model reasoning based on a spectrum sharing model according to an embodiment of the present disclosure. Figure 7 As shown, the process of model reasoning may include the following steps.

[0147] Step S621: The first base station CU obtains first model inference data.

[0148] The first model inference data includes: target feature data of the first base station, target feature data of neighboring base stations of the first base station, and target feature data of the first terminal device.

[0149] The first model inference data is feature data collected by the first base station CU when performing model inference. After the first base station CU receives the first model indication information sent by the OAM node, it deploys or updates the model according to the first model indication information, obtains the spectrum sharing model trained by the OAM node, and the first base station CU triggers local measurement, statistics and recording of the base station to obtain the first model inference data.

[0150] In an embodiment of the present disclosure, the first terminal device can periodically send data required for model training and / or model inference to the first base station CU based on the configured measurement information. The data includes but is not limited to the location information of the first terminal device, the moving speed of the first terminal device, the wireless measurement information of the first terminal device, and timestamp information.

[0151] Step S622: The first base station CU sends a data request message to the second base station CU, and receives the second model inference data sent by the second base station CU.

[0152] The second model inference data includes: target feature data of the second base station, target feature data of neighboring base stations of the second base station, and target feature data of the second terminal device.

[0153] The second model inference data is feature data collected by the second base station CU when performing model inference. The first base station CU sends a data request message to the second base station CU to obtain the feature data collected by the second base station CU to perform model inference operations. Specifically, the data request message may include: identification information of the first base station CU, address information of the first base station CU, reported cell list, and input information.

[0154] The address information of the first base station CU may be the IP address and port address of the first base station CU, and the information is used to indicate that the second base station node will transmit massive AI / ML data through the established user plane channel.

[0155] The reporting cell list may include: cell reporting information, such as cell identification information, synchronization signal and physical broadcast channel block reporting list, synchronization signal and physical broadcast channel block index, etc.; reporting period, used to indicate the average window length of all measurement objects. It should be noted that the cells in the reporting cell list may include all cells under the OAM node.

[0156] The input information is used to indicate data information that needs to be measured and reported by one or more cells under the second base station CU. The input information provided by each cell may include: base station input information, neighboring base station input information and terminal device input information.

[0157] Among them, the base station input information is the input information of the second base station, including the terminal device trajectory prediction, the current / predicted energy efficiency of the second base station, and the current / predicted resource status of the second base station.

[0158] The neighboring base station input information is the input information of the neighboring base stations of the second base station, including the current / predicted energy efficiency of the neighboring base stations of the second base station, the current / predicted resource status of the neighboring base stations of the second base station, and the current spectrum status of the neighboring base stations of the second base station.

[0159] The input information of the terminal device is the input information of the second terminal device, including: location information of the second terminal device, such as the coordinates of the second terminal device, the identification of the service cell, the moving speed of the second terminal device, etc.; wireless measurement information of the second terminal device.

[0160] In some embodiments of the present disclosure, after the first base station CU sends a data request message to the second base station CU, the communication method may further include: the first base station CU receiving a data request reply message sent by the second base station CU in response to the data request message.

[0161] After receiving the data request message sent by the first base station CU, the second base station CU uses a data request reply message to indicate the measurement object for the request message, start the measurement and provide the corresponding measurement object information of the measurement information. The data request reply message may include: the identifier of the second base station CU, the address information of the second base station CU, and the critical diagnosis identifier. The critical diagnosis identifier is used to indicate which information in the received information is not understood or lost.

[0162] Among them, the address information of the second base station CU can be the IP address and port second of the second base station CU, which is used to indicate that massive AI / ML data will be transmitted between base station nodes through the established user plane channel.

[0163] It can be seen that in some embodiments of the present disclosure, the data request message may include the address information of the first base station CU, and the data request reply message may include the address information of the second base station CU. Therefore, the first base station CU establishes a user plane channel with the second base station CU based on the address information of the first base station CU and the address information of the second base station CU to transmit data through the user plane channel.

[0164] In some embodiments of the present disclosure, the data request reply message includes a data request failure and a request retransmission time. Also, after the first base station CU receives the data request reply message sent by the second base station CU in response to the data request message, the communication method may further include: the first base station CU parses the data request reply message, determines that the data request fails, and sends the data request message to the second base station CU according to the request retransmission time.

[0165] If the measurement object in the data request reply message sent by the first base station CU cannot be provided, the data request reply message sent by the second base station CU to the first base station CU is a data request failure message carrying the reason for failure, and the information contained includes but is not limited to: the identifier of the second base station CU, the reason for failure, critical diagnostic identification information and request retransmission time. Among them, the failure reason can be a wireless network layer reason, a transport layer reason, a protocol reason, etc.; the critical diagnostic identification information is when part of the received message is not understood or lost, or the message contains a logical error, this identifier is used to indicate the information that is not understood or lost; the request retransmission time is used to indicate the time to re-initiate the request.

[0166] After the first base station CU receives the data request reply message sent by the second base station CU, it parses the data request reply message, determines that the data request fails, and sends the data request message again according to the requested retransmission time, so that the second base station CU obtains the second model inference data required for model inference.

[0167] Step S623: The first base station CU performs model inference based on the spectrum sharing model and the first model inference data and the second model inference data to obtain a model inference result.

[0168] The first base station CU performs a model reasoning function based on the first model reasoning data obtained by itself and the second model reasoning data obtained by the second base station CU, and generates an AI / ML model reasoning output result including prediction information and decision information.

[0169] Step S624: The first base station CU implements spectrum sharing communication with the second base station CU according to the model inference result.

[0170] The first base station CU and the second base station CU implement spectrum sharing communication based on the model reasoning results. For example, the information list of recommended spectrum sharing cells, each spectrum sharing cell enters the corresponding spectrum sharing state, and the terminal equipment in the spectrum sharing cell switches to the corresponding cell carrying traffic according to the recommended candidate cell information list, realizing network spectrum sharing under the premise of ensuring the terminal user experience.

[0171] The communication method provided by the embodiment of the present disclosure deploys the model reasoning function on the base station CU, so that the base station CU has the model reasoning function, can make accurate predictions on the base station load and spectrum trend and generate model reasoning output information, so that the network performs spectrum sharing operations or the terminal performs corresponding switching operations based on the decision information generated by the model reasoning, thereby improving the spectrum sharing benefits while ensuring the user experience; and, by transmitting address information between the base station CUs, establishing a transmission channel for transmitting data required for AI / ML model training and reasoning, thereby realizing the interaction of AI / ML data between base stations, so that the spectrum sharing base station CU performs model reasoning functions based on the data information reported by each base station.

[0172] Figure 8 A flow chart of a communication method according to another embodiment of the present disclosure is shown. Figure 8 The communication method provided in the embodiment can be performed by Figure 1 The OAM nodes in the Figure 8 As shown, the communication method specifically includes the following steps S810 to S830.

[0173] Step S810, the OAM node receives a first model request message sent by the first base station CU, and sends a first request response message to the first base station CU in response to the first model request message, so that the first base station CU obtains the first model training data according to the first request response message and sends it to the OAM node.

[0174] Step S820: The OAM node receives a second model request message sent by the second base station CU, and sends a second request response message to the second base station CU in response to the second model request message, so that the second base station CU obtains second model training data according to the second request response message and sends it to the OAM node.

[0175] In step S830, the OAM node performs model training according to the first model training data and the second model training data to obtain a trained spectrum sharing model, and sends a model indication message carrying parameters of the spectrum sharing model to the first base station CU and the second base station CU so that the first base station CU and the second base station CU obtain the spectrum sharing model.

[0176] The communication method provided by the embodiment of the present disclosure trains a spectrum sharing model through information interaction between a base station and an OAM node, and deploys the model training on the OAM node. The spectrum sharing model can be used to realize spectrum sharing between base stations, and at least to a certain extent, solve the problems existing in the existing spectrum sharing technology.

[0177] Considering that the principle of solving the problem by the communication method embodiment executed by the OAM node is similar to that of the communication method embodiment executed by the first base station CU, the implementation of the communication method embodiment executed by the OAM node can refer to the implementation of the communication method embodiment executed by the first base station CU, and the repeated parts will not be repeated.

[0178] Specific embodiments are listed below to illustrate the communication method provided in the embodiments of the present disclosure. Fig. 9 An interaction diagram of a communication method according to an embodiment of the present disclosure is shown. Fig. 9 The interaction process between the OAM node, CU1, DU1, UE1, CU2, DU2 and UE2 is shown. Fig. 9 As shown, the interactive process of the communication method may include the following steps.

[0179] Step S901, CU1 sends a model request message to the OAM node, triggering the OAM node to perform AI / ML offline model training to obtain relevant AI / ML model functions. The model request message includes the AI / ML function and model reasoning type used to indicate the request.

[0180] Step S902: The OAM node receives the model request message, determines the network spectrum optimization range and sets the first base station as the anchor base station. The OAM node uses the stored historical data as training data according to the model request message, performs offline model training, and sends a model reply message to CU1.

[0181] Step S903: CU2 sends a model request message to the OAM node to obtain the relevant AI / ML model function, wherein the model request message includes the requested AI / ML function and the model reasoning type.

[0182] Step S904, the OAM node receives the model request message, and sets the second base station as an auxiliary base station according to the model request message to provide the first base station with AI / ML data required by the model. The OAM node sends a model reply message to CU2.

[0183] Step S905, CU1 receives the model reply message, measures and collects the first base station feature input information and the neighboring base station feature input information according to the model feature input variable indication information, and sends a measurement configuration message to DU1 to collect the terminal feature input information.

[0184] Step S906, DU1 receives the measurement configuration message and selects a suitable UE1 for measurement triggering. If a UE is not in the designated area of ​​the first base station or the serving PLMN is not in the PLMN list managed by the first base station, the UE will not report the measurement data.

[0185] Step S907, UE1 performs corresponding measurements according to the measurement configuration message, collects the indicated measurement information, such as UE location information, etc., and completes recording and collection of corresponding measurement information.

[0186] Step S908: UE1 sends the collected relevant measurement information for model training to CU1 via a UE measurement reporting message.

[0187] It should be noted that, according to the configuration information, UE1 periodically sends measurement information to CU1 for use in model training and / or model inference.

[0188] Step S909: CU1 receives the measurement information reported by UE1, and sends the measurement statistics of the relevant information to the OAM node according to the model feature input variable information indication information in the model reply message.

[0189] Step S910, CU2 receives the model reply message, measures and collects the second base station feature input information and the neighboring base station feature input information according to the model feature input variable indication information, and sends a measurement configuration message to DU2 to collect the terminal feature input information.

[0190] Step S911, DU2 receives the measurement configuration message and selects a suitable UE2 for measurement triggering. If a UE is not in the designated area of ​​the second base station or the serving PLMN is not in the PLMN list managed by the second base station, the UE will not report the measurement data.

[0191] Step S912: UE2 performs corresponding measurements according to the measurement configuration message, collects the indicated measurement information, such as UE location information, etc., and completes recording and collection of corresponding measurement information.

[0192] Step S913: UE2 sends the collected relevant measurement information for model training to CU2 via a UE measurement reporting message.

[0193] It should be noted that, according to the configuration information, UE2 periodically sends measurement information to CU2 for use in model training and / or model inference.

[0194] Step S914: CU2 receives the measurement information reported by UE2, and sends the measurement statistics of the relevant information to the OAM node according to the model feature input variable information indication information in the model reply message.

[0195] In step S915, the OAM node receives the data sent by CU1 and CU2, processes and segments the data, and forms a training data set, a verification data set, and a test data set. OAM performs online model training based on the above data sets, and adjusts parameters and optimizes features through the model training, verification, and testing process to obtain a more accurate AI / ML model.

[0196] In step S916, the OAM node sends the information of the AI / ML model that has been trained, verified, and tested online to CU1 via a model deployment / update message.

[0197] In step S917, the OAM node sends the information of the AI / ML model that has been trained, verified, and tested online to CU2 via a model deployment / update message.

[0198] Step S918, CU1 receives the model deployment / update message from the OAM node, triggers the local measurement, statistics and recording process of the base station according to the latest model indication information, and sends a data request message to CU2 to obtain the data collected by CU2 for model reasoning operations.

[0199] Step S919, CU2 receives the data request message, and indicates through a data reply message that the measurement of the requested measurement object can be successfully started and the corresponding measurement object information of the measurement information can be provided.

[0200] Step S920: CU2 sends an input data update message to CU1 to report the measurement information requested by the first base station. The measurement information reported by CU2 includes UE2 measurement reporting data.

[0201] In step S921, CU1 performs a model reasoning function based on the characteristic input information of its own measurement statistics, the collected UE1 measurement report data, and the measurement data reported by CU2, and generates an AI / ML model reasoning output including prediction information and decision information for the network to analyze or execute operations.

[0202] Step S922: Implement network spectrum sharing according to the indication information output by the model reasoning.

[0203] The communication method of the disclosed embodiment deploys the model training and model reasoning functions to the OAM node and the CU of the spectrum sharing base station respectively; the OAM node is responsible for collecting the measurement data information of each node in the network, performing offline and / or online training, and deploying or updating the trained model to each base station CU; the base station CU has a model reasoning function, which can make accurate predictions of the base station load and spectrum trend and generate model reasoning output information, so that the network performs spectrum sharing operations or the terminal performs corresponding switching operations based on the decision information generated by the model reasoning, thereby improving the spectrum sharing benefits while ensuring the user experience. In addition, by transmitting address information between the CUs of each base station and establishing a transmission channel for transmitting the data required for AI / ML model training and reasoning, the interaction of AI / ML data between base stations is realized, enabling the centralized management entity of the spectrum sharing base station to perform model reasoning functions based on the data information reported by each base station.

[0204] Fig.10 FIG. 1 is a schematic diagram showing the structure of a communication device according to an embodiment of the present disclosure. The communication device can be applied to a first base station CU, such as Fig.10 As shown, the communication device 1000 may include: a first message transceiving unit 1010 , a data acquiring unit 1020 and a model acquiring unit 1030 .

[0205] Among them, the first message transceiver unit 1010 can be used to: send a first model request message to the OAM node, and receive a first request response message sent by the OAM node in response to the first model request message; the data acquisition unit 1020 can be used to: obtain first model training data according to the first request response message, send the first model training data to the OAM node, so that the OAM node performs model training according to the first model training data and the second model training data, obtains a trained spectrum sharing model, and sends a first model indication message carrying parameters of the spectrum sharing model to the first base station CU, wherein the first base station CU and the second base station CU belong to base station CUs within the target spectrum optimization range, and the second model training data is sent by the second base station CU to the OAM node; the model acquisition unit 1030 can be used to: obtain the spectrum sharing model according to the first model indication message.

[0206] In some embodiments of the present disclosure, the first request response message includes: training request indication information and training data provider list information, and the training request indication information includes: model feature input variable information. And, the data acquisition unit 1020 can also be used to: obtain the training feature data of the first base station and the training feature data of the neighboring base stations of the first base station according to the model feature input variable information and the training data provider list information; generate measurement configuration information according to the model feature input variable information, and send the measurement configuration information to the first terminal device through the first base station distribution unit DU, so that the first terminal device obtains the training feature data of the first terminal device according to the measurement configuration information, and sends it to the first base station CU, wherein the first terminal device is within the coverage of the first base station, or the first terminal device is in the PLMN list managed by the first base station; send the training feature data of the first base station, the training feature data of the neighboring base stations of the first base station, and the training feature data of the first terminal device to the OAM node.

[0207] In some embodiments of the present disclosure, the first message transceiver unit 1010 can also be used to: send a first model request message to the OAM node, so that the OAM node determines the target spectrum optimization range according to the first model request message, sets the first base station as the anchor base station, and sends a first request response message to the first base station CU; receive the first request response message sent by the OAM node.

[0208] In some embodiments of the present disclosure, the first request response message includes: anchor base station indication information. And the first message transceiving unit 1010 may also be used to: determine that the first base station is an anchor base station according to the anchor base station indication information.

[0209] In some embodiments of the present disclosure, the second model training data is acquired by the second base station CU according to the second request response message sent by the OAM node, and the second request response message is sent by the OAM node to the second base station CU in response to the second model request message sent by the second base station CU; wherein the second model training data includes: training feature data of the second base station, training feature data of the neighboring base stations of the second base station, and training feature data of the second terminal device, and the second terminal device is within the coverage range of the second base station, or the second terminal device is in the PLMN list managed by the second base station.

[0210] In some embodiments of the present disclosure, the second model request message includes: identification information of the second base station CU, and the identification information of the second base station CU is used by the OAM node to set the second base station as an auxiliary base station; and, the second request response message includes: auxiliary base station indication information and anchor base station information, and the auxiliary base station indication information is used by the second base station CU to determine that the second base station is an auxiliary base station, and the anchor base station information is used by the second base station CU to determine the anchor base station.

[0211] In some embodiments of the present disclosure, the second base station CU may obtain the spectrum sharing model according to a second model indication message carrying parameters of the spectrum sharing model sent by the OAM node.

[0212] In some embodiments of the present disclosure, the first model indication message includes: a list of auxiliary base stations. And, the communication device 1000 may also include a model reasoning unit 1040, which can be used to: determine the second base station CU according to the list of auxiliary base stations; perform model reasoning through the second base station CU based on the spectrum sharing model to obtain a model reasoning result, so that the first base station and the second base station can achieve spectrum sharing communication according to the model reasoning result.

[0213] In some embodiments of the present disclosure, the model inference unit 1040 can also be used to: obtain first model inference data, wherein the first model inference data includes target feature data of the first base station, target feature data of the neighboring base stations of the first base station, and target feature data of the first terminal device; send a data request message to the second base station CU, and receive second model inference data sent by the second base station CU, wherein the second model inference data includes the target feature data of the second base station, target feature data of the neighboring base stations of the second base station, and target feature data of the second terminal device; based on the spectrum sharing model, perform model inference according to the first model inference data and the second model inference data to obtain a model inference result; and based on the model inference result, realize spectrum sharing communication with the second base station CU.

[0214] In some embodiments of the present disclosure, the model reasoning unit 1040 may also be used to: receive a data request reply message sent by the second base station CU in response to the data request message.

[0215] In some embodiments of the present disclosure, the data request message includes the address information of the first base station CU, and the data request reply message includes the address information of the second base station CU. In addition, the model reasoning unit 1040 can also be used to: establish a user plane channel with the second base station CU according to the address information of the first base station CU and the address information of the second base station CU, so as to transmit data through the user plane channel.

[0216] In some embodiments of the present disclosure, the data request reply message includes a data request failure and a request retransmission time. Also, the model reasoning unit 1040 may be further configured to: parse the data request reply message, determine that the data request failed, and send a data request message to the second base station CU according to the request retransmission time.

[0217] Fig.11 FIG. 2 shows a schematic diagram of the structure of a communication device according to another embodiment of the present disclosure. The communication device can be applied to an OAM node, such as Fig.11 As shown, the communication device 1100 may include: a second message sending and receiving unit 1110 and a model training unit 1120.

[0218] Among them, the second message transceiver unit 1110 can be used to: receive a first model request message sent by the first base station CU, and send a first request response message to the first base station CU in response to the first model request message, so that the first base station CU obtains the first model training data according to the first request response message, and sends it to the OAM node; and receive a second model request message sent by the second base station CU, and send a second request response message to the second base station CU in response to the second model request message, so that the second base station CU obtains the second model training data according to the second request response message, and sends it to the OAM node. The model training unit 1120 can be used to: perform model training according to the first model training data and the second model training data to obtain a trained spectrum sharing model, and send a model indication message carrying the parameters of the spectrum sharing model to the first base station CU and the second base station CU, so that the first base station CU and the second base station CU obtain the spectrum sharing model.

[0219] Since the principle of solving the problem in the communication device embodiment is similar to that in the above method embodiment, the implementation of the communication device embodiment can refer to the implementation of the above method embodiment, and the repeated parts will not be repeated.

[0220] Fig.12 FIG. 1 shows a structural block diagram of an electronic device in an embodiment of the present disclosure. It should be noted that: Fig.12 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0221] like Fig.12 As shown, the electronic device 1200 includes a central processing unit (CPU) 1201, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1202 or a program loaded from a storage part 1208 into a random access memory (RAM) 1203. In the RAM 1203, various programs and data required for the operation of the electronic device 1200 are also stored. The CPU 1201, the ROM 1202, and the RAM 1203 are connected to each other via a bus 1204. An input / output (I / O) interface 1205 is also connected to the bus 1204.

[0222] The following components are connected to the I / O interface 1205: an input section 1206 including a keyboard, a mouse, etc.; an output section 1207 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1208 including a hard disk, etc.; and a communication section 12012 including a network interface card such as a LAN card, a modem, etc. The communication section 12012 performs communication processing via a network such as the Internet. A drive 1210 is also connected to the I / O interface 1205 as needed. A removable medium 1211, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1210 as needed, so that a computer program read therefrom is installed into the storage section 1208 as needed.

[0223] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part 12012, and / or installed from a removable medium 1211. When the computer program is executed by a central processing unit (CPU) 1201, the above-mentioned functions defined in the system of the present disclosure are executed.

[0224] It should be noted that the computer-readable medium shown in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, terminal device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, a terminal device or a device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, a terminal device or a device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0225] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0226] The units involved in the embodiments described in the present disclosure may be implemented by software or by hardware. The described units may also be arranged in a processor, for example, may be described as: a processor includes a sending unit, an acquiring unit, a determining unit and a first processing unit. Among them, the names of these units do not constitute a limitation on the unit itself under certain circumstances. For example, the first message transceiver unit may also be described as "a unit that sends a first model request message to an OAM node and receives a first request response message sent by the OAM node in response to the first model request message."

[0227] As another aspect, the present disclosure further provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiments; or may exist independently without being assembled into the electronic device. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the method described in the following embodiments. For example, the electronic device may implement the following Figure 2 The steps shown.

[0228] According to one aspect of the present disclosure, a computer program product or a computer program is provided, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in various optional implementations of the above-mentioned embodiments.

[0229] It should be understood that any number of elements in the drawings of the present disclosure is for illustration rather than limitation, and any naming is only for distinction rather than having any limiting meaning.

[0230] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0231] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A communication method, characterized in that: The method comprises: The first base station centralized unit CU sends a first model request message to the operation maintenance and management OAM node, and receives a first request response message sent by the OAM node in response to the first model request message; The first base station CU obtains first model training data according to the first request response message, and sends the first model training data to the OAM node, so that the OAM node performs model training according to the first model training data and the second model training data, obtains a trained spectrum sharing model, and sends a first model indication message carrying parameters of the spectrum sharing model to the first base station CU; wherein the first base station CU and the second base station CU belong to base station CUs within a target spectrum optimization range, and the second model training data is sent by the second base station CU to the OAM node; Acquiring, by the first base station CU, the spectrum sharing model according to the first model indication message; The first model indication message includes: a list of secondary base stations; The method also includes: the first base station CU determines the second base station CU according to the auxiliary base station list; the first base station CU performs model reasoning through the second base station CU based on the spectrum sharing model to obtain a model reasoning result, so that the first base station and the second base station can achieve spectrum sharing communication according to the model reasoning result.

2. The method according to claim 1, characterized in that The first request response message includes: training request indication information and training data provider list information, and the training request indication information includes: model feature input variable information; The first base station CU acquires first model training data according to the first request response message, and sends the first model training data to the OAM node, including: The first base station CU acquires the training feature data of the first base station and the training feature data of neighboring base stations of the first base station according to the model feature input variable information and the training data provider list information; The first base station CU generates measurement configuration information according to the model feature input variable information, and sends the measurement configuration information to the first terminal device through the first base station distribution unit DU, so that the first terminal device obtains the training feature data of the first terminal device according to the measurement configuration information, and sends it to the first base station CU; wherein the first terminal device is within the coverage of the first base station, or the first terminal device is in the public land mobile network PLMN list managed by the first base station; The first base station CU sends the training feature data of the first base station, the training feature data of the neighboring base stations of the first base station, and the training feature data of the first terminal device to the OAM node.

3. The method according to claim 1, characterized in that The first base station centralized unit CU sends a first model request message to an operation, maintenance and management OAM node, and receives a first request response message sent by the OAM node in response to the first model request message, including: The first base station CU sends the first model request message to the OAM node, so that the OAM node determines the target spectrum optimization range according to the first model request message, sets the first base station as an anchor base station, and sends the first request response message to the first base station CU; The first base station CU receives the first request response message sent by the OAM node.

4. The method according to claim 3, characterized in that The first request response message includes: anchor base station indication information; Wherein, after the first base station CU receives the first request response message sent by the OAM node, the method further includes: The first base station CU determines that the first base station is the anchor base station according to the anchor base station indication information.

5. The method according to claim 1, characterized in that The second model training data is acquired by the second base station CU according to a second request response message sent by the OAM node, and the second request response message is sent by the OAM node to the second base station CU in response to a second model request message sent by the second base station CU; The second model training data includes: training feature data of the second base station, training feature data of the neighboring base stations of the second base station and training feature data of the second terminal device, and the second terminal device is within the coverage of the second base station, or the second terminal device is in the PLMN list managed by the second base station.

6. The method according to claim 5, characterized in that The second model request message includes: identification information of the second base station CU, where the identification information of the second base station CU is used by the OAM node to set the second base station as an auxiliary base station; And, the second request response message includes: auxiliary base station indication information and anchor base station information, the auxiliary base station indication information is used by the second base station CU to determine the second base station as the auxiliary base station, and the anchor base station information is used by the second base station CU to determine the anchor base station.

7. The method according to claim 1, characterized in that The second base station CU may obtain the spectrum sharing model according to a second model indication message sent by the OAM node and carrying the parameters of the spectrum sharing model.

8. The method according to claim 7, characterized in that The first base station CU performs model reasoning through the second base station CU based on the spectrum sharing model to obtain a model reasoning result, including: The first base station CU acquires first model inference data; wherein the first model inference data includes: target feature data of the first base station, target feature data of neighboring base stations of the first base station, and target feature data of the first terminal device; The first base station CU sends a data request message to the second base station CU, and receives second model inference data sent by the second base station CU; wherein the second model inference data includes: target feature data of the second base station, target feature data of neighboring base stations of the second base station, and target feature data of the second terminal device; The first base station CU performs model reasoning based on the spectrum sharing model and the first model reasoning data and the second model reasoning data to obtain a model reasoning result; The first base station CU realizes spectrum sharing communication with the second base station CU according to the model inference result.

9. The method according to claim 8, characterized in that After the first base station CU sends a data request message to the second base station CU, the method further includes: The first base station CU receives a data request reply message sent by the second base station CU in response to the data request message.

10. The method according to claim 9, characterized in that The data request message includes the address information of the first base station CU, and the data request reply message includes the address information of the second base station CU; Wherein, the method further comprises: The first base station CU establishes a user plane channel with the second base station CU according to the address information of the first base station CU and the address information of the second base station CU, so as to transmit data through the user plane channel.

11. The method according to claim 9, characterized in that The data request reply message includes data request failure and request resend time; Wherein, after the first base station CU receives the data request reply message sent by the second base station CU in response to the data request message, the method further includes: The first base station CU parses the data request reply message, determines that the data request fails, and sends the data request message to the second base station CU according to the request retransmission time.

12. The method according to any one of claims 1 to 11, characterized in that: The spectrum sharing model is an artificial intelligence / machine learning model.

13. A communication method, characterized in that: The method comprises: The OAM node receives a first model request message sent by the first base station CU, and sends a first request response message to the first base station CU in response to the first model request message, so that the first base station CU obtains the first model training data according to the first request response message and sends it to the OAM node; The OAM node receives a second model request message sent by the second base station CU, and sends a second request response message to the second base station CU in response to the second model request message, so that the second base station CU acquires second model training data according to the second request response message and sends it to the OAM node; The OAM node performs model training according to the first model training data and the second model training data to obtain a trained spectrum sharing model, sends a first model indication message carrying parameters of the spectrum sharing model to the first base station CU, and sends a second model indication message carrying parameters of the spectrum sharing model to the second base station CU, so that the first base station CU and the second base station CU obtain the spectrum sharing model; Among them, the first model indication message includes a list of auxiliary base stations, so that the first base station CU determines the second base station CU according to the list of auxiliary base stations, and performs model reasoning through the second base station CU based on the spectrum sharing model to obtain model reasoning results, so that the first base station and the second base station can realize spectrum sharing communication according to the model reasoning results.

14. A communication device, characterized in that: Applied to a first base station CU, the device includes: A first message transceiving unit, configured to send a first model request message to an OAM node, and receive a first request response message sent by the OAM node in response to the first model request message; A data acquisition unit, used to acquire first model training data according to the first request response message, send the first model training data to the OAM node, so that the OAM node performs model training according to the first model training data and the second model training data, obtains a trained spectrum sharing model, and sends a first model indication message carrying parameters of the spectrum sharing model to the first base station CU; wherein the first base station CU and the second base station CU belong to base station CUs within the target spectrum optimization range, and the second model training data is sent by the second base station CU to the OAM node; A model acquisition unit, configured to acquire the spectrum sharing model according to the first model indication message; Wherein, the first model indication message includes: a list of auxiliary base stations; The device also includes: a model reasoning unit, used to determine the second base station CU according to the auxiliary base station list; and based on the spectrum sharing model, perform model reasoning through the second base station CU to obtain a model reasoning result, so that the first base station and the second base station can achieve spectrum sharing communication according to the model reasoning result.

15. A communication device, characterized in that: Applied to an OAM node, the device comprises: A second message transceiver unit is configured to receive a first model request message sent by a first base station CU, and send a first request response message to the first base station CU in response to the first model request message, so that the first base station CU obtains first model training data according to the first request response message, and sends the first model training data to the OAM node; and receive a second model request message sent by a second base station CU, and send a second request response message to the second base station CU in response to the second model request message, so that the second base station CU obtains second model training data according to the second request response message, and sends the second model training data to the OAM node; a model training unit, configured to perform model training according to the first model training data and the second model training data, obtain a trained spectrum sharing model, send a first model indication message carrying parameters of the spectrum sharing model to the first base station CU, and send a second model indication message carrying parameters of the spectrum sharing model to the second base station CU, so that the first base station CU and the second base station CU obtain the spectrum sharing model; Among them, the first model indication message includes a list of auxiliary base stations, so that the first base station CU determines the second base station CU according to the list of auxiliary base stations, and performs model reasoning through the second base station CU based on the spectrum sharing model to obtain model reasoning results, so that the first base station and the second base station can realize spectrum sharing communication according to the model reasoning results.

16. An electronic device, characterized in that: include: one or more processors; A storage device configured to store one or more programs, which, when executed by the one or more processors, enables the one or more processors to implement the communication method as described in any one of claims 1 to 12, or implement the communication method as described in claim 13.

17. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the computer program implements the communication method according to any one of claims 1 to 12, or implements the communication method according to claim 13.

Images