Network intelligent energy-saving method, network intelligent energy-saving device, medium and electronic equipment

Abstract

The application provides a network intelligent energy-saving method, a network intelligent energy-saving device, a computer readable storage medium and an electronic device, relates to the technical field of communication, and can realize model training and model reasoning through multi-terminal signaling interaction, so that the base station can be controlled by the management end to accurately collect data of the terminal, precise energy saving is realized, network energy consumption is reduced, and network energy efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of communication technology, and more specifically, to a network intelligent energy-saving method, a network intelligent energy-saving device, a computer-readable storage medium, and an electronic device. Background Technology

[0002] To reduce the energy consumption of 5G base station equipment, save on 5G network operating costs, and achieve energy conservation and emission reduction in the telecommunications industry, existing publicly available solutions involve: using artificial intelligence (AI) / machine learning (ML) to uniformly configure energy-saving strategies for 4G / 5G base stations at the network level; constructing a base station service load prediction model based on historical service data of cells within a selected network area; generating network energy-saving strategies; and then executing energy-saving tasks for each base station according to these strategies. However, the energy-saving accuracy of these solutions is not high.

[0003] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this application, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0004] The purpose of this application is to provide a network intelligent energy-saving method, a network intelligent energy-saving device, a computer-readable storage medium, and an electronic device, which can realize model training and model inference through multi-terminal signaling interaction. This facilitates the base station to accurately collect data from the terminal under the control of the management end, thereby achieving precise energy saving, reducing network energy consumption, and improving network energy efficiency.

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

[0006] According to one aspect of this application, a network intelligent energy-saving method is provided, the method comprising:

[0007] The first base station receives model update information and sends a first request message to the second base station according to the energy consumption optimization range to obtain auxiliary data required for model inference from the second base station; wherein, the first request message is used to provide feedback through the control plane and / or user plane, and the first request message includes the first base station node ID; the first base station establishes a preset interface to realize the interaction of AI / ML data;

[0008] After receiving the first request message, if the second base station needs to provide feedback through the first preset interface, the first request message includes the first base station node ID, measurement information indication, measurement indication, reporting characteristics, and reporting period. If the second base station needs to provide feedback through the second preset interface, the first request message also includes the first base station IP address and port address.

[0009] The second base station receives a first request message from the first base station, and replies with a message indicating the measurement object requested by the model inference data to initiate the measurement, and obtains the measurement object information, including the second base station IP address and port address, message type, first base station measurement ID, second base station measurement ID, and critical diagnostic IE;

[0010] If the second base station is unable to provide measurement results, the second base station sends a first failure message to the first base station; wherein the first failure message includes the second base station ID, a description of the failure reason, and a retransmission waiting time;

[0011] The second base station reports a first reply message to the first base station, which includes the requested measurement information. The first reply message includes message type, first base station measurement ID, second base station measurement ID, measurement result feedback, terminal measurement result feedback including terminal location information, terminal moving speed, terminal wireless measurement information, and timestamp, and base station measurement result feedback including base station energy consumption status, base station energy efficiency, base station wireless resource status, and timestamp.

[0012] In one exemplary embodiment of this application, the method further includes:

[0013] The first base station generates the dataset required for model inference based on its own periodic measurement reports, first reply messages, and periodic measurement data from the terminal. The periodic measurement data from the terminal includes RRM measurement information and MDT measurement information. The RRM measurement information includes RSRP, RSRQ, and timestamps. The MDT measurement information includes terminal location information, terminal movement speed, and timestamps.

[0014] The first base station inputs the dataset required for model inference into the latest model so that the latest model can output the corresponding inference results; the inference results are used as the basis for network analysis and include prediction information or energy-saving decision information.

[0015] The first base station generates a second interface message based on the inference results; the inference results include energy-saving analysis identifier, root cause, recommended energy-saving cell ID list, recommended candidate cell ID list, recommended energy-saving operation, load prediction, base station energy efficiency prediction, and resource status prediction.

[0016] The first base station sends the second interface message to the second base station through the preset interface control plane and / or user plane;

[0017] The first and second base stations infer network energy-saving actions based on the model. If the output is a handover strategy, a cell is selected for the terminal before the terminal hands over.

[0018] In one exemplary embodiment of this application, the method further includes:

[0019] The management terminal receives energy consumption parameters periodically reported by each base station within the preset control range corresponding to the management terminal;

[0020] When an energy consumption parameter greater than a preset threshold is detected, the management terminal determines the energy consumption optimization range based on the base station corresponding to the energy consumption parameter greater than the preset threshold; the energy consumption optimization range includes the first base station and the second base station.

[0021] In one exemplary embodiment of this application, the method further includes:

[0022] The management terminal trains a management terminal model based on the historical dataset and obtains the model configuration information corresponding to the trained management terminal model. The historical dataset is divided into an offline training dataset, an offline validation dataset, and an offline test dataset. The historical dataset includes historical measurement data reported by each base station within the energy consumption optimization range.

[0023] The management terminal sends model configuration information to the first base station, including anchor base station indication information, a list of second base station IDs within the energy consumption optimization range, model index, and feature input information.

[0024] The management terminal sends model configuration information, including model index and feature input information, to the second base station.

[0025] The model index includes model use cases, model categories, and model parameters; the model parameters include weights, biases, learning rates, iteration counts, and SVM support vectors; the feature input information includes feature input statistics from the terminal and feature input from the base station; the feature input statistics from the terminal include terminal location information, terminal movement speed, and terminal wireless measurement information; the terminal wireless measurement information includes RSRP and RSRQ; and the feature input from the base station includes base station energy consumption status, base station energy efficiency, and base station wireless resource status.

[0026] In one exemplary embodiment of this application, the method further includes:

[0027] After receiving the model configuration information, the first base station identifies the local base station as the anchor base station, and calls and updates the latest model according to the model configuration information, as well as performs measurement configuration for each terminal in the designated area of ​​the first base station; among which, the measurement configuration includes MDT measurement configuration and RRM measurement configuration;

[0028] The first base station sends a first data collection request to each terminal within the reachable area;

[0029] Each terminal periodically responds to the first data acquisition request by reporting the real-time first terminal measurement data it has collected to the first base station, and performs specific measurement operations based on the MDT and RRM measurement configuration and stores the measurement results; wherein, the terminal measurement data includes terminal location information, terminal moving speed, timestamp, terminal RSRP, terminal RSRQ, and terminal SINR;

[0030] The first base station reports the first base station statistical data and the first terminal measurement data to the management terminal through the model training input message. The model training input message includes the anchor base station identifier and the required training data. The required training data includes the required terminal training data and the required base station training data. The required terminal training data includes the terminal location information, the terminal moving speed, the terminal wireless measurement information, and the timestamp. The required base station training data includes the base station energy consumption status, the base station energy efficiency, and the base station wireless resource status.

[0031] In one exemplary embodiment of this application, the method further includes:

[0032] After receiving the model configuration information, the second base station identifies the local base station as the auxiliary base station. Based on the feature input information indicated in the model configuration information, it sends the received second terminal measurement data and second base station statistical data to the management terminal through the model training input message. The model training input message includes the auxiliary base station node indication and the required training data. The required training data includes the required terminal training data and the required base station training data. The required terminal training data includes terminal location information, terminal movement speed, terminal wireless measurement information, and timestamp. The required base station training data includes base station energy consumption status, base station energy efficiency, and base station wireless resource status.

[0033] In one exemplary embodiment of this application, the method further includes:

[0034] The management terminal receives statistical data from the first base station, measurement results from the first terminal, statistical data from the second base station, and measurement data from the second terminal.

[0035] The management system generates training datasets, validation datasets, and test datasets based on the statistical data of the first base station, the measurement results of the first terminal, the statistical data of the second base station, and the measurement data of the second terminal.

[0036] The management system updates the management model online based on the training dataset, validation dataset, and test dataset.

[0037] The management terminal obtains the model update information corresponding to the updated management terminal model;

[0038] The management terminal sends model update information to the first base station and the second base station; the model update information includes the model index, which includes model use cases, model category, and model parameters.

[0039] In one exemplary embodiment of this application, the method further includes:

[0040] After performing energy-saving operations, the first base station sends a first feedback message to the management terminal; wherein, the first feedback message is used to indicate the first network data required for detecting the performance of the latest model, the first network data is used to optimize the latest model, and the first network data includes energy efficiency values; the first feedback message includes base station energy consumption status, base station energy efficiency, and base station radio resource status;

[0041] After performing energy-saving operations, the second base station sends a second feedback message to the management terminal. The second feedback message is used to indicate the second network data required for testing the performance of the second base station model. The second network data is used to optimize the second base station model and includes energy efficiency values. The second feedback message includes base station energy consumption status, base station energy efficiency, and base station radio resource status.

[0042] In one exemplary embodiment of this application, the method further includes:

[0043] When the network energy-saving status is detected to meet the preset stable state, the management terminal sends a stop message to the first base station and the second base station to indicate the exit of the energy-saving mechanism; the stop message includes network energy-saving suspension indication information and trigger reason information, and the trigger reason includes set threshold comparison information.

[0044] According to one aspect of this application, a network-based intelligent energy-saving device is provided, comprising:

[0045] The first base station is used to receive model update information and send a first request message to the second base station according to the energy consumption optimization range to obtain auxiliary data required for model inference from the second base station; wherein, the first request message is used to provide feedback through the control plane and / or user plane, and the first request message includes the first base station node ID; the first base station establishes a preset interface to realize the interaction of AI / ML data;

[0046] The second base station is used to, after receiving the first request message, if the second base station needs to provide feedback through the first preset interface, the first request message includes the first base station node ID, measurement information indication, measurement indication, reporting characteristics, and reporting period. If the second base station needs to provide feedback through the second preset interface, the first request message also includes the first base station IP address and port address.

[0047] The second base station is used to receive the first request message from the first base station, reply with a message indicating the measurement object requested by the model inference data, so as to start the measurement, obtain the measurement object information, and indicate the IP address and port address of the second base station, the message type, the measurement ID of the first base station, the measurement ID of the second base station, and the critical diagnostic IE;

[0048] If the second base station cannot provide measurement results, the second base station is used to send a first failure message to the first base station; wherein the first failure message includes the second base station ID, a description of the failure reason, and a retransmission waiting time;

[0049] The second base station is used to report a first reply message to the first base station, which includes the measurement information requested. The first reply message includes message type, first base station measurement ID, second base station measurement ID, measurement result feedback, terminal measurement result feedback including terminal location information, terminal moving speed, terminal wireless measurement information, and timestamp, and base station measurement result feedback including base station energy consumption status, base station energy efficiency, base station wireless resource status, and timestamp.

[0050] In one exemplary embodiment of this application, wherein:

[0051] The first base station is used to generate the dataset required for model inference based on the base station's own periodic measurement reports, first reply messages, and periodic measurement data of the terminal; wherein, the periodic measurement data of the terminal includes RRM measurement information and MDT measurement information. The RRM measurement information includes RSRP, RSRQ, and timestamp, and the MDT measurement information includes terminal location information, terminal moving speed, and timestamp.

[0052] The first base station is used to input the dataset required for model inference into the latest model so that the latest model can output the corresponding inference results; wherein, the inference results are used as the basis for network analysis, and the inference results include prediction information or energy-saving decision information;

[0053] The first base station is used to generate a second interface message based on the inference results; wherein, the inference results include energy-saving analysis identifier, root cause, recommended energy-saving cell ID list, recommended candidate cell ID list, recommended energy-saving operation, load prediction, base station energy efficiency prediction, and resource status prediction;

[0054] The first base station is used to send the second interface message to the second base station through a preset interface control plane and / or user plane;

[0055] The first and second base stations are used to infer network energy-saving actions based on the model. If the output is a handover strategy, a cell is selected for the terminal before the terminal hands over.

[0056] In one exemplary embodiment of this application, wherein:

[0057] The management terminal receives energy consumption parameters periodically reported by each base station within the preset control range corresponding to the management terminal;

[0058] When an energy consumption parameter greater than a preset threshold is detected, the management terminal determines the energy consumption optimization range based on the base station corresponding to the energy consumption parameter greater than the preset threshold; the energy consumption optimization range includes the first base station and the second base station.

[0059] In one exemplary embodiment of this application, wherein:

[0060] The management terminal is used to train the management terminal model based on the historical dataset and obtain the model configuration information corresponding to the trained management terminal model. The historical dataset is divided into an offline training dataset, an offline validation dataset, and an offline test dataset. The historical dataset includes historical measurement data reported by each base station within the energy consumption optimization range.

[0061] The management terminal is used to send model configuration information to the first base station, including anchor base station indication information, a list of second base station IDs within the energy consumption optimization range, model index, and feature input information.

[0062] The management terminal is used to send model configuration information, including model index and feature input information, to the second base station;

[0063] The model index includes model use cases, model categories, and model parameters; the model parameters include weights, biases, learning rates, iteration counts, and SVM support vectors; the feature input information includes feature input statistics from the terminal and feature input from the base station; the feature input statistics from the terminal include terminal location information, terminal movement speed, and terminal wireless measurement information; the terminal wireless measurement information includes RSRP and RSRQ; and the feature input from the base station includes base station energy consumption status, base station energy efficiency, and base station wireless resource status.

[0064] In one exemplary embodiment of this application, wherein:

[0065] The first base station is used to determine the local base station as the anchor base station after receiving the model configuration information, and to call and update the latest model according to the model configuration information, as well as to perform measurement configuration for each terminal within the designated area of ​​the first base station; wherein, the measurement configuration includes MDT measurement configuration and RRM measurement configuration.

[0066] The first base station is used to send the first data collection request to each terminal within the reachable area;

[0067] Each terminal is used to periodically respond to the first data acquisition request, report the collected real-time first terminal measurement data to the first base station, and perform specific measurement operations based on the MDT and RRM measurement configuration and store the measurement results; wherein, the terminal measurement data includes terminal location information, terminal moving speed, timestamp, terminal RSRP, terminal RSRQ, and terminal SINR;

[0068] The first base station is used to report the statistical data of the first base station and the measurement data of the first terminal to the management terminal through the model training input message. The model training input message includes the anchor base station identifier and the required training data. The required training data includes the required terminal training data and the required base station training data. The required terminal training data includes the terminal location information, the terminal moving speed, the terminal wireless measurement information, and the timestamp. The required base station training data includes the base station energy consumption status, the base station energy efficiency, and the base station wireless resource status.

[0069] In one exemplary embodiment of this application, wherein:

[0070] The second base station is used to identify the local base station as the auxiliary base station after receiving the model configuration information, and to send the received second terminal measurement data and second base station statistical data to the management terminal through the model training input message according to the feature input information indicated in the model configuration information. The model training input message includes auxiliary base station node indication and required training data. The required training data includes required terminal training data and required base station training data. The required terminal training data includes terminal location information, terminal movement speed, terminal wireless measurement information, and timestamp. The required base station training data includes base station energy consumption status, base station energy efficiency, and base station wireless resource status.

[0071] In one exemplary embodiment of this application, wherein:

[0072] The management terminal is used to receive statistical data from the first base station, measurement results from the first terminal, statistical data from the second base station, and measurement data from the second terminal.

[0073] The management terminal is used to generate training datasets, validation datasets, and test datasets based on the statistical data of the first base station, the measurement results of the first terminal, the statistical data of the second base station, and the measurement data of the second terminal.

[0074] The management terminal is used to update the management terminal model online based on the training dataset, validation dataset, and test dataset.

[0075] The management terminal is used to obtain the model update information corresponding to the updated management terminal model.

[0076] The management terminal is used to send model update information to the first base station and the second base station; the model update information includes the model index, which includes model use cases, model category, and model parameters.

[0077] In one exemplary embodiment of this application, wherein:

[0078] The first base station is used to send a first feedback message to the management terminal after performing energy-saving operations; wherein, the first feedback message is used to indicate the first network data required for detecting the performance of the latest model, the first network data is used to optimize the latest model, and the first network data includes energy efficiency values; the first feedback message includes base station energy consumption status, base station energy efficiency, and base station radio resource status;

[0079] The second base station is used to send a second feedback message to the management terminal after performing energy-saving operations; wherein, the second feedback message is used to indicate the second network data required for detecting the performance of the second base station model, the second network data is used to optimize the second base station model, and the second network data includes energy efficiency values; the second feedback message includes base station energy consumption status, base station energy efficiency, and base station radio resource status.

[0080] In one exemplary embodiment of this application, wherein:

[0081] When the network energy-saving status is detected to meet the preset stable state, the management terminal sends a stop message to the first base station and the second base station to indicate the exit of the energy-saving mechanism; the stop message includes network energy-saving pause indication information and trigger reason information, and the trigger reason includes set threshold comparison information.

[0082] According to one aspect of this application, a computer-readable storage medium is provided, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the method of any one of the above.

[0083] According to one aspect of this application, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform the method of any of the above by executing the executable instructions.

[0084] The exemplary embodiments of this application may have some or all of the following beneficial effects:

[0085] In an example embodiment of the network intelligent energy-saving method provided in this application, a first base station receives model update information and sends a first request message to a second base station according to the energy consumption optimization range to obtain auxiliary data required for model inference from the second base station; wherein, the first request message is used to provide feedback through the control plane and / or user plane, and the first request message includes the first base station node ID; the first base station establishes a preset interface to realize the interaction of AI / ML data; after receiving the first request message, if the second base station needs to provide feedback through the first preset interface, the first request message includes the first base station node ID, measurement information indication, measurement indication, reporting characteristics, and reporting period; if the second base station needs to provide feedback through the second preset interface, the first request message also includes the first base station IP address and port address; the second base station receives the first request message from the first base station and... The model inference data response message indicates the requested measurement object to initiate measurement, obtain measurement object information, and indicates the second base station IP address and port address, message type, first base station measurement ID, second base station measurement ID, and critical diagnostic IE. If the second base station cannot provide measurement results, it sends a first failure message to the first base station, including the second base station ID, a description of the failure reason, and a retransmission wait time. The second base station then reports a first response message to the first base station, reporting the requested measurement information. The first response message includes the message type, first base station measurement ID, second base station measurement ID, measurement result feedback, terminal measurement result feedback including terminal location information, terminal movement speed, terminal wireless measurement information, and a timestamp, and base station measurement result feedback including base station energy consumption status, base station energy efficiency, base station wireless resource status, and a timestamp. This multi-terminal signaling interaction enables model training and model inference, facilitating precise data collection from the terminal by the base station under the control of the management end, thereby achieving precise energy saving, reducing network energy consumption, and improving network energy efficiency.

[0086] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0087] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0088] Figure 1 This illustration schematically shows a diagram of an applicable network system according to one embodiment of the present application;

[0089] Figure 2 A flowchart illustrating a network-based intelligent energy-saving method according to an embodiment of this application is shown schematically.

[0090] Figure 3 A flowchart illustrating another embodiment of a network-based intelligent energy-saving method according to this application is shown schematically;

[0091] Figure 4 This schematic diagram illustrates a structural block diagram of a network-based intelligent energy-saving device according to one embodiment of the present application;

[0092] Figure 5 The schematic diagram illustrates the structure of a computer system suitable for implementing the electronic devices of the present application. Detailed Implementation

[0093] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make this application more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of the embodiments of this application. However, those skilled in the art will recognize that the technical solutions of this application can be practiced with one or more of the specific details omitted, or other methods, components, apparatus, steps, etc., can be employed. In other instances, well-known technical solutions are not shown or described in detail to avoid obscuring various aspects of this application.

[0094] Furthermore, the accompanying drawings are merely illustrative of this application and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0095] Please see Figure 1 , Figure 1 A schematic diagram of an applicable network system according to one embodiment of this application is shown. Figure 1As shown, the network system may include terminal 114, access network device 110, access network device 113, terminal 124, access network device 120, access network device 123, and management terminal (OAM) 130. Access network device 110 and access network device 113 may belong to different mobile communication systems, and similarly, access network device 120 and access network device 123 may belong to different mobile communication systems. Examples of mobile communication systems include: Global System for Mobile Communication (GSM), Evolved Universal Terrestrial Radio Access (E-UTRA), Universal Mobile Telecommunications System (UMTS) and its evolutions, Long Term Evolution (LTE) and various versions based on LTE evolution, 5th Generation (5G) communication systems, and next-generation communication systems such as New Radio (NR). Furthermore, the above-mentioned communication systems can also be applied to future-oriented communication technologies, and all are applicable to the technical solutions provided in the embodiments of this application.

[0096] For example, access network devices 110 and 120 can be evolved Node B (eNB) base stations in a 4G mobile communication system, or next generation Node B (gNB) base stations in a 5G mobile communication system. For ease of description, the access network devices in a 4G mobile communication system can be referred to as 4G access network devices or 4G base stations, and the access network devices in a 5G mobile communication system can be referred to as 5G access network devices or 5G base stations.

[0097] Access network devices 110 and 120 can be access network devices in a stand-alone (SA) network architecture. Access network device 110 can provide wireless network coverage 111, access network device 113 can provide wireless network coverage 112, access network device 120 can provide wireless network coverage 121, and access network device 123 can provide wireless network coverage 122. It can be understood that when access network device 113 is an access network device for a 5G mobile communication system and access network device 110 is an access network device for a 4G mobile communication system, since the spectrum of 5G mobile communication systems is generally higher than that of 4G mobile communication systems, the range of wireless network coverage 112 is smaller than the range of wireless network coverage 111, and is within the range of wireless network coverage 111. Similarly, when access network device 123 is an access network device for a 5G mobile communication system and access network device 120 is an access network device for a 4G mobile communication system, since the spectrum of 5G mobile communication systems is generally higher than that of 4G mobile communication systems, the range of wireless network coverage 122 is smaller than the range of wireless network coverage 121, and is within the range of wireless network coverage 121.

[0098] Terminals 114 and 124 can support multiple mobile communication systems, such as 4G and 5G mobile communication systems. Terminals 114 and 124 can be distributed in... Figure 1In the network system shown, devices can be stationary or mobile. In some embodiments of this application, terminals 114 and 124 can be mobile devices, mobile stations, mobile units, wireless units, remote units, user agents, mobile clients, etc. More specifically, terminal 114 can be user equipment (UE), handheld terminals, laptops, subscriber units, cellular phones, smartphones, wireless data cards, personal digital assistant (PDA) computers, tablet computers, wireless modems, handheld devices, laptop computers, cordless phones, wireless local loop (WLL) stations, machine type communication (MTC) terminals, or other devices capable of accessing the network. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices running iOS, Android, Microsoft, or other operating systems. The aforementioned portable electronic devices can also be other portable electronic devices, such as laptops with touch-sensitive surfaces (e.g., touch panels). It should also be understood that in some other embodiments of this application, terminal 114 may not be a portable electronic device, but a desktop computer with a touch-sensitive surface (e.g., a touch panel). This application does not specifically limit the type of electronic device.

[0099] For example, Figure 1The architecture shown can be implemented as follows: Access network device 110 can be understood as a first base station, and access network device 120 can be understood as a second base station. The first base station receives model update information and sends a first request message to the second base station according to the energy consumption optimization range to obtain auxiliary data required for model inference from the second base station; wherein, the first request message is used to provide feedback through the control plane and / or user plane, and the first request message includes the first base station node ID; the first base station establishes a preset interface to realize the interaction of AI / ML data; after receiving the first request message, if the second base station needs to provide feedback through the first preset interface, the first request message includes the first base station node ID, measurement information indication, measurement indication, reporting features, and reporting period; if the second base station needs to provide feedback through the second preset interface, the first request message also includes the first base station IP address and port address; the second base station receives the first request message from the first base station and replies with message indications through model inference data. For the requested measurement object, the measurement is initiated, and measurement object information is obtained, including the IP address and port address of the second base station, message type, measurement ID of the first base station, measurement ID of the second base station, and critical diagnostic IE. If the second base station cannot provide measurement results, the second base station sends a first failure message to the first base station. The first failure message includes the second base station ID, a description of the failure reason, and a retransmission wait time. The second base station reports a first reply message to the first base station, reporting the requested measurement information. The first reply message includes the message type, measurement ID of the first base station, measurement ID of the second base station, measurement result feedback, terminal measurement result feedback including terminal location information, terminal movement speed, terminal wireless measurement information, and timestamp, and base station measurement result feedback including base station energy consumption status, base station energy efficiency, base station wireless resource status, and timestamp.

[0100] Please see Figure 2 , Figure 2 A flowchart illustrating a network-based intelligent energy-saving method according to an embodiment of this application is shown. Figure 2 As shown, the network intelligent energy-saving method may include steps S210 to S250.

[0101] Step S210: The first base station receives model update information and sends a first request message to the second base station according to the energy consumption optimization range to obtain auxiliary data required for model inference from the second base station. The first request message is used for feedback through the control plane and / or user plane, and includes the first base station node ID. The first base station establishes a preset interface to realize the interaction of AI / ML data.

[0102] Step S220: After receiving the first request message, if the second base station needs to provide feedback through the first preset interface, the first request message includes the first base station node ID, measurement information indication, measurement indication, reporting characteristics, and reporting period. If the second base station needs to provide feedback through the second preset interface, the first request message also includes the first base station IP address and port address.

[0103] Step S230: The second base station receives the first request message from the first base station, and responds with a message indicating the measurement object requested by the model inference data to start the measurement, and obtains the measurement object information, indicating the second base station IP address and port address, message type, first base station measurement ID, second base station measurement ID, and critical diagnostic IE.

[0104] Step S240: If the second base station cannot provide measurement results, the second base station sends a first failure message to the first base station. The first failure message includes the second base station ID, a description of the failure reason, and a retransmission wait time.

[0105] Step S250: The second base station reports a first reply message to the first base station, including the requested measurement information. The first reply message includes message type, first base station measurement ID, second base station measurement ID, measurement result feedback, terminal measurement result feedback including terminal location information, terminal moving speed, terminal wireless measurement information, and timestamp, and base station measurement result feedback including base station energy consumption status, base station energy efficiency, base station wireless resource status, and timestamp.

[0106] Implementation Figure 2 The method shown can realize model training and model inference through multi-terminal signaling interaction, which is conducive to enabling base stations to accurately collect data from terminals under the control of the management terminal, thereby achieving precise energy saving, reducing network energy consumption, and improving network energy efficiency.

[0107] The steps described above in this example implementation will now be explained in more detail.

[0108] As an optional embodiment, the method further includes: the management terminal receiving energy consumption parameters periodically reported by each base station within a preset control range corresponding to the management terminal; when an energy consumption parameter greater than a preset threshold is detected, the management terminal determines the energy consumption optimization range based on the base station corresponding to the energy consumption parameter greater than the preset threshold; wherein the energy consumption optimization range includes the first base station and the second base station.

[0109] As can be seen, by implementing this optional embodiment, the model can be updated based on the signaling interaction between the terminal and the first base station and the signaling interaction between the first base station and the management terminal. The management terminal can further synchronize the updated model to each base station within the energy consumption optimization range, so that each base station can perform energy-saving operations based on the updated model, thereby achieving precise network energy saving.

[0110] The Operation and Maintenance (OAM) terminal is used to implement operation and maintenance management functions for base stations, and the energy consumption optimization range may include one or more base stations. Furthermore, a preset threshold is used to limit the maximum energy consumption. When the energy consumption parameter exceeds this preset threshold, it indicates that the energy consumption is too high and energy saving is needed, which can trigger the OAM terminal to determine the energy consumption optimization range. The preset threshold can be represented as a constant, and can be manually set or obtained by looking up a table; this embodiment does not impose any limitations.

[0111] In addition, optionally, when no energy consumption parameter greater than a preset threshold is detected, the above method may further include: repeatedly executing the above technical solution until an energy consumption parameter greater than the preset threshold is detected.

[0112] As an optional embodiment, the method further includes: the management terminal training a management terminal model based on a historical dataset, and obtaining the model configuration information corresponding to the trained management terminal model; wherein, the historical dataset is used to be divided into an offline training dataset, an offline verification dataset, and an offline test dataset, and the historical dataset includes historical measurement data reported by each base station within the energy consumption optimization range; the management terminal sends the model configuration information to the first base station and the second base station.

[0113] As can be seen, implementing this optional embodiment can achieve model consistency between the base station and the management terminal, thereby facilitating more accurate network energy saving.

[0114] The management model can be an AI / ML model. The model configuration information sent by the management terminal to the first base station may include: first base station indication information, second base station identifier list, model index information, and indication information of the model to be input.

[0115] Specifically, the first base station indication information corresponds to a Boolean / enumerated type. This information is used to instruct the base station receiving the model configuration information to act as the first base station and to instruct it to collect terminal measurement data from the second base station. The second base station identifier list includes identifiers (e.g., IDs) corresponding to each second base station. The model index information includes model use case information, model category information, and model parameters. The model use case information indicates various use case information corresponding to the Boolean / enumerated type (e.g., energy-saving use case information, load use case information, terminal trajectory prediction use case information, etc.). The model category information, indicating the model's processing method for training, includes one or more of the following: logistic regression information, decision tree information, support vector machine information, random forest information, etc. The model parameters include weight values, bias terms, learning rate, number of iterations, and the support vectors corresponding to the support vector machine.

[0116] The indication information of the input model is used to indicate the fields of the data that the first base station needs to collect, such as terminal location, terminal moving speed, terminal reference signal receiving power (RSRP), terminal signal receiving quality (RSRQ), terminal received signal strength indicator (RSSI), signal to interference plus noise ratio (SINR), base station energy consumption status (e.g., active, high, low, inactive), base station energy efficiency expressed as the ratio of data transmission rate to energy consumption, and base station radio resource status. Among them, the base station radio resource status can be used to indicate the usage of physical resource blocks by each service in the downlink and uplink in each cell and each single sideband (SSB) area, as well as the usage of resources at the granularity by the physical downlink control channel (PDCCH) scheduled in the downlink and uplink.

[0117] As an optional embodiment, the method further includes: after receiving the model configuration information, the first base station determines the local base station as the anchor base station, and calls and updates the latest model according to the model configuration information, and performs measurement configuration on each terminal within the designated area of ​​the first base station; wherein, the measurement configuration includes MDT measurement configuration and RRM measurement configuration; the first base station sends a first data acquisition request to each terminal within the reachable area; each terminal periodically responds to the first data acquisition request and reports the collected real-time first terminal measurement data to the first base station, and performs specific measurement operations based on the MDT and RRM measurement configuration and stores the measurement results; wherein, the terminal measurement data includes terminal location information, terminal moving speed, timestamp, terminal RSRP, terminal RSRQ, and terminal SINR; the first base station reports the first base station statistical data and the first terminal measurement data to the management terminal.

[0118] As can be seen, implementing this optional embodiment can achieve precise configuration and signaling interaction between the terminal and the first base station, enabling timely collection of terminal data.

[0119] The anchor base station is used to collect data from neighboring base station nodes required for model training. The MDT measurement configuration includes terminal location information, terminal movement speed, and timestamps. The latest model can be an AI / ML model; similarly, the second base station model can be an AI / ML model. Furthermore, the statistical data for the first base station may include: the first base station identifier, the first base station energy consumption status, the first base station energy efficiency, and the first base station radio resource status, etc., which are not limited in this embodiment.

[0120] Furthermore, the reachable area of ​​the first base station can be understood as the area covered by the first base station, and different first base stations correspond to different reachable areas. Specifically, the first base station sends a first data collection request to each terminal within the reachable area, including sending the data collection request to each terminal within the reachable area based on the Minimization of Drive-Tests (MDT) and / or RRM (Radio Resource Management Measurement) measurement method, to trigger each terminal to respond to the first data collection request. The first data collection request may include the following fields: trigger period, recording period, terminal location information, and terminal movement speed.

[0121] Furthermore, the first terminal measurement data may include terminal ID, terminal location information, enumerated values ​​of various parameters corresponding to the terminal, terminal serving cell ID, terminal moving speed, timestamp, RSRP, RSRQ, SINR, etc. The terminal location information may be Global Navigation Satellite System (GNSS) location information, radio frequency (RF) fingerprint information, etc., and is not limited in this embodiment.

[0122] As an optional embodiment, the method further includes: after receiving the model configuration information, the second base station determines the local base station as the auxiliary base station, and calls and updates the second base station model according to the model configuration information, and performs measurement configuration for each terminal within the designated area of ​​the second base station; wherein, the measurement configuration includes MDT measurement configuration and RRM measurement configuration; the second base station sends a second data acquisition request to each terminal within the reachable area; each terminal periodically responds to the second data acquisition request and reports the collected real-time terminal measurement data to the second base station, and performs specific measurement operations based on the MDT and RRM measurement configuration and stores the measurement results; wherein, the terminal measurement data includes terminal location information, terminal movement speed, timestamp, terminal RSRP, terminal RSRQ, and terminal SINR; the second base station reports the second base station statistical data and the second terminal measurement data to the management terminal.

[0123] As can be seen, implementing this optional embodiment can achieve precise configuration and signaling interaction between the terminal and the second base station, enabling timely collection of terminal data.

[0124] The statistical data of the second base station may include: the second base station identifier, the second base station energy consumption status, the second base station energy efficiency, the second base station radio resource status, etc., which are not limited in this application embodiment.

[0125] Furthermore, the reachable area of ​​the second base station can be understood as the area covered by the second base station, and different second base stations correspond to different reachable areas. Specifically, the second base station sends a second data collection request to each terminal within the reachable area, including sending the data collection request to each terminal within the reachable area based on the Minimization of Drive-Tests (MDT) and / or RRM (Radio Resource Management Measurement) measurement method, to trigger each terminal to respond to the second data collection request. The second data collection request may include the following fields: trigger period, recording period, terminal location information, and terminal movement speed.

[0126] Furthermore, the second terminal measurement data may include terminal ID, terminal location information, enumerated values ​​of various parameters corresponding to the terminal, terminal serving cell ID, terminal moving speed, timestamp, RSRP, RSRQ, SINR, etc. The terminal location information may be Global Navigation Satellite System (GNSS) location information, radio frequency (RF) fingerprint information, etc., and is not limited in this embodiment.

[0127] As an optional embodiment, the method further includes: the management terminal receiving statistical data from a first base station, measurement results from a first terminal, statistical data from a second base station, and measurement data from a second terminal; the management terminal generating a training dataset, a validation dataset, and a test dataset based on the statistical data from the first base station, measurement results from the first terminal, statistical data from the second base station, and measurement data from the second terminal; the management terminal updating the management terminal model online according to the training dataset, validation dataset, and test dataset; the management terminal obtaining model update information corresponding to the updated management terminal model; and the management terminal sending the model update information to the first base station and the second base station.

[0128] As can be seen, implementing this optional embodiment can enable timely model updates, thereby facilitating precise network energy saving.

[0129] The model update information only carries the changed information that needs to be updated. Specifically, the update information can be used to characterize the differences between the management terminal model before and after the update. In addition, the management terminal updates the management terminal model online based on the training dataset, validation dataset, and test dataset, including: training the management terminal model online using the training dataset, validating the inference performance of the trained management terminal model using the validation dataset, and testing the validated management terminal model using the test dataset.

[0130] In step S210, the first base station receives model update information and sends a first request message to the second base station according to the energy consumption optimization range to obtain auxiliary data required for model inference from the second base station; wherein, the first request message is used to provide feedback through the control plane and / or user plane, and the first request message includes the first base station node ID; the first base station establishes a preset interface to realize the interaction of AI / ML data.

[0131] The measurement result feedback includes, but is not limited to, terminal measurement result feedback and base station measurement result feedback. Furthermore, according to the energy consumption optimization range, a first request message is sent to the second base station, including: sending a first Xn request message to each second base station corresponding to the second base station identifier list, containing the first base station ID, feedback method (e.g., periodic feedback), feedback period (e.g., 10ms), measurement indication information, measurement identifier indication, reporting characteristics, and reporting period. The first Xn request message corresponds to the Xn interface, which is used for interconnection between NG-RAN nodes. After receiving the first Xn request message, the second base station can send a first reply message through the Xn-C interface based on control plane network elements and / or user plane network elements. The first reply message includes the data required as indicated in the first Xn request message.

[0132] In step S220, after the second base station receives the first request message, if the second base station needs to provide feedback through the first preset interface, the first request message includes the first base station node ID, measurement information indication, measurement indication, reporting characteristics, and reporting period. If the second base station needs to provide feedback through the second preset interface, the first request message also includes the first base station IP address and port address.

[0133] In step S230, the second base station receives a first request message from the first base station, and responds with a message indicating the measurement object requested by the model inference data to initiate measurement, and obtains measurement object information, including the second base station IP address and port address, message type, first base station measurement ID, second base station measurement ID, and critical diagnostic IE.

[0134] In step S240, if the second base station cannot provide measurement results, the second base station sends a first failure message to the first base station; wherein the first failure message includes the second base station ID, a description of the failure reason, and a retransmission waiting time.

[0135] In step S250, the second base station reports a first reply message to the first base station, which includes the measurement information requested. The first reply message includes message type, first base station measurement ID, second base station measurement ID, measurement result feedback, terminal measurement result feedback including terminal location information, terminal moving speed, terminal wireless measurement information, and timestamp, and base station measurement result feedback including base station energy consumption status, base station energy efficiency, base station wireless resource status, and timestamp.

[0136] The measurement information indication includes, but is not limited to, terminal measurement information indication and base station measurement information indication. The measurement indication refers to the information used by the transmitting base station to instruct the receiving base station to start, stop, and add measurement information during the measurement process. The reporting feature, when the measurement indication is "start," indicates the measurement object requested by the second base station for each location in the bitmap. The reporting period indicates the reporting period for periodic measurements.

[0137] As an optional embodiment, the method further includes: if the first request message includes the IP address and port address of the first base station, after receiving the first request message, the second base station sends a first reply message back to the first base station through a second preset interface (e.g., the Xn-U interface).

[0138] As can be seen, implementing this optional embodiment can realize signaling interaction between the first base station and the second base station, thereby facilitating precise network energy saving.

[0139] Before the second base station receives the first request message, the process further includes: the first base station sending a first Xn request message to each second base station corresponding to the second base station identifier list, containing the first base station ID, feedback method (e.g., periodic feedback), feedback period (e.g., 10ms), measurement indication information, measurement identifier indication, reporting characteristics, reporting period, first base station IP, and first base station port address. The first Xn request message corresponds to the Xn interface, which is used for interconnection between NG-RAN nodes. After receiving the first Xn request message, the second base station can feed back the second inference requirement data through the Xn-U interface based on control plane network elements and / or user plane network elements.

[0140] As an optional embodiment, the method further includes: if the second base station cannot provide measurement results, the second base station sends a first failure message to the first base station; wherein the first failure message includes the second base station node ID, a description of the failure reason, and a waiting time for retransmission.

[0141] As can be seen, implementing this optional embodiment can promptly provide a description of the failure when the data required by the first base station cannot be provided.

[0142] The failure reason description may include, but is not limited to, reasons at the wireless network layer, reasons at the transport layer, and reasons at the protocol layer. The retransmission wait time indicates the time required to resend the request.

[0143] As an optional embodiment, the method further includes: a first base station generating a dataset required for model inference based on its own periodic measurement reports, a first reply message, and periodic measurement data from the terminal; wherein the periodic measurement data from the terminal includes RRM measurement information and MDT measurement information, the RRM measurement information includes RSRP, RSRQ, and timestamp, and the MDT measurement information includes terminal location information, terminal moving speed, and timestamp; the first base station inputs the dataset required for model inference into the latest model so that the latest model outputs the corresponding inference result; wherein the inference result is used as the basis for network analysis, and the inference result includes prediction information or energy-saving decision information; the first base station generates a second interface message based on the inference result; wherein the inference result includes energy-saving analysis identifier, root cause, recommended energy-saving cell ID list, recommended candidate cell ID list, recommended energy-saving operation, load prediction, base station energy efficiency prediction, and resource status prediction; the first base station sends the second interface message to the second base station through a preset interface control plane and / or user plane; the first base station and the second base station execute network energy-saving actions based on the model inference output, and if the output is a handover strategy, then a cell is selected for the terminal before the terminal handover.

[0144] The inference results include: energy-saving analysis identifiers, energy consumption area identifiers, cause description information, a list of recommended energy-saving cell identifiers, a list of recommended candidate cell identifiers, recommended energy-saving operations, traffic load prediction information, base station energy efficiency trend prediction information, and base station resource status prediction information, etc., which are not limited in this embodiment. Specifically, the energy-saving analysis identifier indicates the energy-saving operation based on AI / ML; the energy consumption area identifier indicates the location of the suggested optimization area; the cause description information indicates the cause of energy consumption; the list of recommended energy-saving cell identifiers includes the IDs of one or more energy-saving cells; the list of recommended candidate cell identifiers includes the IDs of one or more candidate cells; the recommended energy-saving operation indicates the energy-saving strategy that can be adopted and the corresponding energy-saving operation; the traffic load prediction information provides prediction information related to traffic load trends; the base station energy efficiency trend prediction information provides predictions related to base station energy efficiency trends; and the base station resource status prediction information provides predictions related to base station resource status. Furthermore, energy-saving decision information may include cell activation, cell deactivation, symbol shutdown, channel shutdown, etc., which are not limited in this embodiment.

[0145] As an optional embodiment, the method further includes: a first base station sending a first feedback message to a management terminal after performing an energy-saving operation; wherein the first feedback message is used to indicate first network data required for detecting the performance of the latest model, the first network data is used to optimize the latest model, and the first network data includes an energy efficiency value; and a second base station sending a second feedback message to a management terminal after performing an energy-saving operation; wherein the second feedback message is used to indicate second network data required for detecting the performance of the second base station model, the second network data is used to optimize the second base station model, and the second network data includes an energy efficiency value.

[0146] As can be seen, implementing this optional embodiment can promptly feed back energy-saving information to the management terminal, so that the management terminal can update the management terminal model in a timely manner for the next precise network energy saving.

[0147] The energy efficiency value can be compared with the base station energy efficiency prediction indication information in the model inference output message to determine the model accuracy and realize the model training, testing, and verification process on the OAM side. The first feedback message / second feedback message may include base station energy consumption status, base station energy efficiency, base station radio resource status, etc., which are not limited in the embodiments of this application.

[0148] As an optional embodiment, the method further includes: when the network energy-saving state is detected to meet the preset stable state, the management terminal sends a stop message to the first base station and the second base station to indicate the exit of the energy-saving mechanism; wherein, the stop message includes network energy-saving suspension indication information and trigger reason information.

[0149] As can be seen, by implementing this optional embodiment, network energy saving can be stopped when the feedback information meets the preset termination conditions, i.e. when the network energy saving state is stable, so as to save network resources.

[0150] After the management terminal sends a stop message to the first base station and the second base station to indicate the exit of the energy-saving mechanism, the above method may further include: the first base station and the second base station may exit the energy-saving mechanism and suspend signaling transmission and data transmission related to model training and model inference.

[0151] Please see Figure 3 , Figure 3 A flowchart illustrating another embodiment of a network-based intelligent energy-saving method according to this application is shown. Figure 3 As shown, the network intelligent energy-saving method may include steps S300 to S340. In this application, the step numbers are only used as unique representations of the steps and do not limit the order in which the steps are executed.

[0152] Step S300: The management terminal receives the energy consumption parameters periodically reported by each base station within the preset control range corresponding to the management terminal; when an energy consumption parameter greater than the preset threshold is detected, the management terminal determines the energy consumption optimization range based on the base station corresponding to the energy consumption parameter greater than the preset threshold; wherein, the energy consumption optimization range includes the first base station and the second base station.

[0153] Step S302: The management terminal trains a management terminal model based on the historical dataset and obtains the model configuration information corresponding to the trained management terminal model. The historical dataset is divided into an offline training dataset, an offline validation dataset, and an offline test dataset. The historical dataset includes historical measurement data reported by each base station within the energy consumption optimization range. The management terminal sends model configuration information to the first base station, including anchor base station indication information, a list of second base station IDs within the energy consumption optimization range, a model index, and feature input information. The management terminal sends model configuration information to the second base station, including a model index and feature input information. The model index includes model use cases, model categories, and model parameters. The model parameters include weights, biases, learning rates, iteration counts, and SVM support vectors. The feature input information includes feature input statistics from the terminal and feature inputs from the base station. The feature input statistics from the terminal include terminal location information, terminal movement speed, and terminal wireless measurement information. The terminal wireless measurement information includes RSRP and RSRQ. The feature inputs from the base station include base station energy consumption status, base station energy efficiency, and base station wireless resource status.

[0154] Step S304: After receiving the model configuration information, the first base station determines the local base station as the anchor base station, and calls and updates the latest model according to the model configuration information, and performs measurement configuration for each terminal in the area specified by the first base station; wherein, the measurement configuration includes MDT measurement configuration and RRM measurement configuration; the first base station sends a first data acquisition request to each terminal in the reachable area.

[0155] Step S306: Each terminal periodically responds to the first data acquisition request by reporting the collected real-time first terminal measurement data to the first base station, and performs specific measurement operations based on the MDT and RRM measurement configuration and stores the measurement results; wherein, the terminal measurement data includes terminal location information, terminal moving speed, timestamp, terminal RSRP, terminal RSRQ, and terminal SINR.

[0156] Step S308: The first base station reports the first base station statistical data and the first terminal measurement data to the management terminal through the model training input message; wherein, the model training input message includes the anchor base station identifier and the required training data, the required training data includes the required terminal training data and the required base station training data, the required terminal training data includes the terminal location information, the terminal moving speed, the terminal wireless measurement information and the timestamp, and the required base station training data includes the base station energy consumption status, the base station energy efficiency and the base station wireless resource status.

[0157] Step S310: After receiving the model configuration information, the second base station determines the local base station as the auxiliary base station and inputs information according to the features indicated in the model configuration information.

[0158] Step S312: Each terminal periodically responds to the second data acquisition request by reporting the collected real-time terminal measurement data to the second base station, and performs specific measurement operations based on the MDT and RRM measurement configuration and stores the measurement results; wherein, the terminal measurement data includes terminal location information, terminal moving speed, timestamp, terminal RSRP, terminal RSRQ, and terminal SINR.

[0159] Step S314: The second base station sends the received measurement data from the second terminal and the statistical data from the second base station to the management terminal through the model training input message; wherein, the model training input message includes the auxiliary base station node indication and the required training data. The required training data includes the required terminal training data and the required base station training data. The required terminal training data includes the terminal location information, the terminal moving speed, the terminal wireless measurement information, and the timestamp. The required base station training data includes the base station energy consumption status, the base station energy efficiency, and the base station wireless resource status.

[0160] Step S316: The management terminal receives the statistical data of the first base station, the measurement results of the first terminal, the statistical data of the second base station, and the measurement data of the second terminal; it generates a training dataset, a validation dataset, and a test dataset based on the statistical data of the first base station, the measurement results of the first terminal, the statistical data of the second base station, and the measurement data of the second terminal; it updates the management terminal model online according to the training dataset, the validation dataset, and the test dataset; and the management terminal obtains the model update information corresponding to the updated management terminal model.

[0161] Step S318: The management terminal sends the model update information to the first base station and the second base station; wherein, the model update information includes the model index, and the model index includes model use cases, model category, and model parameters.

[0162] Step S320: The first base station receives model update information and sends a first request message to the second base station according to the energy consumption optimization range to obtain auxiliary data required for model inference from the second base station; wherein, the first request message is used to provide feedback through the control plane and / or user plane, and the first request message includes the first base station node ID; the first base station establishes a preset interface to realize the interaction of AI / ML data.

[0163] Step S322: After receiving the first request message, if the second base station needs to provide feedback through the first preset interface, the first request message includes the first base station node ID, measurement information indication, measurement indication, reporting characteristics, and reporting period. If the second base station needs to provide feedback through the second preset interface, the first request message also includes the first base station IP address and port address.

[0164] Step S324: The second base station receives the first request message from the first base station, and responds with a message indicating the measurement object requested by the model inference data to start the measurement, and obtains the measurement object information, indicating the second base station IP address and port address, message type, first base station measurement ID, second base station measurement ID, and critical diagnostic IE.

[0165] Step S326: If the second base station cannot provide measurement results, the second base station sends a first failure message to the first base station; wherein the first failure message includes the second base station ID, a description of the failure reason, and a retransmission wait time. The second base station reports a first reply message to the first base station, reporting the requested measurement information; the first reply message includes message type, first base station measurement ID, second base station measurement ID, measurement result feedback, terminal measurement result feedback including terminal location information, terminal moving speed, terminal wireless measurement information, and timestamp, and base station measurement result feedback including base station energy consumption status, base station energy efficiency, base station wireless resource status, and timestamp.

[0166] Step S328: The first base station generates the dataset required for model inference based on its own periodic measurement reports, first reply messages, and periodic measurement data from the terminal; the dataset required for model inference is input into the latest model so that the latest model outputs the corresponding inference results. The inference results are used as the basis for network analysis and include prediction information or energy-saving decision information; the first base station generates a second interface message based on the inference results; the inference results include energy-saving analysis identifier, root cause, recommended energy-saving cell ID list, recommended candidate cell ID list, recommended energy-saving operation, load prediction, base station energy efficiency prediction, and resource status prediction.

[0167] Step S330: The first base station sends the second interface message to the second base station through the preset interface control plane and / or user plane.

[0168] Step S332: The first base station and the second base station infer the output of the model and execute network energy-saving actions. If the output is a handover strategy, a cell is selected for the terminal before the terminal handover.

[0169] Step S334: After performing energy-saving operation, the first base station sends a first feedback message to the management terminal; wherein, the first feedback message is used to indicate the first network data required for detecting the performance of the latest model, the first network data is used to optimize the latest model, and the first network data includes energy efficiency value; the first feedback message includes base station energy consumption status, base station energy efficiency, and base station radio resource status.

[0170] Step S338: After performing energy-saving operation, the second base station sends a second feedback message to the management terminal; wherein, the second feedback message is used to indicate the second network data required for detecting the performance of the second base station model, the second network data is used to optimize the second base station model, and the second network data includes energy efficiency values; the second feedback message includes base station energy consumption status, base station energy efficiency, and base station radio resource status.

[0171] Step S340: When the network energy-saving state is detected to meet the preset stable state, the management terminal sends a stop message to the first base station and the second base station to indicate the exit of the energy-saving mechanism; wherein, the stop message includes network energy-saving pause indication information and trigger reason information, and the trigger reason includes set threshold comparison information.

[0172] It should be noted that steps S300 to S340 are different from... Figure 2 For the specific implementation details of steps S300 to S340, please refer to the examples shown. Figure 2 The steps and their embodiments shown are not repeated here.

[0173] It is evident that implementation Figure 3 The method shown can realize model training and model inference through multi-terminal signaling interaction, which is conducive to enabling base stations to accurately collect data from terminals under the control of the management terminal, thereby achieving precise energy saving, reducing network energy consumption, and improving network energy efficiency.

[0174] Please see Figure 4 , Figure 4 This schematically illustrates a structural block diagram of a network-based intelligent energy-saving device according to one embodiment of the present application. The network-based intelligent energy-saving device 400 and... Figure 2 The methods shown correspond to those described, such as Figure 4 As shown, the network-connected intelligent energy-saving device 400 includes:

[0175] The first base station 402 is used to receive model update information and send a first request message to the second base station 403 according to the energy consumption optimization range, so as to obtain the auxiliary data required for model inference from the second base station 403; wherein, the first request message is used to provide feedback through the control plane and / or user plane, and the first request message includes the node ID of the second base station 403 and the measurement result feedback; the first base station 402 establishes a preset interface to realize the interaction of AI / ML data;

[0176] The second base station 403 is used to, after receiving the first request message, if the second base station 403 needs to provide feedback through the first preset interface, the first request message includes the node ID of the first base station 402, measurement information indication, measurement indication, reporting characteristics, and reporting period; if the second base station 403 needs to provide feedback through the second preset interface, the first request message also includes the IP address and port address of the first base station 402.

[0177] The second base station 403 is used to receive a first request message from the first base station 402, reply with a message indicating the measurement object requested by model inference data to start measurement, and obtain measurement object information to indicate the IP address and port address of the second base station 403, message type, measurement ID of the first base station 402, measurement ID of the second base station 403, and critical diagnostic IE.

[0178] If the second base station 403 cannot provide measurement results, the second base station 403 is used to send a first failure message to the first base station 402; wherein the first failure message includes the second base station 403 ID, a description of the failure reason, and a retransmission waiting time;

[0179] The second base station 403 is used to report a first reply message to the first base station 402, which includes the measurement information requested. The first reply message includes message type, measurement ID of the first base station 402, measurement ID of the second base station 403, measurement result feedback, measurement result feedback of the terminal 404 including the location information of the terminal 404, the moving speed of the terminal 404, the wireless measurement information of the terminal 404, and a timestamp. The base station measurement result feedback includes the base station energy consumption status, base station energy efficiency, base station wireless resource status, and a timestamp.

[0180] It is evident that implementation Figure 4 The device shown can achieve model training and model inference through multi-terminal signaling interaction, which is conducive to enabling the base station to accurately collect data from the terminal 404 under the control of the management terminal 401, thereby achieving precise energy saving, reducing network energy consumption, and improving network energy efficiency.

[0181] In one exemplary embodiment of this application, wherein:

[0182] The first base station 402 is used to generate the dataset required for model inference based on the base station's own periodic measurement reporting, first reply message and periodic measurement data of terminal 404; wherein, the periodic measurement data of terminal 404 includes RRM measurement information and MDT measurement information. The RRM measurement information includes RSRP, RSRQ and timestamp. The MDT measurement information includes terminal 404 location information, terminal 404 moving speed and timestamp.

[0183] The first base station 402 is used to input the dataset required for model inference into the latest model so that the latest model can output the corresponding inference results; wherein, the inference results are used as the basis for network analysis, and the inference results include prediction information or energy-saving decision information;

[0184] The first base station 402 is used to generate a second interface message based on the inference results; wherein, the inference results include energy-saving analysis identifier, root cause, recommended energy-saving cell ID list, recommended candidate cell ID list, recommended energy-saving operation, load prediction, base station energy efficiency prediction, and resource status prediction;

[0185] The first base station 402 is used to send the second interface message to the second base station 403 through a preset interface control plane and / or user plane;

[0186] The first base station 402 and the second base station 403 are used to infer the output from the model and execute network energy-saving actions. If the output is a handover strategy, then a cell is selected for terminal 404 before the handover.

[0187] In one exemplary embodiment of this application, wherein:

[0188] The management terminal 401 receives the energy consumption parameters periodically reported by each base station within the preset control range corresponding to the management terminal 401;

[0189] When an energy consumption parameter greater than a preset threshold is detected, the management terminal 401 determines the energy consumption optimization range based on the base station corresponding to the energy consumption parameter greater than the preset threshold; wherein, the energy consumption optimization range includes the first base station 402 and the second base station 403.

[0190] As can be seen, by implementing this optional embodiment, the model can be updated based on the signaling interaction between the terminal and the first base station and the signaling interaction between the first base station and the management terminal. The management terminal can further synchronize the updated model to each base station within the energy consumption optimization range, so that each base station can perform energy-saving operations based on the updated model, thereby achieving precise network energy saving.

[0191] In one exemplary embodiment of this application, wherein:

[0192] Management terminal 401 is used to train the management terminal 401 model based on the historical dataset and obtain the model configuration information corresponding to the trained management terminal 401 model. The historical dataset is divided into an offline training dataset, an offline validation dataset, and an offline test dataset. The historical dataset includes historical measurement data reported by each base station within the energy consumption optimization range.

[0193] The management terminal 401 is used to send model configuration information to the first base station 402, including anchor base station indication information, a list of second base station 403 IDs within the energy consumption optimization range, model index, and feature input information.

[0194] The management terminal 401 is used to send model configuration information, including model index and feature input information, to the second base station 403.

[0195] The model index includes model use cases, model categories, and model parameters; the model parameters include weights, biases, learning rates, iteration counts, and SVM support vectors; the feature input information includes feature input statistics from the terminal and feature input from the base station; the feature input statistics from the terminal include terminal location information, terminal movement speed, and terminal wireless measurement information; the terminal wireless measurement information includes RSRP and RSRQ; and the feature input from the base station includes base station energy consumption status, base station energy efficiency, and base station wireless resource status.

[0196] As can be seen, implementing this optional embodiment can achieve model consistency between the base station and the management terminal, thereby facilitating more accurate network energy saving.

[0197] In one exemplary embodiment of this application, wherein:

[0198] The first base station 402 is used to determine the local base station as the anchor base station after receiving the model configuration information, and to call and update the latest model according to the model configuration information, as well as to perform measurement configuration on each terminal 404 within the area specified by the first base station 402; wherein, the measurement configuration includes MDT measurement configuration and RRM measurement configuration.

[0199] The first base station 402 is used to send a first data collection request to each terminal 404 within the reachable area;

[0200] Each terminal 404 is used to periodically respond to the first data acquisition request by reporting the collected real-time first terminal measurement data to the first base station 402, and to perform specific measurement operations based on the MDT and RRM measurement configuration and store the measurement results; wherein, the terminal measurement data includes terminal location information, terminal moving speed, timestamp, terminal RSRP, terminal RSRQ, and terminal SINR;

[0201] The first base station 402 is used to report the statistical data of the first base station 402 and the measurement data of the first terminal to the management terminal 401 through the model training input message; wherein, the model training input message includes the anchor base station identifier and the required training data, the required training data includes the required terminal training data and the required base station training data, the required terminal training data includes the terminal location information, the terminal moving speed, the terminal wireless measurement information and the timestamp, and the required base station training data includes the base station energy consumption status, the base station energy efficiency and the base station wireless resource status.

[0202] As can be seen, implementing this optional embodiment can achieve precise configuration and signaling interaction between the terminal and the first base station, enabling timely collection of terminal data.

[0203] In one exemplary embodiment of this application, wherein:

[0204] The second base station 403 is used to determine the local base station as the auxiliary base station after receiving the model configuration information, and to send the received second terminal measurement data and the statistical data of the second base station 403 to the management terminal 401 through the model training input message according to the feature input information indicated in the model configuration information. The model training input message includes the auxiliary base station node indication and the required training data. The required training data includes the required terminal training data and the required base station training data. The required terminal training data includes the terminal location information, the terminal moving speed, the terminal wireless measurement information, and the timestamp. The required base station training data includes the base station energy consumption status, the base station energy efficiency, and the base station wireless resource status.

[0205] As can be seen, implementing this optional embodiment can achieve precise configuration and signaling interaction between the terminal and the second base station, enabling timely collection of terminal data.

[0206] In one exemplary embodiment of this application, wherein:

[0207] Management terminal 401 is used to receive statistical data from the first base station 402, measurement results from the first terminal, statistical data from the second base station 403, and measurement data from the second terminal;

[0208] Management terminal 401 is used to generate training dataset, verification dataset and test dataset based on statistical data of the first base station 402, measurement results of the first terminal, statistical data of the second base station 403 and measurement data of the second terminal;

[0209] Management terminal 401 is used to update the management terminal 401 model online based on the training dataset, validation dataset, and test dataset;

[0210] Management terminal 401 is used to obtain the model update information corresponding to the updated management terminal 401 model;

[0211] The management terminal 401 is used to send model update information to the first base station 402 and the second base station 403; wherein, the model update information includes the model index, and the model index includes model use cases, model category, and model parameters.

[0212] As can be seen, implementing this optional embodiment can enable timely model updates, thereby facilitating precise network energy saving.

[0213] In one exemplary embodiment of this application, wherein:

[0214] The first base station 402 is used to send a first feedback message to the management terminal 401 after performing energy-saving operation; wherein, the first feedback message is used to indicate the first network data required for detecting the performance of the latest model, the first network data is used to optimize the latest model, and the first network data includes energy efficiency value; the first feedback message includes base station energy consumption status, base station energy efficiency, and base station radio resource status;

[0215] The second base station 403 is used to send a second feedback message to the management terminal 401 after performing energy-saving operations; wherein, the second feedback message is used to indicate the second network data required for detecting the performance of the second base station 403 model, the second network data is used to optimize the second base station 403 model, and the second network data includes energy efficiency values; the second feedback message includes base station energy consumption status, base station energy efficiency, and base station radio resource status.

[0216] As can be seen, implementing this optional embodiment can promptly feed back energy-saving information to the management terminal, so that the management terminal can update the management terminal model in a timely manner for the next precise network energy saving.

[0217] In one exemplary embodiment of this application, wherein:

[0218] When the network energy-saving status is detected to meet the preset stable state, the management terminal 401 is used to send a stop message to the first base station 402 and the second base station 403 to indicate the exit of the energy-saving mechanism; wherein, the stop message includes network energy-saving pause indication information and trigger reason information, and the trigger reason includes set threshold comparison information.

[0219] As can be seen, by implementing this optional embodiment, network energy saving can be stopped when the feedback information meets the preset termination conditions, i.e. when the network energy saving state is stable, so as to save network resources.

[0220] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0221] Since the functional modules of the network intelligent energy-saving device in the example embodiments of this application correspond to the steps of the example embodiments of the network intelligent energy-saving method described above, for details not disclosed in the device embodiments of this application, please refer to the embodiments of the network intelligent energy-saving method described above in this application.

[0222] Please see Figure 5 , Figure 5 A schematic diagram of the structure of a computer system suitable for implementing the electronic device of the present application is shown.

[0223] It should be noted that, Figure 5 The computer system 500 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0224] like Figure 5 As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 502 or programs loaded from storage section 508 into random access memory (RAM) 503. The RAM 503 also stores various programs and data required for system operation. The CPU 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0225] The following components are connected to I / O interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to I / O interface 505 as needed. A removable medium 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 510 as needed so that computer programs read from it can be installed into storage section 508 as needed.

[0226] Specifically, according to embodiments of this application, the processes described in the above-described flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the computer program is executed by central processing unit (CPU) 501, it performs the various functions defined in the methods and apparatus of this application.

[0227] In another aspect, this application also provides a computer-readable medium, which may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into the electronic device. The computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the methods described in the above embodiments.

[0228] It should be noted that the computer-readable medium shown in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0229] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0230] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0231] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the foregoing claims.

Claims

1. A network-based intelligent energy-saving method, characterized in that, include: The first base station receives model update information and sends a first request message to the second base station according to the energy consumption optimization range to obtain auxiliary data required for model inference from the second base station; wherein, the first request message is used to provide feedback through the control plane and / or user plane, and the first request message includes the first base station node ID; the first base station establishes a preset interface to realize the interaction of AI / ML data; After receiving the first request message, if the second base station needs to provide feedback through the first preset interface, the first request message includes the first base station node ID, measurement information indication, measurement indication, reporting characteristics, and reporting period. If the second base station needs to provide feedback through the second preset interface, the first request message also includes the first base station IP address and port address. The second base station receives the first request message from the first base station, and responds with a message indicating the measurement object requested by the model inference data to initiate measurement, and obtains measurement object information to indicate the second base station IP address and port address, message type, first base station measurement ID, second base station measurement ID, and critical diagnostic IE; If the second base station is unable to provide measurement results, the second base station sends a first failure message to the first base station; wherein the first failure message includes the second base station ID, a description of the failure reason, and a retransmission waiting time; The second base station reports a first reply message to the first base station, including the requested measurement information. The first reply message includes a message type, a first base station measurement ID, a second base station measurement ID, and measurement result feedback. The measurement result feedback includes terminal measurement result feedback and base station measurement result feedback. The terminal measurement result feedback includes terminal location information, terminal movement speed, terminal wireless measurement information, and a timestamp. The base station measurement result feedback includes base station energy consumption status, base station energy efficiency, base station wireless resource status, and a timestamp.

2. The method according to claim 1, characterized in that, The method further includes: The first base station generates the dataset required for model inference based on its own periodic measurement reports, the first reply message, and the periodic measurement data of the terminal; wherein, the periodic measurement data of the terminal includes RRM measurement information and MDT measurement information, the RRM measurement information includes RSRP, RSRQ, and timestamp, and the MDT measurement information includes terminal location information, terminal moving speed, and timestamp; The first base station inputs the dataset required for model inference into the latest model so that the latest model outputs the corresponding inference result; wherein, the inference result is used as the basis for network analysis, and the inference result includes prediction information or energy-saving decision information; The first base station generates a second interface message based on the inference result; wherein, the inference result includes energy-saving analysis identifier, root cause, recommended energy-saving cell ID list, recommended candidate cell ID list, recommended energy-saving operation, load prediction, base station energy efficiency prediction, and resource status prediction; The first base station sends the second interface message to the second base station through a preset interface control plane and / or user plane; The first and second base stations infer network energy-saving actions based on the model. If the output is a handover strategy, a cell is selected for the terminal before the terminal hands over.

3. The method according to claim 1, characterized in that, The method further includes: The management terminal receives energy consumption parameters periodically reported by each base station within the preset control range corresponding to the management terminal; When an energy consumption parameter greater than a preset threshold is detected, the management terminal determines the energy consumption optimization range based on the base station corresponding to the energy consumption parameter greater than the preset threshold; wherein the energy consumption optimization range includes the first base station and the second base station.

4. The method according to claim 1, characterized in that, The method further includes: The management terminal trains a management terminal model based on a historical dataset and obtains the model configuration information corresponding to the trained management terminal model; wherein, the historical dataset is used to be divided into an offline training dataset, an offline verification dataset, and an offline test dataset, and the historical dataset includes historical measurement data reported by each base station within the energy consumption optimization range; The management terminal sends model configuration information to the first base station, including anchor base station indication information, a list of second base station IDs within the energy consumption optimization range, a model index, and feature input information. The management terminal sends model configuration information, including model index and feature input information, to the second base station. The model index includes model use cases, model categories, and model parameters; the model parameters include weights, biases, learning rates, iteration counts, and SVM support vectors; the feature input information includes feature input statistics from the terminal and feature input from the base station; the feature input statistics from the terminal include terminal location information, terminal movement speed, and terminal wireless measurement information; the terminal wireless measurement information includes RSRP and RSRQ; and the feature input from the base station includes base station energy consumption status, base station energy efficiency, and base station wireless resource status.

5. The method according to claim 4, characterized in that, The method further includes: After receiving the model configuration information, the first base station determines the local base station as the anchor base station, and calls and updates the latest model according to the model configuration information, and performs measurement configuration for each terminal in the designated area of ​​the first base station; wherein, the measurement configuration includes MDT measurement configuration and RRM measurement configuration; The first base station sends a first data collection request to each terminal within the reachable area; Each terminal periodically responds to the first data acquisition request by reporting the collected real-time first terminal measurement data to the first base station, and performs specific measurement operations based on the MDT and RRM measurement configuration and stores the measurement results; wherein, the first terminal measurement data includes terminal location information, terminal moving speed, timestamp, terminal RSRP, terminal RSRQ, and terminal SINR; The first base station reports the first base station statistical data and the first terminal measurement data to the management terminal through a model training input message; wherein, the model training input message includes the anchor base station identifier and the required training data, the required training data includes the required terminal training data and the required base station training data, the required terminal training data includes the terminal location information, the terminal moving speed, the terminal wireless measurement information, and the timestamp, and the required base station training data includes the base station energy consumption status, the base station energy efficiency, and the base station wireless resource status.

6. The method according to claim 5, characterized in that, The method further includes: After receiving the model configuration information, the second base station identifies the local base station as the auxiliary base station and, according to the feature input information indicated in the model configuration information, sends the received second terminal measurement data and second base station statistical data to the management terminal through a model training input message. The model training input message includes an auxiliary base station node indication and required training data. The required training data includes required terminal training data and required base station training data. The required terminal training data includes terminal location information, terminal movement speed, terminal wireless measurement information, and a timestamp. The required base station training data includes base station energy consumption status, base station energy efficiency, and base station wireless resource status.

7. The method according to claim 6, characterized in that, The method further includes: The management terminal receives the statistical data of the first base station, the measurement data of the first terminal, the statistical data of the second base station, and the measurement data of the second terminal; The management terminal generates a training dataset, a validation dataset, and a test dataset based on the statistical data of the first base station, the measurement data of the first terminal, the statistical data of the second base station, and the measurement data of the second terminal. The management terminal updates the management terminal model online based on the training dataset, validation dataset, and test dataset. The management terminal obtains the model update information corresponding to the updated management terminal model; The management terminal sends the model update information to the first base station and the second base station; wherein, the model update information includes a model index, and the model index includes model use cases, model categories, and model parameters.

8. The method according to claim 2, characterized in that, The method further includes: After performing energy-saving operations, the first base station sends a first feedback message to the management terminal; wherein, the first feedback message is used to indicate the first network data required for detecting the performance of the latest model, the first network data is used to optimize the latest model, and the first network data includes energy efficiency values; the first feedback message includes base station energy consumption status, base station energy efficiency, and base station radio resource status. After performing energy-saving operations, the second base station sends a second feedback message to the management terminal; wherein, the second feedback message is used to indicate the second network data required for detecting the performance of the second base station model, the second network data is used to optimize the second base station model, and the second network data includes energy efficiency values; the second feedback message includes base station energy consumption status, base station energy efficiency, and base station radio resource status.

9. The method according to claim 1, characterized in that, The method further includes: When the network energy-saving status is detected to meet the preset stable state, the management terminal sends a stop message to the first base station and the second base station to indicate the exit of the energy-saving mechanism; wherein, the stop message includes network energy-saving pause indication information and trigger reason information, and the trigger reason includes set threshold comparison information.

10. A network-connected intelligent energy-saving device, characterized in that, include: The first base station is used to receive model update information and send a first request message to the second base station according to the energy consumption optimization range to obtain auxiliary data required for model inference from the second base station; wherein, the first request message is used to provide feedback through the control plane and / or user plane, and the first request message includes the first base station node ID; the first base station establishes a preset interface to realize the interaction of AI / ML data; The second base station is configured to, upon receiving the first request message, if the second base station needs to provide feedback through the first preset interface, then the first request message includes the first base station node ID, measurement information indication, measurement indication, reporting characteristics, and reporting period; if the second base station needs to provide feedback through the second preset interface, then the first request message also includes the first base station IP address and port address. The second base station is used to receive the first request message from the first base station, reply with a message indicating the measurement object requested by model inference data to start the measurement, and obtain measurement object information to indicate the IP address and port address of the second base station, message type, measurement ID of the first base station, measurement ID of the second base station, and critical diagnostic IE; If the second base station is unable to provide measurement results, the second base station is used to send a first failure message to the first base station; wherein the first failure message includes the second base station ID, a description of the failure reason, and a retransmission waiting time; The second base station is used to report a first reply message to the first base station, including the requested measurement information. The first reply message includes a message type, a first base station measurement ID, a second base station measurement ID, and measurement result feedback. The measurement result feedback includes terminal measurement result feedback and base station measurement result feedback. The terminal measurement result feedback includes terminal location information, terminal movement speed, terminal wireless measurement information, and a timestamp. The base station measurement result feedback includes base station energy consumption status, base station energy efficiency, base station wireless resource status, and a timestamp.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1-9.

12. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to execute the method of any one of claims 1-9 by executing the executable instructions.

Images