Base station energy saving control method, storage medium and electronic device

Abstract

The embodiment of the invention provides a base station energy-saving control method, a storage medium and an electronic device, and the method comprises the steps: carrying out the model training according to historical user behavior data, obtaining a user behavior prediction model, and enabling the user behavior prediction model to have a time characteristic; inputting the user behavior data at the current moment into a user behavior prediction model to obtain a predicted behavior result at the next moment; and adjusting the energy-saving strategy of the base station based on the predicted behavior result at the next moment. According to the invention, the problems of poor energy-saving effect and poor user experience of the base station in the prior art are solved, and the effects of improving the energy-saving effect of the base station and enhancing the user experience are achieved.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of communications, and more particularly, to a control method, a storage medium, and an electronic device for base station energy saving. Background Art

[0002] There is a dedicated network in wireless communication. This kind of network does not need to be in a working state all the time and only needs to enter the working state when there are specific user terminals that need services. In addition, there is generally a "tidal phenomenon" in a normal network, so the demand for network occupancy is different at different times. During low-demand periods, the base station can adopt energy-saving measures to reduce the power consumption of the base station.

[0003] In the related art, energy saving is generally based on the load situation of the base station. When the load is low, it will be prepared to enter the energy-saving state. If there is a brief abnormal increase in the load at this time, the base station will recalculate the energy-saving time. However, this method is greatly interfered by the load situation and may also cause the base station to be unable to enter the energy-saving state all the time, and it is impossible to wake up the base station in advance to cope with the sudden arrival of users and service impacts, resulting in a poor user experience.

[0004] In summary, there are still problems in the related art such as poor energy-saving effect of the base station and poor user experience. Summary of the Invention

[0005] Embodiments of the present invention provide a control method, a storage medium, and an electronic device for base station energy saving to at least solve the problems of poor energy-saving effect of the base station and poor user experience in the related art.

[0006] According to an embodiment of the present invention, a control method for base station energy saving is provided, including: training a model based on historical user behavior data to obtain a user behavior prediction model, where the user behavior prediction model has a time characteristic; inputting the user behavior data at the current moment into the user behavior prediction model to obtain a predicted behavior result at the next moment; and adjusting the base station energy-saving strategy based on the predicted behavior result at the next moment.

[0007] According to another embodiment of the present invention, a computer-readable storage medium is further provided. A computer program is stored in the computer-readable storage medium, where the computer program is configured to execute the steps in any one of the above method embodiments when running.

[0008] According to another embodiment of the present invention, an electronic device is further provided, including a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0009] Through the present invention, a control method for base station energy saving is provided. By training a model based on historical user behavior data, a user behavior prediction model is obtained. The user behavior prediction model has time characteristics. The user behavior data at the current moment is input into the user behavior prediction model to obtain the predicted behavior result at the next moment. Based on the predicted behavior result at the next moment, the base station energy saving strategy is adjusted. The problem of poor base station energy saving effect and poor user experience in the related art is solved, and the effect of improving the base station energy saving effect and enhancing the user experience is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 is a hardware structure block diagram of a computer terminal for a control method of base station energy saving according to an embodiment of the present invention;

[0011] Figure 2 is a flowchart of a control method for base station energy saving according to an embodiment of the present invention;

[0012] Figure 3 is a structure block diagram of a control system for base station energy saving according to a scenario embodiment of the present invention;

[0013] Figure 4 is a schematic diagram of the training process of a user behavior model learning module according to a scenario embodiment of the present invention;

[0014] Figure 5 is a schematic diagram of the working process of a control system for base station energy saving according to a scenario embodiment of the present invention;

[0015] Figure 6 is a schematic diagram of the working principle of a control system for base station energy saving according to a scenario embodiment of the present invention;

[0016] Figure 7 is a grayscale map of the probability of high-speed trains entering at different time periods according to a scenario embodiment of the present invention;

[0017] Figure 8 is a schematic diagram of the process principle of a de - energy - saving strategy according to a scenario embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] In the following, embodiments of the present invention will be described in detail with reference to the drawings and in conjunction with the embodiments.

[0019] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above - mentioned drawings are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence.

[0020] The method embodiments provided in the embodiments of the present application can be executed on a mobile terminal, a computer terminal or a similar computing device. Taking running on a computer terminal as an example, Figure 1It is a hardware structure block diagram of a computer terminal for a base station energy-saving control method according to an embodiment of the present invention. As Figure 1 shown, the computer terminal may include one or more ( Figure 1 only one is shown in Figure 1 ) processors 102 (the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data. Among them, the above computer terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those of ordinary skill in the art can understand that Figure 1 the structure shown in Figure 1 is only schematic and does not limit the structure of the above computer terminal. For example, the computer terminal may further include more or fewer components than

[0021] shown in

[0022] or have a different configuration from

[0023] shown in Figure 2 The memory 104 can be used to store computer programs. For example, software programs and modules of application software, such as the computer program corresponding to the base station energy-saving control method in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, implements the above method. The memory 104 may include a high-speed random access memory and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some instances, the memory 104 may further include a memory remotely set relative to the processor 102, and these remote memories may be connected to the computer terminal through a network. Examples of the above network include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof. Figure 2 shown, the process includes the following steps:

[0024] Step S202: Perform model training based on historical user behavior data to obtain a user behavior prediction model, where the user behavior prediction model has time characteristics.

[0025] In an exemplary embodiment, performing model training based on historical user behavior data to obtain a user behavior prediction model includes: obtaining user behavior data for a preset historical period and classifying the user behavior data for the preset historical period to obtain classified user behavior data; using the classified user behavior data as training data for model training to obtain a user behavior prediction model.

[0026] In the actual implementation process, the user behavior prediction model has time characteristics. Model training is performed based on historical user behavior data to obtain the user behavior prediction model. According to the user behavior data at the current moment, the user behavior prediction model is used to predict the user behavior data at the next moment, that is, the predicted behavior result.

[0027] In an exemplary embodiment, classifying the user behavior data for the preset historical period to obtain classified user behavior data includes: identifying the network type corresponding to the user behavior data for the preset historical period; determining the characteristic parameters of different user behaviors within the preset historical period according to the network type; classifying the user behavior data for the preset historical period based on the characteristic parameters of different user behaviors within the preset historical period to obtain classified user behavior data.

[0028] In the actual implementation process, the above-mentioned preset historical period is a randomly selected historical time period to obtain the historical user behavior data for that time period. Among them, the network type may include different application scenarios such as high-speed rail scenarios. The characteristic parameters of different user behaviors at least include user behavior characteristics (such as speed characteristics, etc.) and base station resource usage characteristics.

[0029] In an exemplary embodiment, using the classified user behavior data as training data for model training to obtain a user behavior prediction model includes: adding data labels based on the time sequence and base station resource characteristics corresponding to each behavior in the classified user behavior data to obtain training data; performing model training based on the training data to obtain a user behavior prediction model.

[0030] In the actual implementation process, when the data (i.e., user behavior data) reaches a certain amount (more data is required for the first time), weights are assigned to the data in chronological order, and then data identification and training are performed. According to the obtained statistical information of user behavior, the data is classified, the interfering data is removed by data cleaning, and then these data are trained according to the preset classification to obtain the behavior characteristics of users and the base station resource usage characteristics within a specific time period.

[0031] Step S204: Input the user behavior data at the current moment into the user behavior prediction model to obtain the predicted behavior result at the next moment.

[0032] Step S206: Adjust the base station energy-saving strategy based on the predicted behavior result at the next moment.

[0033] In an exemplary embodiment, after adjusting the base station energy-saving strategy based on the predicted behavior result at the next moment, it further includes: determining the accuracy of the predicted behavior result at the next moment based on the strategy implementation result corresponding to the adjusted base station energy-saving strategy; triggering the update training of the user behavior prediction model when the accuracy of the predicted behavior result at the next moment is less than the preset threshold and the number of times that the accuracy of the predicted behavior result fails to meet the standard exceeds the preset number of times.

[0034] In the actual implementation process, the determination of accuracy includes the degree of compliance of the energy-saving result and the degree of compliance of the behavior prediction. As described above, an update is performed only when it does not meet the standard and reaches a certain number of times. In the actual implementation process, a preset update period can also be set, that is, the user behavior prediction model is updated every once in a while to ensure the accuracy of the energy-saving strategy.

[0035] In an exemplary embodiment, adjusting the base station energy-saving strategy based on the predicted behavior result at the next moment includes: judging the energy-saving strategy or the de - energy-saving strategy of the base station at the next moment based on the predicted behavior result at the next moment, and changing the energy-saving state or energy-saving type of the base station according to the energy-saving strategy or the de - energy-saving strategy.

[0036] In an exemplary embodiment, adjusting the base station energy-saving strategy based on the predicted behavior result at the next moment includes: in the case of receiving an external energy-saving instruction, judging whether the user behavior data at the current moment meets the condition of entering energy-saving. If it does not meet, requesting confirmation information from an external device. Entering energy-saving if the confirmation information is received, otherwise not entering energy-saving; in the case of entering energy-saving, determining the energy-saving state or energy-saving type of the base station based on the predicted behavior result at the next moment and the user behavior data at the current moment.

[0037] In the actual implementation process, if it is to enter energy-saving, first check whether there is a setting command. If so, then check whether the real-time data of the user behavior meets the condition of entering the energy-saving state. If it does not meet, return an alarm and request confirmation information. Enter the energy-saving process if the confirmation information is received, otherwise do not enter; then look at the predicted data of the user behavior and combine it with the real-time data of the user behavior to judge which level of energy-saving to enter.

[0038] In an exemplary embodiment, adjusting the base station energy saving strategy based on the predicted behavior result at the next moment includes: pre-regulating the base station to enter the power-saving state according to the predicted behavior result at the next moment; judging whether the base station exits the energy saving according to the user behavior data at the current moment, and exiting the energy saving when the user behavior data at the current moment is consistent with the predicted behavior result at the next moment; and when the user behavior data at the current moment is inconsistent with the predicted behavior result at the next moment, not exiting the energy saving and maintaining the power-saving state.

[0039] In the actual implementation process, exiting the energy saving strategy is not a one-step complete exit from energy saving: it should be preset to the corresponding power-saving state according to the predicted data of user behavior, and then according to the real-time data of user behavior, judge what the next user behavior will be. For example, if both the number of handover users and the number of online users are gradually increasing, it means that users are gradually entering this cell, then the energy saving state should be exited; otherwise, it can be temporarily maintained in the power-saving state.

[0040] In the actual implementation process, the above power-saving state is to enter the pre-power-saving state. When it is determined according to the real-time user data that power saving is required, quickly execute the power-saving action according to the pre-power-saving state, so as to efficiently complete the power saving and better ensure the user experience.

[0041] Through the above steps, by performing model training based on historical user behavior data, a user behavior prediction model is obtained. The user behavior prediction model has time characteristics; inputting the user behavior data at the current moment into the user behavior prediction model to obtain the predicted behavior result at the next moment; adjusting the base station energy saving strategy based on the predicted behavior result at the next moment. This solves the problems of poor base station energy saving effect and poor user experience in the related technology, and achieves the effects of improving the base station energy saving effect and enhancing the user experience.

[0042] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc), and includes several instructions for causing a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in various embodiments of the present invention.

[0043] In this embodiment, a control device for base station energy saving is also provided. This device is used to implement the above-mentioned embodiments and preferred implementation manners, and those that have been described will not be repeated here. As used below, the term "module" can be a combination of software and / or hardware that realizes a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0044] An embodiment of the present invention provides a control device for base station energy saving, including: a training module, configured to perform model training based on historical user behavior data to obtain a user behavior prediction model, and the user behavior prediction model has time characteristics; a prediction module, configured to input the user behavior data at the current moment into the user behavior prediction model to obtain a predicted behavior result at the next moment; an adjustment module, configured to adjust the base station energy saving strategy based on the predicted behavior result at the next moment.

[0045] In the actual implementation process, each module of the above-mentioned control device for base station energy saving is used to execute the specific steps of the control method for base station energy saving in the above-mentioned embodiment, which will not be repeated here.

[0046] It should be noted that the above-mentioned each module can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited thereto: the above-mentioned modules are all located in the same processor; or, the above-mentioned each module is located in different processors in any combination form.

[0047] An embodiment of the present invention also provides a computer-readable storage medium, in which a computer program is stored. Wherein, the computer program is configured to execute the steps in any one of the above-mentioned method embodiments when running.

[0048] In an exemplary embodiment, the above-mentioned computer-readable storage medium may include, but is not limited to: USB flash drive, read-only memory (ROM), random access memory (RAM), mobile hard disk, magnetic disk or optical disc, etc., various media that can store computer programs.

[0049] An embodiment of the present invention also provides an electronic device, including a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above-mentioned method embodiments.

[0050] In an exemplary embodiment, the above-mentioned electronic device may further include a transmission device and an input / output device. Wherein, the transmission device is connected to the above-mentioned processor, and the input / output device is connected to the above-mentioned processor.

[0051] For the specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary embodiments, and details thereof will not be repeated herein.

[0052] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present invention can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. They can be implemented by program codes executable by the computing device, so that they can be stored in a storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a different order than here, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module to implement. In this way, the present invention is not limited to any specific combination of hardware and software.

[0053] In order to enable those skilled in the art to better understand the technical solution of the present invention, the following will be described in conjunction with scenario embodiments.

[0054] Scenario Embodiment 1

[0055] In this scenario embodiment, a control system for base station energy saving is provided. Figure 3 It is a structural block diagram of the control system for base station energy saving according to the scenario embodiment of the present invention, as Figure 3 shown, including: a behavior recognition module, a policy control module, a user behavior model learning module, an energy saving execution module, an input interface module, and an output interface module.

[0056] The behavior recognition module mainly recognizes the behavior of the users of the present network service and judges the specific behavior of the users; the user behavior model learning module can clean the user behavior data recognized by the behavior recognition module, train and obtain the specific behavior of the users of the present network within a certain period of time, and indicate the trained model to the policy control module, which can more accurately judge the specific energy saving strategy or de - energy saving strategy of the base station at the next moment, making the behavior of the base station more targeted, so as to improve the energy saving effect while ensuring that the network can provide services normally, timely and accurately.

[0057] According to the above - mentioned control system for base station energy saving, the behavior recognition module is used to recognize and judge user behavior. When a user behavior that meets the requirements is recognized, the data will be reported to the user behavior model learning module and the policy control module at the same time. The user behavior model learning module makes a more accurate prediction of user behavior based on the received reported data.

[0058] After receiving the data, the user behavior model learning module will store the data. When the number of data reaches a certain amount, it will classify, analyze, train the data and make a prediction of user behavior. The prediction results are respectively transmitted to the behavior recognition module and the policy control module.

[0059] In the scenario embodiment of the present invention, the user behavior model learning module trains the user behavior prediction model. Figure 4 It is a schematic diagram of the training process of the user behavior model learning module according to the scenario embodiment of the present invention. As Figure 4 shown, it is necessary to determine whether the energy-saving result is the same as the energy-saving prediction result. If they are not the same, the number of times needs to be recorded. When the number of times of difference reaches the preset threshold, the user behavior prediction model should be updated in a timely manner. If they are the same, it is necessary to further determine whether the energy-saving result meets the prediction. If it meets the expectation, save the model version, update and train the model according to the current user behavior data, and at the same time, the update period of the model should be set. When it is determined that the update period is met, the model is trained and updated. If the energy-saving prediction result does not meet the expectation, it is necessary to count the deviation of the prediction expectation, and jointly decide whether to update and train the model with the number of times of deviation from the prediction result. When performing the update training, the old version of the model should be saved first, and the old version of the model is trained with the current data to obtain the new version of the model.

[0060] The policy control module receives the data reporting information transmitted by the behavior recognition module, identifies the data reporting information, and decides whether to perform energy saving according to the pre-configured side policy parameters. If it is decided that energy saving is required, a command is sent to the energy saving execution module to perform the energy saving action; this module can also receive the preset information of the input interface module for deciding whether to perform energy saving at this site; this module can also receive the control policy parameter configuration and forced execution command set manually; this module can also send the decision information of this time to the output interface module.

[0061] The energy saving execution module is responsible for executing the energy saving command sent by the policy control module and feeding back the execution result to the policy control module.

[0062] The output interface module is used to interact with other sites and send the decision-related information of this site to the adjacent sites.

[0063] The input interface module is responsible for parameter configuration, forced command input, information transmitted from neighboring cells, and feedback of site-related status.

[0064] Figure 5 It is a schematic diagram of the working process of the base station energy-saving control system according to the scenario embodiment of the present invention. As Figure 5 shown, the working principle process of the cooperation between different modules of the base station energy-saving control system is as follows:

[0065] Step S501, the behavior recognition module extracts behavior recognition information and performs preliminary data cleaning.

[0066] The behavior recognition module mainly identifies user behaviors by combining the network attributes of the base station itself, classifies them according to user behaviors, and simultaneously counts the resource usage of the base station and some user-related information, and periodically transmits these statistical information to the policy control module and the user behavior model learning module.

[0067] Step S502: The user behavior model learning module annotates, copies, and cleans the data again, and then trains and learns to obtain a user behavior prediction model.

[0068] After the user behavior model learning module obtains the statistical information of the behavior recognition module, it classifies the data, cleans the data to eliminate interference data, and then trains these data according to the preset classification to obtain the user behavior characteristics, base station resource usage characteristics, etc. within a specific time period, so as to further obtain a user behavior prediction model. According to the model, the user behavior in the future period can be predicted in advance, and the prediction information is output to the policy control module.

[0069] Step S503: The policy control module makes a decision by combining the prediction model data, the behavior recognition module data, and the input interface module data.

[0070] After receiving the data transmitted by the behavior recognition module, the policy control module first judges which sites need to save energy or exit energy saving, and then judges which type of energy saving to perform. If the user behavior model learning module outputs prediction information, it needs to be combined with the prediction information for judgment. In addition, the policy control module needs to respond in a timely manner to the energy-saving prediction information of the user behavior model learning module in order to provide services more timely. The policy control module transmits the decision information to the energy-saving execution module, and the energy-saving execution module saves energy or exits energy saving according to the instruction. At the same time, the policy control module synchronizes the decision information to the user behavior model learning module.

[0071] Step S504: After receiving the decision information, the user behavior model learning module, on the one hand, judges whether the user behavior prediction model is accurate according to the decision information, and on the other hand, fits the real-time behavior data transmitted by the behavior recognition module to judge whether it meets the expectations. If it does not meet the expectations, the model needs to be further trained.

[0072] Step S505: The user can input information in the input interface module according to needs, or obtain data and status information from the output interface module.

[0073] According to the above-mentioned Scenario Embodiment 1 of the present invention, in the actual implementation process, the user behavior model learning module may not be adopted. Data is obtained through the output interface module, and the data is trained in the offline "user behavior model learning module" to obtain a user behavior prediction model. Input and settings are performed in the human-computer interaction module according to the obtained prediction model to achieve the same purpose. The human-computer interaction module can also be added. After the user behavior model learning module completes the training of the user behavior prediction model, the user can observe the relevant information about the base station's expected entry into and exit from the energy-saving state in the human-computer interaction module, and can input commands to control the relevant content of entering and exiting the energy-saving state in case of an emergency.

[0074] Scenario Embodiment 2

[0075] According to the base station energy-saving control system provided in the above Scenario Embodiment 1, in this scenario embodiment, taking the high-speed rail scenario of the network type as an example, the working principle and process of the base station energy-saving control system are introduced.

[0076] Figure 6 It is a schematic diagram of the working principle of the base station energy-saving control system according to the scenario embodiment of the present invention. As Figure 6 shown, the behavior recognition module, the user behavior model learning module, and the policy control module are used as the main modules, and the high-speed rail scenario in the general network of the network type is taken as an example for description. The working principle is as follows:

[0077] Behavior recognition module:

[0078] First, identify the high-speed rail scenario in the general network of the network type. In this scenario, generally, the base station only needs to provide services at a certain moment. Therefore, the following indicators are mainly identified: handover-in users, handover-out users, and the number of online users; other indicators include access users, user speed recognition, and the usage rate of uplink and downlink resources, etc.

[0079] When both the number of handover-in users and the number of online users increase simultaneously, it means that users are gradually entering the cell. However, it may also be that users along the high-speed rail are handed over into this cell. It is necessary to further identify the speed of the handover-in users. If they are low-speed users, these users need to be removed from the handover-in users and the number of online users. In addition, the access users also need to be removed from the number of online users. Only when both the number of handover-in users and the number of online users increase after cleaning, does it indicate that this high-speed rail train is gradually entering this cell.

[0080] The switching out of users and the decrease in the number of online users mean that users are gradually leaving the cell; however, due to the access of users along the high-speed rail, the number of online users may not be cleared to zero. Therefore, it is necessary to clean the data. By identifying the user speed, these users can be removed. If the number of online users after cleaning is 0 or very small (usually within 10), it means that this high-speed rail has left the cell. At this time, it is also necessary to judge the uplink and downlink resource utilization rates. When the uplink and downlink resource utilization rates are very high, it means that there are still users who need to be served. At this time, it is not appropriate to enter the energy-saving state. After the uplink and downlink resource utilization rates decrease or all online users are migrated out of this cell, the energy-saving state can be entered.

[0081] Through the above data collection, the time trajectory of this high-speed rail's activities in this cell can be finally obtained, so that the time period during which this cell should provide services can be obtained.

[0082] User behavior model learning module:

[0083] The user behavior model learning module will continuously obtain the data of the behavior recognition module. When the data reaches a certain amount (more data is required for the first time), according to the time sequence, certain weights are assigned to the data, and then the data is identified, classified, and trained. Finally, the time map of the services that need to be provided in this cell and the corresponding user behavior characteristics will be obtained. Since the time when the high-speed rail passes through this cell is different, each time period that needs to provide services will be a grayscale map according to the probability of the high-speed rail entering. Figure 7 It is the grayscale map of the high-speed rail entry probability in different time periods according to the scenario embodiment of the present invention. As Figure 7 shown, different energy-saving strategies can be selected in combination with the grayscale map. For example: 0% of the time period can enter the station-level energy saving; less than 0.01% of the time period can enter the carrier-level energy saving, and only the basic carriers are retained for wide-area coverage; less than 0.5% of the time period can perform BWP-level energy saving; less than 10% can perform symbol-level energy saving; 30% and above completely exit the energy saving.

[0084] In addition, the user behavior model learning module needs to receive the energy-saving strategy results feedback by the policy control module and monitor the real-time data of the behavior recognition module to judge the accuracy of the prediction results. If the number of times with a large error accumulates to a certain extent, it is necessary to trigger data identification and training again to obtain an updated user behavior prediction model and the corresponding user behavior characteristics to ensure the real-time effectiveness of the model.

[0085] Policy control module:

[0086] The information for the policy control module to make decisions mainly comes from three aspects: First, the real-time data of the behavior recognition module; second, the prediction data of the user behavior model learning module; third, the information transmitted by the input interface module, including the setting information and the information transmitted from neighboring cells.

[0087] Figure 8 It is a schematic diagram of the process principle of the energy-saving exit strategy according to the embodiment of the present invention scenario. As Figure 8 shown, if it is energy-saving exit, first look at the predicted data of the user behavior model learning module, that is, the prediction result. If the prediction result indicates that energy-saving exit is required, directly exit energy-saving according to the data or enter partial energy-saving. Otherwise, look at the data of the input interface module. If there is a setting command for energy-saving exit or an energy-saving exit indication transmitted from the neighboring cell, energy-saving exit is also required. Otherwise, based on the real-time data transmitted by the behavior recognition module, judge whether the data exceeds the set energy-saving exit threshold. If it does, exit energy-saving; otherwise, do not exit energy-saving. Through relevant processing such as the prediction result indication, the cell can enter the energy-saving exit state in advance and make preparations for providing services in advance to ensure better service for users.

[0088] The energy-saving exit strategy does not completely exit energy-saving in one step: it should be pre-set to the corresponding energy-saving exit state according to the predicted data of the user behavior model learning module. At this time, based on the real-time data transmitted by the behavior recognition module, judge what the next user behavior will be. For example, if both the number of handover-in users and the number of online users are gradually increasing, it means that users are gradually entering this cell, then the energy-saving exit state should be further exited. Conversely, it can be temporarily maintained in this energy-saving exit state.

[0089] After completing the energy-saving exit, the result of the energy-saving exit needs to be fed back to the user behavior model learning module (i.e., the learning module). The learning module judges whether the real-time data and the user prediction data match. If they match, the current process ends. Otherwise, it is necessary to further judge whether the handover-in user rate and the change rate of the number of online users are higher than the prediction data. If not, the current process ends. If so, it means that the prediction data is low and cannot meet the actual user needs. Therefore, it is necessary to adjust the energy-saving level of the energy-saving exit, and then feed back the judgment result and the adjustment result to the learning module for the training and update of the next user behavior prediction model.

[0090] If it is energy-saving entry, first look at whether there is a setting command in the input interface module. If there is, then look at whether the real-time data of the behavior recognition module conforms to the energy-saving entry state. If it does not conform, return an alarm and request confirmation information. After receiving the confirmation information, enter the energy-saving entry process; otherwise, do not enter. Then look at the predicted data of the user behavior model learning module and combine it with the real-time data of the behavior recognition module to judge which level of energy-saving needs to be entered.

[0091] In summary, the control method for base station energy saving provided by the embodiments of the present invention obtains a user behavior prediction model through big data analysis and training, which is used to predict the behavior patterns of users within the base station. The base station judges that it can enter the energy-saving state and wake up in advance according to the behavior patterns of users, so as to achieve the purpose of base station energy saving while ensuring the provision of normal services. The base station automatically performs energy saving and de-energy saving according to the identified user behavior, while improving the energy saving and service effects.

[0092] The foregoing are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A control method for base station energy saving, characterized in that, it includes: Performing model training based on historical user behavior data to obtain a user behavior prediction model, and the user behavior prediction model has time characteristics; Inputting the user behavior data at the current moment into the user behavior prediction model to obtain the predicted behavior result at the next moment; Adjusting the base station energy saving strategy based on the predicted behavior result at the next moment.

2. The method according to claim 1, characterized in that, the performing model training based on historical user behavior data to obtain a user behavior prediction model includes: Obtaining the user behavior data in a preset historical period, and classifying the user behavior data in the preset historical period to obtain classified user behavior data; Using the classified user behavior data as training data for model training to obtain the user behavior prediction model.

3. The method according to claim 2, characterized in that, classifying the user behavior data in the preset historical period to obtain classified user behavior data includes: Identifying the network type corresponding to the user behavior data in the preset historical period; Determining the characteristic parameters of different user behaviors in the preset historical period according to the network type; Classifying the user behavior data in the preset historical period based on the characteristic parameters of different user behaviors in the preset historical period to obtain the classified user behavior data.

4. The method according to claim 2, characterized in that, the using the classified user behavior data as training data for model training to obtain the user behavior prediction model includes: Adding data labels based on the time sequence and base station resource characteristics corresponding to each behavior in the classified user behavior data to obtain the training data; Performing model training based on the training data to obtain the user behavior prediction model.

5. The method according to claim 1, characterized in that, after adjusting the base station energy saving strategy based on the predicted behavior result at the next moment, the method further includes: Determining the accuracy of the predicted behavior result at the next moment based on the strategy implementation result corresponding to the adjusted base station energy saving strategy; Triggering the updated training of the user behavior prediction model when the accuracy of the predicted behavior result at the next moment is less than a preset threshold and the number of times that the accuracy of the predicted behavior result fails to meet the standard exceeds a preset number of times.

6. The method according to claim 1, characterized in that, the adjusting the base station energy saving strategy based on the predicted behavior result at the next moment includes: Judging the energy saving strategy or the de - energy saving strategy of the base station at the next moment based on the predicted behavior result at the next moment, and changing the energy saving state or energy saving type of the base station according to the energy saving strategy or the de - energy saving strategy.

7. The method according to claim 1, characterized in that, the adjusting the base station energy saving strategy based on the predicted behavior result at the next moment includes: In the case of receiving an external energy-saving instruction, determine whether the user behavior data at the current moment meets the condition for entering the energy-saving state. If not, request confirmation information from an external device. If the confirmation information is received, enter the energy-saving state; otherwise, do not enter the energy-saving state. In the case of entering the energy-saving state, determine the energy-saving state or energy-saving type of the base station based on the predicted behavior result at the next moment and the user behavior data at the current moment.

8. The method according to claim 1, wherein, the adjustment of the base station energy-saving strategy based on the predicted behavior result at the next moment includes: According to the predicted behavior result at the next moment, pre-regulate the base station to enter the energy-saving exit state; Judge whether the base station exits the energy-saving state according to the user behavior data at the current moment. If the user behavior data at the current moment is consistent with the predicted behavior result at the next moment, exit the energy-saving state; if the user behavior data at the current moment is inconsistent with the predicted behavior result at the next moment, do not exit the energy-saving state and maintain the energy-saving exit state.

9. A computer-readable storage medium, wherein, a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the method described in any one of claims 1 to 8 is implemented.

10. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, when the processor executes the computer program, the method described in any one of claims 1 to 8 is implemented.

Images