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

Abstract

The present disclosure provides a base station energy-saving method, device, electronic device, and storage medium, relating to the field of communication technologies. The method includes obtaining first traffic data of a target base station at the current moment and within a specified time period before the current moment; dividing the first traffic data into multiple data sets according to a preset time length, where each data set includes multiple traffic value data; obtaining the maximum traffic value data in each data set, and arranging the maximum traffic value data in each data set in time sequence as second traffic data; inputting the second traffic data into a pre-trained traffic prediction model to obtain a first prediction result; inputting the first prediction result into a pre-trained base station energy-saving model to obtain a second prediction result; and matching an energy-saving strategy for the target base station according to the second prediction result. The present disclosure can intelligently select an energy-saving strategy for the base station without manual intervention, and maximize the energy-saving effect on the premise of not affecting the normal use of the base station.

Description

Background Art

[0002] With the development of communication technology, base stations are deployed on a large scale. Therefore, the energy consumption of base stations has become an important issue that needs to be solved urgently. Generally, there are two energy-saving modes for base stations: centralized energy-saving and distributed energy-saving. Centralized energy-saving is to implement energy-saving solutions at the network management level, while distributed energy-saving is to implement energy-saving solutions at the network element. Distributed energy-saving is more in line with actual operation and maintenance conditions because the scenarios corresponding to each base station are not completely consistent. At the same time, distributed energy-saving delegates part of the decision-making capabilities to the base station, which can reduce the pressure on the network management backend server and reduce decision-making delays.

[0003] However, currently, both centralized and distributed energy saving need to rely on manual switch settings, which have the defects of limited energy saving effect, low degree of automation and lack of intelligence.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The present disclosure provides a base station energy saving method, device, electronic device and storage medium, which at least to a certain extent overcome the problem that the base station in the related art cannot intelligently select an energy saving solution.

[0006] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0007] According to one aspect of the present disclosure, a base station energy saving method is provided, including:

[0008] Obtain the first flow data of the target base station at the current time and within a specified time period before the current time;

[0009] According to a preset time length, the first flow data is divided into a plurality of data sets, each data set including a plurality of flow value data;

[0010] Obtain the maximum flow value data in each data set, and arrange the maximum flow value data in each data set in time sequence as the second flow data;

[0011] Inputting the second traffic data into a pre-trained traffic prediction model to obtain a first prediction result, wherein the first prediction result is used to determine the traffic usage at a time to be predicted, wherein the time to be predicted is a time after the current time has passed the preset time length;

[0012] Inputting the first prediction result into a pre-trained base station energy-saving model to obtain a second prediction result, where the second prediction result is used to match an energy-saving strategy;

[0013] Match the energy-saving strategy of the target base station according to the second prediction result.

[0014] In an embodiment of the present disclosure, when the first prediction result is the traffic prediction value of the target base station at the to-be-predicted moment, the second prediction result is an energy-saving strategy identifier, wherein the traffic prediction value is obtained by predicting with the traffic prediction model, and there is a preset mapping relationship between the energy-saving strategy identifier and the energy-saving strategy.

[0015] In an embodiment of the present disclosure, the matching of the energy-saving strategy of the target base station according to the second prediction result specifically includes:

[0016] Match the energy-saving strategy corresponding to the energy-saving strategy identifier according to the energy-saving strategy identifier.

[0017] In an embodiment of the present disclosure, when the first prediction result is the base station utilization rate of the target base station at the to-be-predicted moment, the second prediction result is the base station energy-saving rate, wherein the base station utilization rate is calculated based on the traffic prediction value, the traffic prediction value is obtained by predicting with the traffic prediction model, and there is a preset mapping relationship between the base station energy-saving rate and the energy-saving strategy.

[0018] In an embodiment of the present disclosure, the matching of the energy-saving strategy of the target base station according to the second prediction result specifically includes:

[0019] Match the energy-saving strategy corresponding to the base station energy-saving rate according to the base station energy-saving rate.

[0020] In an embodiment of the present disclosure, the energy-saving strategy includes at least one of the following:

[0021] Symbol off strategy, channel off strategy, carrier off strategy, and base station sleep strategy.

[0022] In an embodiment of the present disclosure, the traffic prediction model is trained based on the long short-term memory (LSTM) model, and the base station energy-saving model is trained based on the deep neural network (DNN) model.

[0023] According to another aspect of the present disclosure, there is provided a base station energy-saving device, including:

[0024] A first acquisition module, configured to acquire first traffic data of a target base station at the current moment and within a specified time period before the current moment;

[0025] A division module, configured to divide the first traffic data into a plurality of data sets according to a preset time length, and each data set includes a plurality of traffic value data;

[0026] A second acquisition module, configured to acquire the maximum traffic value data in each dataset, and arrange the maximum traffic value data in each dataset in time sequence as the second traffic data;

[0027] A first prediction module, configured to input the second traffic data into a pre-trained traffic prediction model to obtain a first prediction result, where the first prediction result is used to determine the traffic usage at a to-be-predicted moment, and the to-be-predicted moment is the moment after the preset time length from the current moment;

[0028] A second prediction module, configured to input the first prediction result into a pre-trained base station energy-saving model to obtain a second prediction result, where the second prediction result is used to match an energy-saving strategy; and

[0029] A strategy matching module, configured to match the energy-saving strategy of the target base station according to the second prediction result.

[0030] According to another aspect of the present disclosure, there is provided an electronic device, including: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the above base station energy-saving method by executing the executable instructions.

[0031] According to another aspect of the present disclosure, there is provided a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above base station energy-saving method is implemented.

[0032] The base station energy-saving method, device, electronic device and storage medium provided by the embodiments of the present disclosure first collect historical traffic data of a base station within a certain period of time, construct input data according to a certain period according to the characteristics of the traffic data, then substitute the input data into a pre-trained traffic prediction model to predict the traffic usage, and based on the predicted traffic usage, match the energy-saving strategy of the base station according to a pre-trained base station energy-saving model. Thus, the present disclosure can intelligently select an energy-saving strategy for the base station without manual intervention, and maximize the energy-saving effect without affecting the normal use of the base station.

[0033] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The accompanying drawings here are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure. Obviously, the accompanying drawings in the following description are only some embodiments of the present disclosure, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0035] Figure 1 A schematic diagram showing the structure of a base station energy-saving system in an embodiment of the present disclosure;

[0036] Figure 2 A flowchart showing a base station energy-saving method in an embodiment of the present disclosure;

[0037] Figure 3 A schematic diagram showing a method for training a traffic prediction model in an embodiment of the present disclosure;

[0038] Figure 4 A schematic diagram showing a method for training a base station energy-saving model in an embodiment of the present disclosure;

[0039] Figure 5 A schematic diagram showing a base station energy-saving device in an embodiment of the present disclosure;

[0040] Figure 6 A schematic diagram showing a base station energy-saving system in an embodiment of the present disclosure;

[0041] Figure 7 A schematic diagram showing the operation process of a base station energy-saving system in an embodiment of the present disclosure; and

[0042] Figure 8 A block diagram showing the structure of an electronic device in an embodiment of the present disclosure. Detailed implementation manners

[0043] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more complete and comprehensive, and will fully convey the concept of the example embodiments to those skilled in the art. The features, structures, or characteristics described may be combined in any suitable manner in one or more embodiments.

[0044] In addition, the drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0045] The solution provided by the present disclosure first obtains the first traffic data of the target base station at the current moment and within a specified time period before the current moment; subsequently, according to a preset time length, the first traffic data is divided into multiple data sets, and each data set includes multiple traffic value data; after obtaining multiple data sets, the maximum traffic value data in each data set is obtained, and the maximum traffic value data in each data set is arranged in time sequence as the second traffic data; next, the second traffic data is input into a pre-trained traffic prediction model to obtain a first prediction result, and the first prediction result is used to determine the traffic usage at the moment to be predicted, where the moment to be predicted is the moment after the current moment by the preset time length; the first prediction result is input into a pre-trained base station energy-saving model to obtain a second prediction result, and the second prediction result is used to match the energy-saving strategy; finally, according to the second prediction result, the energy-saving strategy of the target base station is matched.

[0046] The following will describe this exemplary embodiment in detail with reference to the accompanying drawings and embodiments.

[0047] First, an embodiment of the present disclosure provides a base station energy-saving method, which can be executed by any electronic device with computing and processing capabilities.

[0048] Figure 1 The flowchart of a base station energy-saving method in an embodiment of the present disclosure is shown. As Figure 1 shown, the base station energy-saving method provided in the embodiment of the present disclosure includes the following steps:

[0049] S102, obtain the first traffic data of the target base station at the current moment and within a specified time period before the current moment.

[0050] It should be noted that the target base station in the present disclosure is the base station to which the energy-saving strategy is to be matched, and can be 4G, 5G base stations, etc. in different deployment scenarios.

[0051] It should be noted that the application scenario of the base station in the embodiments of the present invention is not limited, and the base station can be a base station device in any communication system. The communication system includes but is not limited to GSM (Global System of Mobile, global mobile communication) system, CDMA (Code Division Multiple Access, code division multiple access) system, WCDMA (Wideband Code Division Multiple Access, wideband code division multiple access) system, GPRS (General Packet Radio Service, general packet radio service), LTE (Long Term Evolution, long term evolution) system, LTE FDD (Frequency Division Duplex, frequency division duplex) system, LTE TDD (Time Division Duplex, time division duplex) system, UMTS (Universal Mobile Telecommunication System, universal mobile communication system), WiMAX (Worldwide Interoperability for Microwave Access, global interoperability for microwave access) communication system, 5G (5th-Generation, fifth-generation mobile communication technology) or NR (New Radio, new radio) communication system

[0052] It should be noted that the first traffic data can be historical traffic data of the target base station arranged in time sequence.

[0053] In some embodiments, the first traffic data of the target base station can be obtained through the back-end of the network management server.

[0054] Specifically, the back-end of the network management server will collect various performance files of the base station from time to time and report them, including but not limited to the real-time traffic data of the PDCP (Packet Data Convergence Protocol) layer, RLC (Radio Link Control) layer, and MAC (Media Access Control) layer.

[0055] In some embodiments, since the PDCP layer is close to the service layer and can more accurately reflect the real-time traffic situation of the current base station than the RLC layer and the MAC layer. Therefore, the traffic data of the PDCP layer of the base station can be selected as the first traffic data.

[0056] It should be noted that, such as Figure 2As shown, the downlink traffic of the PDCP layer of the base station reaches its peak during the day and flattens out and approaches 0 at dawn, showing an obvious tidal effect. Moreover, the peak traffic and the trough traffic are relatively close every day. Therefore, for the specified time period before the current moment, it can be taken on a daily basis according to the tidal effect of the traffic and adjusted according to the actual application scenario and the model prediction effect. For example, in some application scenarios, the traffic data for the 7 days before the current moment can be selected.

[0057] S104. Divide the first traffic data into multiple data sets according to a preset time length, and each data set includes multiple traffic value data.

[0058] It should be noted that regarding the preset time length, since the first traffic data exists in the performance file obtained through the back-end of the network management server, the preset time length cannot be less than the sampling interval of the back-end of the network management server.

[0059] Furthermore, not all of the first traffic data is necessarily carried in the performance file. For example, at 11:20, the collected PDCP traffic value is empty (i.e., "NIL"), without data; at 11:22, the collected PDCP traffic value is 21023, not empty, with data. If the preset time length is small, such as 5 minutes, even if it is greater than the back-end sampling interval, there is still a very small probability that all the performance file values collected are "NIL". Therefore, in order to ensure that there is data to read, there should be several performance files with numerical records within a preset time length, and the preset time length can be set accordingly.

[0060] S106. Obtain the maximum traffic value data in each data set, and arrange the maximum traffic value data in each data set in chronological order as the second traffic data.

[0061] It should be noted that the second traffic data is composed of the maximum traffic value data in each of the divided data sets arranged in chronological order. For example, the first traffic data from 11:00 to 12:00 on a certain day is divided into 4 data sets with a preset time length of 15 minutes, that is, the 4 data sets obtained after division respectively include the traffic value data from 11:00 to 11:15, 11:15 to 11:30, 11:30 to 11:45, and 11:45 to 12:00. At this time, obtain the maximum traffic value data in each data set and arrange the maximum traffic value data in each data set in chronological order to obtain the second traffic data.

[0062] S108. Input the second traffic data into a pre-trained traffic prediction model to obtain a first prediction result, where the first prediction result is used to determine the traffic usage at the moment to be predicted, and the moment to be predicted is the moment after a preset time length from the current moment.

[0063] It should be noted that the traffic prediction model can be trained through an LSTM (Long Short-Term Memory) model. The LSTM model is a neural network model used to process and predict time series data. It can selectively retain or discard part of the information through a built-in forgetting gate and information enhancement gate to process a series of data containing time information and predict the results. Compared with ordinary neural networks, the LSTM model has greater advantages in processing time series data.

[0064] Specifically, as Figure 3 shown, the training process of the traffic prediction model is as follows:

[0065] S302. Construct a traffic prediction model training set. As mentioned above, since the time intervals of the performance files reported by the base station are not completely consistent and there are some missing data, the reported performance files need to be processed according to a certain preset time length before construction. The processing process is as follows: Read the traffic data from the performance files, divide the traffic data into multiple data sets according to the preset time length (for example, 15 minutes), and obtain the maximum value of the traffic data in each data set. Thus, the daily traffic data can be split into 24×4 data sets. Arrange the maximum values of the traffic data in each data set in time series, and finally a traffic array of 24×4 per day can be obtained as the training set.

[0066] For example, assuming that 15 minutes is used as the preset time length to divide the data sets, and the 12 performance files reported by the base station from 11:00 to 11:15 are used as a data set. In this data set, 3 of the performance files are missing PDCP traffic data, and the highest PDCP traffic value recorded in the remaining 9 performance files is 17582. Then, 17582 is taken as the maximum value of the traffic data in this data set. By analogy, the traffic data in any time period can be processed to obtain a complete traffic prediction model training set.

[0067] S304. Based on the constructed traffic prediction model training set, train the LSTM model to obtain a trained traffic prediction model.

[0068] Specifically, the traffic data within a specified time period before a certain moment in the complete training set can be used as the model input, and the traffic data at this moment can be used as the model output to train the LSTM model. After training is completed, the traffic prediction model can be obtained.

[0069] It should be noted that during the training process of the traffic prediction model, the preset time length selected in step 1 is the same as the preset time length in S106 (for example, both are 15 minutes), and the specified time period in step 2 is the same as the specified time period in S102 (for example, both are 7 days).

[0070] It should be noted that the moment to be predicted is related to the preset time length. Specifically, the moment to be predicted is the moment after the preset time length has passed from the current moment. For example, when the preset time length is 15 minutes, the moment to be predicted is the moment 15 minutes after the current moment.

[0071] It should be noted that according to different energy-saving requirements of the target base station, the first prediction result can be presented in different forms.

[0072] In some embodiments, the first prediction result can be the traffic prediction value at the moment to be predicted of the target base station, where the traffic prediction value can be directly obtained by outputting from the traffic prediction model.

[0073] In other embodiments, the first prediction result can be the base station utilization rate at the moment to be predicted of the target base station. The base station utilization rate can be calculated through the traffic prediction value. Specifically, the base station utilization rate = (traffic prediction value / base station full-load traffic value) × 100%.

[0074] S110, input the first prediction result into the pre-trained base station energy-saving model to obtain the second prediction result, and the second prediction result is used to match the energy-saving strategy.

[0075] It should be noted that the base station energy-saving model can be trained through a DNN model. Those skilled in the art can understand that other neural network models (such as CNN) can also be used to train the base station energy-saving model and achieve the same technical effect, and the present disclosure does not make any limitations in this regard. However, due to the relatively simple usage scenario and relatively few parameters in the embodiments of the present disclosure, the use of the DNN model can maximize the training efficiency while ensuring accuracy.

[0076] Specifically, as Figure 4 shown, the training process of the base station energy-saving model is as follows:

[0077] S402. Construct a training set for the base station energy-saving model. According to the relevant 3GPP protocols, a base station can have two states: a non-energy-saving state and an energy-saving state. Among them, the energy-saving state means that in the case of non-peak traffic, some functions of the cell or network element are turned off or resource usage is restricted to reduce energy consumption. The non-energy-saving state means the normal operating state of the base station. Based on these two states, several different energy-saving solutions can be derived. For example: (1) When the traffic at the current moment is lower than the sleep threshold, the base station migrates users to other cells and enters the sleep state. For example, the sleep threshold can be set as the highest daily traffic value in seven days × 0.05. (2) When the traffic at the current moment is lower than the energy-saving threshold, the base station enters the energy-saving state and selects carrier shutdown, channel shutdown, or symbol shutdown according to its own situation. For example, the energy-saving threshold can be set as the highest daily traffic value in seven days × 0.45. (3) When the traffic at the current moment is higher than the energy-saving threshold and lower than the high-energy-consumption warning threshold, the base station maintains its original state unchanged. For example, the high-energy-consumption warning threshold can be set as the highest daily traffic value in seven days × 0.95. (4) When the traffic at the current moment is higher than the high-energy-consumption warning threshold, the model determines that a new traffic peak may occur at the next moment, and the base station sends a warning message to the integrated network management background. Subsequently, the integrated network management determines whether the base station is overloaded and whether some users need to be migrated to neighboring cells.

[0078] It should be noted that since the specific scenarios faced by each base station are different (such as shopping malls, stations, high-rise buildings, etc.), the specific threshold setting method should be tested according to the actual application scenario. In actual applications, the best solution corresponding to the highest daily traffic value in seven days × 0.45 may not necessarily be carrier shutdown / channel shutdown / symbol shutdown. It may be that the base station enters the sleep state (assuming the target base station is a small-capacity base station for supplementary heating and there is a large-capacity macro base station nearby). The situation of macro base stations or home base stations can be inferred by analogy. Therefore, the base station needs to verify through experiments to obtain the corresponding relationship between the current traffic value and the energy-saving strategy.

[0079] It should be noted that the embodiments of the present disclosure need to collect training data according to the actual application scenario. Therefore, different application scenarios (such as office buildings and stations, which are two completely different scenarios) will train different base station energy-saving models. For example, the traffic value in an office building should have a peak during working hours and a trough during non-working hours; an office building is a multi-story office building, and the base station signal may be affected by the base station signals of several upper and lower floors. The traffic value in a station will have a peak during holidays; in addition, since the station is relatively open, the degree of interference from other base station signals is completely different from that of an office building. Therefore, even for the same traffic value, the energy-saving strategies used in different places are not the same. For this reason, multiple different base station energy-saving models can be trained to meet the requirements in different application scenarios.

[0080] In some embodiments, by experimentally verifying the energy-saving strategies of the target base station, multiple sets of data of (current traffic, energy-saving strategy) can be obtained, thereby constructing a complete training set.

[0081] It should be noted that in the embodiments of the present disclosure, for the convenience of training the base station energy-saving model, different energy-saving strategies can be represented by different energy-saving strategy identifiers (such as -1, 0, 1, 2), that is, a mapping relationship is established between the energy-saving strategy and the energy-saving strategy identifier.

[0082] In other embodiments, since the base station energy-saving rate is also closely related to the energy-saving strategy, by training the base station energy-saving model to fit the base station energy-saving rate, a more accurate selection of the energy-saving strategy can be made.

[0083] It should be noted that the base station energy-saving rate is the ratio of the capacity that can be saved by the base station under the current load. The theoretical calculation formula of the base station energy-saving rate is: base station energy-saving rate = 1 - (current traffic value / base station full-load traffic value) * 100% - reserved idle rate. However, in fact, due to different application scenarios of the base station, there is a certain gap between the actual base station energy-saving rate and the theory. Therefore, the actual base station energy-saving rate can be obtained through experimental verification.

[0084] Furthermore, in the embodiments of the present disclosure, by experimentally verifying the base station energy-saving rate of the target base station, multiple sets of data of (base station utilization rate, base station energy-saving rate) can also be obtained, thereby constructing a complete base station energy-saving model training set.

[0085] S404, Based on the constructed base station energy-saving model training set, train the DNN model to obtain a trained base station energy-saving model.

[0086] It should be noted that when the first prediction result in S108 is the traffic prediction value at the time to be predicted for the target base station, the second prediction result in S110 is the energy-saving strategy identifier, where the energy-saving strategy identifier has a mapping relationship with the energy-saving strategy.

[0087] It should be noted that when the first prediction result in S108 is the base station utilization rate at the time to be predicted for the target base station, the second prediction result in S110 is the base station energy-saving rate, where the base station energy-saving rate has a mapping relationship with the energy-saving strategy.

[0088] S112, According to the second prediction result, match the energy-saving strategy of the target base station.

[0089] It should be noted that the energy-saving strategies include but are not limited to symbol shutdown strategy, channel shutdown strategy, carrier shutdown strategy, and base station sleep strategy.

[0090] It should be noted that the energy-saving strategy of the target base station is matched based on the second prediction result in S110. Since there is a mapping relationship between the energy-saving strategy identifier and the base station energy-saving rate and the energy-saving strategy, the energy-saving strategy can be matched through either the energy-saving strategy identifier or the base station energy-saving rate.

[0091] It is worth noting that when using the base station energy-saving rate to match the energy-saving strategy, since the mapping relationship between the base station energy-saving rate and the energy-saving strategy usually cannot be fully matched. For example, according to the mapping relationship between the base station energy-saving rate and the energy-saving strategy, when the base station energy-saving rate is 10%, energy-saving strategy A is selected; when the base station energy-saving rate is 15%, energy-saving strategy B is selected. If the base station energy-saving rate of the target base station is 13%, there is no corresponding energy-saving strategy. At this time, to ensure that the base station has sufficient capacity for use, the energy-saving strategy corresponding to the base station energy-saving rate that is closest to and less than the base station energy-saving rate of the target base station can be matched, that is, energy-saving strategy A. At the same time, to achieve precise optimization of the energy-saving strategy, the energy-saving strategy A can be preset manually for further adjustment to make the energy-saving rate reach 13%.

[0092] By combining the traffic prediction model and the base station energy-saving model, the present disclosure can directly obtain the best energy-saving plan for the base station at the next moment, and at the same time retain the ability to accept the overall scheduling of the network management in the energy-saving plan, which is more convenient to operate, more in line with the actual application scenario, and the decision-making process of the base station itself.

[0093] For ease of understanding, a specific application example will be provided below to introduce the process of the base station energy-saving method provided by the embodiments of the present disclosure from training to use through the following steps 1 to 7.

[0094] Step 1: Read the performance file reported by the base station, parse the performance file, divide it into multiple data sets according to a preset time length of 15 minutes, and construct a series of time series data based on the highest traffic value in each data set.

[0095] Step 2: Prove through experiments the optimal energy-saving strategies corresponding to different traffic loads of the base station (such as symbol shutdown / carrier shutdown / channel shutdown / sleep / overload reporting), and construct a batch of training data with one-to-one correspondence between traffic values and energy-saving strategies.

[0096] Step 3: Use the time series data constructed in Step 1 to train the LSTM network. After training is completed, a traffic prediction model is obtained.

[0097] Step 4: Use the training data in Step 2 to train the DNN. After training is completed, a base station energy-saving model is obtained.

[0098] Step 5: Obtain the traffic data of the base station at the current moment and the previous 7 days, input it into the traffic prediction model, and the traffic prediction value of the base station at the next moment can be obtained.

[0099] Step 6: Input the traffic prediction value of the base station at the next moment into the energy-saving model to obtain the energy-saving policy identifier of the base station at the next moment. According to the energy-saving policy identifier, the corresponding energy-saving policy can be matched.

[0100] Step 7: At the next moment (15 minutes later), execute the energy-saving policy of the base station at the next moment.

[0101] Based on the same inventive concept, an embodiment of the present disclosure also provides a base station energy-saving device as described in the following embodiments. Since the principle of solving problems in this device embodiment is similar to that in the above method embodiment, the implementation of this device embodiment can refer to the implementation of the above method embodiment, and repeated parts will not be described again.

[0102] Figure 5 The following shows a schematic diagram of a base station energy-saving device in an embodiment of the present disclosure, as Figure 5 shown. The device 500 includes:

[0103] A first acquisition module 501, configured to acquire first traffic data of the target base station at the current moment and within a specified time period before the current moment;

[0104] A division module 502, configured to divide the first traffic data into multiple data sets according to a preset time length, and each data set includes multiple traffic value data;

[0105] A second acquisition module 503, configured to acquire the maximum traffic value data in each data set, and arrange the maximum traffic value data in each data set in time sequence as the second traffic data;

[0106] A first prediction module 504, configured to input the second traffic data into a pre-trained traffic prediction model to obtain a first prediction result, where the first prediction result is used to determine the traffic usage at the moment to be predicted, and the moment to be predicted is the moment after a preset time length from the current moment;

[0107] A second prediction module 505, configured to input the first prediction result into a pre-trained base station energy-saving model to obtain a second prediction result, where the second prediction result is used to match the energy-saving policy; and

[0108] A policy matching module 506, configured to match the energy-saving policy of the target base station according to the second prediction result.

[0109] In some embodiments, when the first prediction result is the traffic prediction value of the target base station at the moment to be predicted, the second prediction result is the energy-saving policy identifier, where the traffic prediction value is predicted by the traffic prediction model, and there is a preset mapping relationship between the energy-saving policy identifier and the energy-saving policy.

[0110] In some embodiments, the policy matching module 506 is specifically configured to:

[0111] Match the energy-saving policy corresponding to the energy-saving policy identifier according to the energy-saving policy identifier.

[0112] In some embodiments, when the first prediction result is the base station utilization rate at the predicted time of the target base station, the second prediction result is the base station energy-saving rate, where the base station utilization rate is calculated based on the traffic prediction value, the traffic prediction value is predicted by the traffic prediction model, and there is a preset mapping relationship between the base station energy-saving rate and the energy-saving policy.

[0113] In some embodiments, the policy matching module 506 is specifically configured to:

[0114] Match the energy-saving policy corresponding to the base station energy-saving rate according to the base station energy-saving rate.

[0115] In some embodiments, the energy-saving policy includes at least one of the following:

[0116] Symbol shutdown policy, channel shutdown policy, carrier shutdown policy, and base station sleep policy.

[0117] In some embodiments, the traffic prediction model is trained based on the long short-term memory (LSTM) model, and the base station energy-saving model is trained based on the deep neural network (DNN) model.

[0118] It should be noted that when the base station energy-saving device provided in the above embodiments is used for base station energy-saving, only the above-mentioned division of each functional module is used for illustration. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the base station energy-saving device provided in the above embodiments and the embodiments of the base station energy-saving method belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0119] Those skilled in the art can understand that various aspects of the present disclosure can be implemented as a system, method, or program product. Therefore, various aspects of the present disclosure can be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects, which can be collectively referred to as "circuitry", "module", or "system" here.

[0120] Based on the same inventive concept, an embodiment of the present disclosure also provides a base station energy-saving system as described in the following embodiments. Since the principle of solving problems in this system embodiment is similar to that of the above method embodiment, the implementation of this system embodiment can refer to the implementation of the above method embodiment, and the repeated parts will not be described again.

[0121] Figure 6 shows a schematic diagram of a base station energy saving system in an embodiment of the present disclosure, as Figure 6 shown, the system 600 includes:

[0122] A data processing module 601. The first part of this module is to parse the performance file read from the network management background. Based on a preset time length (e.g., 15 min), it parses the traffic data and processes it into several data sets. The processing method is as follows: Divide the traffic data of each day into 24×4 data sets, and a series of time series data can be obtained by selecting the maximum traffic value in each data set. For the data processing of the second part of this module, through experiments, the corresponding relationship between the base station traffic value and the energy saving policy identifier (e.g., -1, 0, 1, 2), and the corresponding relationship between the base station utilization rate and the base station energy saving rate need to be constructed. Among them, both the energy saving policy identifier and the base station energy saving rate have a preset mapping relationship with the energy saving policy of the target base station. Of course, according to actual applications, those skilled in the art can only construct one of the corresponding relationships as the training set data, and the embodiments of the present disclosure do not limit this.

[0123] A model training module 602, which includes two parts. The first part obtains the first-stage training data constructed in the data processing module 601 and uses an LSTM network for training. After the training is completed, a traffic prediction model with the traffic characteristics of the target base station will be obtained. The second part obtains the second-stage training data constructed in the data processing module 601 and uses a DNN network for training. After the training is completed, a base station energy saving model will be obtained.

[0124] A model verification module 603. This module first obtains the traffic prediction model and the base station energy saving model in the model training module 602. When using the model verification module 603 to verify or use the traffic prediction model and the base station energy saving model, the traffic array of the current moment and the previous 7 days after being processed can be input into the traffic prediction model to obtain the traffic prediction value of the next moment; then the traffic prediction value of the next moment is input into the base station energy saving model to obtain the corresponding energy saving policy of the next moment, and execute this energy saving policy.

[0125] Those skilled in the art can understand that various aspects of the present disclosure can be implemented as a system, method, or program product. Therefore, various aspects of the present disclosure can be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects, which can be collectively referred to as "circuit", "module", or "system" here.

[0126] Figure 7 shows a schematic diagram of the operation process of a base station energy saving system in an embodiment of the present disclosure.

[0127] Specifically, asFigure 7 As shown in Figure 7 , the data processing module is responsible for parsing and processing the performance files reported by the base station, and constructing the training data of the LSTM model and the training data of the DNN model in the manner of the previous embodiment. The two sets of training data are respectively input into the model training module for training.

[0128] Using the above data, the model training module trains a traffic prediction model with the LSTM model and a base station energy-saving model with the DNN model. After the training is completed, the model files are saved and the model verification module is entered.

[0129] The model verification module first obtains the traffic values of the base station at the current moment and before, uses the traffic prediction model to predict the traffic at the next moment, and then uses the energy-saving model to predict the energy-saving solution at the next moment according to the predicted traffic at the next moment.

[0130] Next, refer to Figure 8 to describe the electronic device 800 according to this embodiment of the present disclosure. Figure 8 The electronic device 800 shown is only an example, and should not impose any limitations on the functions and usage scope of the embodiments of the present disclosure.

[0131] As Figure 8 shown, the electronic device 800 is presented in the form of a general-purpose computing device. The components of the electronic device 800 may include, but are not limited to: at least one of the above processing units 810, at least one of the above storage units 820, and a bus 830 connecting different system components (including the storage unit 820 and the processing unit 810).

[0132] Among them, the storage unit stores program codes, and the program codes can be executed by the processing unit 810, so that the processing unit 810 executes the steps according to various exemplary embodiments of the present disclosure described in the "Exemplary Method" section of this specification. For example, the processing unit 810 may execute the following steps of the above method embodiment: obtaining first traffic data of a target base station at the current moment and within a specified time period before the current moment; dividing the first traffic data into multiple data sets according to a preset time length, and each data set includes multiple traffic value data; obtaining the maximum traffic value data in each data set, and arranging the maximum traffic value data in each data set in time sequence as second traffic data; inputting the second traffic data into a pre-trained traffic prediction model to obtain a first prediction result, where the first prediction result is used to determine the traffic usage amount at a moment to be predicted, and the moment to be predicted is the moment after the current moment after a preset time length; inputting the first prediction result into a pre-trained base station energy-saving model to obtain a second prediction result, where the second prediction result is used to match an energy-saving strategy; and matching an energy-saving strategy of the target base station according to the second prediction result.

[0133] In some embodiments, when the first prediction result is the traffic prediction value at the moment to be predicted for the target base station, the second prediction result is the energy-saving policy identifier, where the traffic prediction value is obtained by predicting with a traffic prediction model, and there is a preset mapping relationship between the energy-saving policy identifier and the energy-saving policy.

[0134] In some embodiments, it specifically includes:

[0135] According to the energy-saving policy identifier, match the energy-saving policy corresponding to the energy-saving policy identifier.

[0136] In some embodiments, when the first prediction result is the base station utilization rate at the moment to be predicted for the target base station, the second prediction result is the base station energy-saving rate, where the base station utilization rate is calculated based on the traffic prediction value, the traffic prediction value is obtained by predicting with a traffic prediction model, and there is a preset mapping relationship between the base station energy-saving rate and the energy-saving policy.

[0137] In some embodiments, it specifically includes:

[0138] According to the base station energy-saving rate, match the energy-saving policy corresponding to the base station energy-saving rate.

[0139] In some embodiments, the energy-saving policy includes at least one of the following:

[0140] Symbol shutdown policy, channel shutdown policy, carrier shutdown policy, and base station sleep policy.

[0141] In some embodiments, the traffic prediction model is trained based on the long short-term memory (LSTM) model, and the base station energy-saving model is trained based on the deep neural network (DNN) model.

[0142] The storage unit 820 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 8201 and / or a cache storage unit 8202, and may further include a read-only storage unit (ROM) 8203.

[0143] The storage unit 820 may also include a program / utilities 8204 having a set (at least one) of program modules 8205. Such program modules 8205 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment.

[0144] The bus 830 may represent one or more of several types of bus structures, including a storage unit bus or a storage unit controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0145] The electronic device 800 can also communicate with one or more external devices 840 (such as a keyboard, a pointing device, a Bluetooth device, etc.), and can also communicate with one or more devices that enable a user to interact with the electronic device 800, and / or communicate with any device that enables the electronic device 800 to communicate with one or more other computing devices (such as a router, a modem, etc.). Such communication can be carried out through the input / output (I / O) interface 850. Moreover, the electronic device 800 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through the network adapter 860. As shown in the figure, the network adapter 860 communicates with other modules of the electronic device 800 through the bus 830. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in combination with the electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0146] Through the description of the above embodiments, those skilled in the art can easily understand that the exemplary embodiments described herein can be implemented by software, or can be implemented by the way of software in combination with necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, and the software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0147] In an exemplary embodiment of the present disclosure, there is also provided a computer-readable storage medium, which can be a readable signal medium or a readable storage medium. A program product capable of implementing the above method of the present disclosure is stored thereon. In some possible implementation manners, various aspects of the present disclosure can also be implemented in the form of a program product, which includes program code. When the program product runs on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary embodiments of the present disclosure described in the "Exemplary Method" section of this specification.

[0148] More specific examples of the computer-readable storage medium in the present disclosure may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0149] In the present disclosure, a computer-readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, in which readable program code is carried. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A readable signal medium may also be any readable medium other than a readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device.

[0150] Optionally, the program code contained on a computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0151] In a specific implementation, the program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on a user computing device, partially on the user device, executed as a stand-alone software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., connected through the Internet using an Internet service provider).

[0152] It should be noted that although several modules or units of a device for action execution are mentioned in the foregoing detailed description, such a division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more of the above-mentioned modules or units may be embodied in one module or unit. Conversely, the features and functions of one module or unit described above may be further divided and embodied by a plurality of modules or units.

[0153] In addition, although the various steps of the methods in the present disclosure are described in a specific order in the drawings, this does not require or imply that these steps must be performed in that specific order, or that all of the steps shown must be performed to achieve the desired result. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution, etc.

[0154] From the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software or by a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (such as a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0155] After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily conceive of other embodiments of the present disclosure. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure, which follow the general principles of the present disclosure and include known common general knowledge or conventional technical means in the technical field not disclosed by the present disclosure. The specification and examples are only regarded as exemplary, and the true scope and spirit of the present disclosure are pointed out by the appended claims.

Claims

1. A base station energy saving method, characterized in that, Including: Obtain the first traffic data of the target base station at the current moment and within a specified time period before the current moment; According to a preset time length, divide the first traffic data into multiple data sets, and each data set includes multiple traffic value data; Obtain the maximum traffic value data in each data set, and arrange the maximum traffic value data in each data set in time sequence as the second traffic data; Input the second traffic data into a pre-trained traffic prediction model to obtain a first prediction result, where the first prediction result is used to determine the traffic usage at the moment to be predicted, and the moment to be predicted is the moment after the current moment by the preset time length; Input the first prediction result into a pre-trained base station energy-saving model to obtain a second prediction result, where the second prediction result is used to match an energy-saving strategy; According to the second prediction result, match the energy-saving strategy of the target base station; According to the second prediction result, matching the energy-saving strategy of the target base station includes: fitting the base station energy-saving rate through training the base station energy-saving model to determine the energy-saving strategy of the target base station; where the base station energy-saving rate is the ratio of the capacity saved under the current load of the base station; When the base station energy-saving rate cannot fully match the existing energy-saving strategies, select the energy-saving strategy closest to and less than the energy-saving rate of the target base station or introduce an artificial adjustment mechanism to determine the corresponding energy-saving strategy.

2. The base station energy saving method according to claim 1, wherein, When the first prediction result is the traffic prediction value at the moment to be predicted of the target base station, the second prediction result is an energy-saving strategy identifier, where the traffic prediction value is predicted by the traffic prediction model, and there is a preset mapping relationship between the energy-saving strategy identifier and the energy-saving strategy.

3. The base station energy saving method according to claim 2, characterized in that, The matching of the energy-saving strategy of the target base station according to the second prediction result specifically includes: According to the energy-saving strategy identifier, match the energy-saving strategy corresponding to the energy-saving strategy identifier.

4. The base station energy saving method according to claim 1, characterized in that, When the first prediction result is the base station usage rate at the moment to be predicted of the target base station, the second prediction result is the base station energy-saving rate, where the base station usage rate is calculated based on the traffic prediction value, the traffic prediction value is predicted by the traffic prediction model, and there is a preset mapping relationship between the base station energy-saving rate and the energy-saving strategy.

5. The base station energy saving method according to claim 4, wherein, The matching of the energy-saving strategy of the target base station according to the second prediction result specifically includes: According to the base station energy-saving rate, match the energy-saving strategy corresponding to the base station energy-saving rate.

6. The base station energy saving method according to claim 1, characterized in that The energy-saving strategy includes at least one of the following: Symbol shutdown strategy, channel shutdown strategy, carrier shutdown strategy, and base station sleep strategy.

7. The base station energy saving method according to claim 1, characterized in that The traffic prediction model is trained based on the long short-term memory (LSTM) model, and the base station energy-saving model is trained based on the deep neural network (DNN) model.

8. An energy-saving device for a base station, characterized in that, Including: A first acquisition module for acquiring the first traffic data of the target base station at the current moment and within a specified time period before the current moment; A division module for dividing the first traffic data into multiple data sets according to a preset time length, and each data set includes multiple traffic value data; A second acquisition module, configured to acquire the maximum traffic value data in each dataset, and arrange the maximum traffic value data in each dataset in time series as second traffic data; A first prediction module, configured to input the second traffic data into a pre-trained traffic prediction model to obtain a first prediction result, where the first prediction result is used to determine the traffic usage at a to-be-predicted moment, and the to-be-predicted moment is the moment after a preset time length from the current moment; A second prediction module, configured to input the first prediction result into a pre-trained base station energy-saving model to obtain a second prediction result, where the second prediction result is used to match an energy-saving strategy; and A strategy matching module, configured to match the energy-saving strategy of the target base station according to the second prediction result; The strategy matching module is further configured to fit the base station energy-saving rate through training the base station energy-saving model to determine the energy-saving strategy of the target base station; where the base station energy-saving rate is the ratio of the capacity saved by the base station under the current load; when the base station energy-saving rate cannot fully match the existing energy-saving strategies, select the energy-saving strategy corresponding to the energy-saving rate that is closest to and less than that of the target base station or introduce an artificial adjustment mechanism to determine the corresponding energy-saving strategy.

9. An electronic device, characterized in that, Comprising: A processor; and A memory, configured to store executable instructions of the processor; Wherein, the processor is configured to execute the base station energy-saving method according to any one of claims 1 to 7 by executing the executable instructions.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the base station energy-saving method according to any one of claims 1 to 7.

Images