CN115038148A - Energy-saving threshold determination method, device, equipment and storage medium - Google Patents
Energy-saving threshold determination method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN115038148A CN115038148A CN202110209109.6A CN202110209109A CN115038148A CN 115038148 A CN115038148 A CN 115038148A CN 202110209109 A CN202110209109 A CN 202110209109A CN 115038148 A CN115038148 A CN 115038148A
- Authority
- CN
- China
- Prior art keywords
- data
- cell
- energy
- scene
- state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 230000015654 memory Effects 0.000 claims description 29
- 238000012549 training Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 12
- 238000005457 optimization Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 7
- 238000010845 search algorithm Methods 0.000 claims description 4
- 230000006403 short-term memory Effects 0.000 claims description 4
- 230000007787 long-term memory Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 abstract description 9
- 230000000694 effects Effects 0.000 abstract description 7
- 230000006870 function Effects 0.000 description 7
- 238000005192 partition Methods 0.000 description 6
- 230000001360 synchronised effect Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 230000004913 activation Effects 0.000 description 3
- 238000011217 control strategy Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000009849 deactivation Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000002542 deteriorative effect Effects 0.000 description 1
- 235000019800 disodium phosphate Nutrition 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
- H04W52/02—Power saving arrangements
- H04W52/0203—Power saving arrangements in the radio access network or backbone network of wireless communication networks
- H04W52/0206—Power saving arrangements in the radio access network or backbone network of wireless communication networks in access points, e.g. base stations
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Design And Manufacture Of Integrated Circuits (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a method, a device, equipment and a storage medium for determining an energy-saving threshold. The method comprises the following steps: acquiring characteristic data of a current set time length of a target cell; predicting the feature data of the next set duration based on the feature data of the current set duration; determining a scene state corresponding to the next set time length of the target cell based on the predicted characteristic data and a cell state base which is constructed in advance; determining a target energy-saving threshold corresponding to the next set time length of the target cell based on the scene state and a pre-constructed energy-saving threshold library; the cell state library comprises corresponding scene states determined based on the engineering parameters of the cells and historical user distribution, and the scene states can indicate first results of space dimension classification corresponding to the cells and second results of continuous classification based on time dimension on the basis of the space dimension classification. The target energy-saving threshold determined by the embodiment of the application can adapt to the change of the user distribution of the target cell based on the time dimension, and the energy-saving effect is better.
Description
Technical Field
The present disclosure relates to wireless communications, and in particular, to a method, an apparatus, a device, and a storage medium for determining an energy saving threshold.
Background
In the case of wireless communication networks, related power saving techniques include: cell off, carrier off, channel off, and symbol off. The principle is that when the network load is light, part of wireless resources are closed, and the rest resources are utilized to provide services, so that the aim of saving energy is fulfilled.
The energy-saving scheme mainly comprises two basic processes: energy-saving activation and energy-saving deactivation, wherein the energy-saving activation means that: the purpose of saving energy is achieved by closing the cell or limiting the use of part of physical resources, and the corresponding base station enters an energy-saving state; the energy-saving deactivation means that: the closed cell is turned on or the use of the restricted physical resources is resumed to satisfy the increased Service demand and QoS (Quality of Service) demand, and the corresponding base station is resumed from the power saving state to the normal state.
In the related art, whether a cell enters an energy saving state is often determined according to a preset energy saving parameter value (i.e., an energy saving threshold). However, the preset energy saving parameter value is determined according to the designated scene, and the energy saving parameter value needs to be manually set after analyzing different network scenes, which results in high labor cost. In addition, the specific scene of the cell changes, which easily causes the problems of unreasonable configuration of the preset energy-saving parameter values, low accuracy and the like, thereby reducing the energy-saving effect of the base station and failing to meet the dynamic environment change and energy-saving requirements.
Disclosure of Invention
In view of this, embodiments of the present application provide a method, an apparatus, a device, and a storage medium for determining an energy saving threshold, which aim to optimize an energy saving effect of a base station.
The technical scheme of the embodiment of the application is realized as follows:
the embodiment of the application provides a method for determining an energy-saving threshold, which comprises the following steps:
acquiring feature data of a current set time length of a target cell, wherein the feature data is used for indicating user distribution of the target cell;
predicting the feature data of the next set duration based on the feature data of the current set duration;
determining a scene state corresponding to the next set time length of the target cell based on the predicted characteristic data and a pre-constructed cell state library;
determining a target energy-saving threshold corresponding to the next set time length of the target cell based on the scene state and a pre-constructed energy-saving threshold library;
the cell state library comprises corresponding scene states determined based on engineering parameters of cells and historical user distribution, the scene states can indicate first results of space dimension classification corresponding to the cells and second results of continuous classification based on time dimension on the basis of the space dimension classification, and the energy-saving threshold library comprises corresponding energy-saving thresholds determined based on historical working parameters of the scene states in the cell state library.
An embodiment of the present application further provides an apparatus for determining an energy saving threshold, including:
the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring characteristic data of a current set time length of a target cell, and the characteristic data is used for indicating user distribution of the target cell;
the prediction module is used for predicting the characteristic data of the next set time length based on the characteristic data of the current set time length;
a first determining module, configured to determine, based on the predicted feature data and a pre-constructed cell state library, a scene state of the target cell corresponding to the next set duration;
a second determining module, configured to determine, based on the scene state and a pre-constructed energy saving threshold library, a target energy saving threshold corresponding to the next set duration for the target cell;
the cell state library comprises corresponding scene states determined based on engineering parameters of cells and historical user distribution, the scene states can indicate first results of space dimension classification corresponding to the cells and second results of continuous classification based on time dimension on the basis of the space dimension classification, and the energy-saving threshold library comprises corresponding energy-saving thresholds determined based on historical working parameters of the cell state library under various scene states.
An embodiment of the present application further provides an energy saving threshold determining device, including: a processor and a memory for storing a computer program capable of running on the processor, wherein the processor, when running the computer program, is configured to perform the steps of the method according to an embodiment of the present application.
The embodiment of the present application further provides a storage medium, where a computer program is stored on the storage medium, and when the computer program is executed by a processor, the steps of the method in the embodiment of the present application are implemented.
According to the technical scheme provided by the embodiment of the application, the feature data of the next set time length is predicted based on the feature data of the current set time length of the target cell, and the scene state corresponding to the next set time length of the target cell is determined based on the predicted feature data and a pre-constructed cell state library, wherein the scene state can indicate a first result of space dimension classification corresponding to the cell and a second result of continuous classification based on the time dimension on the basis of the space dimension classification, so that the predicted scene state of the target cell is a classification result in the space dimension and the time dimension; and then, a target energy-saving threshold corresponding to the next set time length of the target cell is determined based on the scene state and a pre-constructed energy-saving threshold library, so that the determined target energy-saving threshold can adapt to the change of the user distribution of the target cell based on the time dimension, and further, the energy-saving control strategy can be changed based on the change of the user distribution, and the energy-saving effect is better.
Drawings
Fig. 1 is a schematic flowchart of an energy saving threshold determining method according to an embodiment of the present application;
FIG. 2 is a flowchart illustrating an exemplary threshold determination method applied in the present application;
fig. 3 is a schematic structural diagram of an energy saving threshold determining apparatus according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of an energy saving threshold determining device according to an embodiment of the present application.
Detailed Description
The present application will be described in further detail with reference to the following drawings and examples.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the description of the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.
In the related art, the energy saving state of a cell is often determined according to a preset energy saving parameter value, the preset energy saving parameter configuration is often set only based on the scene classification (for example, scenes of hospitals, factories, schools, and the like) of the cell in the spatial dimension, the granularity of the scene classification is coarse, the configuration precision of the energy saving parameter value is low, the dynamic environment change cannot be adapted, and the energy saving effect is limited.
Based on this, in various embodiments of the present application, a method for performing personalized configuration on an energy saving threshold of a cell based on classification results of the cell in two dimensions, namely time and space, is provided, so that a determined energy saving threshold not only adapts to a scene classification requirement of the cell in the space dimension, but also adapts to a change of user distribution in the time dimension, and thus an energy saving control strategy can be changed based on the change of the user distribution, and an energy saving effect is better.
The embodiment of the present application provides a method for determining an energy saving threshold, which may be applied to a base station, and may also be applied to a network device (for example, an energy saving operation and maintenance platform) in communication connection with the base station, and this is not specifically limited in this application. As shown in fig. 1, the method for determining the energy saving threshold includes:
And 102, predicting the characteristic data of the next set time length based on the characteristic data of the current set time length.
And 103, determining a scene state corresponding to the next set time length of the target cell based on the predicted characteristic data and a cell state base which is constructed in advance.
And 104, determining a target energy-saving threshold corresponding to the next set time length of the target cell based on the scene state and a pre-constructed energy-saving threshold library.
In the embodiment of the application, the cell state library includes corresponding scene states determined based on engineering parameters of a cell and historical user distribution, the scene states can indicate a first result of spatial dimension classification corresponding to the cell and a second result of continuous classification based on time dimension on the basis of the spatial dimension classification, and the energy-saving threshold library includes corresponding energy-saving thresholds determined based on historical working parameters under various scene states in the cell state library.
In the embodiment of the application, feature data of the next set time length is predicted based on feature data of the current set time length of a target cell, and a scene state corresponding to the next set time length of the target cell is determined based on the predicted feature data and a pre-constructed cell state library, wherein the scene state can indicate a first result of space dimension classification corresponding to the cell and a second result of continuous classification based on a time dimension on the basis of the space dimension classification, so that the predicted scene state of the target cell is a classification result in the space dimension and the time dimension; and then, a target energy-saving threshold corresponding to the next set time length of the target cell is determined based on the scene state and a pre-constructed energy-saving threshold library, so that the determined target energy-saving threshold can adapt to the change of the user distribution of the target cell based on the time dimension, and further, the energy-saving control strategy can be changed based on the change of the user distribution, and the energy-saving effect is better.
Illustratively, the characteristic data may be MR (Measurement Report) data of the cell. The set time duration may be reasonably set according to the update requirement of the energy saving threshold, for example, the set time duration may be one hour, or may be a time duration longer than or shorter than one hour, which is not specifically limited in this application. The energy saving threshold can be understood as a parameter value for performing energy saving control.
For example, a cell state base and an energy-saving threshold base may be pre-constructed based on an offline algorithm, feature data of a next set time duration of a target cell is predicted based on an online algorithm, and a scene state of the target cell and a corresponding target energy-saving threshold are determined.
Illustratively, the method for determining the energy-saving threshold further comprises:
and clustering to generate the cell state library based on the engineering parameters of the cell and historical user distribution.
Taking the energy-saving threshold determination method applied to the network device as an example, the network device may collect the engineering parameters of the cell and the historical user distribution, and cluster to generate the cell state library.
Illustratively, clustering generates the cell state library based on the engineering parameters and the historical user distribution of the cells, including:
clustering and dividing each cell based on the engineering parameters of the cell to obtain the first result;
clustering and dividing the user distribution of different set durations of the various types of cell histories obtained by dividing the first result to obtain a second result;
determining the scene state based on the first result and the second result.
Here, the engineering parameters of the cell may include index data indicating the coverage of the cell itself, a geographical scenario, for example, including: station height, downward inclination angle, total number of configured adjacent regions, average station spacing, weak coverage and overlapping coverage. And clustering and dividing the engineering parameters of each cell to obtain a first result representing the scene classification of the cells in the spatial dimension. Here, the clustering partition may employ a K-means clustering algorithm (K-means clustering algorithm) so as to perform a first type of partition on the cells based on the engineering parameters of the cells. Illustratively, the results of the first type of partitioning are shown in table 1:
TABLE 1
| Cell identity | Scene state label |
| Cell-1 | Cluster # scene A |
| Cell-2 | Cluster # scene B |
| Cell-3 | Cluster # scene C |
| … | … |
It should be noted that the engineering parameters of the cell may further include: the indexes such as the coverage scene and the building name, that is, the coverage scene, the building name, and the like of each cell may be considered when performing the clustering of the first-class partitions.
Here, on the basis of the first-class partition, clustering partition may be further performed based on the various types of cells of the first-class partition. Illustratively, for the foregoing scenario a, based on the user distribution of different set durations of history, for example, based on the MR data of the small-scale of the cell history, cluster division is performed to obtain categories 1, 2, 3, and 4 … … of the user distribution, and groups optimized in the same kind are obtained comprehensively, for example, a1, a2, A3, a4 … …, B1, B2, B3, B4 … …, and the like.
Illustratively, the scene state determined based on the first and second results is shown in table 2:
TABLE 2
| Cell identity | User distribution | Scene state label |
| Cell-1 | MR1 | Cluster # scene A1 |
| Cell-1 | MR2 | Cluster # scene A2 |
| Cell-1 | MR3 | Cluster # scene A3 |
| Cell-1 | MR4 | Cluster # scene A4 |
As can be seen from table 2, the scene status can indicate a first result of the spatial dimension classification corresponding to the cell and a second result of the continuous classification based on the time dimension on the basis of the spatial dimension classification.
Therefore, the cell state library can be established by performing the joint clustering analysis on the scene of the cell in the space dimension and the user distribution in the time dimension.
Further, in order to implement experience sharing, the energy-saving threshold of the same optimization group (i.e., the same scene state) may be optimized according to the classification result in the cell state library, so as to reduce the amount of computation, and facilitate the online determination of the energy-saving threshold corresponding to each scene state. Based on this, the energy-saving threshold determination method further includes:
determining a corresponding energy-saving threshold based on historical Performance Management (PM) data and Key Performance Indicator (KPI) data of each scene state in the cell state library, and generating the energy-saving threshold library.
Illustratively, the PM data may include: RRC (Radio Resource Control) maximum connection number, uplink traffic, downlink traffic, uplink PRB (Physical Resource Block) utilization, and downlink PRB utilization. Network KPI data may include: RRC establishment success rate, ERAB (Evolved Radio Access Bearer) establishment success rate, Radio Access rate, Radio drop rate, and handover success rate.
Exemplarily, the determining the corresponding energy saving threshold based on the historical PM data and the network KPI data in each scene state in the cell state library includes:
and determining a limit value of the PM data corresponding to the network KPI data and meeting the setting requirement based on the historical PM data and the network KPI data in each scene state in the cell state library, and taking the limit value of the PM data as a corresponding energy-saving threshold.
Here, the limit value of the PM data at which the network KPI data satisfies the setting requirement means: and solving the parameter value of the cell with the network KPI maintained stable and the highest turn-off load threshold (energy saving is most).
Here, for each scene state in the cell state library generated by the clustering, the PM data that keeps the network KPI stable and saves energy most is solved respectively, and is used as the energy-saving threshold corresponding to the corresponding scene state, so as to construct the energy-saving threshold library. Therefore, the cells can be clustered and divided through the scene of the cells and the clustering analysis of the information related to the user distribution, the division of space and time granularity can be obtained, namely the division of a certain cell under certain user distribution, and the corresponding energy-saving threshold is determined aiming at the similar optimized grouping through analyzing the historical PM data and the network KPI data, so that an energy-saving threshold library is formed, and conditions and basis are provided for the later time-space granularity parameter adaptive optimization.
In some embodiments, the highest turn-off load threshold at which the network KPI starts to deteriorate can be solved by numerical analysis for the historical PM data and network KPI data in each scene state in the cell state library, so as to obtain the corresponding energy-saving threshold. Here, the network KPI may start to deteriorate, that is, a variation difference of the network KPI data reaches a set requirement or the network KPI data is lower than a set threshold, which is not limited in the present application.
In some embodiments, the determining, based on the historical PM data and network KPI data in each scene state in the cell state base, a limit value of the PM data corresponding to the network KPI data meeting a set requirement includes:
training an initial model by taking historical PM data and network KPI data in each scene state in the cell state library as training data to obtain a trained model;
and solving the trained model by adopting an optimization search algorithm to obtain the limit value of the PM data corresponding to the network KPI data meeting the set requirement.
Illustratively, a machine learning algorithm is adopted to mine the relationship between the PM data and the network KPI data so as to solve the highest turn-off load threshold of the network KPI starting to deteriorate. The initial model was established as follows:
Y＝f(X)
the input data X is PM data (RRC maximum connection number, uplink traffic, downlink traffic, uplink PRB utilization, downlink PRB utilization), and the output data Y is network KPI data (RRC establishment success rate, ERAB establishment success rate, wireless connection rate, wireless disconnection rate, handover success rate) predicted by a function f. Illustratively, the data in the input data X is read in row by row and converted into a column matrix as the input of the neural network, and similarly, the output data Y matrix is converted into a column matrix as the output of the neural network.
Here, the RELU function is used as an activation function in the neural network, the output Y of the regression problem contains a plurality of variables, and the design cost function is:
wherein, y i Is sample data, y i ' is the predicted value calculated by the neural network, and l is the number of samples. Namely, the average value of Mean Square Error (MSE) of each variable is used as a loss function, and the cost function of the network is minimized by training the model and adjusting the weight and the bias.
And aiming at the trained model, selecting the highest turn-off load threshold of the network KPI starting to deteriorate by adopting an optimization search algorithm. For example, a genetic algorithm may be employed to search the resulting training model for the highest shutdown load threshold that causes the network KPI to begin deteriorating. Thus, the energy saving threshold corresponding to each scene state can be obtained, which is schematically shown in table 3 below.
TABLE 3
| Cell identity | User distribution | Scene state label | Optimal threshold parameter |
| Cell-1 | MR1 | Cluster # scene A1 | Parameter combination 1: numerical value |
| Cell-1 | MR2 | Cluster # scene A2 | Parameter combination 2: numerical value |
| Cell-1 | MR3 | Cluster # scene A3 | Parameter combination 3: numerical value |
| Cell-1 | MR4 | Cluster # scene A4 | Parameter combination 4: numerical value |
| Cell-2 | MR1 | Cluster # scene B1 | Parameter combination 5: numerical value |
| Cell-2 | MR2 | Cluster # scene B2 | Parameter combination 6: numerical value |
| Cell-2 | MR3 | Cluster # scene B3 | Parameter combination 7: numerical value |
| Cell-2 | MR4 | Cluster # scene B4 | Parameter combination 8: numerical value |
| … | … | … | … |
It can be understood that table 3 illustrates an optimal threshold parameter (i.e., an energy saving threshold) corresponding to each optimized group, so as to determine the optimized group to which the cell belongs based on the predicted user distribution of the cell, and further determine the energy saving threshold corresponding to the cell. The optimal threshold parameters corresponding to each optimization group can be constructed off-line, experience sharing is realized, and on-line computation amount can be reduced, so that an on-line part can directly utilize an experience library obtained by off-line computation, the on-line computation amount is greatly reduced, and the deployment of implementation and application is facilitated.
In some embodiments, predicting the feature data of the next set duration based on the feature data of the current set duration comprises: and predicting the feature data of the current set duration based on a Long Short-Term Memory network (LSTM) to obtain the feature data of the next set duration.
Here, the LSTM has a unique memory mode and a forgetting mode, so that the network can sufficiently mine the time-series characteristics of the time-series data, learn the time dependency relationship between the input data, and predict MR data of the next set duration based on the historical MR data of the set duration.
It is understood that, in practical applications, the classification result of the cell in the spatial dimension, that is, the aforementioned first result (scene A, B, C, etc.), may be determined based on the cell identifier, and the scene state corresponding to the next set duration of the cell may be determined based on the predicted MR data. With reference to table 3, assuming that the Cell identifier is determined to be Cell-1, the predicted user distribution falls into MR2, and the scene state label is determined to be a2, then the energy saving threshold corresponding to parameter combination 2 is selected as the target energy saving threshold.
The present application is described in further detail below with reference to application examples.
As shown in fig. 2, in the present application example, the overall flow includes two parts, that is, offline and online: the offline data mining refers to performing joint clustering analysis on the cell scene and the user distribution by using historical data to form a scene and user distribution state base (which is equivalent to the cell state base). According to each division, the relation between the historical energy-saving threshold parameter and the network KPI and the energy consumption is mined, the parameter which keeps the network KPI stable and has the highest turn-off load threshold (the most energy-saving) is solved, and a parameter library (equivalent to the energy-saving threshold library) corresponding to the scene and the user distribution is formed. The online part is to predict the user distribution of the next time granularity of the cell by using real-time cell data, classify the predicted user distribution according to the prediction result, and find out the scene and the user distribution corresponding to the next time of the cell and the energy-saving threshold corresponding to the scene and the user distribution by combining the cell scene information.
Illustratively, the offline part acquires historical network state data (data such as working parameters, PM, network KPI, MR, overlapping coverage scenes and the like), performs preprocessing such as normalization on the acquired data, constructs a scene and user distribution state library based on joint clustering in space dimension and time dimension, establishes a similar optimization group from space granularity and time granularity, can realize experience sharing, and can reduce the computation amount. Based on a scene and user distribution state library, parameter optimization can be performed through modeling and training, for example, a genetic algorithm is adopted to search a highest turn-off load threshold which enables a network KPI to begin to deteriorate in an obtained training model, and an energy-saving threshold corresponding to each scene state is obtained, so that an individualized energy-saving parameter configuration suggestion is obtained.
It can be understood that, along with the existing network test and the accumulation of experience, the scene and user distribution state library and the scene and user distribution parameter library may be continuously updated, for example, updated based on a set period or updated based on an instruction, so as to meet the control requirement of the energy saving strategy corresponding to the new user distribution, thereby implementing the refined energy saving control.
Illustratively, the online portion may predict the user distribution at the next time granularity (such as the hour level) based on the LSTM. For example, the raw data set may be preprocessed to obtain p training set samples, each sampleThere are T MR data of
The corresponding label is { y (1) ,y (2) ,...,y (i) ,...,y (p) In which x (i) Which represents the number of the i-th sample,
t-th data, y, representing the i-th sample (i) And a label corresponding to the ith sample, namely the predicted MR value of the ith sample at the next moment.
The ith sample x (i) Predicted wireless network MR value y at next time as input to long-short term memory network (i) . Performing a network training on one sample requires T time steps, each time step corresponding to one LSTM cell (LSTM cell). The long-term and short-term memory network adopts a self-circulation method, and only one data can be input into the network at each moment. Illustratively, the historical data and the LSTM model can be used for training the model for predicting 1h in the future every 1h, and the data of the previous 3 hours can be used for predicting the value of 1h in the future. And then, based on the predicted MR data of the next hour, determining the scene type of the target cell in the scene and user distribution state library by combining with the cell identifier, and determining the corresponding target energy-saving threshold based on the energy-saving threshold corresponding to the scene type, so that the base station adjusts the energy-saving strategy based on the newly determined target energy-saving threshold.
In order to implement the method for determining an energy saving threshold in the embodiment of the present application, an apparatus for determining an energy saving threshold is further provided in the embodiment of the present application, where the apparatus for determining an energy saving threshold corresponds to the method for determining an energy saving threshold, and each step in the embodiment of the method for determining an energy saving threshold is also completely applicable to the embodiment of the apparatus for determining an energy saving threshold.
As shown in fig. 3, the apparatus for determining an energy saving threshold includes: an acquisition module 301, a prediction module 302, a first determination module 303, and a second determination module 304. The obtaining module 301 is configured to obtain feature data of a current set duration of a target cell, where the feature data is used to indicate user distribution of the target cell; the prediction module 302 is configured to predict feature data of a next set duration based on the feature data of the current set duration; the first determining module 303 is configured to determine a scene state of the target cell corresponding to the next set duration based on the predicted feature data and a pre-constructed cell state library; the second determining module 304 is configured to determine a target energy saving threshold corresponding to the next set time duration of the target cell based on the scene state and a pre-constructed energy saving threshold library. The cell state library comprises corresponding scene states determined based on engineering parameters of cells and historical user distribution, the scene states can indicate first results of space dimension classification corresponding to the cells and second results of continuous classification based on time dimension on the basis of the space dimension classification, and the energy-saving threshold library comprises corresponding energy-saving thresholds determined based on historical working parameters of various scene states in the cell state library.
In some embodiments, the energy saving threshold determining apparatus further includes: a clustering module 305, configured to cluster and generate the cell state library based on the engineering parameters of the cell and the historical user distribution.
In some embodiments, the clustering module 305 is specifically configured to: clustering and dividing each cell based on the engineering parameters of the cell to obtain the first result; clustering and dividing the user distribution of different set durations of the various types of cell histories obtained by dividing the first result to obtain a second result; determining the scene state based on the first result and the second result.
In some embodiments, the energy saving threshold determining apparatus further includes: the energy saving parameter optimization module 306 determines a corresponding energy saving threshold based on the historical performance management PM data and the network key performance indicator KPI data in each scene state in the cell state base, and generates the energy saving threshold base.
In some embodiments, the energy saving parameter optimization module 306 is specifically configured to: and determining a limit value of the PM data corresponding to the network KPI data and meeting the setting requirement based on the historical PM data and the network KPI data in each scene state in the cell state library, and taking the limit value of the PM data as a corresponding energy-saving threshold.
In some embodiments, the energy saving parameter optimization module 306 is specifically configured to: training an initial model by taking historical PM data and network KPI data in each scene state in the cell state library as training data to obtain a trained model; and solving the trained model by adopting an optimization search algorithm to obtain the limit value of the PM data corresponding to the network KPI data meeting the set requirement.
In some embodiments, the prediction module 302 is specifically configured to: and predicting the characteristic data of the current set duration based on a long-term and short-term memory network to obtain the characteristic data of the next set duration.
In practical applications, the obtaining module 301, the predicting module 302, the first determining module 303, the second determining module 304, the clustering module 305, and the energy saving parameter optimizing module 306 may be implemented by a processor in the energy saving threshold determining device. Of course, the processor needs to run a computer program in memory to implement its functions.
It should be noted that: the energy saving threshold determining apparatus provided in the foregoing embodiment is only illustrated by dividing the program modules when determining the energy saving threshold, and in practical applications, the processing allocation may be completed by different program modules according to needs, that is, the internal structure of the apparatus is divided into different program modules to complete all or part of the processing described above. In addition, the energy saving threshold determining apparatus and the energy saving threshold determining method provided in the above embodiments belong to the same concept, and specific implementation processes thereof are detailed in the method embodiments and are not described herein again.
Based on the hardware implementation of the program module, and in order to implement the method according to the embodiment of the present application, an apparatus for determining an energy saving threshold is further provided in the embodiment of the present application. The energy saving threshold determining device may be a base station, or may be a network device communicatively connected to the base station. Fig. 4 shows only an exemplary structure of the power saving threshold determining apparatus, not a whole structure, and a part of or the whole structure shown in fig. 4 may be implemented as necessary.
As shown in fig. 4, the energy saving threshold determining device 400 provided in the embodiment of the present application includes: at least one processor 401, memory 402, a user interface 403, and at least one network interface 404. The various components in power saving threshold determination device 400 are coupled together by a bus system 405. It will be appreciated that the bus system 405 is used to enable communications among the components of the connection. The bus system 405 includes a power bus, a control bus, and a status signal bus in addition to a data bus. For clarity of illustration, however, the various buses are labeled as bus system 405 in fig. 4.
The user interface 403 may include, among other things, a display, a keyboard, a mouse, a trackball, a click wheel, a key, a button, a touch pad, a touch screen, or the like.
The memory 402 in the embodiments of the present application is used to store various types of data to support the operation of the energy saving threshold determining device. Examples of such data include: any computer program for operating on a power saving threshold determination device.
The method for determining the energy-saving threshold disclosed in the embodiment of the present application may be applied to the processor 401, or implemented by the processor 401. The processor 401 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the method for determining the energy saving threshold may be implemented by an integrated logic circuit of hardware in the processor 401 or by instructions in the form of software. The Processor 401 described above may be a general purpose Processor, a Digital Signal Processor (DSP), or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, or the like. Processor 401 may implement or perform the methods, steps, and logic blocks disclosed in the embodiments of the present application. A general purpose processor may be a microprocessor or any conventional processor or the like. The steps of the method disclosed in the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium located in the memory 402, and the processor 401 reads information in the memory 402, and completes the steps of the power saving threshold determining method provided in the embodiment of the present application in combination with hardware thereof.
In an exemplary embodiment, the energy saving threshold determining Device may be implemented by one or more Application Specific Integrated Circuits (ASICs), DSPs, Programmable Logic Devices (PLDs), Complex Programmable Logic Devices (CPLDs), Field Programmable Gate Arrays (FPGAs), general purpose processors, controllers, Micro Controllers (MCUs), microprocessors (microprocessors), or other electronic components for performing the foregoing methods.
It will be appreciated that the memory 402 can be either volatile memory or nonvolatile memory, and can include both volatile and nonvolatile memory. Among them, the nonvolatile Memory may be a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a magnetic random access Memory (FRAM), a magnetic random access Memory (Flash Memory), a magnetic surface Memory, an optical Disc, or a Compact Disc Read-Only Memory (CD-ROM); the magnetic surface storage may be disk storage or tape storage. Volatile Memory can be Random Access Memory (RAM), which acts as external cache Memory. By way of illustration and not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Enhanced Synchronous Dynamic Random Access Memory (Enhanced DRAM), Synchronous Dynamic Random Access Memory (SLDRAM), Direct Memory (DRmb Access), and Random Access Memory (DRAM). The memories described in the embodiments of the present application are intended to comprise, without being limited to, these and any other suitable types of memory.
In an exemplary embodiment, the present application further provides a storage medium, that is, a computer storage medium, which may specifically be a computer readable storage medium, for example, a memory 402 storing a computer program, where the computer program is executable by a processor 401 of an energy saving threshold determining apparatus to complete the steps described in the method of the present application. The computer readable storage medium may be a ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface Memory, optical disk, or CD-ROM, among others.
It should be noted that: "first," "second," and the like are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.
The technical means described in the embodiments of the present application may be arbitrarily combined without conflict.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. An energy saving threshold determining method is characterized by comprising the following steps:
acquiring feature data of a current set time length of a target cell, wherein the feature data is used for indicating user distribution of the target cell;
predicting the feature data of the next set duration based on the feature data of the current set duration;
determining a scene state corresponding to the next set time length of the target cell based on the predicted characteristic data and a pre-constructed cell state library;
determining a target energy-saving threshold corresponding to the next set time length of the target cell based on the scene state and a pre-constructed energy-saving threshold library;
the cell state library comprises corresponding scene states determined based on engineering parameters of cells and historical user distribution, the scene states can indicate first results of space dimension classification corresponding to the cells and second results of continuous classification based on time dimension on the basis of the space dimension classification, and the energy-saving threshold library comprises corresponding energy-saving thresholds determined based on historical working parameters of the scene states in the cell state library.
2. The method of claim 1, further comprising:
and clustering to generate the cell state library based on the engineering parameters of the cell and historical user distribution.
3. The method of claim 2, wherein clustering generates the cell state base based on engineering parameters and historical user distribution of the cell, and comprises:
clustering and dividing each cell based on the engineering parameters of the cell to obtain the first result;
clustering and dividing the user distribution of different set durations of the various types of cell histories obtained by dividing the first result to obtain a second result;
determining the scene state based on the first result and the second result.
4. The method of claim 1, further comprising:
and determining corresponding energy-saving thresholds based on historical performance management PM data and network key performance indicator KPI data in each scene state in the cell state library, and generating the energy-saving threshold library.
5. The method according to claim 4, wherein the determining the corresponding energy saving threshold based on the historical PM data and network KPI data in each scenario state in the cell state base comprises:
and determining a limit value of the PM data corresponding to the network KPI data meeting the setting requirement based on the historical PM data and the network KPI data in each scene state in the cell state library, and taking the limit value of the PM data as a corresponding energy-saving threshold.
6. The method according to claim 5, wherein the determining, based on the historical PM data and network KPI data in each scene state in the cell state library, the limit value of the PM data corresponding to the network KPI data meeting the setting requirement comprises:
training an initial model by taking historical PM data and network KPI data in each scene state in the cell state library as training data to obtain a trained model;
and solving the trained model by adopting an optimization search algorithm to obtain the limit value of the PM data corresponding to the network KPI data meeting the set requirement.
7. The method of claim 1, wherein predicting the profile of the next set duration based on the profile of the current set duration comprises:
and predicting the characteristic data of the current set duration based on a long-term and short-term memory network to obtain the characteristic data of the next set duration.
8. An energy saving threshold determination apparatus, comprising:
the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring characteristic data of a current set time length of a target cell, and the characteristic data is used for indicating user distribution of the target cell;
the prediction module is used for predicting the characteristic data of the next set time length based on the characteristic data of the current set time length;
a first determining module, configured to determine, based on the predicted feature data and a pre-constructed cell state library, a scene state of the target cell corresponding to the next set duration;
a second determining module, configured to determine, based on the scene state and a pre-constructed energy saving threshold library, a target energy saving threshold corresponding to the next set duration for the target cell;
the cell state library comprises corresponding scene states determined based on engineering parameters of cells and historical user distribution, the scene states can indicate first results of space dimension classification corresponding to the cells and second results of continuous classification based on time dimension on the basis of the space dimension classification, and the energy-saving threshold library comprises corresponding energy-saving thresholds determined based on historical working parameters of the scene states in the cell state library.
9. An energy saving threshold determination device, comprising: a processor and a memory for storing a computer program capable of running on the processor, wherein,
the processor, when executing the computer program, is adapted to perform the steps of the method of any of claims 1 to 7.
10. A storage medium having a computer program stored thereon, the computer program, when executed by a processor, implementing the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110209109.6A CN115038148B (en) | 2021-02-24 | 2021-02-24 | Energy-saving threshold determining method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110209109.6A CN115038148B (en) | 2021-02-24 | 2021-02-24 | Energy-saving threshold determining method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115038148A true CN115038148A (en) | 2022-09-09 |
| CN115038148B CN115038148B (en) | 2024-07-16 |
Family
ID=83117719
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110209109.6A Active CN115038148B (en) | 2021-02-24 | 2021-02-24 | Energy-saving threshold determining method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115038148B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109195170A (en) * | 2018-08-31 | 2019-01-11 | 中国联合网络通信集团有限公司 | Cell capacity-enlarging method, apparatus and storage medium |
| CN109768869A (en) * | 2017-11-06 | 2019-05-17 | 中国移动通信有限公司研究院 | A kind of traffic forecast method, system and computer storage medium |
| CN111010700A (en) * | 2018-10-08 | 2020-04-14 | 上海华为技术有限公司 | Method and device for determining load threshold |
| CN111148118A (en) * | 2019-12-18 | 2020-05-12 | 福建省南鸿通讯科技有限公司 | Flow prediction and carrier turn-off method and system based on time sequence |
| WO2020233198A1 (en) * | 2019-05-22 | 2020-11-26 | 中兴通讯股份有限公司 | Intelligent energy-saving method, base station and computer-readable storage medium |
| CN112243249A (en) * | 2019-07-19 | 2021-01-19 | 大唐移动通信设备有限公司 | LTE new access anchor point cell parameter configuration method and device under 5G NSA networking |
-
2021
- 2021-02-24 CN CN202110209109.6A patent/CN115038148B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109768869A (en) * | 2017-11-06 | 2019-05-17 | 中国移动通信有限公司研究院 | A kind of traffic forecast method, system and computer storage medium |
| CN109195170A (en) * | 2018-08-31 | 2019-01-11 | 中国联合网络通信集团有限公司 | Cell capacity-enlarging method, apparatus and storage medium |
| CN111010700A (en) * | 2018-10-08 | 2020-04-14 | 上海华为技术有限公司 | Method and device for determining load threshold |
| WO2020233198A1 (en) * | 2019-05-22 | 2020-11-26 | 中兴通讯股份有限公司 | Intelligent energy-saving method, base station and computer-readable storage medium |
| CN112243249A (en) * | 2019-07-19 | 2021-01-19 | 大唐移动通信设备有限公司 | LTE new access anchor point cell parameter configuration method and device under 5G NSA networking |
| CN111148118A (en) * | 2019-12-18 | 2020-05-12 | 福建省南鸿通讯科技有限公司 | Flow prediction and carrier turn-off method and system based on time sequence |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115038148B (en) | 2024-07-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113543284A (en) | Energy saving method, base station, control unit and storage medium | |
| Demirkol et al. | Location area planning in cellular networks using simulated annealing | |
| CN117036104B (en) | Intelligent electricity utilization method and system based on electric power Internet of things | |
| CN110662245B (en) | Base station load early warning method and device based on deep learning | |
| CN103218675A (en) | Short-term load prediction method based on clustering and sliding window | |
| CN113039506B (en) | Causal learning-based data center foundation structure optimization method | |
| Jiang et al. | Data-driven cell zooming for large-scale mobile networks | |
| CN110380881B (en) | Network capacity expansion prediction method and device | |
| Sesto-Castilla et al. | Use of Machine Learning for energy efficiency in present and future mobile networks | |
| Miozzo et al. | Coordinated load control of renewable powered small base stations through layered learning | |
| CN112954707A (en) | Energy saving method and device for base station, base station and computer readable storage medium | |
| CN112836911A (en) | Method and device for determining cell energy-saving parameter, electronic equipment and storage medium | |
| Temesgene et al. | Dynamic control of functional splits for energy harvesting virtual small cells: A distributed reinforcement learning approach | |
| Dahmani et al. | Optimization of green RNP problem for LTE networks using possibility theory | |
| US10931107B2 (en) | System and method for management of an electricity distribution grid | |
| CN115038148A (en) | Energy-saving threshold determination method, device, equipment and storage medium | |
| CN112308337A (en) | Prediction method, prediction device and processor for probabilistic short-term load of power system | |
| CN114139871B (en) | Method and device for configuring spare capacity of future power generation, storage medium and terminal equipment | |
| Liu et al. | Learning a multi-agent controller for shared energy storage system | |
| CN112954732B (en) | Network load balancing method, device, equipment and storage medium | |
| CN115238775A (en) | Model construction method | |
| CN104679992B (en) | The design method of Markov model based on customer service usage time | |
| CN117896671B (en) | Intelligent management method and system for Bluetooth AOA base station | |
| Yang et al. | RETRACTED ARTICLE: Scenario Classification of Wireless Network Optimization Based on Big Data Technology | |
| US7477895B2 (en) | Method for determining registration areas in a wireless communication system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |