Energy-saving control methods and devices, electronic equipment and storage media for residential communities

By using deep learning models to predict load information and service migration, low-load cells are identified and energy-saving strategies are formulated, solving the problem of base station energy waste and achieving a balance between base station energy consumption and communication quality.

CN116341801BActive Publication Date: 2026-11-17CHINA MOBILE CHENGDU INFORMATION & TELECOMM TECH CO LTD +1
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202111560068.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-20
Publication Date
2026-11-17
Estimated Expiration
2041-12-20

Smart Images

  • Figure CN116341801B_ABST
    Figure CN116341801B_ABST
Patent Text Reader

Abstract

This disclosure provides a cell energy-saving control method, apparatus, electronic device, and storage medium. The cell energy-saving control method includes: determining, at least based on the cell's current service information, whether a first cell exists within a shared coverage area of ​​multiple cells that satisfies a low-load state; when the first cell is included within the shared coverage area, obtaining an energy-saving strategy based on predicted load information within the shared coverage area and a first model; wherein the first model is a deep learning model trained based on historical load information and wireless communication performance indicators within the shared coverage area; when the energy-saving strategy instructs the first cell to enter an energy-saving state, migrating at least a portion of the first cell's services to at least one second cell within the shared coverage area; after migrating at least a portion of the first cell's services to the second cell, controlling the first cell to enter an energy-saving state by suspending at least a portion of its wireless functions according to the energy-saving strategy.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This disclosure relates to the field of network technology, and in particular to an energy-saving control method and device, electronic equipment and storage medium for a residential community. Background Technology

[0002] Base stations are a crucial component of mobile communications. To ensure communication quality, many areas utilize multi-cell coverage. However, base stations consume energy while operating. Considering environmental protection and reducing unnecessary energy consumption, it is permissible to shut down some unnecessary base stations in certain situations.

[0003] However, shutting down some base stations is contingent on ensuring normal communication for terminals or minimal negative impact. Furthermore, the large differences in network traffic volume between peak and off-peak hours, coupled with continuous equipment operation and energy consumption not dynamically adjusted to varying traffic levels, leads to waste. Summary of the Invention

[0004] This disclosure provides an energy-saving control method and device, electronic equipment and medium for a residential community.

[0005] The first aspect of this disclosure provides an energy-saving control method for a residential community. At ,include:

[0006] Based at least on the current service information of the cell, determine whether there is a first cell that meets the criteria of being in a low-load state within the common coverage area of ​​multiple cells;

[0007] When the first cell is included in the common coverage area, an energy-saving strategy is obtained based on the predicted load information in the common coverage area and a first model; wherein, the first model is a deep learning model trained based on historical load information and wireless communication performance indicators in the common coverage area.

[0008] When the energy-saving strategy instructs the first cell to enter an energy-saving state, at least a portion of the services of the first cell are migrated to at least one second cell in the common coverage area;

[0009] After migrating at least some services from the first cell to the second cell, the first cell is controlled to enter the energy-saving state by suspending at least some of the wireless functions according to the energy-saving strategy.

[0010] Based on the above scheme, determining whether there exists a first cell in a low-load state within the shared coverage area of ​​multiple cells, at least based on the current service information of the cell, includes:

[0011] The current load information of the cell is input into the second model to obtain the predicted load information of the shared coverage cell within a preset time period output by the second model;

[0012] When it is determined from the predicted load information that at least one cell in the common coverage area has a load rate less than or equal to the load rate threshold, it is determined that there is a first cell in a low-load state.

[0013] Based on the above scheme, the method further includes:

[0014] Based on the predicted load information, determine the duration for which the first cell will maintain the low load state;

[0015] The energy-saving strategy obtained based on the predicted load information and the first model within the shared coverage area includes:

[0016] The predicted load information within the common coverage area and the duration of the period are input into the first model to obtain the energy-saving strategy, wherein the energy-saving period included in the energy-saving strategy is located within the duration of the first cell maintaining the low load state.

[0017] Based on the above scheme, the energy-saving strategy instructs the first cell to enter the energy-saving state when the first model determines, based on the predicted load information, that at least one second cell is capable of accepting at least a portion of the service transfer from the first cell.

[0018] Based on the above scheme, the information input to the first model also includes at least one of the following:

[0019] Weather information during the specified duration;

[0020] Regional feature information of the common coverage area;

[0021] Event information of predefined events affecting wireless communication during the duration;

[0022] Information about the cells within the shared coverage area;

[0023] Frequency points or carrier information supported by cells within the public coverage area;

[0024] The utilization rate of Physical Resource Blocks (PRBs) within the shared coverage area.

[0025] Based on the above scheme, the energy-saving strategy includes energy-saving mode parameters that indicate the energy-saving mode;

[0026] The energy-saving method includes at least one of the following:

[0027] The first cell is temporarily shut down;

[0028] Suspend some carriers in the first cell;

[0029] Suspend the establishment of radio frequency channels in part of the first cell;

[0030] Suspend wireless transmission within certain symbols in the first cell.

[0031] Based on the above scheme, the method further includes:

[0032] Determine the actual load information within the common coverage area before the first cell enters the energy-saving state according to the energy-saving strategy, and the wireless communication performance indicators within the common coverage area after entering the energy-saving state;

[0033] The first model is trained online based on the determined actual load information and the communication performance indicators.

[0034] Based on the above scheme, the first model includes: Long Short-Term Memory (LSTM) network.

[0035] A second aspect of this disclosure provides an energy-saving control device for a residential community, comprising:

[0036] The first determining module is used to determine, at least based on the current service information of the cell, whether there is a first cell in the common coverage area of ​​multiple cells that meets the condition of being in a low-load state.

[0037] The module is configured to obtain an energy-saving strategy based on the predicted load information in the common coverage area and a first model when the first cell is included in the common coverage area; wherein, the first model is a deep learning model trained based on historical load information and wireless communication performance indicators in the common coverage area.

[0038] A migration module is used to migrate at least a portion of the services of the first cell to at least one second cell in the common coverage area when the energy-saving strategy indicates that the first cell has entered an energy-saving state.

[0039] The control module is configured to, after migrating at least some services of the first cell to the second cell, control the first cell to enter the energy-saving state by suspending the at least some wireless functions according to the energy-saving strategy.

[0040] Based on the above scheme, the first determining module is specifically used to input the current load information of the cell into the second model to obtain the predicted load information of the shared coverage cell within a preset time period output by the second model; when it is determined from the predicted load information that at least one cell in the shared coverage area has a load rate less than or equal to a load rate threshold, it is determined that there is a first cell in a low load state.

[0041] Based on the above scheme, the device further includes:

[0042] The second determining module is used to determine the duration of the first cell maintaining the low load state based on the predicted load information;

[0043] The obtaining module is specifically used to input the predicted load information and the duration of the common coverage area into the first model to obtain the energy-saving strategy, wherein the energy-saving period included in the energy-saving strategy is located within the duration of the first cell maintaining the low load state.

[0044] Based on the above scheme, the energy-saving strategy instructs the first cell to enter the energy-saving state when the first model determines, based on the predicted load information, that at least one second cell is capable of accepting at least a portion of the service transfer from the first cell.

[0045] Based on the above scheme, the information input to the first model also includes at least one of the following:

[0046] Weather information during the specified duration;

[0047] Regional feature information of the common coverage area;

[0048] Event information of predefined events affecting wireless communication during the duration;

[0049] Information about the cells within the shared coverage area;

[0050] Frequency points or carrier information supported by cells within the public coverage area;

[0051] The utilization rate of Physical Resource Blocks (PRBs) within the shared coverage area.

[0052] Based on the above scheme, the energy-saving strategy includes energy-saving mode parameters that indicate the energy-saving mode;

[0053] The energy-saving method includes at least one of the following:

[0054] The first cell is temporarily shut down;

[0055] Suspend some carriers in the first cell;

[0056] Suspend the establishment of radio frequency channels in part of the first cell;

[0057] Suspend wireless transmission within certain symbols in the first cell.

[0058] Based on the above scheme, the device further includes:

[0059] The third determining module is used to determine the actual load information in the common coverage area before the first cell enters the energy-saving state according to the energy-saving strategy, and the wireless communication performance indicators in the common coverage area after entering the energy-saving state.

[0060] The training module is used to train the first model online based on the determined actual load information and the communication performance indicators.

[0061] Based on the above scheme, the first model includes: Long Short-Term Memory (LSTM) network.

[0062] A third aspect of this disclosure provides an electronic device, the electronic device comprising:

[0063] Memory;

[0064] The processor, connected to the memory, is configured to implement the energy-saving control method for the community provided by any of the aforementioned technical solutions by executing computer-executable instructions stored in the memory.

[0065] A fourth aspect of this disclosure provides a computer storage medium storing computer-executable instructions; when executed by a processor, the computer-executable instructions can implement the energy-saving control method for a residential community provided by any of the aforementioned technical solutions.

[0066] The technical solution provided in this disclosure, after determining that a first cell is in a low-load state, will use a first model to obtain an energy-saving strategy. Only if the energy-saving strategy actually allows the first cell to enter the energy-saving state will the first cell be controlled to enter the energy-saving state according to the energy-saving strategy. This reduces the negative impact on wireless communication caused by directly controlling the first cell to enter the energy-saving state after detecting that the first cell is in a low-load state. Therefore, it effectively balances the energy saving of the base station and the communication quality of wireless communication. Attached Figure Description

[0067] Figure 1 A schematic flowchart illustrating an energy-saving control method for a residential community provided in this embodiment of the present disclosure;

[0068] Figure 2 This is a schematic diagram illustrating the effect of a common coverage area provided in an embodiment of the present disclosure;

[0069] Figure 3 A schematic flowchart illustrating an energy-saving control method for a residential community provided in this embodiment of the present disclosure;

[0070] Figure 4 A schematic flowchart illustrating an energy-saving control method for a residential community provided in this embodiment of the present disclosure;

[0071] Figure 5 A common coverage area division diagram provided in this embodiment of the disclosure;

[0072] Figure 6This is a schematic diagram of the structure of an energy-saving control device for a residential community provided in an embodiment of this disclosure;

[0073] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0074] To gain a more detailed understanding of the features and technical content of this disclosure, the implementation of this disclosure will be described in detail below with reference to the accompanying drawings. The accompanying drawings are for reference and illustration only and are not intended to limit this disclosure.

[0075] like Figure 1 As shown in the embodiments of this disclosure, a method for training a network attack detection model is provided, which may include:

[0076] S110: Based at least on the current service information of the cell, determine whether there is a first cell in the common coverage area of ​​multiple cells that meets the condition of being in a low-load state;

[0077] S120: When the first cell is included in the common coverage area, an energy-saving strategy is obtained based on the predicted load information in the common coverage area and the first model; wherein, the first model is a deep learning model trained based on historical load information and wireless communication performance indicators in the common coverage area.

[0078] S130: When the energy-saving strategy instructs the first cell to enter an energy-saving state, at least a portion of the services of the first cell are migrated to at least one second cell in the common coverage area;

[0079] S140: After migrating at least some services of the first cell to the second cell, the first cell is controlled to enter the energy-saving state by suspending the at least some wireless functions according to the energy-saving strategy.

[0080] The energy-saving control method for cells provided in this disclosure can be applied to the network management equipment of a base station or a macro base station, wherein the current service information indicates at least one of the following:

[0081] The types of services being transmitted in each cell within the shared coverage area;

[0082] The volume of services being transmitted in each cell within the shared coverage area.

[0083] The common coverage area refers to an area where multiple cells are simultaneously covered.

[0084] refer to Figure 2 As shown, cell A and cell M overlap in covering the same area, and cell B and cell M overlap in covering the same area. Therefore, the coverage areas of cell A and cell B are the same.

[0085] There are several ways to determine the shared coverage area; the following are two options:

[0086] Method 1: Based on the deployment information and configuration information of base stations such as base stations, small base stations, and micro base stations, determine the areas covered by these base stations, and determine the common coverage area based on whether there is overlap between the coverage areas of each base station.

[0087] Method 2: The terminal will perform neighbor cell measurements, including: same-frequency cell measurements, different-frequency cell measurements, and different-system cell measurements. Based on the cells measured by the terminal and the cell measurement results, the common coverage area can also be determined. For example, the common coverage area can be determined based on the measurement results reported by the receiving terminal.

[0088] In S110, based on the current service information of the cell, it can be determined whether there is a first cell in a low-load state at the current time and / or predict whether there is a first cell that will enter a low-load state within a preset time period.

[0089] In this embodiment of the disclosure, the first cell can be any type of cell, specifically any cell other than the macro cell formed by the macro base station, such as a small cell.

[0090] When a first cell exists, the first model will be used to determine whether the first cell can be connected to the energy-saving state, and if it is connected to the energy-saving state, what energy-saving parameters will be used.

[0091] Energy-saving status indications and energy-saving parameters can both be included in the energy-saving flow measurement.

[0092] The first model outputs energy-saving strategies based on the predicted load information of each cell within the shared coverage area.

[0093] The predicted load information includes at least one of the following:

[0094] Predicted load factor;

[0095] Forecasted business volume for different business segments, etc.;

[0096] Upstream business information;

[0097] Downstream business information.

[0098] The starting point of this future period can be the current moment or any moment after the current moment.

[0099] The first model was obtained by optimizing and training based on historical load information and wireless communication performance indicators.

[0100] The wireless communication indicators include, but are not limited to, at least one of the following:

[0101] Network performance excellence rate;

[0102] Drop rate;

[0103] Packet loss rate;

[0104] Utilization rate of PRB, etc.

[0105] The first model is trained repeatedly until the optimal wireless communication performance index is reached.

[0106] The energy-saving strategy indication allowing the first cell to enter the energy-saving state may include: when the energy-saving state indication in the energy-saving strategy corresponds to allowing the first cell to enter the energy-saving state, then controlling the first cell to enter the energy-saving state.

[0107] If the energy-saving strategy does not allow the first cell to enter the energy-saving state, the first cell will maintain its current state, that is, it will not enter the energy-saving state, so as to avoid the copper wire quality of some UEs being affected if the first cell enters the energy-saving state.

[0108] Before the first cell enters energy-saving mode, at least some of the services affected when the first cell enters energy-saving mode need to be transferred to other cells (i.e., the second cell) that can accept this part of the services.

[0109] After the first cell enters the energy-saving state, at least some of the wireless functions of the first cell are suspended. The specific energy-saving method can be indicated by the energy-saving parameters in the energy-saving strategy.

[0110] The energy-saving parameters include, but are not limited to, at least one of the following:

[0111] Energy-saving mode parameters indicate the energy-saving mode;

[0112] Energy-saving time information indicates the time when the system enters energy-saving mode.

[0113] For example, in certain situations, if the load rate of the first cell is particularly low, and the load rate of the second cell that overlaps with the first cell is also relatively low and can accommodate all users of the first cell, the energy-saving mode parameter of the energy-saving strategy can instruct the first cell to shut down. In this way, the entire first cell is completely shut down, thereby minimizing the energy consumption of the first cell and maximizing energy saving without affecting user communication.

[0114] In some embodiments, such as Figure 3 As shown, S110 may include:

[0115] S111: Input the current load information of the cell into the second model to obtain the predicted load information of the shared coverage cell within a preset time period output by the second model;

[0116] S112: When it is determined from the predicted load information that at least one cell in the common coverage area has a load rate less than or equal to the load rate threshold, it is determined that there is a first cell in a low load state.

[0117] In this embodiment of the disclosure, the second model will output the load information of each cell in the cell that will be covered in a future period of time (i.e., the predicted load information) based on the current load information of the cell.

[0118] The first model outputs predicted load information, and the network management equipment of the base station can determine whether there is at least one cell in the common coverage area whose load rate is lower than the load rate threshold based on the predicted load information. If there is, it is considered that the low-load state cell can be stored.

[0119] In some embodiments, the first cell in a low-load state includes:

[0120] The first type of low-load cell is one where the load rate of all services in the cell is lower than the first load rate threshold.

[0121] The second type of low-load cell is one where the load rate of some services in the cell is lower than the second load rate threshold.

[0122] If, for the first type of low-load cell, after processing its predicted load information, the first model finds that one or more second cells can fully handle its services, then the first type of low-load cell can be shut down via energy-saving mode parameters. For the second type of low-load cell, if one or more second cells can handle the low-load rate services of the first cell, some services of the second type of low-load cell can be shut down via energy-saving mode parameters of the energy-saving strategy.

[0123] In some embodiments, the method further includes:

[0124] Based on the predicted load information, the duration for which the first cell will maintain the low load state is determined.

[0125] The method can also determine the duration of the first cell's low-load state based on the predicted load information; this duration can also be referred to as the low-load period.

[0126] S120 may include:

[0127] The predicted load information within the common coverage area and the duration of the period are input into the first model to obtain the energy-saving strategy, wherein the energy-saving period included in the energy-saving strategy is located within the duration of the first cell maintaining the low load state.

[0128] When the first model determines the energy-saving strategy based on predicted load information, the energy-saving time indicated by the energy-saving time parameter in the energy-saving strategy can fall within the low-load period, and generally does not... out Beyond the aforementioned low-load period.

[0129] For example, the second model will periodically or based on triggers provide predicted load information for each cell in the common coverage area. When new predicted load information is available, it can be input into the first model to trigger the first model to update the energy-saving strategy, thereby enabling dynamic and timely adjustment of the first cell in the common coverage area that is in an energy-saving state.

[0130] The energy-saving strategy instructs the first cell to enter the energy-saving state when the first model determines, based on the predicted load information, that at least one second cell is capable of accepting at least a portion of the service transfer from the first cell.

[0131] In some embodiments, the second model may also output the cell identifier of the second cell and the service identifier of the accepted service.

[0132] In some embodiments, the information input to the first model further includes at least one of the following:

[0133] Weather information during the specified duration;

[0134] Regional feature information of the common coverage area;

[0135] Event information of predefined events affecting wireless communication during the duration;

[0136] Information about the cells within the shared coverage area;

[0137] The utilization rate of Physical Resource Blocks (PRBs) within the shared coverage area.

[0138] In some cases, weather information can affect communication quality, therefore weather information serves as a reference for privacy in the first model's output energy-saving strategy.

[0139] The regional feature information can reflect whether the corresponding geographical area is a communication hotspot, whether the communication capacity varies greatly, and whether there is a tidal phenomenon in the communication capacity.

[0140] For example, the size and fluctuation of communication capacity differ between densely populated urban areas and less densely populated suburban areas.

[0141] For example, if there is a shopping mall or transportation hub with a large flow of people in the shared coverage area, it is necessary to consider whether one or more communities in such areas should be allowed to enter energy-saving mode.

[0142] The regional feature information indicates that the regional features can be used by the first model to determine whether one or more first cells are allowed to enter an energy-saving state.

[0143] The regional feature information can also be used to indicate the wireless communication environment of the common coverage area, which can indicate whether there are many wireless signal transmission obstructions, etc.

[0144] For example, areas with many tall buildings or trees can affect wireless signal transmission. In such environments, even if a cell has a low load rate, it still cannot be turned off.

[0145] In summary, the regional characteristic information can indicate: regional characteristics related to user traffic in the common coverage area and / or environmental characteristics of wireless communication.

[0146] Event information of predefined events affecting wireless communication during the duration includes:

[0147] Various activities that lead to the gathering of crowds, such as concerts or shopping mall openings;

[0148] Such as global press conferences of one or more companies in the region during a certain period of time.

[0149] Of course, the above are just examples of predefined events, and the actual implementation is not limited to the examples above.

[0150] The cell information includes, but is not limited to, at least one of the following:

[0151] Community signage;

[0152] Predicted load information for the community;

[0153] Cell capability information, which at least indicates the frequency bands, carriers, and / or communication standards supported by the cell.

[0154] In some embodiments, the cell capability information may also indicate: the maximum communication capacity of each cell, or the communication capacity of a cell when no congestion occurs.

[0155] The utilization rate of physical resource blocks (PRBs) within the shared coverage area may include: the uplink utilization rate and / or downlink utilization rate of the PRBs.

[0156] The utilization rate of PRB within the shared coverage area may include: the total utilization rate of PRB in the entire coverage area, and / or the utilization rate of PRB in each cell within the shared coverage area.

[0157] By inputting the above information into the first model, the first model can provide appropriate energy-saving strategies for different communication scenarios, thereby achieving optimized energy saving of the cell under different scenarios.

[0158] In some embodiments, the energy-saving strategy includes energy-saving mode parameters that indicate the energy-saving mode;

[0159] The energy-saving method includes at least one of the following:

[0160] The first cell is temporarily shut down;

[0161] Suspend some carriers in the first cell;

[0162] Suspend the establishment of radio frequency channels in part of the first cell;

[0163] Suspend wireless transmission within certain symbols in the first cell.

[0164] Power consumption of the first cell is reduced by shutting down the first cell, disabling the carrier, shutting down the radio frequency channel, and stopping radio signal transmission within some symbols.

[0165] In some embodiments, the method further includes:

[0166] Determine the actual load information within the common coverage area before the first cell enters the energy-saving state according to the energy-saving strategy, and the wireless communication performance indicators within the common coverage area after entering the energy-saving state;

[0167] The first model is trained online based on the determined actual load information and the communication performance indicators.

[0168] In this embodiment, the first model is repeatedly iterated and trained online to optimize it, so that the energy-saving strategy output by the second model can ensure communication quality while saving energy as much as possible within the common coverage area.

[0169] In some embodiments, the first model includes a long short-term memory network.

[0170] If the first model includes a long short-term memory network, the memory and forgetting characteristics of the long short-term memory network can be used to better explore the correlation between various inputs into the first model, thereby outputting a more suitable energy-saving strategy for the current common coverage scenario.

[0171] As a supplement to the transmission capabilities of macro base stations, the construction of 5G small base stations is an essential component. Due to their short wavelengths and weak penetration, macro base station signals face greater challenges in achieving deep indoor coverage. Indoor micro base stations can effectively fill in coverage gaps and hot spots for macro base stations, making their construction a key focus of 5G network development in the coming years. Small base stations also possess intelligent adjustment capabilities, allowing for dynamic adjustments to signal coverage and network load.

[0172] As the number of small base stations deployed increases, unreasonable deployment of small base stations leads to low energy efficiency in the network system, resulting in waste of resources and energy.

[0173] In small cell power consumption, the remote radio unit (RRU) accounts for a large proportion, and within the RRU, the power amplifier accounts for a significant portion of the power consumption. In actual networks, traffic volume often exhibits a significant tidal effect; when traffic is low, the base station remains operational, resulting in substantial energy waste.

[0174] Reducing ineffective energy consumption is a major direction for energy conservation, but it faces numerous challenges. Network traffic volume varies greatly between peak and off-peak hours, and equipment operates continuously without dynamically adjusting energy consumption according to traffic levels, resulting in waste. However, in a typical network, the characteristics of different scenarios vary considerably. Therefore, automatically identifying various scenarios and formulating matching energy-saving strategies is crucial for energy conservation.

[0175] The energy-saving control method for a residential community provided in this embodiment may include the following steps:

[0176] S1. Identify the cells covered by 5G small base stations;

[0177] Specifically, step S1 includes:

[0178] S11. Algorithm for identifying co-coverage cells: The terminal actively initiates inter-frequency measurements, with a total number of measurements S, a number of successful measurements T, and a measurement success rate O = T / S. Assume a capacity cell has U users, N terminals, M terminals supporting inter-frequency measurements, and a terminal support rate T = M / N. The system sets a terminal support rate threshold X and an inter-frequency measurement success rate Y. If T > X and O > Y, then the cells are considered to have co-coverage relationships; otherwise, coverage holes are considered to exist. (Reference) Figure 4 As shown, data analysis is performed and a model is used for prediction. Low-load cells can be identified through cell common coverage learning and analysis of measurement reports and service information.

[0179] S12. Set the periodic time T, run the co-coverage cell algorithm periodically, refresh the co-coverage cell learning results, and the system automatically establishes frequency band co-coverage relationships within and between stations.

[0180] S2. Analyze massive amounts of measurement reports and business information;

[0181] Wherein, step S2 specifically includes:

[0182] S21. Analyze massive measurement report information and service information.

[0183] The objects of measurement analysis include the serving cell and neighboring cells, and the measured physical values include base station signal, base station code (LAC), cell code (CI), adjacent base station, base station location, base station address, radio access success rate, radio utilization rate, voice (for example, VoLTE) radio access success rate, MR coverage rate, coverage type and / or maximum number of E-RABs for QCI1, network uplink rate, etc.

[0184] Analyzing service information includes traffic flow, time interval, etc. The base station data traffic information includes historical traffic information of a single base station and traffic information of base stations of co-covered cells identified thereby.

[0185] S22. Constructing a traffic map may be as follows:

[0186] Constructing the traffic map intuitively displays the temporal and spatial distribution of mobile data traffic, and generates traffic load data. The load data is one type of the load information in the foregoing embodiment.

[0187] The steps for constructing the traffic map are as follows:

[0188] Perform Voronoi segmentation, regarding the small base station network deployment area as a plane R2, wherein the set of m deployed small base stations is regarded as a point set X={x1,x2,...,x m}, and the traffic load of each base station in a certain period of time is obtained from historical statistical data as L={l1,l2,l3,...,l m}. The polygon set representing the coverage of each small base station is obtained through Voronoi segmentation. Figure 5 A schematic diagram of Voronoi segmentation for a co-coverage area is shown.

[0189] That is, T={T1,T2,T3,...,T M}, the Voronoi polygon of small base station x i is:

[0190] T i ={x∈R2; d(x,xi)<d(x,xj), j∈1,...,m, j≠i}, where d(x,xi) represents the Euclidean distance from any point x on the plane R2 to point xi.

[0191] Perform gridding, on the Figure 5 basis of the shown Voronoi polygon, the traffic is determined.

[0192] Divide the plane R2 into n*n grids of the same size.

[0193] Calculating the traffic load at the grid location of the small base station can include:

[0194] The flow load at each grid point within Ti is:

[0195] Lgrid = li * Area(grid) / Area(Ti), where Area(grid) represents the area of ​​each grid and Area(Ti) represents the area of ​​the Voronoi polymorph Ti. This gives the flow load of all grids within R2.

[0196] Predicting traffic load at the grid can include:

[0197] Based on historical statistical data, the traffic load of each base station grid is calculated. Then, through machine learning algorithms, including using the LSTM model (Long Short-Term Memory Network) for time series prediction and introducing an attention mechanism (Attention Model) to consider long-term seasonal effects (such as those from a year ago and a quarter ago), feature engineering is optimized to capture holiday and weekend effects and build a scenario model.

[0198] S23. Discover energy-efficient cells in the network.

[0199] Low-load cells are identified by comparing the predicted traffic information of the small base station's location with the pre-set load rate threshold in the network management system.

[0200] S3. Predict the time interval of the cell's low service and low load state. This time interval is the aforementioned period of continuous low load state.

[0201] Based on the traffic information of the grid where the small base station is located, the time interval of low service and low load is obtained.

[0202] S4. Service migration in low-load cells. Low-load cells refer to the cells mentioned above that are in a low-load state, which may include:

[0203] S41. When the energy-saving cell is under low service load, the system will migrate the services of the energy-saving cell to its compensation cell and put the energy-saving cell into hibernation. The compensation cell here is the second cell mentioned above, that is, the cell that accepts the users transferred from the first cell.

[0204] S42. Through real-time monitoring, the system can promptly wake up dormant energy-saving cells to ensure network quality during peak service periods.

[0205] S5, base station traffic prediction and reinforcement learning, may include:

[0206] S51. By using historical data from a large number of cells in the network, such as time, load information, neighbor cell relationships, and other external factors such as weather and specific events as input, AI modeling is performed at the cell / cell cluster / region level to predict the cell / cell cluster / region load situation in the future. At the same time, by combining different energy-saving functions (carrier shutdown, channel shutdown, symbol shutdown, etc.), the optimal energy-saving time for different energy-saving functions in the cells within this range is determined.

[0207] Channel shutdown: Reduce power consumption by shutting down (or putting to sleep) some transmit radio frequency channels; the time granularity of shutdown or startup is on the order of seconds.

[0208] Symbol shutdown: When the base station detects that some downlink subframes (symbols) have no data transmission, it shuts down the power amplifier and other radio frequency hardware during this period to reduce static power consumption; the granularity of the function activation time is at the microsecond level.

[0209] The AI ​​model obtained from the aforementioned AI modeling can be a deep learning model based on Long Short Memory Network (LSTM).

[0210] S52 employs AI carrier shutdown threshold optimization technology, which can find load thresholds and performance inflection points to maximize energy savings.

[0211] S53. The system automatically optimizes the shutdown threshold based on traffic prediction and reinforcement learning, and iterates and optimizes it online. While ensuring KPIs are met, it can increase the shutdown duration by more than 10%. This shutdown threshold determines the stop time at which energy saving ends.

[0212] S6. Develop matching energy-saving strategies, which may include:

[0213] S61. Based on the collected scene characteristics, automatically identify various different scenes and formulate matching energy-saving strategies. The higher the threshold for entering shutdown, the better the energy-saving effect.

[0214] S62. It adopts an automated multi-mode energy-saving strategy to achieve multi-standard and multi-band coordination, cross-standard coordinated shutdown within a frequency band, and multi-carrier shutdown between frequency bands.

[0215] S7. Input KPI values ​​and iteratively optimize energy-saving strategies and models, which may include:

[0216] S71. In predictive modeling, it is also necessary to monitor the key performance indicators (KPIs) of the network, and provide feedback to the currently used predictive model based on changes in the KPIs, so as to further iterate the predictive model and ultimately achieve the optimal point for energy saving and system performance. The KPIs include, but are not limited to, at least one of the following:

[0217] 4G network experience excellence rate, the indicator is defined as: 4G network performance excellence rate = (1 - percentage of weak coverage cells) × (1 - percentage of high load cells) × (1 - percentage of high interference cells).

[0218] 4G network performance excellence rate, defined as: 4G network performance excellence rate = (1 - percentage of weak coverage cells) × (1 - percentage of high load cells) × (1 - percentage of high interference cells)

[0219] 5G overall coverage rate = 5G duration retention rate * 5G MR coverage rate

[0220] 5G average download speed: Test download data volume / Test duration

[0221] S72. When AI can accurately predict load changes over a future period, when the load is low, the network side can predict the energy-saving functions that can be implemented during that period and estimate the effect of starting energy saving during that period, thereby effectively improving the energy-saving efficiency during the energy-saving period.

[0222] S73. AI methods can accurately predict the effective time periods for energy-saving applications, thereby reducing the impact on performance KPIs caused by unreasonable energy-saving time period configurations in manual settings. Simultaneously, for key protection scenarios, a pre-defined whitelist can be used to prevent energy-saving functions from being implemented in those scenarios, thus avoiding the impact of energy saving on these scenarios.

[0223] S8. Generate base station instructions using the above strategy and send them to the small base station for execution.

[0224] S81. Based on the energy-saving prediction report, a soft-on / soft-off command for the cell is generated and the strategy is sent to the small base station BBU. The BBU generates adjusted and optimized small base station parameters based on the energy-saving strategy and generates energy-saving shutdown commands that the small base station can recognize, thereby improving the energy-saving effect of the small base station.

[0225] This application proposes to predict service trends by analyzing measurement reports and business information using big data. Combined with base station autonomous learning of shared coverage sites, services in predicted low-load cells are migrated to compensated cells. Simultaneously, based on traffic prediction and reinforcement learning, various scenarios are automatically identified, matching energy-saving strategies are formulated, optimal shutdown thresholds are determined, and iterative optimization is performed. Multi-standard and multi-band coordination, cross-standard coordinated shutdown within a frequency band, and multi-carrier shutdown within a frequency band are implemented, along with differentiated configuration of cell-level automated energy-saving parameters. Based on system-defined KPI values, the goal is to maximize energy savings while meeting these KPI requirements.

[0226] Hardware systems, such as massive MIMO and small base stations, will pose significant energy consumption challenges to large-scale 5G deployment. This energy-saving strategy utilizes AI technology for intelligent energy saving, enabling multi-network collaborative energy saving across different scenarios, sites, and times. Furthermore, it maximizes network energy efficiency while ensuring stable KPIs, achieving an optimal balance between energy consumption and KPIs, making it a crucial independent protection target.

[0227] like Figure 6 As shown in the figure, this disclosure provides an energy-saving control device for a residential community, comprising:

[0228] The first determining module 110 is used to determine, at least based on the current service information of the cell, whether there is a first cell in the common coverage area of ​​multiple cells that meets the condition of being in a low-load state.

[0229] The module 120 is configured to obtain an energy-saving strategy based on the predicted load information in the common coverage area and a first model when the first cell is included in the common coverage area; wherein, the first model is a deep learning model trained based on historical load information and wireless communication performance indicators in the common coverage area.

[0230] The migration module 130 is used to migrate at least a portion of the services of the first cell to at least one second cell in the common coverage area when the energy-saving strategy indicates that the first cell enters an energy-saving state.

[0231] The control module 140 is configured to, after migrating at least some services of the first cell to the second cell, control the first cell to enter the energy-saving state by suspending the at least some wireless functions according to the energy-saving strategy.

[0232] In some embodiments, the first determining module 110, the obtaining module 120, and the migration module 130 may be program modules; after the program module is executed by the processor, it can realize the functions of the above-mentioned modules.

[0233] In other embodiments, the first determining module 110, the obtaining module 120, and the migration module 130 may be hardware-software combined modules; the hardware-software combined modules include, but are not limited to, various programmable arrays; the programmable arrays include, but are not limited to, field-programmable arrays and / or complex programmable arrays.

[0234] In some embodiments, the first determining module 110 is specifically used to input the current load information of the cell into the second model to obtain the predicted load information of the shared coverage cell within a preset time period output by the second model; when it is determined from the predicted load information that at least one cell in the shared coverage area has a load rate less than or equal to a load rate threshold, it is determined that there is a first cell in a low load state.

[0235] In some embodiments, the apparatus further includes:

[0236] The second determining module is used to determine the duration of the first cell maintaining the low load state based on the predicted load information;

[0237] The obtaining module 120 is specifically used to input the predicted load information and the duration of the common coverage area into the first model to obtain the energy-saving strategy, wherein the energy-saving period included in the energy-saving strategy is located within the duration of the first cell maintaining the low load state.

[0238] In some embodiments, the energy-saving strategy instructs the first cell to enter the energy-saving state when the first model determines, based on the predicted load information, that at least one second cell is capable of accepting at least a portion of the service transfer from the first cell.

[0239] In some embodiments, the information input to the first model further includes at least one of the following:

[0240] Weather information during the specified duration;

[0241] Regional feature information of the common coverage area;

[0242] Event information of predefined events affecting wireless communication during the duration;

[0243] Information about the cells within the shared coverage area;

[0244] Frequency points or carrier information supported by cells within the public coverage area;

[0245] The utilization rate of Physical Resource Blocks (PRBs) within the shared coverage area.

[0246] In some embodiments, the energy-saving strategy includes energy-saving mode parameters that indicate the energy-saving mode;

[0247] The energy-saving method includes at least one of the following:

[0248] The first cell is temporarily shut down;

[0249] Suspend some carriers in the first cell;

[0250] Suspend the establishment of radio frequency channels in part of the first cell;

[0251] Suspend wireless transmission within certain symbols in the first cell.

[0252] In some embodiments, the apparatus further includes:

[0253] The third determining module is used to determine the actual load information in the common coverage area before the first cell enters the energy-saving state according to the energy-saving strategy, and the wireless communication performance indicators in the common coverage area after entering the energy-saving state.

[0254] The training module is used to train the first model online based on the determined actual load information and the communication performance indicators.

[0255] In some embodiments, the first model includes a long short-term memory network.

[0256] like Figure 7 As shown, this disclosure provides an electronic device, the electronic device comprising:

[0257] Memory;

[0258] The processor, connected to the memory, is configured to implement the energy-saving control method for the cell provided in any of the foregoing embodiments by executing computer-executable instructions stored in the memory, for example, executing... Figure 1 , Figures 3 to 5 Energy-saving control methods for any residential community shown.

[0259] The electronic device may be the management device of the aforementioned base station or a macro base station.

[0260] like Figure 7 As shown, the electronic device may also include a network interface, which can be used to interact with peer devices via a network.

[0261] This disclosure provides a computer storage medium storing computer-executable instructions. When executed by a processor, these instructions can implement the energy-saving control method for a residential community provided in any of the foregoing embodiments, for example, executing... Figure 1 , Figures 3 to 5 Energy-saving control methods for any residential community shown.

[0262] The technical solutions described in the embodiments of this disclosure can be combined arbitrarily without conflict.

[0263] In the several embodiments provided in this disclosure, it should be understood that the disclosed methods and smart devices can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components may be combined, or integrated into another system, or some features may be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0264] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0265] In addition, each functional unit in the various embodiments of this disclosure can be integrated into a second processing unit, or each unit can be a separate unit, or N or more units can be integrated into a unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0266] The above description is merely a specific embodiment of this disclosure, but the scope of protection of this disclosure is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method of energy saving control of a cell, characterized by, include: Based at least on the current service information of the cell, determine whether there is a first cell that meets the criteria of being in a low-load state within the common coverage area of ​​multiple cells; When the first cell is included in the common coverage area, an energy-saving strategy is obtained based on the predicted load information in the common coverage area and a first model; wherein, the first model is a deep learning model trained based on historical load information and wireless communication performance indicators in the common coverage area; the predicted load information is obtained by time series prediction through a long short-term memory network based on the current traffic load of cells in the common coverage area, and by introducing an attention mechanism to consider seasonal effects. When the energy-saving strategy instructs the first cell to enter an energy-saving state, at least a portion of the services of the first cell are migrated to at least one second cell in the common coverage area; After migrating at least some services from the first cell to the second cell, the first cell is controlled to enter the energy-saving state by suspending at least some wireless functions according to the energy-saving strategy. The information input to the first model also includes at least one of the following: Weather information during the period in which the first cell maintains the low load state; Regional feature information of the common coverage area; Event information of predefined events affecting wireless communication during the duration; Information about the cells within the shared coverage area; The utilization rate of Physical Resource Blocks (PRBs) within the shared coverage area.

2. The method of claim 1, wherein, The step of determining, at least based on the current service information of the cell, whether there exists a first cell in the shared coverage area of ​​multiple cells that meets the criteria of being in a low-load state includes: The current service information of the cell is input into the second model to obtain the predicted load information of the common coverage area within a preset time period output by the second model; When it is determined from the predicted load information that at least one cell in the common coverage area has a load rate less than or equal to the load rate threshold, it is determined that there is a first cell in a low-load state.

3. The method according to claim 2, characterized in that, The method further includes: Based on the predicted load information, determine the duration for which the first cell will maintain the low load state; The energy-saving strategy obtained based on the predicted load information and the first model within the shared coverage area includes: The predicted load information within the common coverage area and the duration of the period are input into the first model to obtain the energy-saving strategy, wherein the energy-saving period included in the energy-saving strategy is located within the duration of the first cell maintaining the low load state.

4. The method according to claim 3, characterized in that, The energy-saving strategy instructs the first cell to enter the energy-saving state when the first model determines, based on the predicted load information, that at least one second cell is capable of accepting at least a portion of the service transfer from the first cell.

5. The method according to any one of claims 1 to 4, characterized in that, The energy-saving strategy includes energy-saving mode parameters that indicate the energy-saving method; The energy-saving method includes at least one of the following: The first cell is temporarily shut down; Suspend some carriers in the first cell; Suspend the establishment of radio frequency channels in part of the first cell; Suspend wireless transmission within certain symbols in the first cell.

6. The method according to any one of claims 1 to 4, characterized in that, The method further includes: Determine the actual load information within the common coverage area before the first cell enters the energy-saving state according to the energy-saving strategy, and the wireless communication performance indicators within the common coverage area after entering the energy-saving state; The first model is trained online based on the determined actual load information and the communication performance indicators.

7. The method according to any one of claims 1 to 4, characterized in that, The first model includes: Long Short-Term Memory (LSTM) network.

8. An energy-saving control device for a residential community, characterized in that, include: The first determining module is used to determine, at least based on the current service information of the cell, whether there is a first cell in the common coverage area of ​​multiple cells that meets the condition of being in a low-load state. The module is configured to obtain an energy-saving strategy based on the predicted load information and a first model within the common coverage area when the first cell is included in the common coverage area; wherein, the first model is a deep learning model trained based on historical load information and wireless communication performance indicators within the common coverage area; the predicted load information is obtained by time-series prediction using a long short-term memory network based on the current traffic load of cells within the common coverage area, and by introducing an attention mechanism to consider seasonal effects. A migration module is used to migrate at least a portion of the services of the first cell to at least one second cell in the common coverage area when the energy-saving strategy indicates that the first cell has entered an energy-saving state. The control module is configured to, after migrating at least some services of the first cell to the second cell, control the first cell to enter the energy-saving state by suspending at least some wireless functions according to the energy-saving strategy. The information input to the first model also includes at least one of the following: Weather information during the period in which the first cell maintains the low load state; Regional feature information of the common coverage area; Event information of predefined events affecting wireless communication during the duration; Information about the cells within the shared coverage area; The utilization rate of Physical Resource Blocks (PRBs) within the shared coverage area.

9. The apparatus according to claim 8, characterized in that, The first determining module is specifically used to input the current service information of the cell into the second model to obtain the predicted load information of the common coverage area within a preset time period output by the second model; when it is determined from the predicted load information that at least one cell in the common coverage area has a load rate less than or equal to a load rate threshold, it is determined that there is a first cell in a low load state.

10. The apparatus according to claim 9, characterized in that, The device further includes: The second determining module is used to determine the duration of the first cell maintaining the low load state based on the predicted load information; The obtaining module is specifically used to input the predicted load information and the duration of the common coverage area into the first model to obtain the energy-saving strategy, wherein the energy-saving period included in the energy-saving strategy is located within the duration of the first cell maintaining the low load state.

11. The apparatus according to claim 10, characterized in that, The energy-saving strategy instructs the first cell to enter the energy-saving state when the first model determines, based on the predicted load information, that at least one second cell is capable of accepting at least a portion of the service transfer from the first cell.

12. The apparatus according to any one of claims 8 to 11, characterized in that, The energy-saving strategy includes energy-saving mode parameters that indicate the energy-saving method; The energy-saving method includes at least one of the following: The first cell is temporarily shut down; Suspend some carriers in the first cell; Suspend the establishment of radio frequency channels in part of the first cell; Suspend wireless transmission within certain symbols in the first cell.

13. The apparatus according to any one of claims 8 to 11, characterized in that, The device further includes: The third determining module is used to determine the actual load information in the common coverage area before the first cell enters the energy-saving state according to the energy-saving strategy, and the wireless communication performance indicators in the common coverage area after entering the energy-saving state. The training module is used to train the first model online based on the determined actual load information and the communication performance indicators.

14. The apparatus according to any one of claims 8 to 11, characterized in that, The first model includes: Long Short-Term Memory (LSTM) network.

15. An electronic device, characterized in that, The electronic device includes: Memory; A processor, connected to the memory, is configured to implement the energy-saving control method for the cell as described in any one of claims 1 to 7 by executing computer-executable instructions stored in the memory.

16. A computer storage medium, characterized in that, The computer storage medium stores computer-executable instructions; after being executed by a processor, the computer-executable instructions can realize the energy-saving control method for the community as described in any one of claims 1 to 7.

Citation Information

Patent Citations

  • Base station energy-saving method and device

    CN112312531A