WO2022257670A1 - 基站节能方法、基站节能系统、基站及存储介质 - Google Patents

基站节能方法、基站节能系统、基站及存储介质 Download PDF

Info

Publication number
WO2022257670A1
WO2022257670A1 PCT/CN2022/091509 CN2022091509W WO2022257670A1 WO 2022257670 A1 WO2022257670 A1 WO 2022257670A1 CN 2022091509 W CN2022091509 W CN 2022091509W WO 2022257670 A1 WO2022257670 A1 WO 2022257670A1
Authority
WO
WIPO (PCT)
Prior art keywords
base station
energy
saving
target cell
load adjustment
Prior art date
Application number
PCT/CN2022/091509
Other languages
English (en)
French (fr)
Inventor
郭琳
马俊青
张波
陈利军
范英鹰
季尹禹
刘蕊
龚和平
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2022257670A1 publication Critical patent/WO2022257670A1/zh

Links

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
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
    • H04W52/02—Power saving arrangements
    • 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

Definitions

  • the embodiments of the present application relate to the technical field of base station energy saving, and in particular, to a base station energy saving method, a base station energy saving system, a base station, and a storage medium.
  • the traditional base station energy-saving solution is to apply preset energy-saving functions according to the current cell load.
  • This solution has pain points such as large investment in manual analysis, fixed energy-saving strategies, large impact on network performance, and lack of forward-looking strategies. It cannot effectively balance user service experience and The relationship between the energy saving effect of the base station.
  • Embodiments of the present application provide a base station energy saving method, a base station energy saving system, a base station, and a storage medium.
  • an embodiment of the present application provides a base station energy-saving method, which is applied to a base station energy-saving system.
  • the base station energy-saving system is connected to a base station.
  • the base station energy-saving method includes: acquiring historical data of a target cell, and the historical data is Historical service data and network configuration data related to the target cell; determining load adjustment policy information according to the historical data and the initial energy-saving threshold, the load adjustment policy information including the target energy-saving threshold based on a time window;
  • the load adjustment strategy information is sent to the base station corresponding to the target cell, so that the base station determines the target energy-saving threshold corresponding to the current time window according to the load adjustment strategy information, and adjusts the target energy-saving threshold corresponding to the current time window.
  • the cell cluster where the user is located in the target cell.
  • an embodiment of the present application provides a base station energy saving method, which is applied to a base station, and the base station is connected to a base station energy saving system.
  • the base station energy saving method includes: receiving load adjustment policy information sent by the base station energy saving system, The load adjustment policy information is obtained by the base station energy saving system according to the historical data of the target cell and the initial energy saving threshold, the historical data is historical service data and network configuration data related to the target cell, and the load adjustment policy
  • the information includes a target energy-saving threshold based on a time window; determining the target energy-saving threshold corresponding to the load adjustment policy information in the current time window; adjusting the cell cluster where the user in the target cell is located according to the target energy-saving threshold corresponding to the current time window .
  • the embodiment of the present application provides a base station energy saving system, including: a network management server, connected to the base station to obtain the historical data of the target cell, the historical data is historical service data and network configuration related to the target cell Data; the centralized self-organizing network system CSON is set to obtain the historical data of the target cell and submit a load adjustment policy information calculation request; the intelligent computing system ICS is set to receive the load adjustment policy information calculation request sent by the CSON and the historical data of the target cell, and is further configured to determine load adjustment policy information according to the historical data and the initial energy-saving threshold and send the load adjustment policy information to the CSON, so that the CSON sends the The load adjustment policy information is delivered to the base station through the network management server.
  • a network management server connected to the base station to obtain the historical data of the target cell, the historical data is historical service data and network configuration related to the target cell Data
  • the centralized self-organizing network system CSON is set to obtain the historical data of the target cell and submit a load adjustment policy information calculation request
  • the embodiment of the present application further provides a system, including at least one processor and a memory configured to communicate with the at least one processor; the memory stores information that can be executed by the at least one processor. Instructions, the instructions are executed by the at least one processor, so that the at least one processor can execute the method for power saving of a base station according to the first aspect.
  • the embodiment of the present application further provides a base station, including at least one processor and a memory configured to communicate with the at least one processor; the memory stores information that can be executed by the at least one processor. Instructions, the instructions are executed by the at least one processor, so that the at least one processor can execute the method for power saving of a base station according to the second aspect.
  • the embodiment of the present application also provides a computer-readable storage medium, the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to make a computer execute the The base station energy saving method or execute the base station energy saving method as described in the second aspect.
  • Fig. 1 is an overall method flowchart of a base station energy saving method of a base station energy saving system provided by an embodiment of the present application;
  • Fig. 2 is a flow chart of obtaining CM data and PM data provided by one embodiment of the present application
  • FIG. 3 is a flow chart of generating load adjustment policy information provided by an embodiment of the present application.
  • FIG. 4 is a flowchart of a method for determining a deployment scenario provided by an embodiment of the present application
  • FIG. 5 is a flowchart of a method for obtaining load adjustment strategy information through a subsequence splitting prediction algorithm provided by an embodiment of the present application
  • FIG. 6 is a flow chart of a method for iterative optimization according to data returned by a base station provided by an embodiment of the present application
  • FIG. 7 is a flowchart of a method for considering the influence of neighboring cells on load adjustment policy information provided by an embodiment of the present application
  • FIG. 8 is an overall method flowchart of a base station energy saving method for a base station provided by an embodiment of the present application.
  • FIG. 9 is a flowchart of a method for adjusting a cell cluster where a user is located according to an embodiment of the present application.
  • FIG. 10 is a network structure diagram of a base station energy saving system provided by an embodiment of the present application.
  • Fig. 11 is a structural diagram of a base station energy-saving system provided by an embodiment of the present application divided by functional modules;
  • FIG. 12 is an overall flowchart of a method for energy saving of a base station in an example of the present application.
  • Fig. 13 is a schematic diagram of adjusting the cell cluster where the user is located in the example of the present application.
  • Fig. 14 is a schematic structural diagram of a system provided by an embodiment of the present application.
  • FIG. 15 is a schematic structural diagram of a base station provided by an embodiment of the present application.
  • the energy-saving means of traditional base stations include symbol off, channel off, carrier off and equipment deep sleep.
  • symbol off is used to turn off the idle symbols of the service channel
  • channel off is used to turn off the idle radio frequency of the radio frequency module.
  • channel carrier cut-off is used to turn off the carrier that does not carry services
  • equipment deep sleep is used to control the radio frequency equipment to enter the dormant state;
  • the above solution has certain limitations, among which, the carrier shutdown and channel shutdown will shrink the cell coverage, and the symbol shutdown will increase the scheduling delay of the service, and the wake-up time from deep sleep will be longer. Pain points such as fixed policies, large impact on network performance, and lack of forward-looking policies.
  • the embodiment of the present application provides a base station energy saving method, a base station energy saving system, a base station and a storage medium.
  • the historical data of the target cell is intelligently analyzed, and the load is calculated in combination with the initial energy saving threshold threshold.
  • Adjusting the policy information can enable the base station to intelligently adjust the users in the target cell based on the target energy-saving threshold in the time window corresponding to the load adjustment policy information, optimize the energy efficiency of the base station, and achieve the best balance between energy consumption and performance.
  • an embodiment of the present application provides a base station energy saving method, which is applied to a base station energy saving system.
  • the base station energy saving system is connected to a base station.
  • the base station energy saving method includes but is not limited to the following steps S100, S200 and S300.
  • Step S100 acquiring historical data of the target cell, the historical data being historical service data and network configuration data related to the target cell.
  • the traditional energy-saving solution Due to the unified regional parameter strategy, no scene recognition, lack of adaptability, the traditional energy-saving solution has a low matching degree with the traffic fluctuation of the real site. It affects user perception, so in actual implementation, it is often carried out by uniformly turning off parameters in designated areas. Because the parameters cannot be set differently and cannot be strongly matched with the traffic of the site, the unreasonable parameter setting may cause some sites to suffer business damage when the traffic is busy and affect network performance, while some sites have no energy-saving effect when the business is idle. to maximize.
  • the above-mentioned traditional energy-saving scheme does not consider business scenarios on the one hand, and does not consider the changing rules of the business in different time windows on the other hand, and cannot perceive the business changes of the target cell in advance.
  • the embodiment of the present application takes into account the business scenarios and historical business change rules of the target cell, and considers the different network configurations of the base stations corresponding to different target cells, and obtains an intelligent energy-saving solution through big data analysis and intelligent calculation, so as to realize different scenarios and different Sites, different times, and multi-standard networks collaborate to save energy.
  • the energy-saving effect is maximized, and the best balance between energy consumption and performance is achieved.
  • the embodiment of the present application needs to collect historical data of the target cell, wherein the historical data mainly includes two parts of data, one part is the data related to service performance in the target cell, and the other part is the network configuration data of the target cell, wherein the NR side and the LTE (Long Term Evolution, long-term evolution) side as an example:
  • KPI Key Performance Indicator, key performance indicators
  • NR side Cell uplink/downlink PRB (Physical Resource Block, physical resource block) average occupancy rate, carrier uplink/downlink PRB time-frequency resource utilization, RRC connection establishment user number, cell uplink/downlink PDCP (Packet Data Convergence Protocol, packet Data aggregation protocol) layer average throughput, etc.;
  • PRB Physical Resource Block, physical resource block
  • RRC connection establishment user number Cell uplink/downlink PDCP (Packet Data Convergence Protocol, packet Data aggregation protocol) layer average throughput, etc.
  • PDCP Packet Data Convergence Protocol, packet Data aggregation protocol
  • the average uplink/downlink PRB utilization rate the number of RRC (Radio Resource Control, radio resource control) connection establishment users, the maximum number of NSA (Non-Standalone, non-independent networking) terminal subscribers who subscribe to 5G, and the user plane PDCP SDU ( Service Data Unit, service data unit) downlink data volume, uplink/downlink business average rate, etc.
  • RRC Radio Resource Control, radio resource control
  • NSA Non-Standalone, non-independent networking
  • Network configuration data includes, but is not limited to, base station configuration, target cell coverage data, application scenarios of the target cell (dense urban areas, suburban areas, etc.), site hardware equipment, and the like.
  • the acquisition of the above data can be obtained by the base station energy-saving system from the base station corresponding to the target cell through the network, or can be connected to other servers to obtain the historical data of the target cell stored in the server, which is not limited here.
  • Step S200 determining load adjustment strategy information according to historical data and an initial energy-saving threshold, where the load adjustment strategy information includes a target energy-saving threshold based on a time window.
  • the base station energy saving system analyzes and calculates the historical data and combines the corresponding initial energy saving thresholds to obtain the load adjustment policy information of the target cell. Since the historical data includes the business data of the target cell within a certain period of time, the change rule of the business in different time windows can be obtained through analysis and calculation, and then the target cell can be intelligently predicted according to the change rule of the business obtained from the analysis and calculation and the initial energy-saving threshold. The business load under different time windows, and formulate the target energy-saving threshold for the time window. In this way, when the base station executes the load adjustment policy information, it can automatically trigger the energy saving optimization policy according to the time window, and actively perform energy saving within the time window without human intervention.
  • the energy-saving threshold often includes a variety of different threshold parameters, such as symbol/carrier/channel shutdown and deep sleep load threshold, uplink/downlink PRB utilization threshold and adjustment upper limit, RRC user number threshold and adjustment upper limit etc. Therefore, in the process of formulating load adjustment policy information, the base station energy saving system needs to consider the impact of the initial energy saving threshold, instead of automatically determining the target energy saving threshold directly based on historical data, where the initial energy saving threshold can be determined by a preset value , such as the initial parameters set manually in the first calculation, can also be determined by iterative optimization in the process of each calculation of load adjustment strategy information by the preset value.
  • a preset value such as the initial parameters set manually in the first calculation
  • the initial energy-saving threshold is equivalent to The target energy-saving threshold threshold, so that the load adjustment policy information obtained by analysis and calculation can fit the service change law of the target cell, maximize the energy-saving effect and achieve the best energy consumption and performance without affecting the user service experience. balance.
  • Step S300 sending the load adjustment policy information to the base station corresponding to the target cell, so that the base station determines the target energy saving threshold corresponding to the current time window according to the load adjustment policy information, and adjusts the target cell according to the target energy saving threshold corresponding to the current time window
  • the cell cluster where the user is located
  • the base station energy saving system sends the load adjustment strategy information obtained through analysis and calculation to the base station corresponding to the target cell, so that the base station executes the load adjustment strategy information.
  • the load adjustment strategy information includes the target energy saving threshold of the discrete time window, and the base station executes the corresponding energy saving strategy according to the target energy saving threshold when the corresponding time window reaches the corresponding time window.
  • the base station adjusts the users in the target cell according to the target energy-saving threshold, so as to complete the matching between the business load and the network capacity, and calculate the network according to the current traffic volume of the base station and the energy consumption under the traffic volume Energy efficiency, and judge the business load based on each cell cluster.
  • some users are concentrated in some cell clusters, so that the rest of the cell clusters can be in an idle state, and the carrier can be turned off or the device can be in deep sleep, thereby reducing the energy of the base station. consumption.
  • the base station During the process of migrating the cell cluster where the user is located, the base station considers the service performance before and after the migration, and selects the appropriate user and the appropriate cell cluster under the premise of ensuring that the user experience is not affected, thereby completing the user migration.
  • "one policy for one area" can be realized, and appropriate load adjustment policy information can be tailored for different target communities, so as to ensure that the energy-saving characteristics can be maximized to match the actual operation conditions of the site.
  • the base station actively saves energy according to the target energy saving threshold based on the time window, which can reduce the energy consumption of the base station on the premise that the key KPI indicators of the base station are basically stable, and achieve a balance between energy saving and performance.
  • CM Configuration Management, configuration management
  • PM Performance Management, performance management
  • Step S110 sending a first request to the base station corresponding to the target cell
  • Step S120 acquiring CM data sent by the base station corresponding to the target cell according to the first request
  • Step S130 sending a second request to the SFTP (Secure File Transfer Protocol, Secure File Transfer Protocol) server;
  • Step S140 acquiring PM data sent by the SFTP server according to the second request.
  • the base station energy-saving system obtains data from the base station and the SFTP server respectively, wherein the CM data of the base station corresponding to the target cell is obtained by sending the first request, and the first request can be triggered manually or automatically.
  • the data management tool uses the data management tool to directly import the CM data of the base station.
  • the CM data of the base station can be automatically obtained at a fixed time point every day; in addition, the PM data of the corresponding base station of the target cell can be obtained through the second request.
  • the PM data is saved in the specified directory of the SFTP server, and the energy-saving system of the base station can capture PM data from the specified directory of the SFTP server at regular intervals. It can be understood that it is also possible to acquire CM data or PM data separately, and the acquisition methods correspond to the above steps S110-S120 and steps S130-S140 respectively.
  • step S200 may include the following steps:
  • Step S210 obtaining an initial energy-saving threshold, which is obtained from the preset load adjustment policy information or obtained through iterative optimization of the preset load adjustment policy information;
  • Step S220 preprocessing the historical data to determine the deployment scenarios of the target cell in different time windows
  • Step S230 obtain load adjustment strategy information according to the initial energy-saving threshold, deployment scenario, and time window-based subsequence splitting prediction algorithm.
  • the initial energy-saving threshold may be given by an initial threshold, or may be given by an iteratively optimized threshold.
  • the initial threshold is used to calculate the first load adjustment strategy information of the target cell, and then the base station can feed back the result of energy-saving adjustment based on the load adjustment strategy information back to the base station energy-saving system to iteratively optimize the initial threshold.
  • the relationship between service load and time in historical data can be analyzed based on the time window to determine the deployment scenarios of the target cell in different time windows (for example, the scenarios include "Business districts, residential areas"; at the same time, there are differences in the traffic models of "weekdays and weekends"), so, based on the initial energy-saving threshold, deployment scenarios and subsequence split prediction algorithm based on time windows, we can get In different business scenarios and different time windows, the business load changes, and the corresponding reasonable target energy saving thresholds under different time windows are obtained.
  • the subsequence splitting prediction algorithm is a kind of time series prediction algorithm. According to the structural model of the time series, various non-random components in the sequence are separated and predicted separately, and finally the predicted values of each part are synthesized into a total forecast result.
  • the time series prediction algorithm is used to summarize the law of service change over time in the historical data, and the load change law of the target cell can be accurately obtained, so as to realize active energy saving in the future time window.
  • the deployment scenario of the target cell includes two aspects, one is where the target cell is applied, and the other is the service law of the target cell that changes with time. Therefore, referring to FIG. 4, the deployment scenario in step S220 can be obtained through the following steps:
  • Step S221 determining the service coverage scenario of the target cell and the service time characteristics of the target cell in different time windows according to the historical data;
  • Step S222 determining the deployment scenarios of the target cell in different time windows according to the service coverage scenarios and service time characteristics.
  • the above-mentioned business coverage scenarios are mainly the scenarios of "business districts and residential areas", while the business time characteristics are mainly the business change rules of "weekdays and weekends”.
  • the operating environment of the target cell can be accurately determined through the service coverage scenario and service time characteristics, which is helpful for accurately generating load adjustment policy information.
  • the load adjustment policy information may be obtained through the following steps.
  • Step S231 acquiring energy-saving prediction task parameters including an initial energy-saving threshold, where the energy-saving prediction task parameters are task parameters for executing the energy-saving prediction task on the target cell;
  • Step S232 taking the service coverage scenario and service time characteristics in the deployment scenario as the first influencing factor, taking the distribution characteristics of holidays as the second influencing factor, and taking the energy saving prediction task parameters as the third influencing factor, according to the first influencing factor, the second influencing factor
  • the impact factor, the third impact factor and the subsequence split forecasting algorithm based on the time window obtain the load adjustment strategy information.
  • the energy-saving prediction task parameter is the parameter used when the base station energy-saving method establishes the energy-saving prediction task for the target cell, and the energy-saving prediction task parameter may include the following parameters:
  • Energy-saving period forecasting algorithm parameters including historical data collection time, forecasted load factors, load forecasting model update time, symbol/carrier/channel shutdown and deep sleep load thresholds;
  • Threshold adjustment parameters including uplink/downlink PRB utilization threshold and adjustment upper limit when adjusting the threshold based on energy-saving timeliness (according to the actual load situation on site) and RRC user number threshold and adjustment upper limit when adjusting the threshold based on energy-saving timeliness (according to the actual site load situation) depends).
  • the subsequence split prediction algorithm based on the time window can be used to obtain the best computing performance and optimization effect
  • the subsequence splitting prediction algorithm based on the time window can split the prediction according to the method of the same day within the week, so as to obtain the prediction result with the cycle period of the week and the time granularity of the day; of course, except for the method of the same day within the week , you can also use the same day within the month, the same time period every day, etc., and examples will not be given here.
  • Step S410 obtaining configuration management CM data and performance management PM data adjusted by the base station according to the load adjustment policy information
  • Step S420 iteratively optimize the target energy saving threshold in the load adjustment strategy information according to the adjusted CM data and PM data, and use the iteratively optimized target energy saving threshold as the initial energy saving threshold.
  • the base station energy-saving system obtains the load adjustment policy information based on the initial energy-saving threshold value and may not be completely suitable for the target cell (for example, the load adjustment policy information calculated for the first time based on the artificially set initial threshold value), then the base station adjusts the policy information according to the load. After the policy information is adjusted, the results are fed back to the base station energy-saving system.
  • the base station energy-saving system strengthens self-learning and iteratively optimizes online based on the full-scenario traffic model, big data analysis of energy-saving effects and KPI trends.
  • the process of determining load adjustment strategy information according to historical data and the initial energy-saving threshold in step S200 may further include the following steps:
  • Step S510 obtain the relationship information between the target cell and the adjacent cell through the background data configuration table of the OMC;
  • Step S520 adjusting the load adjustment policy information according to the historical data, the initial energy-saving threshold and the relationship information.
  • the base station energy-saving system obtains the target cell information and neighbor cell relationship information from the OMC (Operation&Maintenance Center, operation and maintenance center) background data configuration table, including frequency information, PCI (Physical Cell Identifie, physical cell identifier) information and basic neighbor cell level information. Configure information such as coverage relationship, CIO (Cell Individual Offset, cell bias), etc., to analyze how energy-saving strategies such as symbol shutdown, channel shutdown, carrier shutdown, and deep sleep match with corresponding configuration information to improve load forecasting the accuracy rate.
  • OMC Opera&Maintenance Center, operation and maintenance center
  • Configure information such as coverage relationship, CIO (Cell Individual Offset, cell bias), etc.
  • an embodiment of the present application also provides a base station energy saving method, which is applied to a base station, and the base station is connected to the base station energy saving system.
  • the base station energy saving method includes but is not limited to the following steps S600, S700 and S800.
  • Step S600 receiving the load adjustment policy information sent by the base station energy saving system, the load adjustment policy information is obtained by the base station energy saving system according to the historical data of the target cell and the initial energy saving threshold, the historical data is the historical service data and network configuration related to the target cell Data, load adjustment strategy information includes target energy-saving thresholds based on time windows;
  • Step S700 determining the target energy-saving threshold corresponding to the load adjustment policy information in the current time window
  • Step S800 adjusting the cell cluster where the user is located in the target cell according to the target energy saving threshold corresponding to the current time window.
  • the base station receives the load adjustment strategy information sent by the base station energy saving system through the network, and when the time window of the load adjustment strategy information is reached, it adjusts the policy information according to the target energy saving threshold corresponding to the current time window and the distribution of users in the target cell In this case, determine how to adjust the cell cluster where the user is located.
  • the base station After receiving the load adjustment strategy information and reaching the corresponding time window, the base station performs the following steps:
  • Step S810 determining user distribution information of each cell cluster of the target cell
  • Step S820 determine the user distribution adjustment mode according to the network performance of the base station in the current time window and the corresponding target energy-saving threshold
  • Step S830 selecting the first user in the target cell according to the user distribution adjustment method and migrating the first user from the source cell cluster to the target cell cluster.
  • Ee Energy efficiency, which is the energy efficiency of the network
  • v is the second-level traffic of the base station, and the unit is bits/s
  • En is the power consumed by the base station under the premise of providing second-level traffic.
  • the above steps are based on judging the traffic load of each cell cluster separately. When there are a small number of users distributed in all cell clusters, each cell cluster can only start symbol shutdown or channel shutdown. At this time, all users are concentrated in some cell clusters, and other The cell cluster can start the carrier shutdown or deep sleep of the device, and shut down more network hardware to maximize the energy saving effect.
  • the above steps are aimed at but not limited to the scenario of multi-standard multi-cell cluster networking, and can be divided into three sub-steps: frequency layer selection, user selection and user migration.
  • the source cell cluster can activate more basic energy-saving functions by migrating users
  • the target cell cluster will not shut down the activated basic energy saving function due to user migration.
  • All services of the user can be carried by the target cell cluster: comprehensive terminal capabilities, mobility restrictions configured by the core network, network slicing and other factors, only when all services of the user can be carried by the target cell cluster can be used as a candidate for migration user;
  • connection state user migration is divided into connection state user migration and idle state user migration:
  • inter-frequency redirection or inter-system redirection is used.
  • the source cell cluster cell reselection priority should be set to the lowest.
  • the base station adjusts the users based on the load adjustment strategy information, with the goal of improving network energy efficiency, so that the multi-standard and multi-frequency layer networks can work together, complete the matching of traffic load and network capacity, and coordinate the adjustment of the source cell
  • the network capacity of the cluster and the target cell cluster can reduce the overall energy consumption and improve the energy efficiency of the base station.
  • the base station energy saving system includes:
  • the network management server is connected to the base station to obtain historical data of the target cell, and the historical data is historical service data and network configuration data related to the target cell;
  • the centralized self-organizing network system CSON is set to obtain the historical data of the target cell and submit a load adjustment strategy information calculation request;
  • the intelligent computing system ICS is set to receive the load adjustment policy information calculation request sent by CSON and the historical data of the target cell, and is also set to determine the load adjustment policy information according to the historical data and the initial energy-saving threshold threshold and send the load adjustment policy information to CSON, so that the CSON sends the load adjustment policy information to the base station through the network management server.
  • the base station is connected to the network management server, and the network management server, CSON, and ICS are connected in turn; among them, the base station is in a certain radio coverage area, through the mobile communication switching center, and the radio transmission and reception of information transmission between mobile terminals
  • the communication station also reports KPI information, UE (User Equipment, user terminal) status information, channel status information, etc. to the network management server, and performs related energy-saving operations.
  • the relevant energy-saving operation includes two aspects: navigation mode and energy-saving mode.
  • the navigation mode refers to the above-mentioned load adjustment strategy information. After modeling, it can include normal mode, normal mode and economic mode. The normal mode does not execute the above-mentioned load adjustment strategy.
  • the normal mode is to execute the navigation mode for small traffic services, and the economic mode is to enforce the navigation mode for all services; the energy saving mode is the same as before, including sign off, channel off
  • the balance between base station performance and energy saving is achieved by superimposing the navigation mode of the embodiment of the present application on the traditional energy saving mode.
  • the network management server is used to periodically update historical data such as CM data and PM data, push it to CSON through the northbound public process (northbound interface), and update and send it to the base station to implement load adjustment policy information at regular intervals.
  • CSON comprehensively collects network data and conducts precise and intelligent positioning and analysis, and sends the corresponding energy-saving calculation tasks to ICS, and sends the load adjustment policy information calculated by ICS to the network management server , to help wireless operators realize the automation of network planning, configuration and optimization processes, and reduce the need for manual work.
  • the intelligent computing system ICS is a one-stop algorithm and intelligent application development platform built on the Spark computing platform, which can help users quickly develop algorithms for business applications and complete algorithm evaluation and release.
  • the intelligent computing system ICS performs artificial intelligence calculation on the historical data and the initial energy-saving threshold to obtain the load adjustment policy information.
  • the modules can be divided into an energy-saving policy module, an energy-saving evaluation module, an energy-saving execution module, and a service navigation module.
  • the energy-saving strategy module is set to automatically configure the energy-saving strategies and parameters of each cell based on historical data such as site configuration, user characteristics, traffic patterns, coverage characteristics, and energy efficiency curves, and send them to the business navigation module and energy-saving execution module; at the same time Guided by the fluctuation range allowed by the KPI, the energy-saving mode is intelligently iteratively optimized.
  • the energy-saving evaluation module is set to monitor and visualize base station service fluctuations, MR (Measurement Result, measurement report) data changes, energy-saving gains, and KPI in real time.
  • the energy-saving execution module is set to perform basic energy-saving functions, and automatically turns on energy-saving functions within the allowed time period output by the energy-saving policy module, including symbol (time slot) shutdown, channel shutdown, carrier shutdown, and deep sleep of equipment.
  • the business navigation module is set to distinguish the types of navigation systems, including intra-system, inter-system, service identification, ARP (Allocation and Retention Priority, allocation and retention priority), service distribution (all business navigation in economic mode or small traffic in normal mode) business navigation).
  • ARP Allocation and Retention Priority, allocation and retention priority
  • service distribution all business navigation in economic mode or small traffic in normal mode
  • the embodiment of the present application collects the historical data of the target cell, classifies the cell business model, evaluates the KPI, and then uses the intelligent prediction algorithm to generate the cell load prediction model (load adjustment policy information) and sends it out regularly.
  • the target cell is within the time window Execute energy-saving adjustment tasks, periodically select users, migrate users to the desired target neighboring cells, and make decisions to dynamically optimize the initial energy-saving threshold during the energy-saving time period to improve energy-saving effects.
  • the embodiment of the present application guides user behavior and adjusts load distribution through service navigation, thereby creating more cells and energy-saving time windows that meet traditional energy-saving conditions.
  • Step S101 creating an energy-saving task and obtaining parameter configuration
  • Energy-saving period forecasting algorithm parameters including historical data collection time, forecasted load factors, load forecasting model update time, symbol/carrier/channel shutdown and deep sleep load thresholds;
  • Threshold adjustment parameters including uplink/downlink PRB utilization threshold and adjustment upper limit when adjusting the threshold based on energy-saving timeliness (according to the actual load situation on site) and RRC user number threshold and adjustment upper limit when adjusting the threshold based on energy-saving timeliness (according to the actual site load situation) depends).
  • the energy-saving shutdown priority of the frequency point needs to be set.
  • Step S102 collecting historical data
  • the historical data comes from the KPI data pushed by the network management.
  • the evaluation load needs to be collected but not limited to the following KPIs:
  • NR side average cell uplink/downlink PRB occupancy rate, carrier uplink/downlink PRB time-frequency resource utilization, number of RRC connection establishment users, cell uplink/downlink PDCP layer average throughput, etc.;
  • the average uplink/downlink PRB utilization rate the number of RRC connection establishment users, the maximum number of NSA terminal subscribers who subscribe to 5G, the downlink data volume of PDCP SDUs on the user plane, and the average rate of uplink/downlink services.
  • the relationship between the target cell and neighboring cells is also obtained from the background data configuration table, including frequency point information, PCI information, and basic configurations at the neighboring cell level such as coverage relationship, CIO and other information.
  • the acquisition method includes the following steps:
  • CSON actively sends a request to the network management server system to obtain CM data
  • the network management server system returns the CM configuration file to CSON.
  • the acquisition method includes the following steps:
  • CSON periodically fetches PM data from the specified directory of the SFTP server.
  • Step S103 data preprocessing
  • Step S104 load forecasting modeling during energy-saving time period and navigation time period
  • This step includes load evaluation factors.
  • the load decision factors adopted in the energy-saving scheme of the base station include uplink PRB utilization, downlink PRB utilization and the number of RRC users.
  • all three load factors may be used; for channel cut-off, the downlink PRB utilization rate and the number of RRC users are used, and for symbol cut-off, only the downlink PRB utilization rate is currently used.
  • the load evaluation factors may not be completely consistent with the front-end load factors, and additional load factors, such as throughput and throughput rate, may also be considered.
  • the input data is community-level (logical community) KPI data
  • the output data is load forecast data of multiple time windows (default is 1 day) at the community level.
  • the modeling of traffic forecasting based on historical data, three types of cells with positive effect, negative effect and no effect are distinguished, and the second-order exponential smoothing prediction algorithm is used to obtain the prediction model with the best calculation performance and the best effect.
  • CSON calculates the energy-saving solution through ICS, and the interaction process between CSON and ICS is as follows:
  • CSON sends an energy-saving strategy calculation request to ICS, including various parameters for calculating energy-saving solutions, such as various threshold parameters, and triggers ICS to calculate energy-saving solutions;
  • CSON regularly sends a progress query request to ICS, asking ICS whether the calculation of this energy-saving strategy has been completed;
  • the ICS calculation After the ICS calculation is completed, it receives the progress query request sent by CSON, and sends the calculation result to CSON.
  • the CSON can automatically execute the process each time the energy-saving policy calculation request is executed.
  • Step S105 NR cell energy saving or service navigation decision
  • This step determines the traffic load based on each cell cluster.
  • the source cell cluster is L1 and the target cell cluster is L2.
  • L1 belongs to the new system deployment cell cluster
  • L2 belongs to the large-scale commercial cell cluster.
  • this step is not limited to the cooperation between adjacent cell clusters, but can be extended to service navigation across cell clusters.
  • Migrating users also includes three sub-steps: frequency layer selection, user selection, and user migration. To avoid repetition, the contents of these three sub-steps are omitted here. See the above description for details.
  • the execution of this step includes the interaction process between CSON and the network management server, including the following steps:
  • CSON sends a request to the network management server, and sends the calculated load adjustment policy information to the network management server;
  • the network management server After receiving the request, the network management server starts to synchronize the load adjustment policy information
  • the network management server After the synchronization is completed, the network management server returns the synchronization result to CSON.
  • Step S106 updating the database
  • the entire network cells may output requests to change parameters, after the database receives a parameter change request, it needs to update the parameter data in a timely manner: the update of the energy-saving time period, according to the received computing energy-saving strategy, directly execute the update of the database.
  • the above-mentioned update method can be directly executed according to the request; in the controlled mode of the database, a confirmation request for updating the database is output and executed, and then executed after manual confirmation.
  • the database can update the database in batches for the cells so that the new energy-saving strategy takes effect.
  • Step S107 data presentation.
  • CSON can present function-related KPIs in real time. At present, it mainly presents the effect of energy saving or navigation operation in the form of manual statistics, that is, the start time period of data presentation can be manually set, and the shutdown time period of centralized network element output is specially marked. To compare the performance before and after energy saving and before and after business navigation.
  • the base station energy saving method coordinates the functions of the base station, network management server, centralized self-organizing network system, and experimental data intelligent computing system According to the different network configurations and historical business conditions of different sites, intelligently customize the coordinated energy-saving scheduling strategy between sites, frequency bands, and standards, that is, load adjustment policy information.
  • ICS uses AI technology to self-learn historical traffic and network KPIs, accurately predict community load trends, and automatically plan future energy-saving strategies to achieve "one station, one policy" and maximize energy-saving effects.
  • the relationship between the above-mentioned base station and the energy-saving system of the base station adopts a similar C/S (Client-Server, server-client) service architecture model, which can be divided into the server side and the network element side.
  • the server side includes an energy-saving strategy module and an energy-saving evaluation module.
  • the network element side includes an energy-saving execution module and a service navigation module.
  • Each module on the server side provides the modules on the network element side with automatic configuration of energy-saving strategies and parameters for each cell, sends them to the service navigation module and energy-saving execution module, and monitors service fluctuations in base stations, MR data changes, energy-saving gains, and KPI visualization services.
  • the network element side performs carrier-level basic energy saving and service navigation functions, and completes the task steps of intelligent energy saving.
  • the entire device operation process forms a closed loop, and the results of the energy-saving execution module will be fed back to the energy-saving evaluation module, so that it can perceive the implementation effect of the energy-saving strategy in a timely manner, and adjust the rationality of the strategy generation in the next cycle, on the premise of ensuring user experience and network performance
  • predict the network load evaluate the service demand, and match the network demand with the network energy consumption to achieve the best energy efficiency ratio of the entire network and the lowest energy consumption per bit.
  • An embodiment of the present application also provides a system, including at least one processor and a memory for communicating with the at least one processor; the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor , so that at least one processor can execute the foregoing base station energy saving method.
  • the memory 1002 can be used to store non-transitory software programs and non-transitory computer-executable programs.
  • the memory 1002 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one disk memory, a flash memory device, or other non-transitory solid-state storage devices.
  • the memory 1002 includes memory located remotely relative to the control processor 1001, and these remote memories may be connected to the system 1000 through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
  • FIG. 14 does not constitute a limitation to the system 1000, and may include more or less components than shown in the figure, or combine some components, or arrange different components.
  • the embodiment of the present application also provides a computer-readable storage medium, the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are executed by one or more control processors, for example, by the Executed by a control processor 1001 of the above-mentioned method, it can make the above-mentioned one or more control processors execute the energy-saving method of the base station in the above-mentioned method embodiment, for example, execute the method steps S100 to S300 in FIG. 1 described above, and the method in FIG. 2 Method step S110 to step S130, method step S210 to step S230 in Fig. 3, method step S221 to step S222 in Fig. 4, method step S231 to step S232 in Fig. 5, method step S410 and step S420 in Fig. 6 And the method step S510 and step S520 in FIG. 7 .
  • the embodiment of the present application also provides a base station, including at least one processor and a memory for communicating with the at least one processor; the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor , so that at least one processor can execute the foregoing base station energy saving method.
  • the memory 2002 can be used to store non-transitory software programs and non-transitory computer-executable programs.
  • the memory 2002 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one disk memory, a flash memory device, or other non-transitory solid-state storage devices.
  • the memory 2002 includes memory that is remotely located relative to the control processor 2001, and these remote memories can be connected to the base station 2000 through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
  • the device structure shown in FIG. 15 does not limit the base station 2000, and may include more or less components than shown in the figure, or combine some components, or arrange different components.
  • the embodiment of the present application also provides a computer-readable storage medium, the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are executed by one or more control processors, for example, by the Executed by a control processor 2001 of the above-mentioned one or more control processors to execute the method for power saving of the base station in the above-mentioned method embodiment, for example, execute steps S600 to S800 of the method in FIG. 8 described above and the method in FIG. 9 Method steps S810 and S830.
  • the base station energy-saving method provided by the embodiment of the present application has at least the following beneficial effects: the embodiment of the present application obtains the historical data of the target cell, based on the initial energy-saving threshold in the base station energy-saving system and the historical service data in the historical data and the corresponding target
  • the network configuration data of the cell is used to predict the business of big data to obtain the load adjustment strategy information.
  • the load adjustment strategy information includes the target energy-saving threshold of the target cell in different time windows.
  • the target energy-saving threshold takes into account the different time windows. Business conditions and possible energy-saving effects. Therefore, the base station performs user migration according to the target energy-saving threshold corresponding to the current time window, and navigates users to network bearers with higher energy efficiency ratios.
  • the active energy saving of the time window improves the efficiency of the entire network.
  • the device embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cartridges, tape, magnetic disk storage or other magnetic storage devices, or can Any other medium used to store desired information and which can be accessed by a computer.
  • communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media .

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

一种基站节能方法、基站节能系统、基站及存储介质，其中，基站节能方法包括获取目标小区的历史数据(S100)，历史数据为与目标小区相关的历史业务数据和网络配置数据；根据历史数据和初始节能门限阈值确定负荷调整策略信息(S200)，负荷调整策略信息包括基于时间窗口的目标节能门限阈值；将负荷调整策略信息发送到目标小区对应的基站，以使基站根据负荷调整策略信息确定在当前时间窗口对应的目标节能门限阈值，并根据该目标节能门限阈值调整目标小区中用户所在的小区簇(S300)。

Description

基站节能方法、基站节能系统、基站及存储介质
相关申请的交叉引用
本申请基于申请号为202110633421.8、申请日为2021年06月07日的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本申请实施例涉及基站节能技术领域，尤其涉及一种基站节能方法、基站节能系统、基站及存储介质。
背景技术
当前移动通信网络多种通信制式共存，随着基站信号覆盖范围和容量持续增大、多模多制式网络间的耦合，使得基站占据移动网络超过一半以上的能耗，因此对基站进行节能降耗一直是运营商和设备商的一项艰巨挑战。
传统的基站节能方案是根据当前小区负荷的情况套用预设的节能功能，该方案存在人工分析投入大、节能策略固化、网络性能影响大、策略缺乏前瞻性等痛点，无法有效平衡用户业务体验和基站节能效果之间的关系。
发明内容
以下是对本文详细描述的主题的概述。本概述并非是为了限制权利要求的保护范围。
本申请实施例提供了一种基站节能方法、基站节能系统、基站及存储介质。
第一方面，本申请实施例提供了一种基站节能方法，应用于基站节能系统，所述基站节能系统与基站连接，所述基站节能方法包括：获取目标小区的历史数据，所述历史数据为与所述目标小区相关的历史业务数据和网络配置数据；根据所述历史数据和初始节能门限阈值确定负荷调整策略信息，所述负荷调整策略信息包括基于时间窗口的目标节能门限阈值；将所述负荷调整策略信息发送到所述目标小区对应的基站，以使所述基站根据所述负荷调整策略信息确定在当前时间窗口对应的目标节能门限阈值，并根据当前时间窗口对应的目标节能阈值调整所述目标小区中用户所在的小区簇。
第二方面，本申请实施例提供了一种基站节能方法，应用于基站，所述基站与基站节能系统连接，所述基站节能方法包括：接收由所述基站节能系统发送的负荷调整策略信息，所述负荷调整策略信息由所述基站节能系统根据目标小区的历史数据和初始节能门限阈值得到，所述历史数据为与所述目标小区相关的历史业务数据和网络配置数据，所述负荷调整策略信息包括基于时间窗口的目标节能门限阈值；确定所述负荷调整策略信息在当前时间窗口对应的目标节能门限阈值；根据当前时间窗口对应的目标节能门限阈值调整所述目标小区中用户所在的小区簇。
第三方面，本申请实施例提供了一种基站节能系统，包括：网络管理服务器，连接基站以获取目标小区的历史数据，所述历史数据为与所述目标小区相关的历史业务数据和网络配置数据；集中式自组织网络系统CSON，被设置为获取所述目标小区的历史数据并提交负荷调整策略信息计算请求；智能计算系统ICS，被设置为接收所述CSON发送的负荷调整策略信息计算请求和所述目标小区的历史数据，还被设置为根据所述历史数据和初始节能门限阈值确定负荷调整策略信息并将所述负荷调整策略信息发送到所述CSON，以使所述CSON将所述负荷调整策略信息通过所述网络管理服务器下发到基站。
第四方面，本申请实施例还提供了一种系统，包括至少一个处理器和用于与所述至少一个处理器通信连接的存储器；所述存储器存储有能够被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够执行如第一方面的基站节能方法。
第五方面，本申请实施例还提供了一种基站，包括至少一个处理器和用于与所述至少一个处理器通信连接的存储器；所述存储器存储有能够被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够执行如第二方面所述的基站节能方法。
第六方面，本申请实施例还提供了一种计算机可读存储介质，所述计算机可读存储介质存储有计算机可执行指令，所述计算机可执行指令用于使计算机执行如第一方面所述的基站节能方法或执行如第二方面所述的基站节能方法。
本申请的其它特征和优点将在随后的说明书中阐述，并且，部分地从说明书中变得显而易见，或者通过实施本申请而了解。本申请的目的和其他优点可通过在说明书、权利要求书以及附图中所特别指出的结 构来实现和获得。
附图说明
附图用来提供对本申请技术方案的进一步理解，并且构成说明书的一部分，与本申请的示例一起用于解释本申请的技术方案，并不构成对本申请技术方案的限制。
图1是本申请一个实施例提供的基站节能系统的基站节能方法的整体方法流程图；
图2是本申请一个实施例提供的获取CM数据和PM数据的流程图；
图3是本申请一个实施例提供的生成负荷调整策略信息的流程图；
图4是本申请一个实施例提供的确定部署场景的方法流程图；
图5是本申请一个实施例提供的通过子序列拆分预测算法得到负荷调整策略信息的方法流程图；
图6是本申请一个实施例提供的根据基站返回数据进行迭代调优的方法流程图；
图7是本申请一个实施例提供的考虑邻区对负载调整策略信息影响的方法流程图；
图8是本申请一个实施例提供的基站的基站节能方法的整体方法流程图；
图9是本申请一个实施例提供的调整用户所在的小区簇的方法流程图；
图10是本申请一个实施例提供的基站节能系统的组网结构图；
图11是本申请一个实施例提供的基站节能系统按功能模块划分的结构图；
图12是本申请示例的基站节能方法的整体流程图；
图13是本申请示例的用户所在的小区簇进行调整的示意图；
图14是本申请实施例提供的系统的结构示意图；
图15是本申请实施例提供的基站的结构示意图。
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施例仅用以解释本申请，并不用于限定本申请。
当前移动通信网络多种制式(2G、3G、4G和5G)并存，随着移动通信网络的覆盖范围和容量持续增大、多模多制式网络间的耦合，使得基站占据移动网络超过一半以上的能耗；另一方面，5G的射频模块MIMO(Multiple Input Multiple Output，多进多出)数量巨大且输出功率较高，NR(New Radio，新空口)基站的功耗是当前4G基站的数倍，因此，降低基站的能耗有助于降低运营商和设备商的成本。
目前传统基站的节能手段包括符号关断、通道关断、载波关断和设备深度休眠，其中，符号关断用于关断业务信道的空闲符号，通道关断用于关断射频模块的空闲射频通道，载波关断用于关断没有承载业务的载波，设备深度休眠则是用于控制射频设备进入休眠状态；上述节能手段的生效方式均为网元对小区的当前负荷做评估，根据当前负荷的大小套用预设的节能手段。上述方案存在一定的局限性，其中载波关断和通道关断会收缩小区覆盖，符号关断会使业务的调度时延增加，深度休眠唤醒时间较长，同时，还存在人工分析投入大、节能策略固化、网络性能影响大、策略缺乏前瞻性等痛点。
基于此，本申请实施例提供了一种基站节能方法、基站节能系统、基站及存储介质，根据大数据平台和智能算法，对目标小区的历史数据进行智能分析，结合初始节能门限阈值计算得到负荷调整策略信息，可以使得基站基于时间窗口对应负荷调整策略信息中的目标节能门限阈值对目标小区中的用户进行智能调整，优化基站能效，实现能耗与性能的最佳平衡。
参照图1，本申请实施例提供了一种基站节能方法，应用于基站节能系统，基站节能系统与基站连接，基站节能方法包括但不限于以下步骤S100、步骤S200和步骤S300。
步骤S100，获取目标小区的历史数据，历史数据为与目标小区相关的历史业务数据和网络配置数据。
传统的节能方案由于区域参数策略统一，无场景识别，缺乏适应性，与真实站点的话务量波动匹配度低，依靠人工分析海量数据易错、耗时，且无法实时监控性能、故障等，影响用户感知，因此在实际实施时，往往采用在指定区域统一关断参数的方式进行。因为参数无法差异化设置，无法与站点话务强匹配，有可能会因参数设置不合理导致一些站点在话务繁忙时业务受损，影响网络性能，而有些站点在业务闲时，节能效果无法达到最大化。
可以看到，上述传统的节能方案一方面没有考虑业务场景，另一方面没有考虑业务在不同时间窗口的变化规律，无法提前感知目标小区的业务变化，因此只能根据当前的业务负荷生硬套用现有的节能方式。本申请实施例将目标小区的业务场景和历史的业务变化规律考虑进来，并考虑不同目标小区对应的基站的不同网络配置，通过大数据分析和智能计算，得到智能节能方案，实现不同场景、不同站点、不同时间、多制式网络协同节能，在保证网络KPI的基础上，使节能效果最大化，实现能耗与性能的最佳平衡。
因此，本申请实施例需要采集目标小区的历史数据，其中历史数据主要包括两部分数据，一部分是目 标小区中与业务性能相关数据，另一部分是目标小区的网络配置数据，其中以NR侧和LTE(Long Term Evolution，长期演进)侧为例：
业务性能相关的数据包括但不限于如下KPI(Key Performance Indicator，关键性能指标)数据：
NR侧：小区上行/下行PRB(Physical Resource Block，物理资源块)平均占用率、载波上行/下行PRB时频资源利用率、RRC连接建立用户数、小区上行/下行PDCP(Packet Data Convergence Protocol，分组数据汇聚协议)层平均吞吐量等；
LTE侧：上行/下行PRB平均利用率、RRC(Radio Resource Control，无线资源控制)连接建立用户数、NSA(Non-Standalone，非独立组网)终端签约5G的最大用户数、用户面PDCP SDU(Service Data Unit，服务数据单元)下行数据量、上行/下行业务平均速率等。
网络配置数据包括但不限于基站配置、目标小区覆盖数据、目标小区的应用场景(密集城区、郊区等)、站点硬件设备等。
上述数据的获取可以由基站节能系统通过网络向目标小区对应的基站获取，也可以连接其他服务器来获取该服务器中存储有的目标小区的历史数据，在此不作限定。
基于历史数据中的业务性能相关数据和网络配置数据，可以通过大数据分析和智能计算为目标小区制定相应的节能策略。
步骤S200，根据历史数据和初始节能门限阈值确定负荷调整策略信息，负荷调整策略信息包括基于时间窗口的目标节能门限阈值。
本步骤中，由基站节能系统对历史数据进行分析计算，并结合相应的初始节能门限阈值，可以得到目标小区的负荷调整策略信息。由于历史数据包含了一段时间内的目标小区的业务数据，因此通过分析计算可以得到在不同时间窗口业务的变化规律，再根据分析计算所得的业务的变化规律和初始节能门限阈值，智能预测目标小区在不同时间窗口下的业务负荷，并为该时间窗口制定目标节能门限阈值。这样，当基站执行该负荷调整策略信息时，就可以根据时间窗口自动触发节能优化策略，在时间窗口内主动进行节能，无需人为干预。
可以理解的是，节能门限阈值往往包含了多种不同的门限参数，例如符号/载波/通道关断以及深度休眠的负荷门限、上行/下行PRB利用率门限及调整上限、RRC用户数门限及调整上限等。因此，基站节能系统在制定负荷调整策略信息的过程中，需要考虑初始节能门限阈值带来的影响，并非直接根据历史数据自动确定目标节能门限阈值，其中，初始节能门限阈值可以由预设值确定，如首次计算时由人为设定的初始参数，也可以由预设值在每次计算负荷调整策略信息的过程中迭代调优确定，此时初始节能门限阈值就相当于上一次迭代调优中的目标节能门限阈值，从而可以使得分析计算得到的负荷调整策略信息能够贴合目标小区的业务变化规律，在不影响用户业务体验的前提下，最大化节能效果，实现能耗与性能的最佳平衡。
步骤S300，将负荷调整策略信息发送到目标小区对应的基站，以使基站根据负荷调整策略信息确定在当前时间窗口对应的目标节能门限阈值，并根据当前时间窗口对应的目标节能门限阈值调整目标小区中用户所在的小区簇。
基站节能系统将分析计算得到的负荷调整策略信息发送到目标小区对应的基站，以使基站执行该负荷调整策略信息。在一些情况下，负荷调整策略信息包含的是离散时间窗口的目标节能门限阈值，基站到达对应的时间窗口时才按照目标节能门限阈值执行相应的节能策略。
当到达执行节能策略时间段时，基站按照目标节能门限阈值对目标小区内的用户进行调整，以便完成业务负荷和网络容量的匹配，根据基站当前的业务量和该业务量下的耗能计算网络能效，并基于各个小区簇分别判断业务负荷，此时将部分用户集中到部分小区簇，就可以使其余小区簇处于空闲状态，可以对其启动载波关断或者设备深度睡眠，从而降低基站的能耗。在迁移用户所在小区簇的过程中，基站考虑迁移前后的业务性能，在保证不影响用户的使用体验的前提下，选择合适的用户和合适的小区簇，从而完成用户迁移。通过上述方式能够实现“一区一策”，为不同目标小区量身定制合适的负荷调整策略信息，保证节能特性最大化匹配站点实际运行情况。
通过上述步骤，基站根据基于时间窗口的目标节能门限阈值进行主动节能，能够保证基站关键KPI指标基本稳定的前提下，降低基站的能耗，实现节能和性能两者的平衡。
可以理解的是，在ISO中定义了网络管理的五大功能，分别是故障管理、计费管理、配置管理、性能管理和安全管理，与基站节能相关的管理功能一般包括配置管理和性能管理。那么在本申请实施例中，历史数据可以包括CM(Configuration Management，配置管理)数据和/或PM(Performance Management，性能管理)数据，两种数据可以通过不同的途径获取，本申请实施例以获取CM数据和PM数据为例进行说明，参照图2，上述步骤S100中获取历史数据可以包括以下步骤：
步骤S110，向目标小区对应的基站发送第一请求；
步骤S120，获取由目标小区对应的基站根据第一请求发送的CM数据；
步骤S130，向SFTP(Secure File Transfer Protocol，安全文件传送协议)服务器发送第二请求；
步骤S140，获取由SFTP服务器根据第二请求发送的PM数据。
本实施例中，基站节能系统分别从基站和SFTP服务器获取数据，其中，通过发送第一请求获取目标小区对应基站的CM数据，第一请求可以通过手动触发，也可以自动触发，手动触发的情况下，使用数据管理工具直接导入基站的CM数据，自动触发的情况下，可以在每天固定时间点自动获取基站的CM数据；另外还通过第二请求获取目标小区对应基站的PM数据，本实施例中PM数据保存在SFTP服务器的指定目录，基站节能系统可以按照固定的时间间隔从SFTP服务器的指定目录抓取PM数据。可以理解的是，单独获取CM数据或者单独获取PM数据也是可以的，获取方式分别对应上述步骤S110-S120和步骤S130-S140。
通过上述步骤获取历史数据后，需要导入相应的工程参数，并对历史数据进行预处理，以便后续进行分析计算负荷调整策略信息，例如，参照图3，步骤S200可以包括以下步骤：
步骤S210，获取初始节能门限阈值，初始节能门限阈值由预设负荷调整策略信息得到或由预设负荷调整策略信息迭代调优得到；
步骤S220，对历史数据进行预处理以确定目标小区在不同时间窗口内的部署场景；
步骤S230，根据初始节能门限阈值、部署场景和基于时间窗口的子序列拆分预测算法得到负荷调整策略信息。
其中初始节能门限阈值可以由初始门限给出，也可以由迭代调优后的门限给出。初始门限用于计算出目标小区的首个负荷调整策略信息，后续可以由基站将根据负荷调整策略信息执行节能调节的结果反馈回到基站节能系统，对初始门限进行迭代调优。另外，由于历史数据包括了一段时间内的业务数据，因此基于时间窗口来分析历史数据中业务负荷与时间之间的关系，即可判断目标小区在不同时间窗口内的部署场景(例如，场景包括“商务区、居民区”；同时，“工作日、周末”的话务模型也存在差异)，这样，基于初始节能门限阈值、部署场景和基于时间窗口的子序列拆分预测算法，即可得到在不同业务场景、不同时间窗口下的业务负荷变化规律，并得到不同时间窗口下对应的合理的目标节能门限阈值。
可以理解的是，子序列拆分预测算法是时间序列预测算法的一种，根据时间序列的结构模型将序列中各种非随机成分分离出来，分别进行预测，最后将各部分预测值合成总的预测结果。本申请实施例中，采用时间序列预测算法总结历史数据中业务随时间变化的规律，可以准确得到目标小区的负荷变化规律，从而实现未来时间窗口内的主动节能。
上述步骤S220中，目标小区的部署场景包括两方面，一方面是目标小区所应用的地方，一方面是目标小区跟随时间变化的业务规律。因此参照图4，步骤S220中部署场景可以通过以下步骤得到：
步骤S221，根据历史数据确定目标小区的业务覆盖场景以及目标小区在不同时间窗口内的业务时间特性；
步骤S222，根据业务覆盖场景和业务时间特性确定目标小区在不同时间窗口内的部署场景。
上述业务覆盖场景主要是“商务区、居民区”方面的场景，而业务时间特性主要是“工作日、周末”方面的业务变化规律。通过业务覆盖场景和业务时间特性可以准确确定目标小区的运行环境，有助于准确生成负荷调整策略信息。
可以理解的是，在计算负荷调整策略信息过程中，还可以加入其它影响因子来进一步提高负荷调整策略信息的准确程度。例如，参照图5，在上述步骤S230可以通过以下步骤得到负荷调整策略信息。
步骤S231，获取包括初始节能门限阈值的节能预测任务参数，节能预测任务参数为对目标小区执行节能预测任务的任务参数；
步骤S232，以部署场景中的业务覆盖场景和业务时间特性为第一影响因子，以节假日分布特性为第二影响因子，以节能预测任务参数作为第三影响因子，根据第一影响因子、第二影响因子、第三影响因子和基于时间窗口的子序列拆分预测算法得到负荷调整策略信息。
节能预测任务参数是基站节能方法为目标小区建立节能预测任务时所采用的参数，节能预测任务参数可以包括以下参数：
公共任务参数，包括节能时间预测功能开关、调整评估迭代周期和评估迭代次数；
节能时间段预测算法参数，包括历史数据的采集时长，预测的负荷因素，负荷预测模型更新时间，符号/载波/通道关断和深度休眠的负荷门限；
门限调整参数，包括基于节能时效调整门限时上行/下行PRB利用率门限及调整上限(根据现场实际负荷情况而定)和基于节能时效调整门限时RRC用户数门限及调整上限(根据现场实际负荷情况而定)。
基于上述节能预测任务参数以及部署场景中的业务覆盖场景和业务时间特性，再结合节假日因子对预测指标的影响，采用基于时间窗口的子序列拆分预测算法，可以得到计算性能最优、优化效果最好的预测模型，即负荷调整策略信息。其中基于时间窗口的子序列拆分预测算法可以根据周内同天的方式拆分预测，从而得到以周为循环周期、以天为时间粒度的预测结果；当然，除了周内同天的方式外，还可以采用月内同天、每天同一时间段等方式，在此不一一举例。
由于目标小区的业务情况和配置情况并非一成不变，因此在得到负荷调整策略信息后，需要根据基站 优化后的实际情况反馈优化负荷调整策略信息，从而不断迭代调优，得到贴合于目标小区的情况的负荷调整策略信息。参照图6，本申请实施例将负荷调整策略信息发送到目标小区对应的基站后，还包括以下步骤：
步骤S410，获取由基站根据负荷调整策略信息调整后所得到的配置管理CM数据和性能管理PM数据；
步骤S420，根据调整后的CM数据和PM数据对负荷调整策略信息中的目标节能门限阈值进行迭代调优，将迭代调优后的目标节能门限阈值作为初始节能门限阈值。
基站节能系统在基于初始节能门限阈值得到负荷调整策略信息并不一定完全适应于目标小区(例如基于人为设定的初始门限值而首次计算得到的负荷调整策略信息)，那么基站在根据负荷调整策略信息调整后，将结果反馈到基站节能系统，基站节能系统根据全场景话务模型，节能效果和KPI趋势的大数据分析，强化自学习，在线不断迭代优化。例如，每小区每日提取性能数据，利用聚类算法寻求不同门限参数最优调整步长，每日KPI基线优化刷新后监控网络核心KPI(包括建立类、掉话类、切换类、用户体验类等)，在允许浮动范围内，不断迭代预测模型，最终达到节能和系统性能的最佳平衡点。
在一些情况下，对用户所在的小区簇进行调整可以跨小区进行，例如将目标小区的用于迁移至相邻小区，这时负荷调整策略信息需要考虑邻区的情况，从而使得基站在根据负荷调整策略信息调整用户时可以将用户迁移到邻区。参照图7，步骤S200中根据历史数据和初始节能门限阈值确定负荷调整策略信息的过程中还可以包括以下步骤：
步骤S510，通过OMC的后台数据配置表获取目标小区与相邻小区的关系信息；
步骤S520，根据根据历史数据、初始节能门限阈值和关系信息调整负荷调整策略信息。
基站节能系统从OMC(Operation&Maintenance Center，操作维护中心)后台数据配置表中获取目标小区信息和邻区关系信息，包括频点信息、PCI(Physical Cell Identifie，物理小区标识)信息和邻区级的基本配置比如覆盖关系、CIO(Cell Individal Offset，小区偏置)等信息，用于分析符号关断、通道关断、载波关断、深度休眠等节能策略如何与相应的配置信息做匹配，提高负荷预测的准确率。
参照图8，本申请实施例还提供了一种基站节能方法，应用于基站，基站与基站节能系统连接，基站节能方法包括但不限于以下步骤S600、步骤S700和步骤S800。
步骤S600，接收由基站节能系统发送的负荷调整策略信息，负荷调整策略信息由基站节能系统根据目标小区的历史数据和初始节能门限阈值得到，历史数据为与目标小区相关的历史业务数据和网络配置数据，负荷调整策略信息包括基于时间窗口的目标节能门限阈值；
步骤S700，确定负荷调整策略信息在当前时间窗口对应的目标节能门限阈值；
步骤S800，根据当前时间窗口对应的目标节能门限阈值调整目标小区中用户所在的小区簇。
基站通过网络等方式接收基站节能系统发送的负荷调整策略信息，当到达负荷调整策略信息的时间窗口，则根据负荷调整策略信息在当前时间窗口对应的目标节能门限阈值，以及目标小区中用户的分布情况，确定如何调整用户所在的小区簇。参照图9，基站在接收到负荷调整策略信息并到达相应的时间窗口后，执行以下步骤：
步骤S810，确定目标小区的各个小区簇的用户分布信息；
步骤S820，根据在当前时间窗口基站的网络效能和对应的目标节能门限阈值确定用户分布调整方式；
步骤S830，根据用户分布调整方式选择目标小区中的第一用户并将第一用户从源小区簇迁移到目标小区簇。
可以理解的是，在当前时间窗口基站的网络效能由下式得到：
Figure PCTCN2022091509-appb-000001
其中Ee是Energy efficiency，为网络能效；v是基站秒级业务量，单位是bits/s；En是基站在提供秒级业务量前提下所耗费的功率。上述步骤基于各小区簇分别判断话务负荷，当所有小区簇都有少量用户分布时，各小区簇只能启动符号关断或通道关断，此时将所有用户都集中到部分小区簇，其他小区簇就可以启动载波关断或者设备深度休眠，关闭更多的网络硬件，以最大化节能效果。上述步骤针对但不局限于多制式多小区簇组网的场景，可以分为三个子步骤：频层选择、用户选择和用户迁移。
其中，频层选择，实际上对小区簇进行选择，需要考虑源小区簇与目标小区簇间的约束与协作关系，这些关系包括：
(1)源小区簇和目标小区簇之间存在覆盖重叠区域；
(2)源小区簇中存在可以迁移到目标小区簇的用户；
(3)源小区簇通过迁移用户，可以启动更多的基础节能功能；
(4)目标小区簇不会因为用户迁移而关闭已经启动的基础节能功能。
其中，用户选择，是在已经明确源小区簇和目标小区簇的前提下进行，需要确保目标小区簇的用户业 务体验不受影响，需要考虑的主要因素包括：
(1)用户所有业务能够由目标小区簇承载：综合终端能力，核心网配置的移动性限制，网络切片等因素，只有用户所有的业务都能够由目标小区簇承载时，才能作为迁移的备选用户；
(2)用户所有业务体验不受影响：不同小区簇的无线能力不同，对应的不同业务在不同小区簇之间的体验不同，基于能效迁移用户的过程中不需要保证业务体验完全不变，但是要保证用户迁移到目标小区簇后，业务体验满足预期要求。
其中，用户迁移，分为连接态用户迁移及空闲态用户迁移：
(1)针对连接态用户迁移：
根据目标小区簇的不同，选择系统内异频切换或异系统切换。部分场景受限于终端能力或组网场景，采用异频重定向或异系统重定向。
(2)针对空闲态用户分布调整：
对于已经迁移到目标小区簇的用户，为避免通过小区重选再次驻留并接入源小区簇，都应将源小区簇小区重选专用优先级设置为最低。
通过上述步骤方法，基站基于负荷调整策略信息对用户进行调整，以提升网络能效为目标，使多制式、多频层间的网络协同工作，完成话务负荷和网络容量的匹配，协同调整源小区簇和目标小区簇网络容量，实现基站整体的能耗降低和能效提升。
本申请实施例还提供了一种基站节能系统，参照图10，基站节能系统包括：
网络管理服务器，连接基站以获取目标小区的历史数据，历史数据为与目标小区相关的历史业务数据和网络配置数据；
集中式自组织网络系统CSON，被设置为获取目标小区的历史数据并提交负荷调整策略信息计算请求；
智能计算系统ICS，被设置为接收CSON发送的负荷调整策略信息计算请求和目标小区的历史数据，还被设置为根据历史数据和初始节能门限阈值确定负荷调整策略信息并将负荷调整策略信息发送到CSON，以使CSON将负荷调整策略信息通过网络管理服务器下发到基站。
在网络拓扑上看，基站连接网络管理服务器，网络管理服务器、CSON和ICS依次连接；其中，基站在一定的无线电覆盖区域中，通过移动通信交换中心，与移动终端之间进行信息传递的无线电收发信电台，还向网络管理服务器上报KPI信息、UE(User Equipment，用户终端)状态信息、信道状态信息等，并执行相关的节能操作。其中，相关的节能操作包括导航模式和节能模式两个方面，导航模式是指上述负荷调整策略信息，在模式化后可以包括正常模式、普通模式和经济模式，正常模式是不执行上述负荷调整策略信息(不在导航时间窗口内或者基站本身是基础覆盖小区)，普通模式是对小流量业务执行导航模式，经济模式是强制对所有业务执行导航模式；节能方式如前，包括符号关断、通道关断、载波关断和设备深度休眠等，通过在传统的节能方式上叠加本申请实施例的导航模式，实现基站性能和节能的平衡。
网络管理服务器用于定时更新CM数据和PM数据等历史数据，通过北向公共进程(北向接口)推送给CSON，并每隔一段时间更新并下发给基站执行负荷调整策略信息。
CSON以在网用户的通话感受为出发点，全面收集网络数据并进行精确智能定位和分析，并将相应的节能计算任务发送到ICS，并将ICS计算完成的负荷调整策略信息下发到网络管理服务器，帮助无线运营商实现网络规划、配置和优化过程的自动化，减轻对人工的需求。
ICS是构建在Spark计算平台之上，一站式的算法与智能应用开发平台，可帮助用户快速进行业务应用的算法开发，完成算法评估和发布。例如，在本申请实施例中，通过智能计算系统ICS对历史数据和初始节能门限阈值进行人工智能计算，得到负荷调整策略信息。
参照图11，根据上述的网络拓扑和设备对应的功能，可以将按模块划分为节能策略模块、节能评估模块、节能执行模块和业务导航模块。
节能策略模块，被设置为基于站点配置、用户特征、话务规律、覆盖特性、能效曲线等历史数据，自动配置各小区的节能策略及参数，并下发给业务导航模块和节能执行模块；同时在KPI允许的波动范围为导向，智能迭代优化节能模式。
节能评估模块，被设置为对基站业务波动、MR(Measurement Result，测量报告)数据变化、节能增益、KPI的实时监控及可视化。
节能执行模块，被设置为执行基础节能功能，在节能策略模块输出的允许时间段内自主开启节能功能，包括符号(时隙)关断、通道关断、载波关断和设备深度休眠等。
业务导航模块，被设置为区分导航的制式类型，包括制式内、制式间、业务识别、ARP(Allocation and Retention Priority，分配和保留优先级)、业务分发(经济模式全部业务导航或者普通模式小流量业务导航)。
总之，本申请实施例通过采集目标小区的历史数据，进行小区业务模型分类、KPI评估，然后采用智能预测算法生成小区负荷预测模型(负荷调整策略信息)并定时下发，目标小区在时间窗口内执行节能调 整任务，周期性选择用户，将用户迁移至期望目标邻区，并在节能时间段判决进行节能初始门限动态调优，以提高节能效果。本申请实施例通过业务导航的方式，引导用户行为，调整负荷分布，从而创造出更多满足传统节能条件的小区和节能时间窗口。
下面以实际示例对本申请实施例的基站节能方法进行说明：
参照图10至图12，基于基站-网络管理服务器-CSON-ICS的网络拓扑执行如下基站节能方法：
步骤S101，创建节能任务和获取参数配置；
首先确定执行节能任务的目标小区，为避免出现集中式网元性能受限，在一次节能任务中无法对较多小区进行预测，或者对于部分小区(比如基础覆盖小区)不需要开启节能，可以通过人工确定执行节能任务的目标小区的范围。一般包含两种方式：按照频点进行选择，按照站点进行选择。
然后进行节能预测任务参数设置，包括：
公共任务参数，包括节能时间预测功能开关、调整评估迭代周期和评估迭代次数；
节能时间段预测算法参数，包括历史数据的采集时长，预测的负荷因素，负荷预测模型更新时间，符号/载波/通道关断和深度休眠的负荷门限；
门限调整参数，包括基于节能时效调整门限时上行/下行PRB利用率门限及调整上限(根据现场实际负荷情况而定)和基于节能时效调整门限时RRC用户数门限及调整上限(根据现场实际负荷情况而定)。
上述如果通过频点选择目标小区，那么需要设置频点的节能关断优先级，取值越大对应节能关断优先级越高，基础覆盖小区优先级固定为0；如果通过站点选择目标小区，则不需要设置频点的节能关断优先级。
步骤S102，采集历史数据；
历史数据来源于网管推送的KPI数据，本申请所涉及的4G/5G协同节能策略的生成，评估负荷需要采集但不局限于如下KPI：
NR侧：小区上行/下行PRB平均占用率、载波上行/下行PRB时频资源利用率、RRC连接建立用户数、小区上行/下行PDCP层平均吞吐量等；
LTE侧：上行/下行PRB平均利用率、RRC连接建立用户数、NSA终端签约5G的最大用户数、用户面PDCP SDU下行数据量、上行/下行业务平均速率。
另一方面，若考虑将用户迁移到邻区中去，那么还从后台数据配置表中获取目标小区和邻区的关系，包括频点信息、PCI信息和邻区级的基本配置比如覆盖关系、CIO等信息。
对于CM数据来说，获取方式包括以下步骤：
CSON主动向网络管理服务器系统发送请求，获取CM数据；
网络管理服务器系统将CM配置文件返回给CSON。
对于PM数据来说，获取方式包括以下步骤：
网络管理服务器系统上导入性能模板并创建性能数据定时导出任务，把PM数据定时推送到SFTP服务器的指定目录；
CSON周期性从该SFTP服务器的指定目录抓取PM数据。
步骤S103，数据预处理；
先进行业务模型的分类；基于现网话务应用特点，在不同时间段中业务模型存在明显差异，场景包括“商务区、居民区”；同时，“工作日、周末”的话务模型也存在差异，可按照周内同天子序列拆分预测算法，分别进行后续建模和计算。
步骤S104，节能时间段和导航时间段负荷预测建模；
此步骤包括负荷评估因素，目前基站节能方案中采用的负荷判决因素包括上行PRB利用率，下行PRB利用率和RRC用户数。其中对于载波关断和业务导航，这三种负荷因素均可能用到；对于通道关断会用到下行PRB利用率和RRC用户数，对于符号关断目前仅采用下行PRB利用率。在负荷预测中负荷评估因素可不完全和前台的负荷因素保持一致，还可以考虑额外的负荷因素，比如吞吐量、吞吐率等。
另外，采用预测准确度高的模型，输入数据为小区级(逻辑小区)KPI数据，输出数据为小区级多个时间窗口(默认为1天)的负荷预测数据。话务预测在建模时根据历史数据，区分出正效应、负效应以及无效应三类小区，利用二阶指数平滑预测算法，得到计算性能最优、效果最好的预测模型。
此步骤中CSON通过ICS计算节能方案，CSON和ICS的交互过程如下：
CSON向ICS发送节能策略计算请求，包含计算节能方案的各个参数，比如各种门限参数，触发ICS计算节能方案；
CSON定时向ICS发送进度查询请求，询问ICS本次节能策略是否计算结束；
ICS计算结束后接收到CSON发送的进度查询请求，把计算结果发送到CSON。
其中，通过ICS的IP地址，每次节能策略计算请求执行时CSON可以自动执行该过程。
步骤S105，NR小区节能或业务导航判决；
此步骤基于各小区簇分别判断话务负荷，参照图13，假设源小区簇为L1，目标小区簇为L2，L1属于新制式部署小区簇，L2属于大规模商用小区簇。按照网络能效的定义，若对于数据传输速率的表现，L1、L2的用户体验相同，L2能耗远高于L1，则显然L1的网络能效低于L2，那么将用户从L1迁移至L2。另外，本步骤不仅局限于相邻小区簇间的协作，可以延伸到跨小区簇的业务导航，假设邻区也是目标小区簇为L3，那么从L1将用户迁移出去的时候，可以选择L2和L3迁入。迁移用户同样包含了频层选择、用户选择和用户迁移三个子步骤，为避免重复，在此省略这三个子步骤的内容，详细可见上面的说明。
此步骤的执行过程中包括CSON与网络管理服务器的交互过程，包括以下步骤：
CSON向网络管理服务器发送请求，把计算所得的负荷调整策略信息发送到网络管理服务器；
网络管理服务器接收请求后开始同步负荷调整策略信息；
网络管理服务器同步结束后，给CSON返回同步结果。
步骤S106，数据库更新；
由于全网小区都有可能会输出更改参数的请求，因此在数据库收到参数变更请求后，需要及时更新参数数据：节能时间段的更新，根据收到的计算节能策略，直接进行数据库执行更新。
上述更新方式在数据库的自动模式下，可以根据请求直接执行，在数据库的受控模式下，则输出执行数据库更新确认请求，待人工确认后在执行。
当在更改参数的请求较多的情况下，数据库可以对小区分批次进行数据库的更新，以便新的节能策略生效。
步骤S107，数据呈现。
CSON可实时呈现功能相关KPI，目前主要以人工统计的形式呈现节能或导航运行的效果，即可以人工设置数据呈现的起始时间段，对于集中式网元输出的关断时间段做特殊标记，以进行节能前后、业务导航前后的性能对比。
本示例详细介绍了基站节能方法的步骤和基站节能系统各个子系统的功能及其交互过程，基站节能方法通过协调基站、网络管理服务器、集中式自组织网络系统、实验数据智能计算系统的功能，根据不同站点不同的网络配置和历史业务情况，智能定制站点间、频段间和制式间的协同节能调度策略，即负荷调整策略信息。ICS使用AI技术自学习历史话务和网络KPI，准确预测小区负荷趋势，自动规划未来节能策略方式，实现“一站一策”，最大化节能效果。
上述基站与基站节能系统之间的关系采用类似C/S(Client-Server，服务器-客户机)服务架构模式，可以分为服务器侧和网元侧，在服务器侧包含节能策略模块、节能评估模块，网元侧包含节能执行模块、业务导航模块。
服务器侧各个模块向网元侧的模块提供自动配置各小区的节能策略及参数、发给业务导航模块和节能执行模块，同时监测基站业务波动、MR数据变化、节能增益、KPI可视化等服务。
网元侧按照服务器侧模块提供的节能或导航策略，进行载波级基础节能和业务导航功能的执行，完成智能节能的任务步骤。
整个装置运行过程形成闭环，节能执行模块的结果会反馈给节能评估模块，使其对节能策略的执行效果及时感知，并调整下一周期策略生成的合理性，在保证用户体验和网络性能的前提下，预测网络负荷，评估业务需求，让网络需求匹配网络能耗，实现整网能效比最佳，每比特能耗最低。
本申请实施例的还提供了一种系统，包括至少一个处理器和用于与至少一个处理器通信连接的存储器；存储器存储有能够被至少一个处理器执行的指令，指令被至少一个处理器执行，以使至少一个处理器能够执行前述的基站节能方法。
参照图14，以系统1000中的控制处理器1001和存储器1002可以通过总线连接为例。存储器1002作为一种非暂态计算机可读存储介质，可用于存储非暂态软件程序以及非暂态性计算机可执行程序。此外，存储器1002可以包括高速随机存取存储器，还可以包括非暂态存储器，例如至少一个磁盘存储器、闪存器件、或其他非暂态固态存储器件。在一些实施方式中，存储器1002包括相对于控制处理器1001远程设置的存储器，这些远程存储器可以通过网络连接至系统1000。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
本领域技术人员可以理解，图14中示出的装置结构并不构成对系统1000的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
本申请实施例的还提供了一种计算机可读存储介质，该计算机可读存储介质存储有计算机可执行指令，该计算机可执行指令被一个或多个控制处理器执行，例如，被图14中的一个控制处理器1001执行，可使得上述一个或多个控制处理器执行上述方法实施例中的基站节能方法，例如，执行以上描述的图1中的方法步骤S100至步骤S300、图2中的方法步骤S110至步骤S130、图3中的方法步骤S210至步骤S230、图4中的方法步骤S221至步骤S222、图5中的方法步骤S231至步骤S232、图6中的方法步骤S410和步骤S420以及图7中的方法步骤S510和步骤S520。
本申请实施例的还提供了一种基站，包括至少一个处理器和用于与至少一个处理器通信连接的存储器；存储器存储有能够被至少一个处理器执行的指令，指令被至少一个处理器执行，以使至少一个处理器能够执行前述的基站节能方法。
参照图15，以基站2000中的控制处理器2001和存储器2002可以通过总线连接为例。存储器2002作为一种非暂态计算机可读存储介质，可用于存储非暂态软件程序以及非暂态性计算机可执行程序。此外，存储器2002可以包括高速随机存取存储器，还可以包括非暂态存储器，例如至少一个磁盘存储器、闪存器件、或其他非暂态固态存储器件。在一些实施方式中，存储器2002包括相对于控制处理器2001远程设置的存储器，这些远程存储器可以通过网络连接至基站2000。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
本领域技术人员可以理解，图15中示出的装置结构并不构成对基站2000的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
本申请实施例的还提供了一种计算机可读存储介质，该计算机可读存储介质存储有计算机可执行指令，该计算机可执行指令被一个或多个控制处理器执行，例如，被图15中的一个控制处理器2001执行，可使得上述一个或多个控制处理器执行上述方法实施例中的基站节能方法，例如，执行以上描述的图8中的方法步骤S600至步骤S800以及图9中的方法步骤S810和步骤S830。
本申请实施例提供的基站节能方法，至少具有如下有益效果：本申请实施例通过获取目标小区的历史数据，基于基站节能系统中的初始节能门限阈值和上述历史数据中的历史业务数据和对应目标小区的网络配置数据，进行大数据的业务预测而得到负荷调整策略信息，该负荷调整策略信息包含了目标小区在不同时间窗口内的目标节能门限阈值，该目标节能门限阈值考虑了不同时间窗口的业务情况和可能达到的节能效果，因此基站根据当前时间窗口对应的目标节能门限阈值进行用户迁移，将用户导航到能效比更高的网络承载，可以在保证业务情况基本不变的前提下，实现时间窗口的主动节能，提高整网效能。
以上所描述的装置实施例仅仅是示意性的，其中作为分离部件说明的单元可以是或者也可以不是物理上分开的，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。
本领域普通技术人员可以理解，上文中所公开方法中的全部或某些步骤、系统可以被实施为软件、固件、硬件及其适当的组合。某些物理组件或所有物理组件可以被实施为由处理器，如中央处理器、数字信号处理器或微处理器执行的软件，或者被实施为硬件，或者被实施为集成电路，如专用集成电路。这样的软件可以分布在计算机可读介质上，计算机可读介质可以包括计算机存储介质(或非暂时性介质)和通信介质(或暂时性介质)。如本领域普通技术人员公知的，术语计算机存储介质包括在用于存储信息(诸如计算机可读指令、数据结构、程序模块或其他数据)的任何方法或技术中实施的易失性和非易失性、可移除和不可移除介质。计算机存储介质包括但不限于RAM、ROM、EEPROM、闪存或其他存储器技术、CD-ROM、数字多功能盘(DVD)或其他光盘存储、磁盒、磁带、磁盘存储或其他磁存储装置、或者可以用于存储期望的信息并且可以被计算机访问的任何其他的介质。此外，本领域普通技术人员公知的是，通信介质通常包含计算机可读指令、数据结构、程序模块或者诸如载波或其他传输机制之类的调制数据信号中的其他数据，并且可包括任何信息递送介质。
以上是对本申请的实施例进行了说明，但本申请并不局限于上述实施方式，熟悉本领域的技术人员在不违背本申请精神的前提下还可作出种种的等同变形或替换，这些等同的变形或替换均包含在本申请权利要求所限定的范围内。

Claims (14)

  1. 一种基站节能方法，应用于基站节能系统，所述基站节能系统与基站连接，所述基站节能方法包括：
    获取目标小区的历史数据，所述历史数据为与所述目标小区相关的历史业务数据和网络配置数据；
    根据所述历史数据和初始节能门限阈值确定负荷调整策略信息，所述负荷调整策略信息包括基于时间窗口的目标节能门限阈值；以及
    将所述负荷调整策略信息发送到所述目标小区对应的基站，以使所述基站根据所述负荷调整策略信息确定在当前时间窗口对应的目标节能门限阈值，并根据当前时间窗口对应的目标节能门限阈值调整所述目标小区中用户所在的小区簇。
  2. 根据权利要求1所述的基站节能方法，其中，所述历史数据包括配置管理CM数据和/或性能管理PM数据；
    所述获取目标小区的历史数据，包括以下至少之一：
    向所述目标小区对应的基站发送第一请求，并获取由所述目标小区对应的基站根据所述第一请求发送的CM数据；或
    向安全文件传送协议SFTP服务器发送第二请求，并获取由所述SFTP服务器根据所述第二请求发送的PM数据。
  3. 根据权利要求1所述的基站节能方法，其中，所述根据所述历史数据和初始节能门限阈值确定负荷调整策略信息，包括：
    获取初始节能门限阈值，所述初始节能门限阈值由预设负荷调整策略信息得到或由预设负荷调整策略信息迭代调优得到；
    对所述历史数据进行预处理以确定所述目标小区在不同时间窗口内的部署场景；以及
    根据所述初始节能门限阈值、所述部署场景和基于时间窗口的子序列拆分预测算法得到负荷调整策略信息。
  4. 根据权利要求3所述的基站节能方法，其中，所述对所述历史数据进行预处理以确定所述目标小区在不同时间窗口内的部署场景，包括：
    根据所述历史数据确定所述目标小区的业务覆盖场景以及所述目标小区在不同时间窗口内的业务时间特性；
    根据所述业务覆盖场景和所述业务时间特性确定所述目标小区在不同时间窗口内的部署场景。
  5. 根据权利要求4所述的基站节能方法，其中，所述根据所述初始节能门限阈值、所述部署场景和基于时间窗口的子序列拆分预测算法得到负荷调整策略信息，包括：
    获取包括所述初始节能门限阈值的节能预测任务参数，所述节能预测任务参数为对所述目标小区执行节能预测任务的任务参数；
    以所述部署场景中的业务覆盖场景和业务时间特性为第一影响因子，以节假日分布特性为第二影响因子，以所述节能预测任务参数作为第三影响因子，根据所述第一影响因子、所述第二影响因子、所述第三影响因子和所述基于时间窗口的子序列拆分预测算法得到负荷调整策略信息。
  6. 根据权利要求1所述的基站节能方法，其中，将所述负荷调整策略信息发送到所述目标小区对应的基站后，所述基站节能方法还包括：
    获取由所述基站根据所述负荷调整策略信息调整后所得到的配置管理CM数据和性能管理PM数据；
    根据调整后的CM数据和PM数据对所述负荷调整策略信息中的目标节能门限阈值进行迭代调优，将迭代调优后的目标节能门限阈值作为初始节能门限阈值。
  7. 根据权利要求1所述的基站节能方法，其中，所述根据所述历史数据和初始节能门限阈值确定负荷调整策略信息，包括：
    通过操作维护中心的后台数据配置表获取所述目标小区与相邻小区的关系信息；
    根据所述历史数据、初始节能门限阈值和所述关系信息调整所述负荷调整策略信息。
  8. 根据权利要求1所述的基站节能方法，其中，所述根据所述历史数据和初始节能门限阈值确定负荷调整策略信息，包括：
    通过智能计算系统ICS对所述历史数据和初始节能门限阈值进行人工智能计算，得到负荷调整策略信息。
  9. 一种基站节能方法，应用于基站，所述基站与基站节能系统连接，所述基站节能方法包括：
    接收由所述基站节能系统发送的负荷调整策略信息，所述负荷调整策略信息由所述基站节能系统根据目标小区的历史数据和初始节能门限阈值得到，所述历史数据为与所述目标小区相关的历史业务数据和网 络配置数据，所述负荷调整策略信息包括基于时间窗口的目标节能门限阈值；
    确定所述负荷调整策略信息在当前时间窗口对应的目标节能门限阈值；以及
    根据当前时间窗口对应的目标节能门限阈值调整所述目标小区中用户所在的小区簇。
  10. 根据权利要求9所述的基站节能方法，其中，所述根据当前时间窗口对应的目标节能门限阈值调整所述目标小区中用户所在的小区簇，包括：
    确定所述目标小区的各个小区簇的用户分布信息；
    根据在当前时间窗口所述基站的网络效能和对应的目标节能门限阈值确定用户分布调整方式；以及
    根据所述用户分布调整方式选择所述目标小区中的第一用户，并将所述第一用户从源小区簇迁移到目标小区簇。
  11. 一种基站节能系统，包括：
    网络管理服务器，连接基站以获取目标小区的历史数据，所述历史数据为与所述目标小区相关的历史业务数据和网络配置数据；
    集中式自组织网络系统CSON，被设置为获取所述目标小区的历史数据并提交负荷调整策略信息计算请求；以及
    智能计算系统ICS，被设置为接收所述CSON发送的负荷调整策略信息计算请求和所述目标小区的历史数据，还被设置为根据所述历史数据和初始节能门限阈值确定负荷调整策略信息并将所述负荷调整策略信息发送到所述CSON，以使所述CSON将所述负荷调整策略信息通过所述网络管理服务器下发到基站。
  12. 一种系统，包括至少一个处理器和用于与所述至少一个处理器通信连接的存储器；所述存储器存储有能够被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够执行如权利要求1至8中任意一项所述的基站节能方法。
  13. 一种基站，包括至少一个处理器和用于与所述至少一个处理器通信连接的存储器；所述存储器存储有能够被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够执行如权利要求9至10中任意一项所述的基站节能方法。
  14. 一种计算机可读存储介质，所述计算机可读存储介质存储有计算机可执行指令，所述计算机可执行指令用于使计算机执行如权利要求1至8中任意一项所述的基站节能方法或执行如权利要求9至10中任意一项所述的基站节能方法。
PCT/CN2022/091509 2021-06-07 2022-05-07 基站节能方法、基站节能系统、基站及存储介质 WO2022257670A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110633421.8 2021-06-07
CN202110633421.8A CN115515208A (zh) 2021-06-07 2021-06-07 基站节能方法、基站节能系统、基站及存储介质

Publications (1)

Publication Number Publication Date
WO2022257670A1 true WO2022257670A1 (zh) 2022-12-15

Family

ID=84425459

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/091509 WO2022257670A1 (zh) 2021-06-07 2022-05-07 基站节能方法、基站节能系统、基站及存储介质

Country Status (2)

Country Link
CN (1) CN115515208A (zh)
WO (1) WO2022257670A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200022077A1 (en) * 2018-07-12 2020-01-16 P. I. Works TR Bilisim Hizm. San. ve Tic A.S. Energy-Saving Mobile Communication Network
CN111988836A (zh) * 2019-05-22 2020-11-24 中兴通讯股份有限公司 一种智能节能的方法、基站及计算机可读存储介质
CN112566226A (zh) * 2020-12-16 2021-03-26 北京电信规划设计院有限公司 5g基站智能化节能方法
CN112836911A (zh) * 2019-11-25 2021-05-25 中兴通讯股份有限公司 小区节能参数的确定方法、装置、电子设备及存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200022077A1 (en) * 2018-07-12 2020-01-16 P. I. Works TR Bilisim Hizm. San. ve Tic A.S. Energy-Saving Mobile Communication Network
CN111988836A (zh) * 2019-05-22 2020-11-24 中兴通讯股份有限公司 一种智能节能的方法、基站及计算机可读存储介质
CN112836911A (zh) * 2019-11-25 2021-05-25 中兴通讯股份有限公司 小区节能参数的确定方法、装置、电子设备及存储介质
CN112566226A (zh) * 2020-12-16 2021-03-26 北京电信规划设计院有限公司 5g基站智能化节能方法

Also Published As

Publication number Publication date
CN115515208A (zh) 2022-12-23

Similar Documents

Publication Publication Date Title
Rao et al. Optimal caching placement for D2D assisted wireless caching networks
Azari et al. On the latency-energy performance of NB-IoT systems in providing wide-area IoT connectivity
CN110139289B (zh) 一种调度方法及调度系统
WO2021209024A1 (zh) 节能方法、基站、控制单元及存储介质
CN111417180B (zh) 一种协同基站节能管理的网络调度方法及装置
CN105357692A (zh) 一种多网协同的网络优化与节能方法和系统
CN112804739A (zh) 一种基站节能方法、装置、设备及系统、存储介质
CN110149680A (zh) 省电处理方法、服务器、基站及终端设备
CN103249006A (zh) 一种基于多播的网络数据预推送方法
US20220247634A1 (en) Information processing method, apparatus, device and computer readable storage medium
Mi et al. Software-defined green 5G system for big data
CN110278603A (zh) 一种移动终端动态功耗调整的方法
CN105682124B (zh) 一种基于虚拟网络的节能方法
Xu et al. An improved communication resource allocation strategy for wireless networks based on deep reinforcement learning
CN103369640B (zh) 基站节电方法及装置
Chen et al. Device to device networks with cache-enabled and self-sustained mobile helpers
WO2022257670A1 (zh) 基站节能方法、基站节能系统、基站及存储介质
CN113238814A (zh) 基于多用户和分类任务的mec任务卸载系统及优化方法
CN105007594A (zh) Lte-a异构网络中一种联合优化mlb与mro的方法
Tang et al. A distance‐sensitive distributed repulsive sleeping approach for dependable coverage in heterogeneous cellular networks
CN116209046A (zh) 一种实时节能的移动通信方法、装置、网络侧设备及介质
CN103517441A (zh) 一种小区间干扰协调的方法及系统
WO2022127386A1 (zh) 状态迁移的方法、网络设备及存储介质
WO2021208877A1 (zh) 一种网络性能数据的监控方法及相关设备
CN115243349A (zh) 基站节能方法、装置、电子设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22819270

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE