Base station capacity expansion license allocation method and system, electronic equipment and storage medium

Abstract

The invention provides a base station capacity expansion license allocation method and system, electronic equipment and a storage medium, and aims to solve the problem that maximum base station capacity expansion cannot be realized, the method comprises the following steps: obtaining historical service data of all base stations, including access user information and digital modulation mode related parameters; obtaining user feature information and service information on a base station according to the access user information; calculating the flow value per unit of the base station; load prediction is carried out on each base station, whether the base station needs to be expanded is judged according to a prediction result, and the total number of RBs actually allocated to users by the base station needing to be expanded is estimated; according to the relevant parameters of the digital modulation mode of the base station, estimating an increasable predicted flow value of the total number of RBs actually allocated to the user for use; calculating the value corresponding to the increased predicted flow value of each base station; and determining an allocation scheme of the capacity expansion license according to the value corresponding to the increased predicted flow value of each base station. The expansion value can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of communication technology, and in particular to a base station expansion license allocation method, a base station expansion license allocation system, an electronic device and a computer-readable storage medium. Background Art

[0002] With the development of mobile communication technology and the rapid growth of the number of users, the load pressure on base stations continues to increase. This is mainly because users' demand for the network has shifted from simple voice communication to high-traffic, high-speed applications such as video streaming, online games, and virtual reality, which has caused the traffic carried by the network to rise sharply. At the same time, users' expectations for Internet access speeds are getting higher and higher, and long-term network delays and freezes are no longer acceptable. Under high load, base stations are difficult to meet users' demand for high-speed and stable connections, which poses a more severe challenge to network performance. There is an urgent need for capacity expansion and optimization to cope with the growing network pressure and user experience requirements. In order to meet this challenge, network optimization has become the key. Network optimization includes but is not limited to load balancing, cell breathing, frequency optimization and other technical means, aiming to improve the overall performance of the network and user experience.

[0003] At present, the most common method is to expand the capacity of base stations, and capacity expansion requires the allocation of expansion licenses, which are usually purchased by operators from base station hardware manufacturers such as ZTE and Huawei. There are two main methods for allocating expansion licenses: (1) The allocation of expansion licenses is mainly determined based on the load level of the base station. Base stations with higher loads are given priority in obtaining expansion resources; (2) Based on the type of scene where the base station is located (such as residential areas, business areas, industrial areas, etc.), expansion licenses are allocated preferentially to areas with higher scene importance. However, both methods have shortcomings. The shortcoming of the first allocation method is that it only considers the load and does not comprehensively consider the improvement of the overall network value after the expansion. The shortcoming of the second allocation method is that the consideration dimension is too single and does not evaluate the specific improvement value of the expansion to various scenarios. Summary of the invention

[0004] In order to at least solve the problem in the prior art that the base station expansion does not comprehensively consider the improvement of the overall network value after the expansion, and does not evaluate the specific improvement value of the expansion for various scenarios. The present disclosure provides a base station expansion license allocation method, a base station expansion license allocation system, an electronic device, and a computer-readable storage medium, which can accurately realize the expansion value judgment based on the historical access data of the base station and other information, and prioritize the base stations to be expanded, thereby improving the expansion value.

[0005] In a first aspect, the present disclosure provides a method for allocating a base station expansion license,

[0006] Methods include:

[0007] Obtain historical service data of all base stations, including base station access user information and base station digital modulation mode related parameters;

[0008] According to the access user information of the base station, obtain the user characteristic information and the service information on the base station;

[0009] Calculate the value of each unit of traffic of the base station based on the user characteristic information and the service information on the base station;

[0010] Carry out load forecasting for each base station, determine whether the base station needs to be expanded based on the forecast results, and estimate the total number of RBs (Resource Blocks) actually allocated to users for use by the base stations that need to be expanded;

[0011] Estimate the expected traffic value that can be increased by the total number of RBs actually allocated to users based on the parameters related to the base station digital modulation method;

[0012] The value corresponding to the expected traffic value of each base station is calculated based on the per-unit traffic value of the base station and the expected traffic value that can be increased;

[0013] The allocation plan for expansion licenses is determined based on the value corresponding to the expected traffic growth of each base station.

[0014] Furthermore,

[0015] The user characteristic information includes: user importance and importance of the area where the base station is located;

[0016] The service information on the base station includes the traffic generated by the user connecting to the base station, the proportion of the user's access time to the base station, the average consumption of the user in a preset time period, and the proportion of the user's traffic peak time;

[0017] The per-unit flow value of the base station is calculated based on the user characteristic information and the service information on the base station by the following formula (1):

[0018]

[0019] Where P m is the value of each unit of traffic of base station m, N is the number of users accessing base station m in the corresponding time period, K n The traffic generated by user n connecting to base station m, T n is the proportion of time that user n accesses base station m, P n is the average consumption of user n in the preset time period, U n is the importance of user n, W m is the importance of the area where base station m is located, Y nis the proportion of user n’s traffic peak time, L m is the traffic generated during the traffic peak time of base station m.

[0020] Furthermore, the load forecasting for each base station includes:

[0021] Perform data collection, collect historical data and environmental data of the base station, the historical data includes: RB utilization rate, traffic and access user information of base station m, the access user information includes quantity, traffic, consumption price, user type; the environmental data includes: external event information; and mark the data according to the time period corresponding to the data, the corresponding time period includes weekdays, weekends, holidays, and intraday time periods;

[0022] Perform data preprocessing;

[0023] Perform feature engineering on the data, select time features, moving window features and environmental features;

[0024] Select a preset machine learning model;

[0025] Divide the processed data into training sets and test sets, and perform model training on the machine learning model;

[0026] Output the load forecast result of base station m in a certain time period in the future through the trained model;

[0027] The model prediction results are compared with the actual load, and the model recalibration is triggered when deviations are found.

[0028] Furthermore,

[0029] The determining whether the base station needs to be expanded according to the prediction result includes:

[0030] When the predicted PRB (Physical Resource Block) utilization rate and the number of access users of base station m in a specific time period in the future meet the conditions for base station expansion, it means that the base station needs to be expanded; if not, it means that the base station does not need to be expanded;

[0031] The conditions for the base station capacity expansion include that the PRB utilization rate of the base station in a specific time period is ≥ B%, and the number of access users is ≥ C, where B and C are preset values;

[0032] The estimating, for the base station that needs to be expanded, the total number of RBs actually allocated to the user by the base station that needs to be expanded includes:

[0033] After the expansion base station is determined, it is determined that after base station m is allocated and obtains the expansion license, base station m will have the original RB resource quantity R m The number of RB resources increased based on V max, PRB utilization rate of base station m F(A), number of RBs required by the user F(A)*R m ;

[0034] The total number of RBs actually allocated to users by the base station is calculated by the following formula (2): m ;

[0035]

[0036] Furthermore,

[0037] The base station digital modulation mode related parameters include the digital modulation mode used by the base station and the proportion of various digital modulation modes used by the base station;

[0038] The estimated flow value of the total number of RBs actually allocated to users for use estimated based on the parameters related to the base station digital modulation method includes:

[0039] The expected flow rate that can be increased is calculated by the following formula (3):

[0040]

[0041] Among them, L m+z is the estimated traffic value per unit time that can be increased by the total number of RBs actually allocated to users by the base station m to be expanded, y represents the yth digital modulation mode, z represents the maximum value of the digital modulation mode type, H y is the proportion of the yth digital modulation method used by base station m, R represents the amount of RE (Resource Element) resource data that can be scheduled in one time slot of each resource block, Ty is the amount of data that can be transmitted by each RE resource corresponding to the yth digital modulation method used by the base station, and A is a constant.

[0042] Furthermore,

[0043] The value corresponding to the expected traffic value of each base station increased is calculated according to the per-unit traffic value of the base station and the expected traffic value that can be increased, including:

[0044] The value corresponding to the expected traffic value of base station m is calculated by the following formula (4):

[0045] P m+ z = L m+ z * P m , (4)

[0046] Among them, P m+z is the value corresponding to the expected traffic increase of base station m, L m+zP is the total number of RBs actually allocated to users by base station m, and the predicted traffic value per unit time that can be increased. m is the value of each unit of traffic of base station m;

[0047] The allocation scheme of the expansion license is determined according to the value corresponding to the expected traffic value of each base station, including:

[0048] Press P m+z The base stations to be expanded are sorted from high to low in size, and expansion licenses are allocated preferentially to the base stations to be expanded that are ranked higher;

[0049] Expand the capacity of base stations that have obtained expansion licenses.

[0050] In a second aspect, the present disclosure provides a base station expansion license allocation system, the system comprising:

[0051] A first acquisition module, which is configured to acquire historical service data of all base stations, including access user information of the base stations and parameters related to the digital modulation mode of the base stations;

[0052] A second acquisition module, which is configured to acquire user feature information and service information on the base station according to the access user information of the base station;

[0053] A first calculation module, which is configured to calculate the value of each unit flow of the base station according to the user characteristic information and the service information on the base station;

[0054] A first estimation module is configured to perform load prediction for each base station, determine whether the base station needs to be expanded according to the prediction result, and estimate the total number of RBs actually allocated to users for the base station that needs to be expanded;

[0055] A second estimation module is configured to estimate an expected flow value that can be increased by the total number of RBs actually allocated to the user according to parameters related to the digital modulation mode of the base station;

[0056] A second calculation module is configured to calculate the value corresponding to the expected traffic value of each base station according to the per-unit traffic value of the base station and the expected traffic value that can be increased;

[0057] The determination module is configured to determine a distribution plan for the expansion license according to the value corresponding to the expected traffic value of each base station increase.

[0058] Furthermore,

[0059] The user characteristic information includes: user importance and importance of the area where the base station is located;

[0060] The service information on the base station includes the traffic generated by the user connecting to the base station, the proportion of the user's access time to the base station, the average consumption of the user in a preset time period, and the proportion of the user's traffic peak time;

[0061] The first calculation module is specifically configured to calculate the value of each unit flow of the base station by the following formula (1):

[0062]

[0063] Where P m is the value of each unit of traffic of base station m, N is the number of users accessing base station m in the corresponding time period, K n The traffic generated by user n connecting to base station m, T n is the proportion of time that user n accesses base station m, P n is the average consumption of user n in the preset time period, U n is the importance of user n, W m is the importance of the area where base station m is located, Y n is the proportion of user n’s traffic peak time, L m is the traffic generated during the traffic peak time of base station m.

[0064] Furthermore, the first estimation module includes a prediction unit, and the prediction unit is configured to:

[0065] Perform data collection, collect historical data and environmental data of the base station, the historical data includes: RB utilization rate, traffic and access user information of base station m, the access user information includes quantity, traffic, consumption price, user type; the environmental data includes: external event information; and mark the data according to the time period corresponding to the data, the corresponding time period includes weekdays, weekends, holidays, and intraday time periods;

[0066] Perform data preprocessing;

[0067] Perform feature engineering on the data, select time features, moving window features and environmental features;

[0068] Select a preset machine learning model;

[0069] Divide the processed data into training sets and test sets, and perform model training on the machine learning model;

[0070] Output the load forecast result of base station m in a certain period of time in the future through the trained model;

[0071] The model prediction results are compared with the actual load, and the model recalibration is triggered when deviations are found.

[0072] Furthermore,

[0073] The first estimation module further includes a judgment unit and an estimation unit, and the judgment unit is configured as:

[0074] When the PRB utilization rate and the number of access users of the base station m in a specific time period in the future predicted by the prediction unit both meet the conditions for base station capacity expansion, it means that the base station needs to be expanded; if not, it means that the base station does not need to be expanded;

[0075] The conditions for the base station capacity expansion include that the PRB utilization rate of the base station in a specific time period is ≥ B%, and the number of access users is ≥ C, where B and C are preset values;

[0076] The estimation unit is set as:

[0077] After the expansion base station is determined, it is determined that after base station m is allocated and obtains the expansion license, base station m will have the original RB resource quantity R m The number of RB resources increased based on V max , PRB utilization rate of base station m F(A), number of RBs required by the user F(A)*R m ;

[0078] The total number of RBs actually allocated to users by the base station is calculated by the following formula (2): m ;

[0079]

[0080] Furthermore,

[0081] The base station digital modulation mode related parameters include the digital modulation mode used by the base station and the proportion of various digital modulation modes used by the base station;

[0082] The second estimation module is specifically configured as follows:

[0083] The expected flow rate that can be increased is calculated by the following formula (3):

[0084]

[0085] Among them, L m+z is the estimated traffic value per unit time that can be increased by the total number of RBs actually allocated to users by the base station m to be expanded, y represents the yth digital modulation mode, z represents the maximum value of the digital modulation mode type, H y is the proportion of the yth digital modulation method used by base station m, R represents the amount of RE resource data that can be scheduled in one time slot of each resource block, Ty is the amount of data that can be transmitted by each RE resource corresponding to the yth digital modulation method used by the base station, and A is a constant.

[0086] Furthermore,

[0087] The second calculation module is specifically configured as follows:

[0088] The value corresponding to the expected traffic value of base station m is calculated by the following formula (4):

[0089] P m+ z = L m+ z * P m , (4)

[0090] Among them, P m+z is the value corresponding to the expected traffic increase of base station m, L m+z

[0091] P is the total number of RBs actually allocated to users by base station m, and the predicted traffic value per unit time that can be increased. m is the value of each unit flow of base station m;

[0092] The determination module is specifically configured as follows:

[0093] Press P m+z The base stations to be expanded are sorted from high to low in size, and expansion licenses are allocated preferentially to the base stations to be expanded that are ranked higher;

[0094] Expand the capacity of base stations that have obtained expansion licenses.

[0095] In a third aspect, the present disclosure provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the base station expansion license allocation method as described in any one of the first aspects.

[0096] In a fourth aspect, the present disclosure provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the base station expansion license allocation method described in any one of the above-mentioned first aspects is implemented.

[0097] Beneficial effects:

[0098] The base station expansion license allocation method, base station expansion license allocation system, electronic device and storage medium provided by the present disclosure obtain the historical load, access users and other data of all base stations; calculate the value corresponding to each 1Gb of traffic generated by the base station based on the user characteristic information of the base station access user information and the business information on the base station; estimate the load of the base station at a certain time in the future based on machine learning, define the base station with an estimated load exceeding the threshold as the base station to be expanded, and calculate the number of RBs actually allocated to users after the expansion; finally, convert the increased expected traffic based on the number of RBs, and then calculate the value corresponding to the increased expected traffic, and base stations with high value will be expanded first. Therefore, it is possible to achieve a more accurate judgment of the expansion value based on the historical access data and other information of the base station, prioritize the base stations to be expanded, and thus improve the expansion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0099] Figure 1 A flowchart of a method for allocating base station expansion licenses provided in Embodiment 1 of the present disclosure;

[0100] Figure 2 An architecture diagram of a base station expansion license allocation system provided in Embodiment 2 of the present disclosure;

[0101] Figure 3 This is an architecture diagram of an electronic device provided in Embodiment 3 of the present disclosure. DETAILED DESCRIPTION

[0102] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the present disclosure is further described in detail below in conjunction with the drawings and embodiments. It should be understood that the specific embodiments and drawings described herein are only used to explain the present invention, rather than to limit the present invention.

[0103] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence; and, in the absence of conflict, the embodiments in the present disclosure and the features in the embodiments can be arbitrarily combined with each other.

[0104] The terms used in the embodiments of the present disclosure are only for the purpose of describing specific embodiments and are not intended to limit the present disclosure. The singular forms of "a", "said" and "the" used in the embodiments of the present disclosure and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings.

[0105] In the subsequent description, the suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present disclosure, and have no specific meanings. Therefore, "module", "component" or "unit" can be used in a mixed manner.

[0106] At present, capacity expansion requires the allocation of capacity expansion licenses to expand base stations. There are two main ways to allocate capacity expansion license resources to high-load cells: (1) The allocation of capacity expansion licenses is mainly determined based on the load level of the base station. Base stations with higher loads are given priority in obtaining capacity expansion resources; (2) Based on the type of scene where the base station is located (such as residential areas, business areas, industrial areas, etc.), capacity expansion licenses are allocated preferentially to areas with higher scene importance.

[0107] However, there may be deviations in the value assessment of base station capacity expansion using these two methods, and they cannot achieve the maximum expansion value.

[0108] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned problems existing in the prior art are described in detail below with specific embodiments. It is understandable that in the embodiments of the present application, the execution subject can perform some or all of the steps in the embodiments of the present application, and these steps or operations are only examples. The embodiments of the present application can also perform other operations or deformations of various operations. In addition, each step can be performed in a different order presented in the embodiments of the present application, and it is possible not to perform all the operations in the embodiments of the present application. And, the following several specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0109] Figure 1 A flow chart of a method for allocating base station expansion licenses provided in the first embodiment of the present disclosure is shown in FIG. Figure 1 As shown, the method includes:

[0110] Step S101: Acquire historical service data of all base stations, including access user information of the base stations and parameters related to the digital modulation mode of the base stations;

[0111] Step S102: acquiring user feature information and service information on the base station according to the access user information of the base station;

[0112] Step S103: Calculate the value of each unit flow of the base station according to the user characteristic information and the service information on the base station;

[0113] Step S104: performing load prediction for each base station, determining whether the base station needs to be expanded based on the prediction result, and estimating the total number of resource blocks RBs actually allocated to users by the base station that needs to be expanded for use;

[0114] Step S105: estimating the expected flow value that can be increased by the total number of RBs actually allocated to the user according to the parameters related to the digital modulation mode of the base station;

[0115] Step S106: Calculate the value corresponding to the expected traffic value of each base station according to the per-unit traffic value of the base station and the expected traffic value that can be increased;

[0116] Step S107: determining a distribution plan for expansion licenses according to the value corresponding to the expected traffic value of each base station increase.

[0117] In order to allocate expansion licenses to base stations with the highest value improvement when allocating expansion licenses, the embodiments of the present disclosure provide a method for allocating expansion licenses, by calculating the value of each unit of traffic of the base station and the increased expected traffic after obtaining the license, thereby calculating the value corresponding to the increased expected traffic, and base stations with high value are expanded first.

[0118] Specifically, they include:

[0119] Obtain historical service data of all base stations, including the traffic generated by each base station, the access user information of the base station, and the parameters related to the digital modulation mode of the base station; the traffic unit is GB, and the data is collected on an hourly basis, covering different time periods and scenarios (such as peak and non-peak periods), to ensure the comprehensiveness of the data and facilitate subsequent data calculations based on time periods. Use a distributed database system to store and manage the collected data to support large-scale data processing and fast retrieval. Historical service data can be historical service data for certain time periods, such as time periods when users are active. Data for time periods when users are not active may not be collected, such as midnight, to reduce the amount of data.

[0120] Based on the access user information of the base station, obtain the user feature information and the service information on the base station, calculate the user traffic and package information, and use the traffic generated by the user on the base station, the user's corresponding average monthly consumption, and other data to calculate the value of each unit of traffic (e.g. 1Gb) generated by the base station;

[0121] Load forecasting is performed for each base station. Based on machine learning, the load of the base station at a certain time in the future is estimated. Base stations with estimated loads exceeding the threshold are defined as base stations to be expanded, and the subsequent load status of the base stations is determined. The total number of RBs actually allocated to users for the base stations to be expanded is estimated;

[0122] The expected traffic growth is calculated based on the number of RBs, and finally the value corresponding to the expected traffic growth is calculated. Base stations with high value are expanded first. This enables a more accurate judgment of the value of expansion. When there are multiple licenses, the base stations to be expanded can be prioritized, and the corresponding number of top-ranked base stations can be selected to allocate expansion licenses, thereby increasing the value of expansion and improving overall operational efficiency and service quality.

[0123] Furthermore,

[0124] The user characteristic information includes: user importance and importance of the area where the base station is located;

[0125] The service information on the base station includes the traffic generated by the user connecting to the base station, the proportion of the user's access time to the base station, the average consumption of the user in a preset time period, and the proportion of the user's traffic peak time;

[0126] The per-unit flow value of the base station is calculated based on the user characteristic information and the service information on the base station by the following formula (1):

[0127]

[0128] Where P m is the value of each unit of traffic of base station m, N is the number of users accessing base station m in the corresponding time period, K n The traffic generated by user n connecting to base station m, T n is the proportion of time that user n accesses base station m, P n is the average consumption of user n in the preset time period, U n is the importance of user n, W m is the importance of the area where base station m is located, Y n is the proportion of user n’s traffic peak time, L m is the traffic generated during the traffic peak time of base station m.

[0129] Get the traffic K generated by user n connecting to base station m among the N users accessing base station m in the corresponding time period n (GB), the proportion of time that user n accesses base station m, T n , the average consumption of user n in the preset time period P n (Yuan), User n Importance U n , the importance of the area where base station m is located W m , User n traffic peak time ratio Y n .

[0130] T n is defined as the duration of user n accessing base station m / the total access duration of N users accessing base station m. For example, in a certain time period, user n accesses base station m for 2 hours, and the total access duration of N users is 300 hours, then T n =2 / 300.

[0131] Next, the importance of U to users n , Regional Importance m , Traffic peak time ratio Y n Give a more specific explanation.

[0132] (1) User importance U n

[0133] User Importance n It is a weighted indicator to measure user levels and can be defined based on factors such as user type, business needs, and contribution to revenue. In base station expansion decisions, priority is given to allocating resources to base stations where high-value users gather to ensure the network experience of these users. The following is a specific definition and implementation example of user priority.

[0134] User Importance n It can be defined based on three factors: user type, business needs, and package level:

[0135] 1. User Type

[0136] VIP users: For example, enterprise users, who are more important. Usually, these users have higher requirements for network quality and lower tolerance for service interruptions.

[0137] Ordinary users: general consumers, with relatively stable network requirements and lower importance.

[0138] 2. Business requirements

[0139] Key business users: such as users who rely on the network for financial transactions, telemedicine, etc. These users require a stable network environment and are more important.

[0140] Entertainment business users: such as users who mainly use videos and games, are relatively less important.

[0141] 3. Package Level

[0142] Premium package users: users who enjoy higher bandwidth and faster Internet speeds, pay more and are more important.

[0143] Basic package users: users with lower package fees, less important.

[0144] In practice, different types of users can be assigned a specific importance value, starting with 1. Here is an example assignment:

[0145] Table 1: Example of user importance value allocation scheme

[0146] User Type Business requirements Package Level <![CDATA[Importance U n > VIP User (Enterprise) Key business users Premium Package 1.5 VIP User (Enterprise) Entertainment business users Basic Package 1.3 Ordinary user (individual) Key business users Premium Package 1.2 Ordinary user (individual) Entertainment business users Premium Package 1.1 Ordinary user (individual) Entertainment business users Basic Package 1.0

[0147] Assume that there are three types of users in a base station, and their importance is as follows:

[0148] User A: VIP enterprise user, critical business needs, using the premium package. By definition, the priority Un of user A is 1.5.

[0149] User B: ordinary user, using the basic package, mainly for entertainment services, with a priority of 1.0.

[0150] User C: A common user who uses the premium package for remote work, which is a critical business requirement. The priority level is Un 1.2.

[0151] (2) Regional importance W m

[0152] Regional Importance m It is an important indicator for evaluating the strategic value of the area where the base station is located, and is used to prioritize the allocation of resources to more strategic areas in capacity expansion decisions. Different areas (such as commercial areas, transportation hubs, residential areas, etc.) have different requirements and impacts on network quality, so their capacity expansion priorities should also be different.

[0153] Regional Importance m It can be defined based on the following main factors:

[0154] Economic activity level: The higher the level of economic activity in an area, the greater the demand for the network, such as commercial areas, financial centers, etc.

[0155] User density: Areas with high user density usually have greater demand for network resources, such as shopping malls, transportation hubs, etc.

[0156] Serving special groups: Some areas may need to serve important groups, such as government offices, educational institutions, hospitals, etc.

[0157] Strategic network coverage: Certain areas may be important nodes in the operator's network coverage, such as along highways, important transportation hubs, etc.

[0158] Network sensitivity: Certain areas require particularly high network stability, such as hospitals or emergency service agencies.

[0159] The importance weights W of different types of regions are given below m An example is shown in Table 2:

[0160] Table 2: Importance weights W of different types of regions m Example

[0161]

[0162] For example:

[0163] Scenario 1: Commercial area. Assume that a base station A is located in a busy commercial area, which includes banks, enterprises, shopping centers, etc. Since the economic activity level in this area is high, there are many users and they have high requirements for network stability and speed, the W in the area where the base station A is located is mSet the value to 1.5.

[0164] Scenario 2: Transportation hub. Base station B is located in a large transportation hub, such as a train station or airport. Users in such areas are highly mobile, and high-quality network services need to be provided to a large number of users, especially during peak hours. m Set the value to 1.4.

[0165] Scenario 3: Ordinary residential area. Base station C is located in an ordinary residential area. Although the network demand is relatively stable, the overall user requirements for network quality are not as high as those in commercial areas and transportation hubs. Therefore, the W m Set the value to 1.0.

[0166] Scenario 4: Hospital. Base station D is located near the hospital. Considering the high stability and reliability requirements of the hospital’s network, especially in emergency situations, network failures may directly affect patient care and medical services. m Set the value to 1.6.

[0167] User n's traffic peak time ratio Y n

[0168] User n traffic peak time ratio Y n It is an important indicator for evaluating the proportion of user traffic during high-load periods of the network. It reflects the consumption of resources by users when network resources are tight (peak period), and helps operators determine which base stations put more pressure on the network during critical periods, so that the needs of these base stations can be given priority when expanding capacity.

[0169] User n traffic peak time ratio Y n It is the ratio of the traffic generated by users during the high-load period of the base station to their total traffic. The high-load period usually refers to the period when the network usage reaches or approaches full load (such as the working period during the day, the entertainment period at night, etc.). The specific time can be set according to the actual situation.

[0170] The calculation formula is as follows: n = Traffic generated by user n during peak hours / total traffic generated by user n

[0171] Calculate Y n Specific steps:

[0172] Step 1: Monitor the traffic usage of user n at base station m and record user traffic data by time period.

[0173] Step 2: Identify the peak hours of the base station and count the traffic generated by users during these hours.

[0174] Step 3: Calculate the proportion of the user's traffic generated during the peak period to the total traffic, and obtain the user's traffic peak time proportion Yn .

[0175] Definition of peak hours:

[0176] The peak hours should be defined according to the characteristics of the area where the base station is located. The following are the definitions of peak hours for several typical area types:

[0177] Commercial areas: The peak of network usage in commercial areas usually occurs during working hours, especially during the daytime working hours. Peak period: 9:00-18:00 (working days), because commercial users and corporate users use the network more intensively during this period.

[0178] Transportation hubs: The peak user traffic of transportation hubs (such as railway stations and airports) is usually concentrated in the morning and evening peak hours, and the demand for network resources is relatively large. Peak hours: 7:00-9:00, 17:00-19:00 (peak travel hours), during which time periods, user mobility is strong and network load is high.

[0179] Residential areas: The network peak hours in residential areas are mostly concentrated in the evening after get off work, mainly for home users to use the network for entertainment, social activities, etc. Peak period: 18:00-22:00, during which users watch videos, play games and socialize, and the network demand is relatively high.

[0180] Shopping areas / commercial shopping centers: The peak network hours in shopping areas and large shopping centers usually occur in the afternoons and evenings on weekends and holidays, when there are more customers. Peak hours: 14:00-20:00 (weekends / holidays), shoppers and customers use the network more frequently during these hours.

[0181] Hospitals / Emergency Services: Hospitals and other places have a steady demand on the network, but there may be instantaneous peak demand during emergencies or high-load medical services. Peak hours: 24 / 7, especially during emergencies or medical emergencies.

[0182] School / University: The peak hours for educational institutions usually occur during classroom teaching and extracurricular activities. Peak hours: 8:00-16:00 (weekdays), when teachers and students use the network for teaching and learning.

[0183] For example

[0184] Example 1: Commercial Area Base Station

[0185] Base station A is located in a busy commercial area, and the peak hours are defined as 9:00-18:00. The greater the proportion of network traffic used by users during these hours, the higher the Y n The higher the value, the greater the network consumption of the user during high-load periods.

[0186] A user is in a commercial area, and the traffic distribution is as follows:

[0187] Peak traffic: 20Gbit

[0188] Total traffic: 25Gbit

[0189] Then the user's traffic peak time accounts for Y n For: Y n =20 / 25=0.8, users occupy 80% of the traffic during peak hours, so the consumption of network resources is relatively large.

[0190] Example 2: Residential Base Station

[0191] Base station B is located in an ordinary residential area, with a peak period of 18:00-22:00. User B's traffic distribution is as follows:

[0192] Peak traffic: 5Gbit

[0193] Total traffic: 10Gbit

[0194] Then user B's traffic peak time accounts for Y n For: Y n =5 / 10=0.5, which indicates that 50% of user B's traffic is generated during the evening peak hours in residential areas.

[0195] The preset time period can be set according to actual conditions, such as 3, 6 months, 12 months, etc., preferably 6 months, to obtain the average consumption P of user n in the last 6 months. n (Yuan).

[0196] For example, assume that the historical service data of all base stations within 24 hours on September 12, 2024 is obtained, where the relevant information of base station m is as shown in Table 3:

[0197] Table 3: User feature information example

[0198] User <![CDATA[K n (GB)]]> <![CDATA[T n ]]> <![CDATA[P n (Yuan)]]> <![CDATA[U n ]]> <![CDATA[W m ]]> <![CDATA[Y n ]]> 1 3 0.003 128 1.5 1.4 0.4 2 4 0.0032 128 1.3 1.4 0.5 3 6 0.006 168 1.2 1.4 0.7 …… …… …… …… …… …… …… N 7 0.0061 98 1.1 1.4 0.3

[0199] After obtaining the above data, the value of each unit of traffic during the traffic peak time of base station m can be calculated. Based on the data obtained in the above steps, the value corresponding to each 1Gbit of traffic generated by base station m is calculated as P m , the specific formula is as follows, the calculated P m The values ​​are stored in a database to support subsequent calculations.

[0200] Traffic value of user n = K n *P n *U n *T n *Y n;

[0201]

[0202] Considering the importance of the area, the value of 1GB of traffic at base station m is:

[0203]

[0204] Among them, L m It is the traffic generated during the peak traffic time of the base station.

[0205] Formula (1) comprehensively considers traffic, economic contribution, duration, user importance, peak time proportion, and the strategic importance of the region. m With the calculation, operators can prioritize the allocation of expansion resources for base stations with high contribution value and high demand, ensure the rational allocation and use of network resources, and improve the overall network performance and user experience.

[0206] Furthermore, the load forecasting for each base station includes:

[0207] Perform data collection, collect historical data and environmental data of the base station, the historical data includes: RB utilization rate, traffic and access user information of base station m, the access user information includes quantity, traffic, consumption price, user type; the environmental data includes: external event information; and mark the data according to the time period corresponding to the data, the corresponding time period includes weekdays, weekends, holidays, and intraday time periods;

[0208] Perform data preprocessing;

[0209] Perform feature engineering on the data, select time features, moving window features and environmental features;

[0210] Select a preset machine learning model;

[0211] Divide the processed data into training sets and test sets, and perform model training on the machine learning model;

[0212] Output the load forecast result of base station m in a certain time period in the future through the trained model;

[0213] The model prediction results are compared with the actual load, and the model recalibration is triggered when deviations are found.

[0214] Load forecasting for each base station can provide a scientific basis for capacity expansion decisions. For example, when it is predicted that the load of a base station will exceed its capacity in the next few hours, capacity expansion measures can be taken in advance to ensure network stability and user experience.

[0215] The specific process of load forecasting includes:

[0216] Step 1: Data Collection

[0217] Load usually refers to the PRB utilization of the base station and the number of access users. The first task of load forecasting is to collect a large amount of historical data. The data required for the forecasting model usually includes the following categories:

[0218] Historical data: including resource block (RB) utilization rate of base station m, traffic (GB), access user information such as quantity, traffic, consumption price, user type, etc.

[0219] Time factor: The time period corresponding to the data, such as weekdays, weekends, holidays, time periods, etc. It can help the model distinguish the load differences in different time periods.

[0220] Environmental data: external events (such as large events, weather, etc.).

[0221] Step 2: Data Preprocessing

[0222] Clean, standardize, and process the data so that the model can better understand and use the information.

[0223] Missing value processing: For missing data, use the mean or interpolation method to fill it in to ensure that the model is not affected by missing data.

[0224] Normalization: Normalize various types of data (such as RB utilization and number of users) to a similar range, usually between 0 and 1, to make the model training process more stable.

[0225] Step 3: Feature Engineering

[0226] Feature engineering is an important part of dynamic load forecasting. By extracting appropriate features, the prediction performance of the model can be significantly improved.

[0227] Time features: Extract time features (hours, days, days of the week, etc.) to help the model identify patterns, such as the difference in load between weekdays and weekends.

[0228] Moving window feature: Using the sliding window method, historical load data within a period of time is generated as features. For example, the RB utilization rate and number of users in the past 24 hours can be used to predict the load of base station m in the next 1 hour, 6 hours, 12 hours, 24 hours, etc.

[0229] Environmental characteristics: Weather, holidays, major events, etc. are taken as additional characteristics. This information often has a significant impact on the load.

[0230] Step 4: Model selection and training

[0231] Select a suitable machine learning model for training. Common prediction models include:

[0232] LSTM(Long Short-Term Memory Network):

[0233] Applicable scenarios: LSTM is good at processing time series data and can capture the temporal relationship in historical load data. It is particularly suitable for predicting load changes over a long period of time.

[0234] XGBoost (Gradient Boosting Decision Tree):

[0235] Applicable scenarios: XGBoost is suitable for processing a large number of nonlinear features, time series and multidimensional features. It is suitable for processing short-term load forecasting.

[0236] Step 5: Model training

[0237] The data is divided into training set and test set for model training. The training process involves the following steps:

[0238] Data partitioning: The data is divided into 80% training set and 20% test set in chronological order.

[0239] Model training: input historical load data and output load forecast values ​​for future time periods.

[0240] Model Validation: Use metrics like MAE (mean absolute error) or RMSE (root mean square error) to measure the performance of the model on the test set.

[0241] Step 6: Predict the results

[0242] The trained model can output the load forecast results of base station m in a certain time period in the future. For example, the load of base station m in the next 1 hour, 6 hours, 12 hours, and 24 hours. For example:

[0243] Output: RB utilization rate in the next hour, number of users in the next hour.

[0244] Step 7: Real-time adjustment and dynamic optimization

[0245] Real-time monitoring: Compare the model prediction results with the actual load, and trigger model recalibration when deviations are found.

[0246] Adaptive learning: Update data in real time through the model to continuously optimize prediction accuracy, especially in the face of unexpected events (such as holidays and sudden events).

[0247] This model can be used to predict the load changes of base station m in the future and provide a scientific basis for capacity expansion decisions. For example, when it is predicted that the load of a base station will exceed its capacity in the next few hours, capacity expansion measures can be taken in advance to ensure network stability and user experience.

[0248] Implementation Example: Load Forecasting

[0249] Suppose we are predicting the load of a base station in the next 1, 6, 12, and 24 hours. The base station is located in a commercial area and has a higher load during the peak hours on weekdays (such as 9:00-18:00). The following is a complete implementation step:

[0250] 1. Data Collection

[0251] The load data of the base station in the past 24 hours is collected from the historical data of the base station, including RB utilization, number of users, time characteristics, etc. The data example is shown in Table 4 below.

[0252] Table 4: Load data of a base station in the past 24 hours

[0253] time RB Utilization Traffic(GB) Number of users Time characteristics 2024-09-01 09:00 65％ 10GB 120 Working Days 2024-09-01 10:00 50％ 8GB 110 Working Days 2024-09-01 11:00 75％ 12GB 130 Working Days ... ... ... ... ...

[0254] 2. Data Preprocessing

[0255] Normalization: Normalize RB utilization, number of users, etc. to between 0 and 1. For example: RB utilization: 0.75 (meaning 75%)

[0256] 3. Feature Engineering

[0257] We construct features based on the data of the past 24 hours to predict the load of the base station in the next 1, 6, 12, and 24 hours. Features may include:

[0258] RB utilization sequence for the past 24 hours.

[0259] Time characteristics: The current time is 9:00 on a weekday, which is a peak period.

[0260] External events: No special events at this time.

[0261] 4. Select LSTM model for training

[0262] Model architecture: Use the LSTM model to input features such as RB utilization and number of users in the past 24 hours, and output the predicted values ​​for the next 1, 6, 12, and 24 hours. The parameters are as follows: from tensorflow.keras.models import Sequential

[0263] from tensorflow.keras.layers import LSTM,Dense

[0264] model=Sequential()

[0265] model.add(LSTM(units=50,return_sequences=True,input_shape=(X_train.shape[1],X_train.shape[2])))

[0266] model.add(LSTM(units=50))

[0267] model.add(Dense(units=1, 6, 12, 24)#Output the RB utilization forecast for the next 1, 6, 12, and 24 hours. This can be set according to the situation.

[0268] Model training:

[0269] Use 80% of the data for training and 20% for validation.

[0270] Use Adam optimizer and MSE (mean squared error) as the loss function.

[0271] Training process:

[0272] Each round of training outputs the current prediction error. Assume that after 50 rounds of training, the final MAE (mean absolute error) on the test set is 3%.

[0273] Predicting future loads

[0274] After the model training is completed, the latest 24-hour data (or data from other time periods) is input, and the model will predict the load conditions of the base station in the next 1, 6, 12, and 24 hours.

[0275] Prediction results:

[0276] In the next hour, the load of base station m is:

[0277] Predicted RB utilization: 80%

[0278] Estimated number of users: 200

[0279] Next 6 hours:

[0280] Predicted RB utilization: 86%

[0281] Estimated number of users: 190

[0282] Next 12 hours:

[0283] Predicted RB utilization: 95%

[0284] Estimated number of users: 270

[0285] Next 24 hours:

[0286] Predicted RB Utilization: 105%

[0287] Predicted number of users: 280

[0288] The model predicts that the load on base stations will gradually increase in the future and needs to be processed, otherwise it may cause network jams and lead to user complaints.

[0289] Furthermore,

[0290] The determining whether the base station needs to be expanded according to the prediction result includes:

[0291] When the predicted PRB utilization rate and the number of access users of base station m in a specific time period in the future meet the conditions for base station expansion, it means that the base station needs to be expanded; if not, it means that the base station does not need to be expanded;

[0292] The conditions for the base station capacity expansion include that the PRB utilization rate of the base station in a specific time period is ≥ B%, and the number of access users is ≥ C, where B and C are preset values;

[0293] The estimating, for the base station that needs to be expanded, the total number of RBs actually allocated to the user by the base station that needs to be expanded includes:

[0294] After the expansion base station is determined, it is determined that after base station m is allocated and obtains the expansion license, base station m will have the original RB resource quantity R m The number of RB resources increased based on V max , PRB utilization rate of base station m F(A), number of RBs required by the user F(A)*R m ;

[0295] The total number of RBs actually allocated to users by the base station is calculated by the following formula (2): m ;

[0296]

[0297] The values ​​of B and C can be set according to the actual situation. Generally, B is greater than 85% and C is greater than 140. Assume that the expansion conditions of the base station are:

[0298] PRB utilization rate ≥ 90% and the number of access users ≥ 160

[0299] When the predicted PRB utilization rate and the number of access users of base station m in the next hour (which can be 2 hours, 6 hours, 12 hours, 24 hours, etc., determined by yourself) meet the condition, the base station needs to be expanded. If it does not meet the condition, it does not need to be expanded.

[0300] According to the above conditions, after determining the base station to be expanded, assuming that in the next A hours, the PRB utilization rate of the base station m that needs to be expanded is F(A), calculate the estimated unit traffic value of the base station m that needs to be expanded under this PRB state.

[0301] RB resources are the resources of the base station, which are used to allocate to users to process data. The number of RBs in each time slot of the base station is fixed and related to the physical bandwidth. Assume that R m After the allocation of the expansion license is obtained, the base station is in the original R m will increase V max RB resources.

[0302] Assume that the PRB utilization rate of the base station m that needs to be expanded is F(A), and the number of RBs required by the user is F(A)*R m If the number of RBs required by the user is F(A)*R m , which exceeds the total number of RBs of the expanded base station, i.e., R m +V max , indicating that RB is insufficient and the user cannot obtain sufficient RB resources. The maximum amount he can obtain is R m +V mzx RBs. If the number of RBs required by the user is F(A)*R m , which is lower than the total number of RBs of the expanded base station, that is, R m +V max , indicating that RB is sufficient, the user can obtain sufficient RB resources, and the number of RB resources obtained is F(A)*R m .

[0303] The total number of RBs actually allocated to users, counted as S z .

[0304]

[0305] For example, suppose a base station has a R m 100, V max is 50, R m +V max is 150. F(A) is F(24), which takes a value of 125%.

[0306] F(A)*R m =125<150,S z =125.

[0307] Assume R m 100, V max is 50, R m +V max is 150. F(A) is F(12), which takes a value of 160%.

[0308] F(A)*R m =160>150,S z =150.

[0309] Furthermore,

[0310] The base station digital modulation mode related parameters include the digital modulation mode used by the base station and the proportion of various digital modulation modes used by the base station;

[0311] The estimated flow value of the total number of RBs actually allocated to users for use estimated based on the parameters related to the base station digital modulation method includes:

[0312] The expected flow rate that can be increased is calculated by the following formula (3):

[0313]

[0314] Among them, L m+z is the estimated traffic value per unit time that can be increased by the total number of RBs actually allocated to users by the base station m to be expanded, y represents the yth digital modulation mode, z represents the maximum value of the digital modulation mode type, H y is the proportion of the yth digital modulation method used by base station m, R represents the amount of RE resource data that can be scheduled in one time slot of each resource block, Ty is the amount of data that can be transmitted by each RE resource corresponding to the yth digital modulation method used by the base station, and A is a constant.

[0315] In a possible implementation, according to the number of newly added resource blocks that can be actually used and the parameters related to the digital modulation mode of the base station, the predicted additional traffic per unit time of each base station to be expanded is calculated, wherein the predicted additional traffic per unit time is the traffic generated per unit time by the newly added resource blocks that can be actually used. The digital modulation modes used by the base station include: QPSK, 16QAM, 64QAM, and 256QAM modulation modes, and the QPSK, 16QAM, 64QAM, and 256QAM modulation modes used by the base station m account for H m , Q m , A m , D m , namely H1, H2, H3 and H4.

[0316] The specific form of formula 3 is as follows:

[0317] L m+z =(S z *H m *12*7*2+S z *Q m *12*7*4+S z *

[0318] Am *12*7*6+S z *D m *12*7*8)*2000*3600 / 10243;

[0319] The unit of traffic is GB.

[0320] H m , Q m , A m , D m They are the proportions of QPSK (Quadrature Phase Shift Keying), 16QAM (16Quadrature Amplitude Modulation), 64QAM, and 256QAM modulation modes used by base stations; 12*7 means that one RB resource has 12*7 RE resources, and one RB is scheduled in one time slot, so 12*7 RE resources are scheduled in one time slot; 2 means that when the base station uses QPSK modulation mode, one RE can transmit 2 bits; 4 means that when the base station uses 16QAM modulation mode, one RE can transmit 4 bits; 6 means that when the base station uses 64QAM modulation mode, one RE can transmit 6 bits; 8 means that when the base station uses 256QAM modulation mode, one RE can transmit 8 bits; 2000 means that there are 2000 time slots in 1 second; 3600 means that there are 3600 seconds in 1 hour; 1024 to the power of 3 represents the conversion between bit and Gbit.

[0321] Furthermore,

[0322] The value corresponding to the expected traffic value of each base station increased is calculated according to the per-unit traffic value of the base station and the expected traffic value that can be increased, including:

[0323] The value corresponding to the expected traffic value of base station m is calculated by the following formula (4):

[0324] P m+ z = L m+ z * P m , (4)

[0325] Among them, P m+z is the value corresponding to the expected traffic increase of base station m, L m+z P is the total number of RBs actually allocated to users by base station m, and the predicted traffic value per unit time that can be increased. m is the value of each unit flow of base station m;

[0326] The allocation scheme of the expansion license is determined according to the value corresponding to the expected traffic value of each base station, including:

[0327] Press P m+z The base stations to be expanded are sorted from high to low in size, and expansion licenses are allocated preferentially to the base stations to be expanded that are ranked higher;

[0328] Expand the capacity of base stations that have obtained expansion licenses.

[0329] Formula (4) can be used to calculate the value corresponding to the expected traffic growth of each base station, that is, the additional value after the base station is allocated a license. The allocation plan of the expansion license can be determined based on this value. For example, there are currently five base station cells to be expanded, A, B, C, D and E, and their corresponding predicted additional values ​​are 1000, 1200, 1500, 1700 and 1300 respectively. When there is only one expansion license, D can be selected as the target base station. When there are two expansion licenses, D and C can be selected as the target base station cells.

[0330] Allocate the license to the target base station to expand the resource block of the target base station. In this step, allocating the license to the target base station may be importing the license into the data of the target base station or sending the license to the target base station. This application does not impose any specific restrictions on the method of allocating the license. When the license is allocated to the target base station, the scheduling module of the target base station will be automatically increased.

[0331] The present disclosure is described below with an example.

[0332] Assume that there are three base stations A, B, and C to be expanded. F(24) is shown below. All of them have reached the expansion threshold. The above values ​​corresponding to the three base stations are shown in Table 5 below. However, there are only two expansion licenses currently.

[0333] Table 5: Values ​​corresponding to base stations

[0334] parameter Base station A to be expanded Base station B to be expanded Base station C to be expanded <![CDATA[R m ]]> 100(20M bandwidth) 100(20M bandwidth) 50(10M bandwidth) <![CDATA[H m ]]> 20％ 20％ 30％ <![CDATA[Q m ]]> 30％ 20％ 20％ <![CDATA[A m ]]> 30％ 40％ 30％ <![CDATA[D m ]]> 10％ 20％ 20％ <![CDATA[P m ]]> 25 20 30 <![CDATA[V max ]]> 50 50 50 F(24) 110％ 175％ 165％ <![CDATA[S z ]]> 110 150 82.5 <![CDATA[L m+z ]]> 260.2 439.3 223.1 <![CDATA[P m+z ]]> 6505.7 8786.90 6691.6

[0335] S z , L m+z The calculation results are shown in Table 5 above. From the above calculation, we can see that the expansion value of base station A is P m+z The capacity expansion value of base station B is 6505.7 yuan, the capacity expansion value of base station C is 8786.90 yuan, and the capacity expansion value of base station C is 6691.6 yuan. Assuming that there are only 2 capacity expansion licenses, the capacity expansion value of base stations B and C is greater, so they are given priority.

[0336] The disclosed embodiment obtains data such as the historical load and access users of all base stations; based on the information of the users accessing the base stations, the value of each 1Gb of traffic generated by the base stations is calculated using the traffic generated by the users on the base stations and the corresponding average monthly consumption, user importance, regional importance, traffic peak time ratio and other data; based on machine learning, the load of the base stations at a certain time in the future is estimated, and the base stations with estimated loads exceeding the threshold are defined as base stations to be expanded, and the number of RBs actually allocated to users after the expansion is calculated; finally, the expected traffic growth is converted according to the number of RBs, and finally the value corresponding to the expected traffic growth is calculated, and base stations with high value are expanded first. It is possible to achieve a more accurate judgment of the value of expansion, prioritize the base stations to be expanded, and thus improve the value of expansion.

[0337] The second embodiment of the present disclosure also provides a base station expansion license allocation system, such as Figure 2 As shown, the system comprises:

[0338] A first acquisition module 11 is configured to acquire historical service data of all base stations, including access user information of the base stations and parameters related to the digital modulation mode of the base stations;

[0339] A second acquisition module 12, which is configured to acquire user feature information and service information on the base station according to the access user information of the base station;

[0340] A first calculation module 13, which is configured to calculate the value of each unit flow of the base station according to the user characteristic information and the service information on the base station;

[0341] A first estimation module 14 is configured to perform load prediction for each base station, determine whether the base station needs to be expanded according to the prediction result, and estimate the total number of RBs actually allocated to users for use by the base station that needs to be expanded;

[0342] A second estimation module 15 is configured to estimate the expected flow value of the total number of RBs actually allocated to the user for use that can be increased according to the parameters related to the base station digital modulation mode;

[0343] A second calculation module 16 is configured to calculate the value corresponding to the expected traffic value of each base station according to the per-unit traffic value of the base station and the expected traffic value that can be increased;

[0344] The determination module 17 is configured to determine a distribution scheme for the expansion license according to the value corresponding to the expected traffic value of each base station increase.

[0345] Furthermore,

[0346] The user characteristic information includes: user importance and importance of the area where the base station is located;

[0347] The service information on the base station includes the traffic generated by the user connecting to the base station, the proportion of the user's access time to the base station, the average consumption of the user in a preset time period, and the proportion of the user's traffic peak time;

[0348] The first calculation module 13 is specifically configured to calculate the value of each unit flow of the base station by the following formula (1):

[0349]

[0350] Where P m is the value of each unit of traffic of base station m, N is the number of users accessing base station m in the corresponding time period, K n The traffic generated by user n connecting to base station m, T n is the proportion of time that user n accesses base station m, P n is the average consumption of user n in the preset time period, U n is the importance of user n, W m is the importance of the area where base station m is located, Y n is the proportion of user n’s traffic peak time, L m is the traffic generated during the traffic peak time of base station m.

[0351] Furthermore, the first estimation module 14 includes a prediction unit, and the prediction unit is configured as:

[0352] Perform data collection, collect historical data and environmental data of the base station, the historical data includes: RB utilization rate, traffic and access user information of base station m, the access user information includes quantity, traffic, consumption price, user type; the environmental data includes: external event information; and mark the data according to the time period corresponding to the data, the corresponding time period includes weekdays, weekends, holidays, and intraday time periods;

[0353] Perform data preprocessing;

[0354] Perform feature engineering on the data, select time features, moving window features and environmental features;

[0355] Select a preset machine learning model;

[0356] Divide the processed data into training sets and test sets, and perform model training on the machine learning model;

[0357] Output the load forecast result of base station m in a certain time period in the future through the trained model;

[0358] The model prediction results are compared with the actual load, and the model recalibration is triggered when deviations are found.

[0359] Furthermore,

[0360] The first estimation module 14 further includes a judgment unit and an estimation unit, and the judgment unit is configured as:

[0361] When the PRB utilization rate and the number of access users of the base station m in a specific time period in the future predicted by the prediction unit both meet the conditions for base station capacity expansion, it means that the base station needs to be expanded; if not, it means that the base station does not need to be expanded;

[0362] The conditions for the base station capacity expansion include that the PRB utilization rate of the base station in a specific time period is ≥ B%, and the number of access users is ≥ C, where B and C are preset values;

[0363] The estimation unit is set as:

[0364] After the expansion base station is determined, it is determined that after base station m is allocated and obtains the expansion license, base station m will have the original RB resource quantity R m The number of RB resources increased based on V max , PRB utilization rate of base station m F(A), number of RBs required by the user F(A)*R m ;

[0365] The total number of RBs actually allocated to users by the base station is calculated by the following formula (2): m ;

[0366]

[0367] Furthermore,

[0368] The base station digital modulation mode related parameters include the digital modulation mode used by the base station and the proportion of various digital modulation modes used by the base station;

[0369] The second estimation module 15 is specifically configured as follows:

[0370] The expected flow rate that can be increased is calculated by the following formula (3):

[0371]

[0372] Among them, L m+z is the estimated traffic value per unit time that can be increased by the total number of RBs actually allocated to users by the base station m to be expanded, y represents the yth digital modulation mode, z represents the maximum value of the digital modulation mode type, H y is the proportion of the yth digital modulation method used by base station m, R represents the amount of RE resource data that can be scheduled in one time slot of each resource block, Ty is the amount of data that can be transmitted by each RE resource corresponding to the yth digital modulation method used by the base station, and A is a constant.

[0373] Furthermore,

[0374] The second calculation module 16 is specifically configured as:

[0375] The value corresponding to the expected traffic value of base station m is calculated by the following formula (4):

[0376] P m+ z = L m+ z * P m , (4)

[0377] Among them, P m+z is the value corresponding to the expected traffic increase of base station m, L m+z P is the total number of RBs actually allocated to users by base station m, and the predicted traffic value per unit time that can be increased. m is the value of each unit of traffic of base station m;

[0378] The determination module is specifically configured as follows:

[0379] Press P m+z The base stations to be expanded are sorted from high to low in size, and expansion licenses are allocated preferentially to the base stations to be expanded that are ranked higher;

[0380] Expand the capacity of base stations that have obtained expansion licenses.

[0381] The base station expansion license allocation system of the disclosed embodiment is used to implement the base station expansion license allocation method in method embodiment 1, so the description is relatively simple. For details, please refer to the relevant description in the previous method embodiment, which will not be repeated here.

[0382] In addition, if Figure 3 As shown, the third embodiment of the present disclosure further provides an electronic device, including a memory 100 and a processor 200, wherein the memory 100 stores a computer program, and when the processor 200 runs the computer program stored in the memory 100, the processor 200 executes the above-mentioned various possible methods.

[0383] The memory 100 is connected to the processor 200. The memory 100 may be a flash memory, a read-only memory or other memory, and the processor 200 may be a central processing unit or a single-chip microcomputer.

[0384] In addition, an embodiment of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by a processor to perform the above-mentioned various possible methods.

[0385] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, computer program modules or other data). Computer-readable storage media include, but are not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable read only memory), flash memory or other memory technology, CD-ROM (Compact Disc Read-Only Memory), Digital Video Disc (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0386] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present disclosure, but the present disclosure is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and substance of the present disclosure, and these modifications and improvements are also considered to be within the scope of protection of the present disclosure.

Claims

1. A method for allocating base station expansion licenses, characterized in that: The method comprises: Obtain historical service data of all base stations, including base station access user information and base station digital modulation mode related parameters; According to the access user information of the base station, obtain the user characteristic information and the service information on the base station; Calculate the value of each unit of traffic of the base station based on the user characteristic information and the service information on the base station; Carry out load forecasting for each base station, determine whether the base station needs to be expanded based on the forecast results, and estimate the total number of resource blocks (RBs) actually allocated to users for use by the base station that needs to be expanded; Estimate the expected traffic value that can be increased by the total number of RBs actually allocated to users based on the parameters related to the base station digital modulation method; The value corresponding to the expected traffic value of each base station is calculated based on the per-unit traffic value of the base station and the expected traffic value that can be increased; The allocation plan for expansion licenses is determined based on the value corresponding to the expected traffic growth of each base station.

2. The method according to claim 1, characterized in that: The user characteristic information includes: user importance and importance of the area where the base station is located; The service information on the base station includes the traffic generated by the user connecting to the base station, the proportion of the user's access time to the base station, the average consumption of the user in a preset time period, and the proportion of the user's traffic peak time; The per-unit flow value of the base station is calculated based on the user characteristic information and the service information on the base station by the following formula (1): Where P m is the value of each unit of traffic of base station m, N is the number of users accessing base station m in the corresponding time period, K n The traffic generated by user n connecting to base station m, T n is the proportion of time that user n accesses base station m, P n is the average consumption of user n in the preset time period, U n is the importance of user n, W m is the importance of the area where base station m is located, Y n is the proportion of user n’s traffic peak time, L m is the traffic generated during the traffic peak time of base station m.

3. The method according to claim 1, characterized in that The load prediction for each base station includes: Perform data collection, collect historical data and environmental data of the base station, the historical data includes: RB utilization rate, traffic and access user information of base station m, the access user information includes quantity, traffic, consumption price, user type; the environmental data includes: external event information; and mark the data according to the time period corresponding to the data, the corresponding time period includes weekdays, weekends, holidays, and intraday time periods; Perform data preprocessing; Perform feature engineering on the data, select time features, moving window features and environmental features; Select a preset machine learning model; Divide the processed data into training sets and test sets, and perform model training on the machine learning model; Output the load forecast result of base station m in a certain period of time in the future through the trained model; The model prediction results are compared with the actual load, and the model recalibration is triggered when deviations are found.

4. The method according to claim 2, characterized in that: The determining whether the base station needs to be expanded according to the prediction result includes: When the predicted physical resource block PRB utilization rate and the number of access users of base station m in a specific time period in the future meet the conditions for base station expansion, it means that the base station needs to be expanded; if not, it means that the base station does not need to be expanded; The conditions for the base station capacity expansion include that the PRB utilization rate of the base station in a specific time period is ≥ B%, and the number of access users is ≥ B, where A and B are preset values; The estimating, for the base station that needs to be expanded, the total number of RBs actually allocated to the user by the base station that needs to be expanded includes: After the expansion base station is determined, it is determined that after base station m is allocated and obtains the expansion license, base station m will have the original RB resource quantity R m The number of RB resources increased based on V max , PRB utilization rate of base station m F(A), number of RBs required by the user F(A)*R m ; The total number of RBs actually allocated to users by the base station is calculated by the following formula (2): m ; 5. The method according to claim 4, characterized in that The base station digital modulation mode related parameters include the digital modulation mode used by the base station and the proportion of various digital modulation modes used by the base station; The estimated flow value of the total number of RBs actually allocated to users for use estimated based on the parameters related to the base station digital modulation method includes: The expected flow rate that can be increased is calculated by the following formula (3): Among them, L m+z is the estimated traffic value per unit time that can be increased by the total number of RBs actually allocated to users by the base station m to be expanded, y represents the yth digital modulation mode, z represents the maximum value of the digital modulation mode type, H y is the proportion of the yth digital modulation method used by base station m, R represents the amount of resource element RE resource data that can be scheduled in one time slot of each resource block, Ty is the amount of data that can be transmitted by each RE resource corresponding to the yth digital modulation method used by the base station, and A is a constant.

6. The method according to claim 1 or 5, characterized in that: The value corresponding to the expected traffic value of each base station increased is calculated according to the per-unit traffic value of the base station and the expected traffic value that can be increased, including: The value corresponding to the expected traffic value of base station m is calculated by the following formula (4): P m+z =L m+z *P m , (4) Among them, P m+z is the value corresponding to the expected traffic increase of base station m, L m+z P is the total number of RBs actually allocated to users by base station m, and the predicted traffic value per unit time that can be increased. m is the value of each unit flow of base station m; The allocation scheme of the expansion license is determined according to the value corresponding to the expected traffic value of each base station, including: Press P m+z The base stations to be expanded are sorted from high to low in size, and expansion licenses are allocated preferentially to the base stations to be expanded that are ranked higher; Expand the capacity of base stations that have obtained expansion licenses.

7. A base station expansion license allocation system, characterized in that: The system comprises: A first acquisition module, which is configured to acquire historical service data of all base stations, including access user information of the base stations and parameters related to the digital modulation mode of the base stations; A second acquisition module, which is configured to acquire user feature information and service information on the base station according to the access user information of the base station; A first calculation module, which is configured to calculate the value of each unit flow of the base station according to the user characteristic information and the service information on the base station; A first estimation module is configured to perform load prediction for each base station, determine whether the base station needs to be expanded according to the prediction result, and estimate the total number of RBs actually allocated to users for the base station that needs to be expanded; A second estimation module is configured to estimate an expected flow value that can be increased by the total number of RBs actually allocated to the user according to parameters related to the digital modulation mode of the base station; A second calculation module is configured to calculate the value corresponding to the expected traffic value of each base station according to the per-unit traffic value of the base station and the expected traffic value that can be increased; The determination module is configured to determine a distribution plan for the expansion license according to the value corresponding to the expected traffic value of each base station increase.

8. The system according to claim 7, characterized in that The user characteristic information includes: user importance and importance of the area where the base station is located; The service information on the base station includes the traffic generated by the user connecting to the base station, the proportion of the user's access time to the base station, the average consumption of the user in a preset time period, and the proportion of the user's traffic peak time; The first calculation module is specifically configured to calculate the value of each unit flow of the base station by the following formula (1): Where P m is the value of each unit of traffic of base station m, N is the number of users accessing base station m in the corresponding time period, K n The traffic generated by user n connecting to base station m, T n is the proportion of time that user n accesses base station m, P n is the average consumption of user n in the preset time period, U n is the importance of user n, W m is the importance of the area where base station m is located, Y n is the proportion of user n’s traffic peak time, L m is the traffic generated during the traffic peak time of base station m.

9. The system according to claim 7, characterized in that The first estimation module includes a prediction unit, and the prediction unit is configured as: Perform data collection, collect historical data and environmental data of the base station, the historical data includes: RB utilization rate, traffic and access user information of base station m, the access user information includes quantity, traffic, consumption price, user type; the environmental data includes: external event information; and mark the data according to the time period corresponding to the data, the corresponding time period includes weekdays, weekends, holidays, and intraday time periods; Perform data preprocessing; Perform feature engineering on the data, select time features, moving window features and environmental features; Select a preset machine learning model; Divide the processed data into training sets and test sets, and perform model training on the machine learning model; Output the load forecast result of base station m in a certain period of time in the future through the trained model; The model prediction results are compared with the actual load, and the model recalibration is triggered when deviations are found.

10. The system according to claim 9, characterized in that The first estimation module further includes a judgment unit and an estimation unit, and the judgment unit is configured as: When the PRB utilization rate and the number of access users of the base station m in a specific time period in the future predicted by the prediction unit both meet the conditions for base station capacity expansion, it means that the base station needs to be expanded; if not, it means that the base station does not need to be expanded; The conditions for the base station capacity expansion include that the PRB utilization rate of the base station in a specific time period is ≥ B%, and the number of access users is ≥ C, where B and C are preset values; The estimation unit is set as: After the expansion base station is determined, it is determined that after base station m is allocated and obtains the expansion license, base station m will have the original RB resource quantity R m The number of RB resources increased based on V max , PRB utilization rate of base station m F(A), number of RBs required by the user F(A)*R m ; The total number of RBs actually allocated to users by the base station is calculated by the following formula (2): m ; 11. The system according to claim 10, characterized in that The base station digital modulation mode related parameters include the digital modulation mode used by the base station and the proportion of various digital modulation modes used by the base station; The second estimation module is specifically configured as follows: The expected flow rate that can be increased is calculated by the following formula (3): Among them, L m+z is the estimated traffic value per unit time that can be increased by the total number of RBs actually allocated to users by the base station m to be expanded, y represents the yth digital modulation mode, z represents the maximum value of the digital modulation mode type, H y is the proportion of the yth digital modulation method used by base station m, R represents the amount of RE resource data that can be scheduled in one time slot of each resource block, Ty is the amount of data that can be transmitted by each RE resource corresponding to the yth digital modulation method used by the base station, and A is a constant.

12. The system according to claim 7 or 11, characterized in that: The second calculation module is specifically configured as: The value corresponding to the expected traffic value of base station m is calculated by the following formula (4): P m+z =L m+z *P m , (4) Among them, P m+z is the value corresponding to the expected traffic increase of base station m, L m+z P is the total number of RBs actually allocated to users by base station m, and the predicted traffic value per unit time that can be increased. m is the value of each unit flow of base station m; The determination module is specifically configured as follows: Press P m+z The base stations to be expanded are sorted from high to low in size, and expansion licenses are allocated preferentially to the base stations to be expanded that are ranked higher; Expand the capacity of base stations that have obtained expansion licenses.

13. An electronic device, characterized in that: The method comprises a memory and a processor, wherein a computer program is stored in the memory, and when the processor runs the computer program stored in the memory, the processor executes the method for allocating base station expansion licenses according to any one of claims 1 to 6.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for allocating base station expansion licenses according to any one of claims 1 to 6 is implemented.

Images