Base station parameter optimization method, device, equipment, medium and program product

Abstract

The application relates to the technical field of wireless networks, and provides a base station parameter optimization method and device, equipment, a medium and a program product. Wireless key data indicators KPI of a base station to be optimized are acquired. The distribution of log data of a preset training base station is adjusted according to the KPI, and corrected log data is obtained. The corrected log data is input into a pre-trained test model, the base station performance to be predicted of the base station to be optimized is predicted, and the base station operation parameters of the base station to be optimized are optimized based on the prediction result, wherein the test model is trained based on the log data of the preset training base station. In the parameter optimization, the wireless key data indicators KPI that are easy to obtain are selected, the actual parameters of the base station need to be modified, different parameters do not need to be traversed for each base station, the base station parameter optimization is realized in a model estimation manner, and the base station parameter optimization efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of wireless network technology, and in particular to a method, apparatus, device, medium, and program product for optimizing base station parameters. Background Technology

[0002] The specific functions or parameters of a base station exhibit different performance characteristics in different service scenarios. In order to obtain better network performance, it is essential to configure different parameters for different base stations.

[0003] Currently, when optimizing base station networks, optimization personnel typically collect base station parameters and then configure and adjust them based on experience. However, the various influencing factors in wireless communication networks are extremely complex, making it impossible to directly determine the performance of different parameters under different service scenarios through manual experience. If we were to directly try different parameters for each base station and finally determine the optimal parameters based on the actual performance of each base station, the workload would be too enormous, resulting in low efficiency in base station parameter optimization. Summary of the Invention

[0004] This application provides a method, apparatus, device, medium, and program product for optimizing base station parameters, in order to solve the problem of low efficiency in base station parameter optimization.

[0005] Firstly, this application provides a method for optimizing base station parameters, including:

[0006] Obtain the key wireless data indicators (KPIs) of the base station to be optimized;

[0007] Adjust the preset log data distribution of the training base stations according to the KPIs to obtain corrected log data;

[0008] The corrected log data is input into a pre-trained test model to predict the base station performance of the base station to be optimized, and the base station operating parameters of the base station to be optimized are tuned based on the prediction results. The test model is trained based on the log data of a preset training base station.

[0009] In one embodiment, adjusting the preset log data distribution of the training base stations according to KPIs to obtain corrected log data includes:

[0010] Based on KPIs, determine the sample size of log data from the preset training base stations;

[0011] The corrected log data is obtained by sampling based on the sampling quantity.

[0012] In one embodiment, determining the sampling quantity of log data from a preset training base station based on KPIs includes:

[0013] Based on KPIs, the first distribution corresponding to the performance indicators of the communication network is obtained;

[0014] Based on the log data of the preset training base stations, a second distribution corresponding to the first distribution is obtained;

[0015] Based on the first distribution and the second distribution, determine the preset sampling quantity of log data for the training base station.

[0016] In one embodiment, the communication network performance metric is path loss, and determining the sampling quantity of log data for the preset training base station based on the KPI includes:

[0017] Based on KPIs, the first path loss distribution with the first loss interval is obtained;

[0018] Based on the log data of the preset training base stations, a second path loss distribution corresponding to the first path loss distribution is obtained;

[0019] Based on the first path loss distribution and the second path loss distribution, the sampling quantity of log data for the preset training base stations is determined.

[0020] In one embodiment, determining the sampling quantity of the preset training base station log data based on the first path loss distribution and the second path loss distribution includes:

[0021] The target road loss interval is determined based on the ratio between the first road loss distribution and the second road loss distribution within the same road loss interval.

[0022] Based on the distribution of KPIs within the target road loss range and the number of samples of log data from the preset training base stations within the target road loss range, determine the total number of samples of log data from the preset training base stations.

[0023] Based on the total number of samples and the first path loss distribution, the number of samples of log data from the preset training base station in each path loss interval is determined.

[0024] In one embodiment, the test model trains the boosting tree model based on log data from a preset training base station.

[0025] In one embodiment, obtaining the wireless key data indicators (KPIs) of the base station to be optimized includes:

[0026] Based on the gateway, obtain the wireless key data indicators (KPIs) corresponding to the parameters to be optimized of the base station to be optimized;

[0027] Correspondingly, the step of inputting the corrected log data into a pre-trained test model to predict the base station performance of the base station to be optimized, and optimizing the base station operating parameters of the base station to be optimized based on the prediction results, includes:

[0028] The corrected log data is input into a pre-trained test model to obtain the predicted base station performance corresponding to any value of the parameter to be optimized.

[0029] The base station operating parameters are configured to the optimal parameter values ​​corresponding to the predicted base station performance in order to complete the base station parameter optimization.

[0030] Secondly, this application also provides a base station parameter optimization device, comprising:

[0031] The acquisition module is used to acquire the wireless key data indicators (KPIs) of the base station to be optimized.

[0032] The adjustment module is used to adjust the preset log data distribution of the training base stations according to the KPIs to obtain corrected log data;

[0033] The optimization module is used to input the corrected log data into a pre-trained test model, predict the base station performance of the base station to be optimized, and optimize the base station operating parameters of the base station to be optimized based on the prediction results. The test model is trained based on the log data of a preset training base station.

[0034] Thirdly, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement any of the base station parameter optimization methods described above.

[0035] Fourthly, this application provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the base station parameter optimization method as described above.

[0036] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the base station parameter optimization method as described above.

[0037] The base station parameter optimization method, apparatus, device, medium, and program products provided in this application obtain the wireless key data indicators (KPIs) of the base station to be optimized; adjust the log data distribution of a preset training base station according to the KPIs to obtain corrected log data; input the corrected log data into a pre-trained test model to predict the base station performance of the base station to be optimized, and optimize the base station operating parameters of the base station to be optimized based on the prediction results. The test model is trained based on the log data of the preset training base station. In parameter optimization, this application selects easily accessible wireless key data indicators (KPIs), adjusts the log data distribution of the preset training base station to obtain corrected log data, inputs the corrected log data into a pre-trained test model to predict the base station performance of the base station to be optimized, and optimizes the base station operating parameters of the base station to be optimized based on the prediction results. This eliminates the need to modify the actual parameters of the base station or traverse different parameters for each base station, achieving base station parameter optimization through model prediction, thus improving the efficiency of base station parameter optimization. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is one of the flowcharts illustrating the base station parameter optimization method provided in this application;

[0040] Figure 2 This is a weighted diagram of the log data of the preset training base station provided in this application;

[0041] Figure 3 This is a flowchart illustrating the predicted and actual performance provided in this application;

[0042] Figure 4 This is the second flowchart of the base station parameter optimization method provided in this application;

[0043] Figure 5 This is a schematic diagram of the base station parameter optimization device provided in this application;

[0044] Figure 6 This is a schematic diagram of the structure of the electronic device provided in this application. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0046] Figure 1 This is one of the flowcharts illustrating the base station parameter optimization method provided in this application, such as... Figure 1 As shown, this application provides a base station parameter optimization method, including:

[0047] Step S100: Obtain the Key Performance Indicator (KPI) of the base station to be optimized; the Key Performance Indicator (KPI) includes the indicator data of the uplink rate related parameters of the base station to be optimized, which is used to characterize the network performance characteristics of the base station to be optimized.

[0048] KPIs can be obtained through the gateway without having to go to the base station site. This is easier than retrieving large amounts of log data and is related to the base station, thus reflecting the characteristics of each base station to be optimized.

[0049] Step S200: Adjust the preset log data distribution of the training base stations according to the KPI to obtain the corrected log data;

[0050] The preset training base station log data (hereinafter referred to as log data) is the log data of a small batch (at least dozens) of base stations in the preset business scenario. The distribution of the preset training base station log data is adjusted according to the KPI, that is, the log data is sampled from the preset training base station log data according to the KPI data distribution to obtain the corrected log data, so that the corrected log data is more correlated with the base station to be optimized, so as to improve the accuracy of network performance prediction.

[0051] Step S300: Input the corrected log data into a pre-trained test model to predict the base station performance of the base station to be optimized, and optimize the base station operating parameters of the base station to be optimized based on the prediction results. The test model is trained based on the log data of a preset training base station.

[0052] The preset boosting tree model is trained using log data from preset training base stations and actual network performance data from a small batch of base stations. This automatically learns the relationship between different business scenarios, different parameters, and network performance, and builds a test model.

[0053] Specifically, by inputting the corrected log data into the test model, the predicted network performance of the base station to be optimized under various parameters can be obtained from the test model output. By selecting the best predicted network performance, the optimal parameters of the base station to be optimized can be determined, thus completing the base station parameter optimization.

[0054] Understandably, this application selects readily available wireless key data indicators (KPIs) for parameter optimization, adjusts the distribution of log data from a preset training base station to obtain corrected log data, inputs the corrected log data into a pre-trained test model, predicts the performance of the base station to be optimized, and optimizes the base station operating parameters based on the prediction results. This eliminates the need to modify the actual parameters of the base station or traverse different parameters for each base station. By using model prediction, base station parameter optimization is achieved, thus improving the efficiency of base station parameter optimization.

[0055] Based on the above embodiments, as an optional embodiment, the test model trains the boosting tree model based on the log data of a preset training base station.

[0056] This application primarily considers the impact of different parameter modifications on user uplink speed. Therefore, when selecting log data, it considers using log data related to uplink speed for modeling.

[0057] Optionally, the log data includes user behavior data, channel condition data, network load data, base station parameters, and terminal capability data. Specifically, the user behavior data includes uplink user buffer status reports, the channel condition data includes uplink path loss and uplink interference, the network load data includes the number of remaining resource blocks and the number of scheduled users, the base station parameters include uplink waveform parameters, and the terminal capability data includes maximum transmit power.

[0058] In one embodiment, the user's current uplink rate is determined by the smaller of the maximum supported uplink rate and the user's current required uplink rate. The user's current required uplink rate is indicated by the user's Buffer Status Report (BSR), while the schedulable radio resources and bearer efficiency determine the maximum uplink rate. Specifically, schedulable radio resources are related to the number of remaining resource blocks (RBs) and the number of scheduled users. Bearer efficiency is related to channel conditions (path loss, uplink interference), terminal capabilities (maximum transmit power), and base station parameters (uplink waveform parameters). Therefore, the preset log data related to the uplink rate mainly includes the following aspects:

[0059] User behavior: Uplink BSR (lsgbsr);

[0060] Channel conditions: uplink path loss (ulPl), uplink interference (puschNi);

[0061] Network load: Remaining RBs (restRBNum), number of scheduled users (schdUeNum);

[0062] Base station parameters: Uplink waveform parameters (para);

[0063] Terminal capability: Maximum transmit power (pcMax).

[0064] Optionally, the boosting tree model uses the XGBoost model, also known as Extreme Gradient Boosting Tree, which is an implementation of the boosting algorithm. It can be used to solve classification or regression problems and has excellent results. The algorithm of the XGBoost model is to continuously add trees and perform feature splits to grow a tree. Each time a tree is added, it learns a new function to fit the residual of the previous prediction. After training, k trees are obtained. To predict the score of a sample, based on the features of the sample, it will fall into a corresponding leaf node in each tree. Each leaf node corresponds to a score. Finally, the scores of each tree are added together to get the predicted value of the sample.

[0065] The cost function of the XGBoost model is:

[0066]

[0067] Where n is the number of training samples, l is the loss for a single sample, and y′ i y represents the model's prediction of the training samples. i For the true label values ​​of the training samples, ∑ k Ω(f k The regularization term defines the complexity of the model.

[0068] y i y' = f(x), where x is the input value, specifically a 7-dimensional vector (i.e., x = [x_ulPl, x_lcgBsr, x_schdUeNum, x_restRbNum, x_puschNi, x_pcMax, x_para]), which represents the seven log data related to the uplink rate mentioned above. f(x) is the output value of the model (i.e., the predicted uplink rate), y' = f(x), where x is the input value, x is the input value, x is the input value, and f(x) ... i This represents the actual upload speed.

[0069] Figure 2 This is a weighted diagram of the log data of the preset training base stations provided in this application, such as... Figure 2As shown, log data is used as training samples, and the actual performance of the base station is used as the label of the training samples to train the XGBoost model and obtain the test model. After the model is trained, the weight values ​​can be automatically output through feature_importance: the feature_importance attribute is an attribute in the XGBoost model used to represent the importance of each feature in the model.

[0070] Figure 3 This is a flowchart illustrating the predicted and actual performance provided in this application, as shown below. Figure 3 As shown, after the test model is trained, test samples are extracted from the log data and input into the test model. The predicted performance output by the test model is compared with the actual performance of the test samples. Figure 3 As shown, the test model trained based on the XGBoost model can predict the base station performance relatively accurately, thus improving the accuracy of parameter optimization.

[0071] Understandably, this application creatively introduces the XGBoost model, which is used in image classification, medical simulation prediction, and industry, into the field of base station parameter optimization. It does not require modifying the actual parameters of the base station, nor does it require traversing different parameters for each base station. It achieves base station parameter optimization through model prediction, thereby improving the efficiency of base station parameter optimization.

[0072] Based on the above embodiments, as an optional embodiment, obtaining the wireless key data indicators (KPIs) of the base station to be optimized includes:

[0073] Step S100: Obtain the wireless key data indicators (KPIs) corresponding to the parameters to be optimized of the base station to be optimized based on the gateway.

[0074] Specifically, collecting base station log data requires on-site testing. Collecting log data from each base station for parameter optimization would be extremely labor-intensive and inefficient. Obtaining KPIs through a gateway is much easier.

[0075] KPIs include cell uplink UE Throughput rate distribution, PUSCH PRB utilization distribution, QPSK uplink transmission count distribution, 16QAM uplink transmission count distribution, 64QAM uplink transmission count distribution, 256QAM uplink transmission count distribution, PUSCH SINR distribution, inter-area uplink service user path loss distribution, SRS NI distribution, cell uplink UE Throughput data transmission time, and cell uplink UE... Throughput data volume, number of PRBs used in uplink DTCH, traffic of successful uplink DTCH transmissions, number of slots scheduled for uplink DTCH, number of uplink schedulable slots in the cell, total number of uplink scheduled users in the cell, average uplink interference level in the cell, number of uplink DTXs for different modulation schemes, number of uplink residual packets for different modulation schemes, number of TBs of successful uplink reception for different modulation schemes, number of TBs of failed uplink reception for different modulation schemes, number of PRBs occupied by uplink PRACH, number of PRBs used in uplink DCCH, number of PRBs used in uplink DTCH, number of available PRBs for PUSCH, number of PRBs used in PUSCH, number of PRBs used in PUCCH, distribution of uplink transmissions in QPSK, distribution of uplink transmissions in 16QAM, distribution of uplink transmissions in 64QAM, distribution of uplink transmissions in 256QAM, cumulative path loss value of uplink service users, cumulative path loss count of uplink service users.

[0076] Depending on the parameters to be optimized, the extracted KPIs will also differ. For example, if the waveform adaptive function parameters of a certain base station need to be modified, considering only the modification of the waveform adaptive function parameters, the effective received signal power at the base station is obtained by subtracting the path loss from the terminal's maximum transmit power. The effective signal power is then divided by the uplink interference to obtain the SINR. When the terminal's maximum transmit power and interference remain unchanged, SINR is only related to path loss. The higher the SINR, the higher the carrying efficiency, which affects the uplink rate. Therefore, if the waveform parameters of a certain base station need to be modified, the KPI is the path loss distribution of uplink service users in different intervals.

[0077] Correspondingly, the step of inputting the corrected log data into a pre-trained test model to predict the base station performance of the base station to be optimized, and optimizing the base station operating parameters of the base station to be optimized based on the prediction results, includes:

[0078] Step S310: Input the corrected log data into the pre-trained test model to obtain the predicted base station performance corresponding to any parameter value of the parameter to be optimized.

[0079] For example, if the parameter to be optimized is the waveform adaptive function parameter, the waveform adaptive function parameter has three values: auto, CP-OFDM, and DFT-S-OFDM. By inputting the log data related to the path loss distribution into the network performance prediction model, the gain corresponding to the three values ​​of auto, CP-OFDM, and DFT-S-OFDM output by the network performance prediction model can be obtained.

[0080] Step S320: Configure the base station operating parameters to the optimal parameter values ​​corresponding to the predicted base station performance to complete the base station parameter optimization.

[0081] Understandably, this application uses KPIs to adjust log data, that is, it maps log data through KPIs. Compared with parameter optimization methods that directly optimize parameters based on whether KPIs are stable or whether log data is reasonable, this application creatively optimizes parameters through KPI mapping, which can reduce the difficulty of data acquisition, eliminate the need to traverse all data for judgment, and improve the efficiency of parameter optimization.

[0082] Based on the above embodiments, as an optional embodiment, the step of adjusting the preset log data distribution of the training base stations according to KPIs to obtain corrected log data includes:

[0083] Step S210: Based on KPIs, determine the preset number of log data samples for the training base stations;

[0084] The existing network has a huge number of base stations, and obtaining base station log data requires on-site network surveys. Therefore, obtaining large amounts of base station log data for model correction is extremely difficult and time-consuming. Considering the feasibility of large-scale implementation across the entire network, this application proposes a method based on KPI mapping to replace base station log data.

[0085] Step S220: Sampling is performed based on the sampling quantity to obtain the corrected log data.

[0086] Optionally, determining the sample size of the training base station's log data based on KPIs includes:

[0087] Step S211: Based on KPIs, obtain the first distribution corresponding to the performance indicators of the communication network;

[0088] Step S212: Based on the log data of the preset training base stations, obtain the second distribution corresponding to the first distribution;

[0089] Step S213: Determine the preset number of samples of log data from training base stations based on the first distribution and the second distribution.

[0090] In one embodiment, the communication network performance metric is path loss, and determining the sampling quantity of log data for the preset training base station based on the KPI includes:

[0091] Step S211: Based on KPI, obtain the first path loss distribution with the first loss interval;

[0092] Taking road loss as an example, the first road loss distribution of the KPI under multiple preset intervals can be a road loss distribution with an interval of 5dB, such as: dict_pl_kpi={[-90--95]:20%,[-95--100]:50%,[-100--105]:30%}.

[0093] Step S212: Based on the log data of the preset training base stations, obtain the second path loss distribution corresponding to the first path loss distribution;

[0094] Taking path loss as an example, the second indicator distribution of log data in multiple path loss intervals can be the path loss distribution of base station log data at 5dB intervals, such as dict_pl_log={[-90~95]:50% (5000 records), [-95~100]:30% (3000 records), [-100~105]:20% (2000 records)}.

[0095] Step S213: Based on the first path loss distribution and the second path loss distribution, determine the preset sampling quantity of log data for the training base station.

[0096] In one embodiment, determining the sampling quantity of the preset training base station log data based on the first path loss distribution and the second path loss distribution includes:

[0097] Step S2131: Determine the target road loss interval based on the ratio between the first road loss distribution and the second road loss distribution within the same road loss interval;

[0098] Specifically, the two distributions obtained by dict_pl_log and dict_pl_kpi are divided to obtain the following distributions: [-90~95]: 20% / 50% = 0.4, [-95~100]: 50% / 30% = 1.67, [-100~105]: 30% / 20% = 1.5. The preset performance range with the largest ratio is [-95~100].

[0099] Step S2132: Based on the distribution of KPIs in the target path loss interval and the number of samples of log data of the preset training base station in the target path loss interval, determine the total number of samples of log data of the preset training base station.

[0100] The total number of log data samples is obtained by dividing the number of base station log data samples in the preset path loss interval with the largest ratio by the proportion of KPI. The total number of samples = 3000 / 50% = 6000.

[0101] Step S2133: Based on the total number of samples and the first path loss distribution, determine the number of samples of the log data of the training base station in each path loss interval.

[0102] Calculate the number of samples for each road loss interval and sample the log data. The sampled log data will then match the KPI distribution. For example: [-90~95]: 6000*20%=1200; [-95~100]: 6000*50%=3000; [-100~105]: 6000*30%=1800.

[0103] Specifically, 1200 log entries are randomly sampled from 5000 entries in the [-90~95] range, 3000 log entries are sampled from 3000 entries in the [-95~100] range (no sampling), and 1800 log entries are randomly sampled from 2000 entries in the [-100~105] range. The final corrected base station log data is as follows: dict_pl_log={[-90~95]:20% (1200 entries), [-5~100]:50% (3000 entries), [-100~105]:30% (1800 entries)}.

[0104] It is understandable that when modifying base station parameters, this application takes into account the accuracy and timeliness of model establishment, as well as the operability of large-scale implementation across the entire network. By mapping base station log data through KPIs, the goal of efficient and accurate parameter optimization is achieved, thereby improving the accuracy and efficiency of parameter optimization.

[0105] The experimental results of this application are illustrated below.

[0106] Experiment 1

[0107] If the waveform adaptive function parameters of a base station need to be modified, the KPI is path loss distribution. The waveform adaptive function parameters are divided into auto, CP-OFDM, and DFT-S-OFDM. CP uplink supports dual streams, making it suitable for areas with a high concentration of users in the cell center. For edge users, DFT-S-OFDM is used; there is no power loss, but it only supports single streams, making it suitable for areas with weak coverage.

[0108] The waveform adaptive function parameters were modified for three base stations in the existing network. The gain change data before and after the modification, as well as the predicted gain change value output by the network performance prediction model, are shown in the table below.

[0109]

[0110]

[0111] Experiment 2

[0112] The PRB randomization function parameter is mainly used to solve interference problems in low-load scenarios. The optimal parameters will differ under different interference and load distributions, so it mainly corrects the distribution of interference and load. This parameter field is named `reserved250` and has three dimensions. The first dimension controls whether to enter the uplink enhanced PRB randomization optimization function based on the uplink cell coordination allocation strategy; 0 indicates off, 1 indicates on. The second dimension represents the interval exponent of the randomized RB starting positions when uplink enhanced PRB randomization is enabled. That is, the number of RBs between the randomized RB starting positions is 2 raised to the power of this parameter. Currently, the default value is set to 3, meaning there are 8 randomized RB starting positions within the three segments of the cell. The third dimension is the uplink PRB utilization threshold. The uplink enhanced PRB randomization optimization function is only performed when the utilization of this slot is lower than this threshold.

[0113] The PRB randomization function parameters were modified in three base stations on the existing network. The gain change data before and after the modification, as well as the predicted gain change value output by the network performance prediction model, are shown in the table below.

[0114]

[0115] Due to the complexity of the current network environment, modifying parameters based on conventional principles may not accurately reflect the actual situation. For example, in Industrial Park 11580000, disabling the uplink enhanced PRB randomization optimization function actually increased the uplink speed. This is likely because there are many edge users, requiring larger continuous RBs for output transmission, and PRB randomization may negatively impact performance. Therefore, the verification results show that this algorithm can evaluate the effect of parameter modifications without altering the parameters themselves, demonstrating high accuracy in gain assessment.

[0116] Figure 4 This is the second flowchart illustrating the base station parameter optimization method provided in this application, as shown below. Figure 4 As shown, the base station parameter optimization method provided in this application includes a training process and a testing process. The data source for model building is base station log data, and the data source used for parameter optimization is KPI. That is, a model is first trained using log data from a small batch of base stations. For base stations that need parameter optimization, the base station's KPI is first obtained, and then the KPI is used to adjust the distribution of log data in various dimensions of the model to obtain a new set of log data. The optimal parameters are then obtained using the new log data.

[0117] In a network, certain functions or parameters of a base station may exhibit different performance characteristics under different service scenarios. Therefore, configuring different parameters for different base stations to achieve better network performance is essential. This application can automatically learn the relationship between different service scenarios, parameters, and network performance. This model is then used to estimate the performance of other base stations under different parameters, thereby achieving parameter optimization. Considering the feasibility of large-scale implementation across the entire network, this application uses readily available data metrics for parameter optimization, while also incorporating the advantages of log data. Therefore, it is highly likely to be used by our company, other operators, and suppliers, and has excellent prospects for promotion in the ZTE region. Furthermore, since this application does not impose many restrictions on the underlying structure and logic, it also has good application prospects in other vendor regions.

[0118] The base station parameter optimization device provided in this application is described below. The base station parameter optimization device described below can be referred to in correspondence with the base station parameter optimization method described above.

[0119] Figure 5 This is a schematic block diagram of the base station parameter optimization device provided in this application, such as... Figure 5 As shown, this application also provides a base station parameter optimization device, including:

[0120] The acquisition module 510 is used to acquire the wireless key data indicators (KPIs) of the base station to be optimized.

[0121] The adjustment module 520 is used to adjust the preset log data distribution of the training base stations according to the KPI to obtain corrected log data;

[0122] The optimization module 530 is used to input the corrected log data into a pre-trained test model to predict the base station performance of the base station to be optimized, and to optimize the base station operating parameters of the base station to be optimized based on the prediction results. The test model is trained based on the log data of a preset training base station.

[0123] It should be noted that the base station parameter optimization device provided by the present invention can execute the base station parameter optimization method described in any of the above embodiments during specific operation, and has the technical effects corresponding to the method. This embodiment will not elaborate on this.

[0124] Figure 6 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 6As shown, the electronic device may include: a processor 610, a communications interface 620, a memory 630, and a communication bus 640, wherein the processor 610, the communications interface 620, and the memory 630 communicate with each other via the communication bus 640. The processor 610 can call logical instructions in the memory 630 to execute a base station parameter optimization method, which includes:

[0125] Obtain the key wireless data indicators (KPIs) of the base station to be optimized;

[0126] Adjust the preset log data distribution of the training base stations according to the KPIs to obtain corrected log data;

[0127] The corrected log data is input into a pre-trained test model to predict the base station performance of the base station to be optimized, and the base station operating parameters of the base station to be optimized are tuned based on the prediction results. The test model is trained based on the log data of a preset training base station.

[0128] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0129] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the base station parameter optimization method provided by the above methods, the method including:

[0130] Obtain the key wireless data indicators (KPIs) of the base station to be optimized;

[0131] Adjust the preset log data distribution of the training base stations according to the KPIs to obtain corrected log data;

[0132] The corrected log data is input into a pre-trained test model to predict the base station performance of the base station to be optimized, and the base station operating parameters of the base station to be optimized are tuned based on the prediction results. The test model is trained based on the log data of a preset training base station.

[0133] Furthermore, this application also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the base station parameter optimization method provided by the methods described above, the method comprising:

[0134] Obtain the key wireless data indicators (KPIs) of the base station to be optimized;

[0135] Adjust the preset log data distribution of the training base stations according to the KPIs to obtain corrected log data;

[0136] The corrected log data is input into a pre-trained test model to predict the base station performance of the base station to be optimized, and the base station operating parameters of the base station to be optimized are tuned based on the prediction results. The test model is trained based on the log data of a preset training base station.

[0137] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0138] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for optimizing base station parameters, characterized in that, include: Obtain the key wireless data indicators (KPIs) of the base station to be optimized; Adjust the preset log data distribution of the training base stations according to the KPIs to obtain corrected log data; The corrected log data is input into a pre-trained test model to predict the base station performance of the base station to be optimized, and the base station operating parameters of the base station to be optimized are tuned based on the prediction results. The test model is trained based on the log data of a preset training base station.

2. The base station parameter optimization method according to claim 1, characterized in that, The step of adjusting the preset log data distribution of the training base stations according to KPIs to obtain corrected log data includes: Based on KPIs, determine the sample size of log data from the preset training base stations; The corrected log data is obtained by sampling based on the sampling quantity.

3. The base station parameter optimization method according to claim 2, characterized in that, The step of determining the sampling quantity of log data for the preset training base station based on KPIs includes: Based on KPIs, the first distribution corresponding to the performance indicators of the communication network is obtained; Based on the log data of the preset training base stations, a second distribution corresponding to the first distribution is obtained; Based on the first distribution and the second distribution, determine the preset sampling quantity of log data for the training base station.

4. The base station parameter optimization method according to claim 3, characterized in that, The communication network performance metric is path loss. The determination of the sampling quantity of log data for the preset training base stations based on KPIs includes: Based on KPIs, the first path loss distribution with the first loss interval is obtained; Based on the log data of the preset training base stations, a second path loss distribution corresponding to the first path loss distribution is obtained; Based on the first path loss distribution and the second path loss distribution, the sampling quantity of log data for the preset training base stations is determined.

5. The base station parameter optimization method according to claim 4, characterized in that, The step of determining the sampling quantity of log data for the preset training base station based on the first path loss distribution and the second path loss distribution includes: The target road loss interval is determined based on the ratio between the first road loss distribution and the second road loss distribution within the same road loss interval. Based on the distribution of KPIs within the target road loss range and the number of samples of log data from the preset training base stations within the target road loss range, determine the total number of samples of log data from the preset training base stations. Based on the total number of samples and the first path loss distribution, the number of samples of log data from the preset training base station in each path loss interval is determined.

6. The base station parameter optimization method according to claim 1, characterized in that, The test model is trained on a preset boosting tree model based on log data from a preset training base station.

7. The base station parameter optimization method according to claim 1, characterized in that, The acquisition of the wireless key data indicators (KPIs) of the base station to be optimized includes: Based on the gateway, obtain the wireless key data indicators (KPIs) corresponding to the parameters to be optimized of the base station to be optimized; Correspondingly, the step of inputting the corrected log data into a pre-trained test model to predict the base station performance of the base station to be optimized, and optimizing the base station operating parameters of the base station to be optimized based on the prediction results, includes: The corrected log data is input into a pre-trained test model to obtain the predicted base station performance corresponding to any value of the parameter to be optimized. The base station operating parameters are configured to the optimal parameter values ​​corresponding to the predicted base station performance in order to complete the base station parameter optimization.

8. A base station parameter optimization device, characterized in that, include: The acquisition module is used to acquire the wireless key data indicators (KPIs) of the base station to be optimized. The adjustment module is used to adjust the preset log data distribution of the training base stations according to the KPIs to obtain corrected log data; The optimization module is used to input the corrected log data into a pre-trained test model, predict the base station performance of the base station to be optimized, and optimize the base station operating parameters of the base station to be optimized based on the prediction results. The test model is trained based on the log data of a preset training base station.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the base station parameter optimization method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the base station parameter optimization method as described in any one of claims 1 to 7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the base station parameter optimization method as described in any one of claims 1 to 7.

Images