CN111368384B - Method and device for predicting antenna engineering parameters - Google Patents

Abstract

The present application provides a method and device for predicting or correcting antenna engineering parameters. The method includes: first, based on an engineering parameter set including engineering parameters of the first device antenna in a first geographical area, and a set of engineering parameters including configuration data of the first device antenna Perform model training on the configuration data set and the measurement report set uploaded by the terminal to the antenna of the first device to obtain an antenna engineering parameter prediction model, and then send the configuration data of the antenna of the second device in the second geographical area and the terminal to the second device. The measurement report data uploaded by the antenna is input into the antenna engineering parameter prediction model to obtain the prediction result of the engineering parameter of the second device antenna or correct the engineering parameter of the second device antenna. This method enables more accurate prediction and correction of antenna engineering parameters at lower cost.

Description

Method and device for predicting antenna engineering parameters

technical field

The invention relates to the field of communication technologies, in particular to a training method and device for an antenna engineering parameter prediction model, and a method and device for predicting engineering parameters based on the antenna engineering parameter prediction model.

Background technique

Long Term Evolution (Long Term Evolution, LTE) is the long-term evolution of the Universal Mobile Telecommunications System (UMTS) technical standard formulated by the 3rd Generation Partnership Project (3rd Generation Partnership Project, 3GPP). As the scale of LTE network construction continues to expand and the number of LTE users increases dramatically, the accuracy of base station engineering parameters ("engineering parameters" may be referred to as "engineering parameters" in this article) becomes more and more important in daily network optimization and adjustment. The engineering parameters of the base station refer to parameters related to the radio frequency antenna of the base station in the wireless network planning, for example, the longitude and latitude, azimuth angle, downtilt angle of the antenna and so on. The accuracy of the engineering parameters of the base station is related to the accuracy of network data and network coverage, and has an extremely important impact on user terminal perception and network problem analysis.

However, the traditional verification of the accuracy of engineering parameters has always been a major shortcoming in network optimization. One of the existing methods is to test, analyze and verify the accuracy of engineering parameters by manually going to the station. However, this method of manually checking engineering parameters will have some artificially introduced errors, and it is difficult to ensure the timeliness and integrity of the engineering parameter data, so the checking efficiency is low and the accuracy is poor. Another existing practice is to install a positioning device at the base station, such as a global positioning system (Global Positioning System, GPS) device, obtain the position information of the base station radio frequency antenna through the positioning device and send it to the network management device, and then the network management device is based on the base station radio frequency antenna. The location information of the generated work parameters. However, this method will bring a large equipment overhead, and some base station equipment may be installed indoors. In this case, the positioning equipment has low indoor positioning accuracy, resulting in poor accuracy of the resulting work parameters.

SUMMARY OF THE INVENTION

The embodiments of the present invention provide a method and device for predicting engineering parameters of an antenna, which can overcome the defects of the prior art and realize a relatively accurate and low-cost generation scheme of engineering parameters.

In a first aspect, an embodiment of the present invention provides a method for predicting engineering parameters of an antenna. The method includes a method for training an antenna engineering parameter prediction model. The method specifically includes: acquiring a first engineering parameter set, a first configuration data set, and The first measurement report data set uploaded by the terminal to the first device antenna; wherein the first engineering parameter set includes the engineering parameters of the first device antenna, and the engineering parameters of the first device antenna include the first device at least one of antenna location data (eg, longitude, latitude, altitude, etc. of the first device antenna) and attitude data (eg, downtilt, azimuth, etc. of the first device antenna); the first set of configuration data Including the configuration data of the first device antenna, the configuration data of the first device antenna represents the configuration information of the network parameters of the first device antenna; the measurement report (Measurement Report, MR in the first measurement report data set) ) data includes the location data of the terminal (for example, the longitude, latitude, altitude, etc. of the terminal) and signal received power (Reference Signal Received Power, RSRP) data; wherein, the first device antenna may be within a first geographic area Any device antenna among the multiple device antennas; perform model training according to the engineering parameters of the first device antenna, the configuration data of the first device antenna, and the first measurement report data set, and obtain an antenna operating parameter prediction model (It may be referred to as an engineering parameter prediction model in this paper); the antenna engineering parameter prediction model is used to output the second device according to the second configuration data set and the second measurement report data set uploaded by the terminal to the antenna of the second device. The engineering parameters of the antenna. Wherein, the second device antenna may be a device antenna in a second geographic area, and the second geographic area may be different from the first geographic area.

It can be seen that the embodiment of the present invention can build a model for predicting the operating parameters of the device antenna based on ready-made sample data (eg, MR data, configuration data, operating parameter data, etc.) by means of model training. In this way, the subsequent application of the model will enable to obtain the predicted operating parameters of the equipment antenna with high reliability based on the MR data and configuration data. Therefore, implementing the technical solutions of the embodiments of the present invention can overcome the defects of the prior art, effectively reduce the acquisition cost of the equipment antenna engineering parameters, and improve the accuracy of the engineering parameters.

In this embodiment of the present invention, the first geographic area represents a geographic location range where one or more device antennas corresponding to the sample data used for model training are located. If the sample data used for model training corresponds to multiple device antennas, the multiple device antennas may be referred to as multiple device antennas in the first geographic area, and so on, the first device antennas in this embodiment may be referred to as is the first device antenna in the first geographical area, other device antennas in the multiple device antennas except the first device antenna may be referred to as other device antennas in the first geographical area, and so on.

Based on the first aspect, in a first implementation manner, the antenna engineering parameter prediction model includes an antenna engineering parameter generation model (herein may simply be referred to as an engineering parameter generation model); the engineering parameters based on the antenna of the first device , Perform model training on the configuration data of the first device antenna and the first measurement report data set to obtain an antenna operating parameter prediction model, including: according to the configuration data of the first device antenna and the first measurement report data set to obtain the first sample feature data of the first device antenna; the first sample feature data includes a plurality of cells belonging to the first device antenna or a remote radio unit (Radio Remote Unit, RRU) Signal received power data, and position data of the terminal corresponding to each signal received power data in the plurality of signal received power data; according to the configuration data of the first device antenna, the antenna type of the first device antenna is obtained; according to Model training is performed on the engineering parameters of the first device antenna and the first feature set to obtain the antenna engineering parameter generation model; the first feature set includes the first sample feature data and the first device antenna. Antenna type, the antenna engineering parameter generation model is used to output engineering parameters according to the input first feature set.

Wherein, the antenna engineering parameter generation model is, for example, a neural network (Neural Networks, NN) algorithm model. For example, the training process of the antenna parameter generation model can be expressed by the following formula:

(Latitude, Longtitude)=NN(Feature ₁₀₅₆ , AntennaType, W _nn1 )

Among them, Latitude represents the latitude value in the industrial parameter of the device antenna, Longtitude represents the longitude value in the industrial parameter of the device antenna, NN represents the neural network algorithm, Feature ₁₀₅₆ represents the first sample feature data, AntennaType represents the type of the device antenna, W _nn1 represents the model parameters in the work parameter generation model.

For different device antennas, the corresponding Latitude, Longtitude, Feature ₁₀₅₆ , and AntennaType data are also different. According to these data as the input data of the model generation model, the model can be trained to calculate W _nn1 (for example, this sample The gradient descent method can be used to calculate W _nn1 ) in , so as to obtain the trained model for generating parameters.

It can be seen that the embodiment of the present invention can perform data extraction based on ready-made sample data (such as MR data, configuration data, engineering parameter data, etc.), and obtain the input data of the engineering parameter generation model (such as first sample feature data, device antenna, etc.). The type data of the equipment antenna, the working parameter data of the equipment antenna, etc., so as to train the working parameter prediction model for predicting the equipment antenna based on these data (the working parameter prediction model here can be regarded as the working parameter generation model), and the application of this model will It can realize the generation of the engineering parameters of the device antenna, thereby obtaining the predicted engineering parameters with higher reliability. Therefore, the implementation of the embodiment of the present invention can be adapted to various sample data scenarios to effectively perform data screening, and improve the efficiency and accuracy of model training. Therefore, the accuracy of the subsequent prediction of the working parameters of the equipment antenna based on the model can be improved, and the cost of obtaining the working parameters of the equipment antenna can be effectively reduced.

Based on the first implementation of the first aspect, in a possible embodiment, when the amount of MR data in the training set is large, it occupies a large amount of memory, resulting in a large amount of MR data in a common antenna cell. However, the cell lists of common antennas corresponding to different device antennas are also different, and the number of cells is not fixed. In order to better perform model training (eg, avoid overfitting, improve computing speed and efficiency), the embodiment of the present invention may design a unified training data template for the MR data corresponding to the common antenna cells of different equipment antennas. The MR data corresponding to the common antenna cells of each device antenna may be screened and merged based on the training data template to obtain sample feature data of the common antenna cells of each device antenna (herein may be referred to as first sample feature data).

In a process of obtaining first sample feature data of a device antenna, the obtaining first sample feature data according to the configuration data of the first device antenna and the first measurement report data set includes: according to the The configuration data of the first device antenna is used to determine the cell or RRU belonging to the first device antenna; from the measurement report data set, the measurement report data corresponding to the cell or RRU is determined; according to the measurement corresponding to the cell or RRU Feature extraction is performed on the report data to obtain the first sample feature data.

For example, in a specific implementation, the value range of the RSRP value of each cell is {1, 4, 7, ..., 97}. For a single device antenna, such as device antenna 1, the value of device antenna 1 can be In the MR data of the shared-antenna cell, the RSRP value of the serving cell (for example, cell 1) is a certain predetermined value (for example, the predetermined value is 7, of course, any other value can also be selected) User Equipment (User Equipment) of all MR data. , the longitude and latitude of the UE) are averaged to obtain the center point, then the center point can be approximately regarded as the possible longitude and latitude position of the base station equipment. Then, from the center point, the rays are uniformly extended in multiple directions (eg, 8 directions). Then, from the MR data of each cell in the common antenna cell of the device antenna 1, it can be found that when the RSRP values of the serving cells are 1, 4, 7, ..., 97 (33 in total), the values of the rays from each direction can be obtained. The position with the same value is the closest MR data (if there is no such MR data, all 0s can be used to construct an MR data instead). In this way, 8×33=264 sets of MR data are found in total. Then, for each of the 264 sets of low-dimensional MR data, two of the features such as the longitude where the UE is located, the latitude where the UE is located, the height where the UE is located, and the angle of arrival (AOA) of the antenna of cell 1 are selected and extracted. One or more pieces of first sample characteristic data forming a common antenna cell. For example, when the four features of the longitude where the UE is located, the latitude where the UE is located, the altitude where the UE is located, and the AOA of cell 1 are simultaneously extracted, 4×264=1056 sub-features are generated, and the first sub-feature composed of such 1056 sub-features is generated. The sample feature data can be recorded as "Feature ₁₀₅₆ ".

It can be seen that the implementation of this embodiment can adapt to various sample data scenarios, improve the operation speed and efficiency of the model training process, and thereby train a better model for generating work parameters.

Based on the first aspect, in the second embodiment, in addition to the antenna working parameter generation model, the antenna working parameter prediction model also includes an antenna working parameter correction model (herein, it may also be referred to as a working parameter correction model); in this case , the engineering parameter set further includes engineering parameters of at least one other device antenna; the configuration data set further includes configuration data of the at least one other device antenna; the at least one other device antenna represents the multiple device antennas device antennas other than the first device antenna in ;

Correspondingly, after the model training is performed according to the engineering parameters of the antenna of the first device and the first feature set, and the model for generating the engineering parameters of the antenna is obtained, the method further includes: according to the configuration data set and the measurement A data set is reported, and second sample characteristic data of the first device antenna is obtained, where the second sample characteristic data includes a plurality of received signal power data of a cell or an RRU of the at least one other device antenna, and the plurality of The position data of the terminal corresponding to each signal received power data in the signal received power data and the signal received power data of the cell or RRU of the antenna of the first device; Prediction results of engineering parameters and prediction results of engineering parameters of the at least one other device antenna; perform model training according to the engineering parameter set and the second feature set to obtain the antenna engineering parameter correction model; the second feature set Including the second sample feature data, the prediction result of the engineering parameter of the first device antenna, and the prediction result of the engineering parameter of the at least one other device antenna, the antenna engineering parameter correction model is used according to the input second. The feature set outputs engineering parameters.

Wherein, the antenna working parameter correction model is, for example, a neural network (Neural Networks, NN) algorithm model. For example, the training process of the work parameter correction model can be expressed by the following formula:

(Latitude, Longtitude)=NN((Feature _{join_i} ,Feature _{basic_i} ),W _nn2 )

Among them, Latitude represents the latitude value in the industrial parameter of the device antenna i, Longtitude represents the longitude value in the industrial parameter of the device antenna i, NN represents the neural network algorithm, Feature _{join_i} represents the second sample feature data of the top N antenna of the device antenna, Feature _{basic_i} represents the predicted operating parameters of the (1+Top N) antenna group of the device antenna i, and W _nn2 represents the model parameters in the operating parameter correction model.

Therefore, for different device antennas, the corresponding data of Latitude, Longtitude, Feature _{join_i} and Feature _{basic_i} are also different. According to these data as the input data of the correction model, the model can be trained to calculate W _nn2 (for example, In this example, the gradient descent method can be used to calculate W _nn2 ), so as to obtain a trained model for correcting the working parameters.

It can be seen that the embodiment of the present invention can not only perform data extraction based on ready-made sample data (such as MR data, configuration data, engineering parameter data, etc.) The model obtains the prediction result of the engineering parameter of the first device antenna and the prediction result of the engineering parameter of the at least one other device antenna, thereby forming the input data of the antenna engineering parameter correction model (for example, the second sample feature data, the first The prediction result of the engineering parameters of a device antenna and the prediction result of the engineering parameters of the at least one other device antenna, the engineering parameter set, etc.), so as to further train the antenna engineering parameter correction model based on these data. The model can further correct the predicted working parameters of the working parameter generation model, so as to obtain the predicted working parameters with higher reliability. The antenna working parameter prediction model in this embodiment includes an antenna working parameter generation model and an antenna working parameter correction model. By applying this model, the working parameters of the device antenna can be generated and corrected, thereby obtaining predicted working parameters with high reliability. Implementing the embodiments of the present invention can be adapted to various sample data scenarios to effectively perform data screening, and improve the efficiency and accuracy of model training. Therefore, the accuracy of subsequent prediction of the working parameters of the equipment antenna based on the model can be improved, and the cost of obtaining the working parameters of the equipment antenna can be effectively reduced.

Based on the second implementation manner of the first aspect, in a possible embodiment, the at least one other device antenna is top N device antennas surrounding the first device antenna, and the top N device antennas represent all N device antennas most related to the first device antenna among the plurality of device antennas, where N is an integer greater than or equal to 1.

For example, in some embodiments, the top N antennas are among the plurality of device antennas around the first device antenna, the N device antennas that are the most peripheral in spatial distance from the first device antenna. In still other embodiments, the top N antennas are the N device antennas with the largest degree of signal overlap with the first device antenna among the plurality of device antennas surrounding the first device antenna; in still other embodiments, the top N antennas It is the N device antennas with which the terminal switches the most times among the surrounding multiple device antennas of the first device antenna, and so on.

The embodiment of the present invention performs the training of the working parameter correction model based on the most relevant N device antennas of the first device antenna, which is beneficial to the prediction of the first device antenna based on the working parameters of the most relevant N device antennas of the first device antenna Work parameters are corrected, so as to train a better work parameter correction model, and improve the budget speed and prediction accuracy of the work parameter correction model. Therefore, implementing this embodiment is beneficial to improve the reliability of the prediction result of the work parameter prediction model (ie, including the work parameter generation model and the work parameter correction model) as a whole.

Based on the second implementation manner of the first aspect, in a possible embodiment, the obtaining second sample feature data according to the configuration data set and the measurement report data set includes: according to the configuration data set, Determine the cells or RRUs that belong to the first device antenna, and the cells or RRUs that belong to the top N device antennas; set the cells or RRUs of the top N device antennas according to the first measurement report data set The cell or RRU of any device antenna corresponds to at least one measurement report data, and each measurement report data in the at least one measurement report data includes the signal received power data of the cell or RRU of any device antenna, the Signal received power data of a cell of a device antenna or RRU and position data of the terminal; extract features according to the cells of the top N device antennas or the measurement report data corresponding to the cells of each device antenna in the RRU or the RRU, and obtain the second sample feature data.

The second sample feature data represents the measurement features (or joint antenna measurement features) presented by different device antennas in the same UE measurement (ie, in the same UE geographic location).

For example, in a specific implementation, for the top N antennas of any device antenna, such as the topN antennas of device antenna A, a cell is determined according to the shared antenna cell list of each adjacent antenna of the device antenna A, and the cell is called A neighboring cell of device antenna A, then N neighboring cells will be determined correspondingly to N neighboring cells, and such N neighboring cells may be referred to as the top N neighboring cells of device antenna A. For example, for the first adjacent antenna in the top N antennas of the device antenna A, if there are multiple cells in the common antenna cell of the first adjacent antenna, exemplarily select the first adjacent antenna at a low latitude of the first adjacent antenna. In the MR data, the cell with the largest number of occurrences is used as the adjacent cell corresponding to the first adjacent antenna. By analogy, the neighboring cells corresponding to each neighboring antenna in the top N antennas, that is, the top N neighboring cells of the device antenna A can be determined respectively. Then, for any adjacent cell in the top N adjacent cells of any device antenna, for example, any adjacent cell in the top N adjacent cells of the device antenna A, M pieces of MR data can be selected from the multiple MR data of the device antenna A , and associate the M pieces of MR data with the neighboring cell. Wherein, any one of the M pieces of MR data includes the measurement feature information of the adjacent cell (for example, the RSRP of the adjacent cell, the AOA of the adjacent cell, etc.), exemplarily, the M pieces of low-dimensional MR data The RSRP values of the neighboring cells may be different, and the M is an integer greater than or equal to 1. In this way, the second sample feature data of the device antenna A can be extracted from the respective low-dimensional MR data of the M low-dimensional MR data, and each sample feature data can include the positioning information of the UE (such as the longitude where the UE is located). , the latitude where the UE is located, the altitude where the UE is located, etc.), two or more of the RSRP of the serving cell, the AOA of the serving cell, the RSRP of the neighboring cell, the AOA of the neighboring cell, and so on. That is, based on the above description, M low-dimensional MR data associated with any one of the topN neighboring cells may be determined, and M second sample feature data may be determined based on the M low-dimensional MR data. For the convenience of description, the M second sample feature data of the top N neighboring cells of the device antenna i may be recorded as "Featurejoin_i", and the device antenna i is any device antenna among the plurality of device antennas.

It can be seen that the implementation of this embodiment can adapt to various sample data scenarios, improve the operation speed and efficiency of the model training process, and thereby train a better model for correcting the working parameters.

Based on the first implementation manner and the second implementation manner of the first aspect, in a possible embodiment, before performing model training, feature information selection may also be performed on the MR data corresponding to the common antenna cell of the device antenna, so as to obtain Low-dimensional MR data for common antenna cells.

That is, in an optional embodiment, feature information selection of K dimensions is performed on MR data corresponding to each cell with a common antenna to obtain low-dimensional MR data, where K is an integer greater than or equal to 2.

For example, in some embodiments, the feature information of the K dimensions to be selected includes two or more of the following: location information where the UE is located, for example, including the longitude where the UE is located, the latitude where the UE is located, and optionally It also includes the height of the UE, etc.; the respective IDs of the cells with the same antenna, such as the ID of cell 1 (for example, cell 1 is the serving cell), the ID of cell 2... the ID of cell J, etc., J is an integer greater than or equal to 1 . The respective RSRPs of cells with a common antenna, for example, the RSRP of cell 1, the RSRP of cell 2, the RSRP of cell N, and so on. The respective AOAs of cells with a common antenna, such as the AOA of cell 1, the AOA of cell 2, the AOA of cell N, and so on.

Correspondingly, in this case, the obtaining the first sample feature data according to the configuration data of the first device antenna and the measurement report data set specifically includes: determining according to the configuration data of the first device antenna A cell or RRU belonging to the antenna of the first device; from the measurement report data set, determine the low-dimensional measurement report data corresponding to the cell or RRU; perform the measurement according to the low-dimensional measurement report data corresponding to the cell or RRU Feature extraction to obtain the first sample feature data.

Correspondingly, in this case, the obtaining the second sample feature data according to the configuration data set and the measurement report data set specifically includes: determining, according to the configuration data set, the antenna belonging to the first device. Cells or RRUs, and cells or RRUs belonging to the top N device antennas; according to the first measurement report data set, set the cells or RRUs of the topN device antennas or any device antenna in the RRU, respectively Corresponding to at least one low-dimensional measurement report data, each low-dimensional measurement report data in the at least one low-dimensional measurement report data includes signal received power data of a cell or RRU of any device antenna, the first The signal reception power data of the cell or RRU of the device antenna and the location data of the terminal. Feature extraction is performed according to the cells of the top N device antennas or the measurement report data corresponding to the cells or RRUs of each device antenna in the RRU to obtain the second sample feature data.

It can be seen that implementing the embodiments of the present invention is beneficial to reduce computational complexity and improve model training efficiency.

In a second aspect, an embodiment of the present invention provides a method for predicting engineering parameters of an antenna. The method includes a method for predicting engineering parameters based on an antenna engineering parameter prediction model. The method specifically includes: acquiring a second configuration data set, and sending the terminal to the A second measurement report data set uploaded by the second device antenna; wherein the second configuration data set includes configuration data of the second device antenna, and the configuration data of the second device antenna represents the configuration data of the second device antenna. Configuration information of network parameters; the measurement report data in the second measurement report data set includes the location data of the terminal (for example, the terminal's longitude, latitude, altitude, etc.) and signal received power data (referred to as RSRP data); wherein , the second device antenna may be any device antenna among multiple device antennas in the second geographic area; the second configuration data set and the second measurement report data set are input into the antenna operating parameter prediction model ( This paper may be referred to as the engineering parameter prediction model) to obtain the prediction result of the engineering parameters of the antenna of the second device; wherein, the antenna engineering parameter prediction model is based on the first engineering parameter set, the first configuration data set, and the terminal Obtained by training the first measurement report data set uploaded to the first device antenna; the engineering parameters of the second device antenna include the location data of the first device antenna (such as the longitude, latitude, altitude of the second device antenna) at least one of altitude, etc.) and attitude data (eg, downtilt, azimuth, etc. of a second device antenna); the first device antenna may be any of a plurality of device antennas within the first geographic area , the second geographic area may be different from the first geographic area.

The first engineering parameter set includes engineering parameters of the first device antenna, and the engineering parameters of the first device antenna include at least one of position data and attitude data of the first device antenna; the first device antenna A configuration data set includes configuration data of the first device antenna, where the configuration data of the first device antenna represents configuration information of network parameters of the first device antenna; measurement report data in the first measurement report data set The location data and signal reception power data of the terminal are included; the first device antenna is any device antenna among multiple device antennas in the first geographic area.

It can be seen that in the embodiment of the present invention, a pre-trained antenna operating parameter prediction model can be input into the model based on ready-made sample data (such as MR data, configuration data, etc.), and a more accurate operating parameter of the device antenna can be obtained. Ref. Therefore, implementing the technical solutions of the embodiments of the present invention can overcome the defects of the prior art, effectively reduce the acquisition cost of the equipment antenna engineering parameters, and improve the accuracy of the engineering parameters.

In the embodiment of the present invention, the second geographic area represents the geographic location range where one or more device antennas that need to predict the industrial parameters are located in practical applications. If there are multiple device antennas corresponding to the sample data used for the prediction of industrial parameters, the multiple device antennas may be referred to as multiple device antennas in the second geographical area, and so on, the second device antennas in this embodiment may be referred to as The device antenna is the second device antenna in the second geographical area, and other device antennas in the multiple device antennas except the second device antenna may be referred to as other device antennas in the second geographical area. Wherein, the geographic location range represented by the second geographic area may be different from the geographic location range represented by the first geographic area, that is, the second device antenna of the second geographic area may be different from the first geographic area. A device antenna, the plurality of device antennas of the second geographic area may be different from the plurality of device antennas of the first geographic area.

Based on the second aspect, in the first embodiment, the antenna engineering parameter prediction model includes an antenna engineering parameter generation model (herein, it may also be referred to as an engineering parameter generation model); the second configuration data set and the The second measurement report data set is input into the antenna engineering parameter prediction model, and the prediction result of the engineering parameters of the second device antenna is obtained, including: according to the second measurement report data set and the configuration of the second device antenna data to obtain the first sample feature data of the second device antenna; the first sample feature data of the second device antenna includes multiple signals belonging to the cell or remote radio unit RRU of the second device antenna receiving power data, and position data of the terminal corresponding to each signal receiving power data in the plurality of signal receiving power data; obtaining the antenna type of the second device antenna according to the configuration data of the second device antenna; The first sample feature data and the antenna type of the second device antenna are input into the antenna engineering parameter generation model to obtain a first prediction result of the engineering parameters of the second device antenna.

The antenna engineering parameter generation model is obtained by performing model training according to engineering parameters of the first device antenna in the first geographical area and a first feature set; wherein the first feature set includes all The first sample feature data of the first device antenna and the antenna type of the first device antenna, and the first sample feature data of the first device antenna includes multiple cells or RRUs belonging to the first device antenna. pieces of signal received power data, and position data of the terminal corresponding to each signal received power data in the plurality of signal received power data.

Wherein, the antenna engineering parameter generation model is, for example, a neural network (Neural Networks, NN) algorithm model. For example, the training process of the antenna parameter generation model can be expressed by the following formula:

(Latitude, Longtitude)=NN(Feature ₁₀₅₆ , AntennaType, W _nn1 )

Among them, Latitude represents the latitude value in the industrial parameter of the device antenna, Longtitude represents the longitude value in the industrial parameter of the device antenna, NN represents the neural network algorithm, Feature ₁₀₅₆ represents the first sample feature data, AntennaType represents the type of the device antenna, W _nn1 represents the model parameters in the work parameter generation model.

For different equipment antennas, the corresponding Feature ₁₀₅₆ and AntennaType data are also different. Inputting these data into the engineering parameter generation model can obtain the engineering parameter prediction results of different equipment antennas.

It can be seen that in this embodiment of the present invention, data extraction can be performed based on ready-made sample data (such as MR data, configuration data, etc.), and input data (such as first sample feature data, device antenna type data, etc.) of the work parameter generation model can be obtained. etc., thereby input to the trained engineering parameter prediction model (the engineering parameter prediction model here can be regarded as the engineering parameter generation model) based on these data, can realize the predicted engineering parameter of the generation equipment antenna. Therefore, implementing the embodiment of the present invention can It is suitable for various sample data scenarios to effectively extract data, improve the accuracy of predicting the working parameters of the second equipment antenna based on the model, and effectively reduce the acquisition cost of the second equipment antenna working parameters.

Based on the first implementation manner of the second aspect, in a possible embodiment, similar to the process of acquiring the first sample characteristic data in the first implementation manner of the first aspect, the reporting data according to the second measurement The process of obtaining the first sample feature data from the configuration data of the second device antenna includes: determining a cell or RRU belonging to the second device antenna according to the configuration data of the second device antenna; In the second measurement report data set, determine the measurement report data corresponding to the cell or the RRU; perform feature extraction according to the measurement report data corresponding to the cell or the RRU to obtain the first sample feature data.

It can be seen that the implementation of this embodiment can adapt to various sample data scenarios, improve the operation speed and efficiency of the model prediction process, so as to obtain prediction parameters with high accuracy.

Based on the second aspect, in the second embodiment, the antenna engineering parameter prediction model includes an antenna engineering parameter correction model (also referred to as an engineering parameter correction model in this document) in addition to the antenna engineering parameter generation model; the The second configuration data set further includes configuration data of at least one other device antenna; the at least one other device antenna represents a device antenna other than the second device antenna among the plurality of device antennas; the obtaining the second device antenna After obtaining the first prediction result of the engineering parameters of the device antenna, the method further includes: obtaining second sample feature data of the second device antenna according to the second configuration data set and the second measurement report data set, The second sample feature data of the second device antenna includes multiple signal received power data of the cell or RRU of the at least one other device antenna, and a terminal corresponding to each signal received power data in the multiple signal received power data The location data of the second device antenna and the signal received power data of the cell or RRU of the second device antenna; according to the antenna engineering parameter generation model, obtain the prediction result of the engineering parameters of the at least one other device antenna; The characteristic data, the first prediction result of the engineering parameter of the second device antenna, and the prediction result of the engineering parameter of the at least one other device antenna are input into the antenna engineering parameter correction model, and the engineering parameter of the second device antenna is obtained. The second prediction result for the parameter.

Wherein, the antenna engineering parameter correction model is obtained by performing model training according to a first engineering parameter set and a second feature set in the first geographical area; wherein, the first engineering parameter set includes the first engineering parameter set. Engineering parameters of the device antenna and engineering parameters of at least one other device antenna in the first geographical area; the second feature set includes second sample feature data of the first device antenna, The prediction result of engineering parameters and the prediction result of engineering parameters of at least one other device antenna in the first geographical area; the second sample characteristic data includes the cell of at least one other device antenna in the first geographical area or Multiple signal received power data of the RRU, as well as the location data of the terminal corresponding to each signal received power data in the multiple signal received power data, and the signal received power data of the cell of the first device antenna or the RRU; The prediction result of the engineering parameter of a device antenna and the prediction result of the engineering parameter of at least one other device antenna in the first geographical area are obtained according to the antenna engineering parameter generation model.

Wherein, the antenna working parameter correction model is, for example, a neural network (Neural Networks, NN) algorithm model. For example, the training process of the work parameter correction model can be expressed by the following formula:

(Latitude, Longtitude)=NN((Feature _{join_i} ,Feature _{basic_i} ),W _nn2 )

Among them, Latitude represents the latitude value in the industrial parameter of the device antenna i, Longtitude represents the longitude value in the industrial parameter of the device antenna i, NN represents the neural network algorithm, Feature _{join_i} represents the second sample feature data of the top N antenna of the device antenna, Feature _{basic_i} represents the predicted operating parameters of the (1+Top N) antenna group of the device antenna i, and W _nn2 represents the model parameters in the operating parameter correction model.

Therefore, for different equipment antennas, the corresponding data of Feature _{join_i} and Feature _{basic_i} are also different. According to these data input into the engineering parameter correction model, the predicted operating parameters of the equipment antenna can be predicted by the engineering parameter correction model. Further corrections are made to obtain a more reliable prediction result of the industrial parameters.

It can be seen that the embodiment of the present invention can not only perform data extraction based on ready-made sample data (such as MR data, configuration data, etc.), but also obtain the prediction of engineering parameters of the second device antenna according to the first implementation manner of the second aspect The results and the prediction results of the engineering parameters of the at least one other device antenna, thereby forming the input data of the antenna engineering parameter correction model (for example, the second sample feature data, the prediction results of the engineering parameters of the first device antenna and all The prediction result of the engineering parameters of the antenna of at least one other device, etc.), so as to input these data into the antenna engineering parameter correction model to further correct the predicted engineering parameters of the engineering parameter generation model, so as to obtain higher reliability. degree of forecasting parameters. The antenna working parameter prediction model in this embodiment includes an antenna working parameter generation model and an antenna working parameter correction model. By applying this model, it is possible to generate and correct the working parameters of the antenna of the second device, thereby obtaining a predicted working parameter with high reliability. . Implementing the embodiments of the present invention can be adapted to various sample data scenarios to effectively perform data screening, and improve the efficiency and accuracy of model training. Therefore, the accuracy of the subsequent prediction of the working parameters of the antenna of the second device based on the model can be improved, and the cost of obtaining the working parameters of the antenna of the second device can be effectively reduced.

Based on the second implementation manner of the second aspect, in a possible embodiment, the at least one other device antenna is top N device antennas surrounding the second device antenna, and the top N device antennas represent all N device antennas most related to the second device antenna among the plurality of device antennas, where N is an integer greater than or equal to 1.

For example, in some embodiments, the top N antennas are among the plurality of device antennas around the second device antenna, the N device antennas in the periphery with the most spatial distance from the second device antenna. In still other embodiments, the top N antennas are the N device antennas with the largest degree of signal overlap with the second device antenna among the surrounding multiple device antennas of the second device antenna; in still other embodiments, the top N antennas Among multiple device antennas surrounding the second device antenna, the N device antennas with which the terminal switches the most times, and so on.

In this embodiment of the present invention, based on the most relevant N equipment antennas of the second equipment antenna, the operating parameter correction model can be used to predict the operating parameters of the second equipment antenna, which is beneficial to the N equipment antennas based on the most relevant second equipment antennas. The predicted working parameters of the first equipment antenna are corrected, and the budget speed and prediction accuracy of the working parameter correction model are improved. Therefore, implementing this embodiment is beneficial to improve the reliability of the prediction result of the work parameter prediction model (ie, including the work parameter generation model and the work parameter correction model) as a whole.

Based on the second implementation manner of the second aspect, in a possible embodiment, similar to the process of acquiring the second sample feature data in the second implementation manner of the first aspect, the second configuration data set and The process of obtaining the second sample feature data from the second measurement report data set may include: determining, according to the second configuration data set, a cell or RRU belonging to the antenna of the second device, and a cell or RRU belonging to the top N devices The cell or RRU of the antenna; according to the second measurement report data set, the cell of the top N device antennas or the cell or RRU of any device antenna in the RRU is set to correspond to at least one measurement report data, and the at least one measurement Each measurement report data in the report data includes the signal received power data of the cell of any device antenna, the signal received power data of the cell of the second device antenna, and the position data of the terminal; according to the top N Feature extraction is performed on the cell of the device antenna or the measurement report data corresponding to the cell of each device antenna or the RRU in the RRU to obtain the second sample feature data.

It can be seen that the implementation of this embodiment can adapt to various sample data scenarios, improve the operation speed and efficiency of the model-based work parameter prediction process, and obtain work parameter prediction results with high reliability.

Based on the first implementation manner and the second implementation manner of the first aspect, in a possible embodiment, when performing engineering parameter prediction based on the model, feature information selection may also be performed on the MR data corresponding to the common antenna cell of the device antenna , so as to obtain low-dimensional MR data of the co-antenna cell.

That is, in an optional embodiment, feature information selection of K dimensions is performed on MR data corresponding to each cell with a common antenna to obtain low-dimensional MR data, where K is an integer greater than or equal to 2.

Correspondingly, in this case, the obtaining the first sample feature data according to the second measurement report data set and the configuration data of the antenna of the second device specifically includes: according to the configuration of the antenna of the second device data, determine the cell or RRU belonging to the second device antenna; from the second measurement report data set, determine the low-dimensional measurement report data corresponding to the cell or RRU; according to the low-dimensional corresponding to the cell or RRU Feature extraction is performed on the measurement report data of the first sample to obtain the first sample feature data.

Correspondingly, in this case, the obtaining the second sample feature data according to the second configuration data set and the second measurement report data set specifically includes: determining the affiliation according to the second configuration data set The cell or RRU of the second device antenna, and the cell or RRU belonging to the top N device antennas; according to the second measurement report data set, set any one of the cells or RRUs of the top N device antennas The cell or RRU of the device antenna corresponds to at least one low-dimensional measurement report data, and each measurement report data in the at least one low-dimensional measurement report data includes signal received power data of the cell of any device antenna, the Signal received power data of the cell of the second device antenna and position data of the terminal; feature extraction is performed according to the cell of the top N device antennas or the low-dimensional measurement report data corresponding to the cell of each device antenna in the RRU or the RRU , to obtain the second sample feature data.

It can be seen that the implementation of the embodiments of the present invention is beneficial to reduce the computational complexity and improve the efficiency of the model-based prediction of labor and participation.

In a third aspect, an embodiment of the present invention provides a computing device, where the computing device includes: a data acquisition module and a model training module. in:

A data acquisition module, configured to acquire a first engineering parameter set, a first configuration data set in a first geographical area, and a first data set uploaded by the terminal to the first device antenna among the plurality of device antennas in the first geographical area A measurement report data set; wherein the first engineering parameter set includes engineering parameters of the first device antenna, and the engineering parameters of the first device antenna include at least one of position data and attitude data of the first device antenna One; the first configuration data set includes configuration data of the antenna of the first device, and the configuration data of the antenna of the first device represents the configuration information of the network parameters of the antenna of the first device; the first measurement report data The centralized measurement report data includes position data and signal received power data of the terminal; the first device antenna is any device antenna among the multiple device antennas;

A model training module, configured to perform model training according to the engineering parameters of the first device antenna, the configuration data of the first device antenna, and the measurement report data set to obtain an antenna engineering parameter prediction model; the antenna engineering parameter prediction The model is used to output the data of the second device antenna according to the second configuration data set in the second geographical area by the terminal and the second measurement report data set uploaded by the terminal to the second device antenna in the second geographical area. engineering parameters.

Each functional module of the computing device can be specifically used to implement the method described in the first aspect.

In a fourth aspect, an embodiment of the present invention provides yet another computing device, including: a data acquisition module and an industrial parameter prediction module. in:

a data acquisition module, configured to acquire a second configuration data set in a second geographical area and a second measurement report data set uploaded by the terminal to the second device antenna in the second geographical area; wherein the second configuration data The set includes configuration data of the second device antenna, and the configuration data of the second device antenna represents configuration information of network parameters of the second device antenna; the measurement report data in the second measurement report data set includes the position data and signal received power data of the terminal; the second device antenna is any device antenna among multiple device antennas in the second geographic area;

an engineering parameter prediction module, configured to input the second configuration data set and the second measurement report data set into an antenna engineering parameter prediction model, and obtain a prediction result of the engineering parameters of the antenna of the second device; wherein, the The antenna engineering parameter prediction model is based on the first engineering parameter set in the first geographical area, the first configuration data set, and the first device antenna uploaded by the terminal to the first device antenna among the plurality of device antennas in the first geographical area. The measurement report data set is obtained by training; the engineering parameters of the second device antenna include at least one of position data and attitude data of the first device antenna.

Each functional module of the computing device can be specifically used to implement the method described in the second aspect.

In a fifth aspect, an embodiment of the present invention provides a computing device for training an antenna working parameter prediction model, where the computing device includes: a memory and a processor. The memory is used to store sample data and program codes, and the processor is used to execute the program codes to implement the method described in the first aspect.

In a sixth aspect, an embodiment of the present invention provides a computing device for predicting operating parameters based on an antenna operating parameter prediction model, the computing device including: a memory and a processor. The memory is used to store sample data and program codes, and the processor is used to execute the program codes to implement the method described in the second aspect

In a seventh aspect, an embodiment of the present invention provides a system, where the system includes the device described in the third aspect and the device described in the fourth aspect; or, the system includes the device described in the fifth aspect and the device described in the fourth aspect. The device described in the sixth aspect. in:

The computing device for training an antenna engineering parameter prediction model is specifically used to obtain a first engineering parameter set, a first configuration data set in a first geographical area, and a terminal to a plurality of devices in the first geographical area. The first measurement report data set uploaded by the first device antenna in the antenna; perform model training according to the engineering parameters of the first device antenna, the configuration data of the first device antenna, and the first measurement report data set, and obtain an antenna working parameter prediction model; inputting the antenna working parameter prediction model to the computing device for predicting the working parameter based on the antenna working parameter prediction model;

The computing device for predicting the working parameters based on the antenna working parameter prediction model is specifically used to obtain the second configuration data set in the second geographical area, the first configuration data set uploaded by the terminal to the antenna of the second device in the second geographical area. Two measurement report data sets; inputting the second configuration data set and the second measurement report data set into an antenna engineering parameter prediction model to obtain a prediction result of engineering parameters of the second device antenna.

In an eighth aspect, an embodiment of the present invention provides a non-volatile computer-readable storage medium; the computer-readable storage medium is used to store an implementation code of the method of the first aspect (or the second aspect). When the program code is executed by a computing device, the computing device is used for the method provided by any possible design of the first aspect (or the second aspect).

In a ninth aspect, an embodiment of the present invention provides a computer program product. The computer program product includes program instructions which, when executed by a computing device, perform the method provided by any possible design of the aforementioned first aspect (or second aspect). The computer program product can be a software installation package, which can be downloaded and stored in the computing device under the condition that the method provided by any of the possible designs of the first aspect (or the second aspect) needs to be used. The computer program product is executed on a processor to implement the method of the first aspect (or the second aspect).

It can be seen that the embodiment of the present invention can train a model for predicting industrial parameters, through the trained model for predicting industrial parameters (such as the predicting model for industrial parameters in this embodiment), based on ready-made sample data (such as MR data, configuration data, etc.) are input into the model to generate and correct the working parameters of the equipment antenna, so as to obtain the predicted working parameters with high reliability. Therefore, implementing the technical solutions of the embodiments of the present invention can overcome the defects of the prior art, effectively reduce the acquisition cost of the equipment antenna engineering parameters, and improve the accuracy of the engineering parameters.

Description of drawings

In order to more clearly describe the technical solutions in the embodiments of the present invention or the background technology, the accompanying drawings required in the embodiments or the background technology of the present invention will be described below.

1 is a schematic diagram of a system architecture provided by an embodiment of the present invention;

2 is a schematic structural diagram of a computing device provided by an embodiment of the present invention;

3 is a schematic structural diagram of a system for determining an operating parameter provided by an embodiment of the present invention;

4 is a schematic diagram of a model training process provided by an embodiment of the present invention;

5 is a schematic diagram of a model-based work parameter prediction process provided by an embodiment of the present invention;

6 is a schematic structural diagram of another work parameter determination system provided by an embodiment of the present invention;

7 is a schematic diagram of another model training process provided by an embodiment of the present invention;

8 is a schematic diagram of another model-based work parameter prediction process provided by an embodiment of the present invention;

FIG. 9 is a schematic structural diagram of another system for determining working parameters provided by an embodiment of the present invention;

10 is a schematic diagram of another model training process provided by an embodiment of the present invention;

11 is a schematic diagram of another model-based work parameter prediction process provided by an embodiment of the present invention;

12 is a schematic flowchart of a model training method provided by an embodiment of the present invention;

FIG. 13 is a schematic flowchart of a model-based method for predicting working parameters according to an embodiment of the present invention;

14 is a schematic flowchart of a training method for an engineering parameter prediction model provided by an embodiment of the present invention;

15 is a schematic diagram of some MR data provided by an embodiment of the present invention;

16 is a schematic diagram of MR data of common antenna cells of some base station antennas according to an embodiment of the present invention;

17 is a schematic diagram of dimension selection of low-dimensional MR data provided by an embodiment of the present invention;

18 is a schematic diagram of a scenario for acquiring first sample feature data provided by an embodiment of the present invention;

19 is a schematic flowchart of an engineering parameter prediction method based on an engineering parameter prediction model provided by an embodiment of the present invention;

20 is a schematic flowchart of another method for predicting an engineering parameter based on an engineering parameter prediction model provided by an embodiment of the present invention;

21 is a schematic structural diagram of another computing device provided by an embodiment of the present invention;

FIG. 22 is a schematic structural diagram of another computing device provided by an embodiment of the present invention.

Detailed ways

The embodiments of the present invention will be described below with reference to the accompanying drawings in the embodiments of the present invention. The terms used in the embodiments of the present invention are only used to explain specific embodiments of the present invention, and are not intended to limit the present invention. In the present invention, "at least one" means one or more, and "plurality" means two or more. "And/or", which describes the relationship of the associated objects, indicates that there can be three kinds of relationships, for example, A and/or B, it can indicate that A exists alone, A and B exist at the same time, and B exists alone, where A, B can be singular or plural. The character "/" generally indicates that the associated objects are an "or" relationship. "At least one item(s) below" or similar expressions thereof refer to any combination of these items, including any combination of single item(s) or plural items(s). For example, at least one item (a) of a, b, or c can represent: a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, c may be single or multiple .

Referring to FIG. 1, FIG. 1 is an exemplary schematic diagram of a system architecture of an embodiment of the present invention. FIG. 1 shows a terminal, a network element device, and a computing device involved in an embodiment of the present invention, wherein the terminal and the network element device can communicate with each other through a communication network of a certain air interface technology. The communication network of a certain air interface technology may include, for example: an existing 2G (2nd Generation, such as Global System for Mobile Communications, GSM) network, Worldwide Interoperability for Microwave Access (Worldwide Interoperability for Microwave Access, WiMAX) ) communication network, 3G (3rdGeneration, such as UMTS) network, Wideband Code Division Multiple Access (WCDMA) network, Time Division-Synchronous Code Division Multiple Access (TD-SCDMA) network, 4G (4th Generation, such as FDD LTE, TDD LTE) network and new radio access technology (New Radio Access Technology, New RAT) network, such as future 4.5G (such as LTE Advanced Pro) network, 5G (5th generation) network, etc. one or more of etc.

Wherein, a network element (Network Element, NE) device is a device used to communicate with one or more terminals (such as terminal 1, terminal 2, and terminal 3 in the figure). Specifically, the network element device may include a radio frequency antenna , which is used to communicate and interact with the terminal through the cell belonging to the radio frequency antenna. In this paper, the radio frequency antenna of the network element equipment may be referred to as the device antenna for short. It should be noted that in different application scenarios, the device antenna may also be called the device antenna. For the base station antenna, cell antenna, RRU antenna. The network element equipment can be a BTS (BaseTransceiver Station) in GSM or CDMA (Code Division Multiple Access), an NB (NodeB) in WCDMA, or an evolved NodeB (evolved Node B, eNB) in LTE, Or a relay station, or a vehicle-mounted device, or an access network device (gNodeB) in a future 5G network, or an access network device in a future evolved Public Land Mobile Network (Public Land Mobile Network, PLMN) network, etc. Herein, for the convenience of description, the technical solution of the present invention will be described by taking the network element device as the base station device as an example in the specific embodiments.

A terminal may include a relay (Relay), and a network element device that can perform data communication can be regarded as a terminal, which will be introduced in the present invention as a terminal in a general sense. A terminal may also be referred to as a mobile station, access terminal, user equipment, subscriber unit, subscriber station, mobile station, remote station, remote terminal, mobile device, user terminal, wireless communication device, user agent, or user equipment, or the like. The terminal can be a mobile phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a tablet computer, a PDA (Personal Digital Assistant, PDA), a wireless communication function Handheld devices or other processing devices connected to wireless modems, in-vehicle devices, wearable devices, and mobile stations in future 5G networks or terminals in future evolved PLMN networks, etc. The terminal can detect signals (such as cell signals) of one or more network element devices in the environment, and communicate and interact with the device antenna (may be referred to as the target device antenna) of the target network element device through the serving cell. Herein, for the convenience of description, the technical solution of the present invention will be described by taking the terminal as a user equipment (User Equipment, UE) as an example.

The cell mentioned in the following embodiments may be a cell corresponding to a base station device (such as a serving cell, a neighboring cell), and a cell may belong to a macro base station or a base station corresponding to a small cell (Small cell). Cells may include: Metro cells, Micro cells, Pico cells, Femto cells, etc. These small cells have the characteristics of small coverage and low transmit power, and are suitable for Provide high-speed data transmission services.

The computing device is used to execute the model training method described in the various embodiments of the present invention, and/or the model-based engineering parameter prediction method (referred to as the engineering parameter prediction method). The computing device may be an independent physical server, or a server cluster (such as a cloud computing service center) composed of multiple physical servers. The computing device may also be deployed at the base station device in the form of a processing chip or a controller, or deployed in a network management system (Network Management System, NMS) or a network element management system (Element Management System, EMS). The input data for the computing device to perform the model training method and/or the model-based prediction method may include data provided by the terminal and the base station device, and these data may be the data uploaded by the terminal to the base station device, and/or, the data configured in the terminal. data, and/or data obtained by measuring the base station equipment, and the like. For example, the data includes, for example, sample data for model training (data in a training set, such as configuration data, measurement report data, traffic statistics, engineering parameters, etc.), such as input data for model prediction (prediction Centralized data, such as configuration data, measurement report data, traffic statistics data, engineering parameters, etc.) and so on.

Referring to FIG. 2 , FIG. 2 is a schematic structural diagram of a computing device 102 according to an embodiment of the present invention. As shown in FIG. The processor 1021, the memory 1022 and the communication interface 1023 may be connected by a bus or in other ways (in FIG. 2, the connection by a bus is taken as an example). in:

Communication interface 1023 may be used to obtain data for model training, and/or data for model prediction. In a specific embodiment, the communication interface 1023 may be used to receive data sent by the terminal and/or the base station device. In yet another embodiment, the communication interface 1023 may be used to receive data transmitted over a wired or wireless network. In yet another embodiment, the communication interface 1023 may be used to obtain data in a non-volatile memory (eg, hard disk, USB flash drive, magnetic disk, flash memory, etc.).

The processor 1021 may be one or more central processing units (Central Processing Units, CPUs). FIG. 2 takes one processor as an example. In the case where the processor 1021 is a CPU, the CPU may be a single-core CPU, or Can be a multi-core CPU. The processor has a computing function and a function of controlling the operation of the computing device 102, and the processor can be configured with one or more of the following functions: executing the model training method described in the embodiments of the present invention; Model prediction method; run the engineering parameter determination system described in the embodiment of the present invention (this may be referred to as the engineering parameter determination system, or the antenna engineering parameter determination system); run the training module described in the embodiment of the present invention, for example, Engineering parameter generation model training module (herein may be referred to as the engineering parameter generation model training module), engineering parameter correction model training module (herein may be referred to as the engineering parameter correction model training module), engineering parameter prediction model training module (this article may referred to as the engineering parameter prediction model training module), etc., to implement the model training method described in the embodiment of the present invention; run the prediction module described in the embodiment of the present invention, for example, the engineering parameter generation module prediction module (this may be referred to as the For the engineering parameter generation module prediction module), the engineering parameter correction model prediction module (this article may be referred to as the engineering parameter correction model prediction module), the engineering parameter prediction model prediction module (this article may be referred to as the engineering parameter prediction model prediction module), etc. etc., to implement the model-based work parameter prediction method described in the embodiments of the present invention.

The memory 1022 includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or portable Read-only memory (Compact Disc Read-Only Memory, CD-ROM), the memory 1022 is used to store relevant program codes and data, and the program codes are, for example, for implementing the model training method and/or the model-based prediction process involved in the embodiment of the present invention. The code instructions of the method of the parameter, the data for example includes the data of the training set, and/or the sample data of the prediction set; also used to store the work parameter determination system, the work parameter determination system can be used to perform the learning of feature combinations through machine learning. , train related models (such as a work parameter generation model, a work parameter correction model, a work parameter prediction model, etc.), and generate or correct the work parameters based on the related models, and so on.

It should be noted that FIG. 2 is only a specific implementation manner of the embodiment of the present invention. In practical applications, the computing device 102 may further include more or less components, which is not limited here.

Various engineering parameter determination systems (may be referred to as an engineering parameter determination system for short, or an antenna engineering parameter determination system) involved in the embodiments of the present invention are further described below.

Referring to FIG. 3 , FIG. 3 shows a system 201 for determining an operating parameter according to an embodiment of the present invention. The system 201 for determining an operating parameter may include at least one of an operating parameter generation model training module 2011 and an operating parameter generation model prediction module 2012 . Among them, the engineering parameter generation model training module 2011 is used to train the engineering parameter generation model (herein may be referred to as the engineering parameter generation model, or as the antenna engineering parameter generation model) based on the sample data of the training set, so as to obtain after training The working parameter generation model. In a possible embodiment, the trained work parameter generation model may also be tested to verify whether the work parameter generation model meets the training target. The work parameter generation model training module 2011 can input the trained work parameter generation model into the work parameter generation model prediction module 2012, and the work parameter generation model training module 2011 can also send the relevant information of the sample feature combination to the work parameter generation model prediction module 2012 . The working parameter generation model prediction module 2012 is used to predict the working parameters based on the sample data of the prediction set, the relevant information of the sample feature combination, and the trained working parameter generation model, etc., to generate the working parameters of the target device antenna.

Referring to FIG. 4, for the training module 2011 of the work parameter generation model, in some embodiments, the sample data of the training set includes feature data X and label data Y of the first geographical area, X includes, for example, a measurement report data set and a configuration data set, Wherein, the measurement report data set includes multiple measurement report (Measurement Report, MR) data. In this paper, the MR data in the sample data can be the MR data reported by the UE to the base station equipment. The MR data is measured by a period or a specific event. measurement/quality measurement/UE internal measurement/UE location measurement/AOA measurement, etc.) as a unit, record the network environment characteristics at a certain time and a certain point in the call process. For example, each MR data may include the location information of the UE (such as geographic information such as the longitude and latitude information of the UE, altitude information, etc.), the signal received power (Reference Signal Received Power, RSRP) data of the serving cell detected by the UE, and the data of the neighboring cells. One or more of RSRP data, serving cell timing advance, UE transmit power headroom, angle of arrival (Angle of arrival, AOA), etc.; the configuration data set includes at least one configuration data. The configuration data may include configuration information of network parameters of the base station device, such as the antenna type of the device antenna, cell list, Physical Cell Identifier (PCI), Physical Random Access Channel (PRACH), and so on. Y includes an engineering parameter set of the radio frequency antenna of at least one base station device, and the engineering parameter set includes at least one engineering parameter (may be referred to as an engineering parameter for short). Here, each engineering parameter includes, for example, the longitude, latitude, azimuth, and downtilt angle of the antenna. at least one of etc.

As shown in FIG. 4 , the training module 2011 of the engineering parameter generation model can pre-build a basic model of the engineering parameter generation model (there is an unknown model parameter W1), and the engineering parameter generation model can be represented by Y=Model(X, W1), where Model represents the model function, and W1 represents the model parameters. Then, the work parameter generation model training module 2011 can use the sample data (MR data, configuration data, work parameters) to perform model training on the work parameter generation model, and calculate the model parameter W1, thereby obtaining the trained work parameter generation model.

Referring to FIG. 5, for the engineering parameter generation model prediction module 2012, in some embodiments, the sample data of the prediction set includes MR data (including UE's positioning information) and configuration data. The work parameter generation model prediction module 2012 includes the work parameter generation model trained by the work parameter generation model training module 2011 . As shown in FIG. 5 , the working parameter generation model prediction module 2012 can extract the data in the prediction set and input it into the trained working parameter generation model, thereby inputting the predicted results of the working parameters of the target device antenna in the second geographical area.

Referring to FIG. 6 , FIG. 6 shows another working parameter determination system 202 according to an embodiment of the present invention. The working parameter determination system 202 may include at least one of the working parameter correction model training module 2021 and the working parameter correction model prediction module 2022 One. Among them, the engineering parameter correction model training module 2021 is used to train the engineering parameter correction model (which may be referred to as the engineering parameter correction model, or the antenna engineering parameter correction model) based on the sample data of the training set, so as to obtain the trained engineering parameter correction model. See Correction Model. In a possible embodiment, a test can also be performed on the trained model for correcting the parameters of the workers to verify whether the model for generating the workers' parameters reaches the training target. The work parameter correction model training module 2021 can input the trained work parameter correction model into the work parameter correction model prediction module 2022, and the work parameter correction model training module 2021 can also send the relevant information of the sample feature combination to the work parameter correction model prediction module 2022 . The working parameter correction model prediction module 2022 is used to predict the working parameters based on the sample data of the prediction set, the relevant information of the sample feature combination, and the trained working parameter correction model, etc., so as to realize the correction of the working parameters of the target device antenna. .

Referring to FIG. 7 , for the training module 2021 of the engineering parameter correction model, in some embodiments, the sample data of the training set includes feature data X and label data Y of the first geographical area, X includes, for example, a measurement report data set, a configuration data set and A low-confidence engineering parameter set, Y includes a high-confidence engineering parameter set of at least one device antenna. Wherein, the engineering parameters in the low-credibility engineering parameter set are, for example, the engineering parameters obtained in a relatively rough manner (eg, the engineering parameters obtained through a manual measurement), and the engineering parameters in the high-confidence engineering parameter set are The parameters are, for example, work parameters obtained in a relatively accurate manner (eg, work parameters obtained by using multiple GPS measurements and multiple manual measurements).

As shown in FIG. 7, the engineering parameter correction model training module 2021 can pre-build a basic model of the engineering parameter correction model (there is an unknown model parameter W2), and the engineering parameter correction model can be represented by Y=Model(X, W2), where Model represents the model function, and W1 represents the model parameters. Then, the engineering parameter correction model training module 2021 can use the sample data (MR data, configuration data, low-credibility engineering parameters and high-confidence engineering parameters) to perform model training on the engineering parameter correction model, and calculate the model parameters. W2, so as to obtain the trained work parameter correction model.

Referring to FIG. 8, for the engineering parameter correction model prediction module 2022, in some embodiments, the sample data of the prediction set includes MR data (including UE's positioning information), configuration data and engineering parameters. The work parameter correction model prediction module 2022 includes the work parameter correction model trained by the work parameter correction model training module 2021 . As shown in FIG. 8 , the engineering parameter correction model prediction module 2022 can extract the data in the prediction set and input it into the trained engineering parameter correction model, thereby inputting the prediction result of the engineering parameters of the target device antenna in the second geographical area, thereby The prediction result can be used to compare with the engineering parameters in the prediction set, so as to realize the correction of the engineering parameters in the prediction set.

In still other embodiments of the present invention, the above two model training and model prediction processes may also be integrated. For example, the above-mentioned work parameter generation model training module 2011 and the above-mentioned work parameter correction model training module 2012 can be implemented comprehensively, and the above-mentioned work parameter generation model prediction module 2012 and the above-mentioned work parameter correction model prediction module 2022 can be comprehensively implemented.

Referring to FIG. 9 , FIG. 9 shows another working parameter determination system 203 according to an embodiment of the present invention. The working parameter determination system 203 may include at least one of the working parameter prediction model training module 2031 and the working parameter prediction model prediction module 2032 One. The engineering parameter prediction model training module 2031 is used to train the engineering parameter prediction model (herein may be referred to as the engineering parameter prediction model, or as the antenna engineering parameter prediction model) based on the sample data of the training set, so as to obtain the trained engineering parameter prediction model. Refer to the prediction model. Wherein, the work parameter prediction model training module 2031 may include a work parameter generation model training module and a work parameter correction model training module, and the work parameter prediction model prediction module 2032 may include a work parameter generation model prediction module and a work parameter correction model prediction module, that is, Said, the engineering parameter prediction model can be regarded as the synthesis of the engineering parameter generation model and the engineering parameter correction model. Therefore, the work parameter prediction model training module 2031 can input the trained work parameter prediction model into the work parameter prediction model prediction module 2032, and the work parameter prediction model training module 2031 can also send the relevant information of the sample feature combination to the work parameter prediction model for prediction Module 2032. The working parameter prediction model prediction module 2032 is used to predict the working parameters based on the sample data of the prediction set, the relevant information of the sample feature combination, and the trained working parameter prediction model, etc., to obtain the predicted results of the working parameters of the target device antenna.

Wherein, in a possible implementation, in the process of training its own model (engineering parameter correction model), the engineering parameter correction model training module can also use the model trained by the engineering parameter generation model training module (engineering parameter generation model) The output data implements its own model for training.

Similarly, in the process of engineering parameter prediction, the engineering parameter generation model prediction module can generate the engineering parameter data, and further input the generated engineering parameter data to the engineering parameter correction model prediction module, and then output the target through the engineering parameter correction model prediction module. The predicted results of the industrial parameters of the device antenna. Therefore, based on the embodiment of FIG. 9 , it is beneficial to obtain a highly reliable work parameter prediction result.

Referring to FIG. 10 , for the training module 2031 of the engineering parameter prediction model, in some embodiments, the sample data of the training set includes feature data X and label data Y of the first geographical area, X includes, for example, a measurement report data set and a configuration data set, Y includes a set of engineering parameters for at least one device antenna.

As shown in FIG. 10 , the engineering parameter prediction model training module 2031 can pre-build a basic model of the engineering parameter prediction model (there is an unknown model parameter W3), and the engineering parameter prediction model can be represented by Y=Model(X, W3), where Model represents the model function, W3 represents the model parameters of the engineering parameter prediction model, and W3 can be determined by the model parameter W1 of the engineering parameter generation model and the model parameter W2 of the engineering parameter correction model. That is to say, in some embodiments, the work parameter prediction model Y=Model(X, W3) can be regarded as the work parameter generation model Y=Model(X, W1) and the work parameter correction model Y=Model(X, W2 ) synthesis. Then, the working parameter prediction model training module 2031 can use the sample data (MR data, configuration data, and working parameters) to perform model training on the working parameter prediction model, and calculate the model parameter W3. In some embodiments, the sample data can be first input into the work parameter generation model training module of the work parameter prediction model training module 2031 to train the work parameter generation model, thereby determining the model parameter W1 of the work parameter generation model. The trained model for generating parameters is obtained. Therefore, the prediction result of the engineering parameters can be outputted to the engineering parameter correction model training module of the engineering parameter prediction model training module 2031 according to the trained engineering parameter generation model. The engineering parameter correction model training module trains the engineering parameter correction model based on the sample data and the prediction results of the engineering parameters to obtain and determine the model parameter W2 of the engineering parameter correction model. Therefore, through the above process, the training of the working parameter prediction model is realized, and the trained working parameter prediction model is obtained.

Referring to FIG. 11 , for the engineering parameter prediction model prediction module 2032, in some embodiments, the sample data of the prediction set includes MR data (including UE's positioning information) and configuration data. The work parameter prediction model prediction module 2032 includes the work parameter prediction model trained by the work parameter prediction model training module 2031 (that is, it can be regarded as including the trained work parameter generation model and the trained work parameter correction model). As shown in FIG. 11 , the engineering parameter prediction model prediction module 2012 can extract the data in the prediction set, input it to the engineering parameter generation model prediction module of the engineering parameter prediction model prediction module 2032, and use the trained engineering parameter generation model to output the engineering parameter generation model. The predicted results of the parameters are sent to the engineering parameter correction model prediction module of the engineering parameter prediction model training module 2031, and the trained engineering parameter correction model is utilized to output the highly reliable engineering parameter prediction results of the target device antenna in the second geographical area. .

It can be understood that in some of the model training and model prediction processes described above, the work parameters may include at least one of the position data and attitude data of the device antenna, the position data of the device antenna such as the longitude and/or latitude of the device antenna, The attitude data of the device antenna, such as the azimuth angle and/or downtilt angle of the device antenna, etc., so, in practical applications, it can be based on different industrial parameter content (for example, according to the longitude and latitude of the antenna, or according to the longitude and latitude and azimuth of the antenna. angle, or based on the latitude and longitude of the antenna, azimuth and downtilt, etc.), and train different models.

Based on the work parameter determination system described above, a model training method provided by an embodiment of the present invention is described below. Please refer to FIG. 12. FIG. 12 is a schematic flowchart of a model training method provided by an embodiment of the present invention. The method may include but is not limited to the following steps:

Step 301: Acquire an engineering parameter set, a configuration data set of at least one device antenna in a first geographical area, and a measurement report data set uploaded by a terminal to a target device antenna in the at least one device antenna in the first geographical area.

In this embodiment of the present invention, the first geographic area represents a geographic location range where one or more device antennas corresponding to the sample data used for model training are located. If there are multiple device antennas corresponding to the sample data used for model training, the multiple device antennas may be referred to as multiple device antennas in the first geographical area, and by analogy, the target device antennas in this embodiment may be referred to as is the target device antenna in the first geographical area, and other device antennas in the plurality of device antennas except the target device antenna may be referred to as other device antennas in the first geographical area.

In some embodiments, the model that needs to be trained in this embodiment of the method includes an engineering parameter generation model. For example, the model can be regarded as the engineering parameter generation model described in the foregoing embodiment of FIG. 4 . In this case:

The engineering parameter set includes the engineering parameters of the target device antenna in the first geographical area (the target device antenna in the first geographical area may also be referred to as the first device antenna). The configuration data set may include configuration data of the target device antenna, and the specific content of the configuration data includes configuration information of network parameters of the target device antenna, such as the antenna type of the target device antenna, cell list and other information.

The measurement report data set includes multiple measurement report data (that is, MR data) uploaded by the terminal to the target device antenna in the first geographical area, and the specific content of the MR data may include the location information of the UE (such as the longitude and latitude information of the UE, Geographic information such as altitude information), the RSRP data of the serving cell detected by the UE, and the RSRP data of the neighboring cell detected by the UE. Optionally, it may further include at least one of the AOA of the serving cell, the AOA of the neighboring cell, the timing advance of the serving cell, the UE transmit power headroom, and the like.

In still other embodiments, when the models to be trained in this embodiment of the method include an engineering parameter generation model and an engineering parameter correction model, the models can be regarded as the engineering parameter prediction model described in the foregoing embodiment of FIG. 10, for example. In case:

The engineering parameter set includes the working parameters of the device antennas of multiple base station devices in the first geographic area, and the working parameters of the device antennas of the multiple base station devices include the working parameters of the target device antenna and the working parameters of at least one other device antenna. , the target device antenna mentioned here can be understood as any device antenna in the multiple base station devices.

The configuration data set may include configuration data of one or more base station devices (that is, including configuration data of the antenna of the target device), and the specific content of the configuration data includes configuration information of network parameters of the base station device, such as the antenna of the device antenna. Type, cell list, etc.

The measurement report data set includes multiple MR data uploaded by the terminal to the antenna of the target device in the first geographical area, and optionally, may also include MR data uploaded by other terminals to the antenna of the at least one other device.

Likewise, the specific content of the MR data may include location information of the UE (such as geographic information such as longitude and latitude information of the UE, altitude information, etc.), RSRP data of the serving cell detected by the UE, and RSRP data of neighboring cells detected by the UE. Optionally, it may further include at least one of the AOA of the serving cell, the AOA of the neighboring cell, the timing advance of the serving cell, the UE transmit power headroom, and the like.

Step 302: Perform model training according to the engineering parameters, configuration data and measurement report data set of the antenna of the target device to obtain an engineering parameter prediction model.

In some embodiments, when the model to be trained in this embodiment of the method includes an engineering parameter generation model, the training process of the model may include the following: obtaining a first Sample feature data (for example, Feature ₁₀₅₆ described in the embodiment of FIG. 14 ), wherein the first sample feature data includes multiple signal received power data of cells belonging to the antenna of the target device, and the multiple The position data of the terminal corresponding to each signal received power data in the signal received power data; the antenna type of the target device antenna is obtained according to the configuration data; the model is carried out according to the engineering parameters of the target device antenna and the first feature set training to obtain the work parameter generation model; wherein, the first feature set includes the first sample feature data and the antenna type of the antenna of the target device, and the work parameter generation model is used according to the input first The feature set outputs engineering parameters.

For the specific training process of the working parameter generation model, reference may also be made to the foregoing embodiment in FIG. 4 and the detailed description of steps 501 to 507 in the following embodiment of FIG. 14 , which are not repeated here for the sake of brevity of the description.

In some embodiments, when the model to be trained in this embodiment of the method includes a work parameter generation model and a work parameter correction model, the training process of the model may include training on the work parameter generation model and on the work parameter correction model. training.

Wherein, the training process of the work parameter generation model may include: obtaining first sample feature data (for example, Feature ₁₀₅₆ described in the embodiment of FIG. 14 ) according to the configuration data and the measurement report data set, wherein, The first sample characteristic data includes a plurality of received signal power data of a cell belonging to the antenna of the target device, and position data of a terminal corresponding to each signal received power data in the plurality of received signal power data; according to the configure data to obtain the antenna type of the target device antenna; perform model training according to the engineering parameters of the target device antenna and a first feature set to obtain the engineering parameter generation model; wherein the first feature set includes the The first sample feature data and the antenna type of the target device antenna, and the engineering parameter generation model is used to output engineering parameters according to the input first feature set.

Similarly, for the specific training process of the work parameter generation model, reference may also be made to the foregoing embodiment in FIG. 4 and the detailed description of steps 501 to 507 in the following embodiment of FIG. 14 .

The training process of the engineering parameter correction model may include: obtaining second sample feature data (for example, Feature _{join_i} described in the embodiment of FIG. 14 ) according to the configuration data set and the measurement report data set, the second sample feature data The sample feature data includes multiple signal received power data of the cell of the at least one other device antenna, and location data of the terminal corresponding to each signal received power data in the multiple signal received power data and the cell of the target device antenna The received signal power data; according to the engineering parameter generation model, the prediction result of the engineering parameter of the target device antenna and the prediction result of the engineering parameter of the at least one other device antenna are obtained; according to the engineering parameter set and the second Model training is performed on the feature set to obtain the engineering parameter correction model; the second feature set includes the second sample feature data, the prediction result of the engineering parameters of the target device antenna, and the engineering parameters of the at least one other device antenna. The parameter prediction result, the engineering parameter correction model is used to output engineering parameters according to the input second feature set.

For the specific training process of the work parameter correction model, reference may also be made to the foregoing embodiment in FIG. 7 and the detailed description of steps 508 to 512 in the following embodiment of FIG. 14 , which are not repeated here for the sake of brevity of the description.

It can be seen that the embodiment of the present invention can build a model for predicting the working parameters of the equipment antenna based on the ready-made sample data (such as MR data, configuration data, working parameter data, etc.) Realize the generation and correction of the working parameters of the equipment antenna, so as to obtain the predicted working parameters with high reliability. Therefore, implementing the technical solutions of the embodiments of the present invention can overcome the defects of the prior art, effectively reduce the acquisition cost of the equipment antenna engineering parameters, and improve the accuracy of the engineering parameters.

Based on the working parameter determination system described above, a model-based working parameter prediction method provided by an embodiment of the present invention is described below. Please refer to FIG. 13. FIG. 13 is a schematic flowchart of a model-based work parameter prediction method provided by an embodiment of the present invention. The method may include but is not limited to the following steps:

Step 401: Acquire the measurement report data set and the configuration data set uploaded by the terminal to the antenna of the target device in the second geographical area.

In the embodiment of the present invention, the second geographic area represents the geographic location range where one or more device antennas that need to predict the industrial parameters are located in practical applications. If there are multiple device antennas corresponding to the sample data used for the prediction of industrial parameters, the multiple device antennas may be referred to as multiple device antennas in the second geographic area, and so on, the target device in this embodiment may be referred to as The antenna is the target device antenna in the second geographical area, and other device antennas in the multiple device antennas except the target device antenna may be referred to as other device antennas in the second geographical area. The geographic location range represented by the second geographic area may be different from the geographic location range represented by the first geographic area, that is, the target device antenna of the second geographic area may be different from the target device antenna of the first geographic area Antennas, the plurality of device antennas of the second geographic area may be different from the plurality of device antennas of the first geographic area.

In some embodiments, the model used for predicting the working parameters required by this method embodiment includes a pre-trained working parameter generation model, for example, the model can be regarded as the working parameter generation model described in the foregoing embodiment of FIG. 5 , In this situation:

The configuration data set may include configuration data of the target device antenna, and the specific content of the configuration data includes configuration information of network parameters of the target device antenna (the target device antenna in the second geographical area may also be referred to as the second device antenna). , such as the antenna type of the target device antenna, the cell list, and so on.

The measurement report data set includes a plurality of MR data uploaded by the terminal to the target device antenna in the second geographical area, and the specific content of the MR data may include the location information of the UE (such as geographic information such as the latitude and longitude information of the UE, altitude information, etc.). ), the RSRP data of the serving cell detected by the UE, and the RSRP data of the neighboring cell detected by the UE. Optionally, it may further include at least one of the AOA of the serving cell, the AOA of the neighboring cell, the timing advance of the serving cell, the UE transmit power headroom, and the like.

In still other embodiments, when the models used for predicting the working parameters required by this embodiment of the method include a pre-trained model for generating the working parameters and a model for correcting the working parameters, the models may be, for example, the model shown in the foregoing embodiment in FIG. 11 . Describes the engineering parameter prediction model, in this case:

The configuration data set may include configuration data of one or more base station devices (that is, including configuration data of the antenna of the target device), and the specific content of the configuration data includes configuration information of network parameters of the base station device, such as the antenna of the device antenna. Type, cell list, etc.

The measurement report data set includes multiple MR data uploaded by the terminal in the second geographic area to the antenna of the target device, and optionally, may also include MR data uploaded by other terminals to the antenna of the at least one other device.

Likewise, the specific content of the MR data may include location information of the UE (such as geographic information such as longitude and latitude information of the UE, altitude information, etc.), RSRP data of the serving cell detected by the UE, and RSRP data of neighboring cells detected by the UE. Optionally, it may further include at least one of the AOA of the serving cell, the AOA of the neighboring cell, the timing advance of the serving cell, the UE transmit power headroom, and the like.

Step 402: Input the measurement report data set and the configuration data set into the trained engineering parameter prediction model, and obtain the prediction result of the engineering parameters of the target device antenna.

In some embodiments, in the case where the model required for predicting the working parameter in the embodiment of the method includes a pre-trained working parameter generating model, the process of predicting the working parameter based on the model may include the following: according to the measurement reporting the data set and the configuration data of the antenna of the target device to obtain first sample feature data; the first sample feature data includes multiple signal received power data of cells belonging to the antenna of the target device, and the multiple The position data of the terminal corresponding to each signal received power data in the signal received power data; the antenna type of the target device antenna is obtained according to the configuration data; the first sample feature data and the target device antenna's antenna type are obtained; The antenna type is input to the trained engineering parameter generation model to obtain a first prediction result of engineering parameters of the target device antenna.

For the specific process of predicting the working parameters based on the working parameter generation model, reference may also be made to the foregoing embodiment in FIG. 5 and the detailed description of steps 601 to 607 in the following embodiment of FIG. 19 .

In some embodiments, in the case that the model required for predicting the working parameter in the embodiment of the method includes a pre-trained model for generating the working parameter and a model for correcting the working parameter, the process of predicting the working parameter based on the model may include: The working parameter generation model generates working parameters and further corrects the generated working parameters based on the working parameter correction model to obtain a final working parameter prediction result.

Wherein, the process of generating the working parameters based on the working parameter generation model may include: obtaining first sample feature data according to the measurement report data set and the configuration data of the antenna of the target device; the first sample feature data includes affiliation Multiple signal received power data of the cell of the target device antenna, and location data of the terminal corresponding to each signal received power data in the multiple signal received power data; according to the configuration data, obtain the target device antenna Antenna type; input the first sample feature data and the antenna type of the target device antenna into the trained engineering parameter generation model to obtain a first prediction result of the engineering parameters of the target device antenna.

Similarly, for the specific process of predicting engineering parameters based on the engineering parameter generation model, reference may also be made to the foregoing embodiment in FIG. 5 and the detailed description of steps 601 to 607 in the following embodiment of FIG. 19 .

Wherein, the specific process of further correcting the generated working parameters based on the working parameter correction model to obtain the final working parameter prediction result may include: obtaining second sample feature data according to the configuration data set and the measurement report data set , the second sample feature data includes multiple signal received power data of the cell of the at least one other device antenna, and the location data of the terminal corresponding to each signal received power data in the multiple signal received power data and the Signal received power data of the cell of the target device antenna; according to the engineering parameter generation model, obtain the prediction result of the engineering parameter of at least one other device antenna in the second geographical area; The first prediction result of the engineering parameter of the target device antenna and the prediction result of the engineering parameter of at least one other device antenna in the second geographical area are input into the trained engineering parameter correction model, and the engineering parameter of the target device antenna is obtained. The second prediction result for the parameter.

For the specific process of working parameter prediction based on the working parameter correction model, reference may also be made to the foregoing embodiment of FIG. 8 and the detailed description of steps 608 to 611 of the following embodiment of FIG. 19 , which are not repeated here for brevity of the description.

It can be seen that in this embodiment of the present invention, a pre-trained model for predicting industrial parameters can be input into the model based on ready-made sample data (such as MR data, configuration data, etc.), so as to realize the generation and correction of the device antenna. engineering parameters, so as to obtain the predicted engineering parameters with high reliability. Therefore, implementing the technical solutions of the embodiments of the present invention can overcome the defects of the prior art, effectively reduce the acquisition cost of the equipment antenna engineering parameters, and improve the accuracy of the engineering parameters.

Based on the above-mentioned working parameter determination system 203 , the following describes a training method of a working parameter prediction model provided by an embodiment of the present invention. The working parameter includes the longitude and latitude of the antenna as an example for description. Please refer to FIG. 14. FIG. 14 is a schematic flowchart of a training method for an engineering parameter prediction model provided by an embodiment of the present invention. The method may include but not limited to the following steps:

Step 501: Acquire an engineering parameter set, a configuration data set, and an MR data set of a first geographic area.

Specifically, the model training module (for example, the engineering parameter generation model training module in the engineering parameter prediction model training module, or a separate engineering parameter generation model training module) can obtain the engineering parameter set of the first geographical area from the sample data of the training set , configuration dataset and MR dataset.

Wherein, the engineering parameter set may include the engineering parameters of the device antennas of one or more base station devices in the first geographical area, and here, each engineering parameter includes, for example, the longitude and latitude of the antenna. The configuration data set includes configuration data of one or more base station devices, and the configuration data may include configuration information of network parameters of the base station devices, such as antenna types of device antennas, cell lists, and the like. The MR dataset includes a plurality of MR data. The multiple MR data may be reported by one or more UEs to the one or more base station devices, and each MR data may include, for example, location information of the UE (such as geographic information such as latitude and longitude information of the UE, altitude information, etc.) , at least two of the RSRP data of the serving cell detected by the UE, the RSRP data of the neighboring cell detected by the UE, the AOA of the serving cell, the AOA of the neighboring cell, the timing advance of the serving cell, the UE transmit power headroom, etc. one or more. For example, a certain UE may periodically (eg, every 10 seconds) upload MR data to the target device antenna in the first geographic area, then a plurality of MRs within a preset time period may be collected from the target device antenna Data is input to the training set.

Step 502: Obtain a list of cells with a common antenna according to the configuration data.

In a specific embodiment, multiple MR data in the MR data set may include MR data related to radio frequency antennas of different base station equipment. In order to extract the MR data respectively associated with each base station equipment from these MR data, the The plurality of MR data are subjected to data extraction and data classification.

In some embodiments, the number of cells owned by different device antennas is also different according to different antenna types and site conditions of the device antennas. Then the model training module can obtain configuration data from the prediction set, classify multiple cells according to the device antenna, and obtain a list of cells belonging to the same device antenna. For example, the cell list of the common antenna of the radio frequency antenna of the base station device 1 (referred to as device antenna 1 for short) may include Cell-1, Cell-2, and so on. The cell list of the common antenna of the radio frequency antenna of the base station device 2 (referred to as device antenna 2 for short) may include Cell-3, Cell-4, Cell-5, and so on.

It should be noted that, in other possible embodiments of the present invention, data extraction and data classification may also be performed on the multiple MR data in other manners. For example, the number of remote radio units (Remote Radio Unit, RRU) owned by different device antennas is also different. A cell can correspond to multiple RRUs, then the model training module can obtain configuration data from the prediction set, classify multiple RRUs according to the device antenna, and obtain a list of RRUs belonging to the same device antenna. The RRU list for the same antenna can contain ID of one or more RRUs. For example, the cell list of the common antennas of the device antennas of the base station device A may include RRU-1, RRU-2, RRU-3, RRU-4, etc. In these embodiments, the subsequent steps may use RRU correspondence The MR report corresponding to the cell replaces the MR data corresponding to the cell, and the specific implementation process of the model training can be similar to the implementation process of the cell, which will not be described in detail later.

Step 503: Obtain MR data corresponding to the cells with the same antenna according to the MR data set and the list of cells with the same antenna.

Specifically, the model training module can classify multiple MR data in the MR data set, and obtain the MR data corresponding to each cell with the same antenna (that is, the same antenna) by correlating the serving cell ID of the MR data with the cell ID of the common antenna (referred to as cell MR data).

For example, in some embodiments, the MR data set includes MR data 1, MR data 2, MR data 3, and MR data 4 as shown in FIG. 15, wherein the serving cell related to MR data 1 is cell 1, and the MR data The serving cell related to 2 is cell 2, the serving cell related to MR data 3 is cell 3, and the serving cell related to MR data 4 is cell 4. If the common antenna cell list of the radio frequency antenna of base station device 1 (ie device antenna 1) includes cell 1 and cell 2, and the common antenna cell list of the radio frequency antenna of base station device 2 (ie device antenna 2) includes cell 3 and cell 4, then These MR data may be associated with the cell IDs in the common-antenna cell list through the serving cell ID of the MR data, so as to obtain MR data corresponding to the common-antenna cells. As shown in FIG. 16 , MR data 1 can be associated with cell 1 of device antenna 1, and MR data 2 can be associated with cell 2 of device antenna 1, so the MR data of the common antenna cell of device antenna 1 can include MR data 1 and MR data 2. MR data 3 may be associated with cell 3 of device antenna 2 , and MR data 4 may be associated with cell 4 of device antenna 2 , so the MR data of the common antenna cell of device antenna 2 may include MR data 3 and MR data 4 .

In particular, in a possible embodiment, if the number of MR data of a common antenna cell of a certain antenna is lower than a certain value, then the MR data of this antenna can be discarded to avoid affecting the accuracy of the trained model and avoiding affecting the Accuracy of predicted parameters.

Step 504: Select feature information for the MR data corresponding to the common-antenna cell to obtain low-dimensional MR data of the common-antenna cell.

In an optional embodiment, in order to reduce computational complexity and improve model training and prediction efficiency, K-dimensional feature information selection is performed on the MR data corresponding to each cell with a common antenna to obtain low-dimensional MR data, where K is greater than or equal to An integer of 2.

For example, in some embodiments, the feature information of the K dimensions to be selected includes two or more of the following:

The location information where the UE is located, such as the longitude where the UE is located, the latitude where the UE is located, and optionally the altitude where the UE is located, etc.;

The respective IDs of cells with a common antenna, such as the ID of cell 1 (eg, cell 1 is a serving cell), the ID of cell 2... the ID of cell J, etc., J is an integer greater than or equal to 1.

The respective RSRPs of cells with a common antenna, for example, the RSRP of cell 1, the RSRP of cell 2, the RSRP of cell N, and so on.

The respective AOAs of cells with a common antenna, such as the AOA of cell 1, the AOA of cell 2, the AOA of cell N, and so on.

For example, as shown in FIG. 17 , FIG. 17 exemplarily shows the low-dimensional MR data 1 obtained after the feature information selection is performed on the MR data 1 described in FIG. 15 . Compared with MR data 1 described in FIG. 15 , MR data 1 reduces the feature information of UE transmit power headroom, serving cell timing advance, etc., so the low-dimensional MR data 1 shown in FIG. 17 is compared with FIG. 15 . The described MR data 1 has fewer data dimensions. It can be understood that, correspondingly, the MR data 2, MR data 3 and MR data 4 described in FIG. 15 can also obtain low-dimensional MR in a similar way to the MR data 1 data. Therefore, implementing this embodiment is conducive to reducing the amount of stored data, improving the computational complexity of subsequent model training and model prediction, and improving computational efficiency.

It should be noted that the above example shown in FIG. 17 is only used to explain the technical solution of the present invention and is not a limitation. In practical applications, lower-dimensional MR data can be selected as required. For example, in an optional embodiment, the low-dimensional MR data can also be designed not to include at least one of the altitude where the UE is located and the AOA of the cell, etc. .

Step 505: Calculate the first sample characteristic data of the common-antenna cell according to the low-dimensional MR data of the common-antenna cell.

In a specific embodiment, when the amount of MR data in the training set is large, it occupies a large amount of memory, which will result in a large amount of MR data in cells with a common antenna. However, the cell lists of common antennas corresponding to different device antennas are also different, and the number of cells is not fixed. In order to better perform model training (eg, avoid overfitting, improve computing speed and efficiency), the embodiment of the present invention may design a unified training data template for the MR data corresponding to the common antenna cells of different equipment antennas. The MR data corresponding to the common antenna cells of each device antenna may be screened and merged based on the training data template to obtain sample feature data of the common antenna cells of each device antenna (herein may be referred to as first sample feature data).

For example, a method of data filtering and merging is introduced below, so that each antenna obtains sample feature data of the same dimension.

Referring to FIG. 18, in a specific implementation, the value range of the RSRP value of each cell is, for example, {1, 4, 7, ..., 97}. For a single device antenna, such as device antenna 1, the value of device antenna 1 In the MR data of the common antenna cell (optional, such as low-dimensional MR data), the RSRP value of the serving cell (for example, cell 1) is a predetermined value (for example, the predetermined value is 7, of course, any other value can be selected. The longitude and latitude of all the MR data of the UE are averaged to obtain the center point, then the center point can be approximately regarded as the possible longitude and latitude position of the base station equipment. Then, starting from the center point, it extends evenly in multiple directions (for example, in the illustration, there are eight directions of east, west, south, north, southeast, northeast, southwest, and northwest, of course, it can also be in other directions) Rays in each direction are exemplarily marked with position points corresponding to RSRPs with values of 1, 4, 7, . Then, in the MR data (optional, such as low-dimensional MR data) of the common antenna cell of the device antenna 1, the RSRP values of the serving cell (the serving cell of the device antenna 1 include, for example, cell 1) are 1, 4, and 7. The latitude and longitude of the UE corresponding to ,...,97 are mapped to the area shown in FIG. 18 , and each circle in the illustration represents the latitude and longitude of the UE corresponding to the cell of a certain value of RSRP.

Then, in an example, from the low-dimensional MR data of each cell in the common-antenna cell of the device antenna 1, it is possible to find that the RSRP values of the serving cells are 1, 4, 7, . . . , 97 (33 in total) When , the MR data closest to the position points with the same value of rays in each direction are respectively (if there is no such MR data, all 0s can be used to construct an MR data instead). That is to say, for a cell with an RSRP value of 1, 8 most suitable MR data are found corresponding to 8 directions respectively; for a cell with an RSRP value of 4, 8 most suitable MR data corresponding to 8 directions are also found. Appropriate MR data; for a cell with an RSRP value of 7, the 8 most suitable MR data are also found corresponding to 8 directions... For a cell with an RSRP value of 97, the pages correspond to 8 directions to search The 8 most suitable MR data are obtained, and so on. In this way, a total of 8×33=264 sets of low-dimensional MR data are found.

Then, for each of the 264 sets of low-dimensional MR data, two or more of the features such as the longitude where the UE is located, the latitude where the UE is located, the height where the UE is located, and the AOA of cell 1 are selected to form a common antenna cell. The first sample feature data of . For example, when the four features of the longitude where the UE is located, the latitude where the UE is located, the altitude where the UE is located, and the AOA of cell 1 are simultaneously extracted, 4×264=1056 sub-features are generated (for the convenience of description, such 1056 sub-features are The constituted first sample feature data may be recorded as "Feature ₁₀₅₆ "); for another example, when only two features of the longitude where the UE is located and the latitude where the UE is located are extracted, 2×264=528 sub-features are generated.

For example, the first sample characteristic data table of the common antenna cell of the device antenna 1 generated according to the above method is shown in Table 1:

Table 1

It should be noted that the above examples are only used to explain the technical solutions of the present invention and are not intended to limit them.

Step 506: Obtain the antenna type of the device antenna according to the configuration data.

Specifically, the model training module can obtain the type (AntennaType) of the device antenna to which the common antenna cell belongs according to the configuration data in the configuration data set. For example, the antenna type of the device antenna 1 can be obtained according to the configuration data of the device antenna 1. The configuration data of the device obtains the antenna type of Antenna 2, and so on.

Step 507: Train an engineering parameter generation model according to the engineering parameter set, the first sample feature data and the antenna type.

In a specific embodiment, the model training module can use a machine learning algorithm to build an engineering parameter generation model (for example, a neural network algorithm model), according to the working parameters of the equipment antenna in the training set, the first sample feature obtained by step 505. The data and the type of device antenna obtained through step 506 are used to perform model training on the work parameter generation model.

For example, the training process of the work parameter generation model can be expressed by the following formula:

(Latitude, Longtitude)=NN(Feature ₁₀₅₆ , AntennaType, W _nn1 )

Among them, Latitude represents the latitude value in the industrial parameter of the device antenna, Longtitude represents the longitude value in the industrial parameter of the device antenna, NN represents the neural network algorithm, Feature ₁₀₅₆ represents the first sample feature data obtained by the above-mentioned Table 1, AntennaType represents the type of device antenna, and W _nn1 represents the model parameters in the work parameter generation model.

Therefore, the model training module conducts model training based on a large number of sample data in the training set. For different device antennas, the corresponding Latitude, Longtitude, Feature ₁₀₅₆ , and AntennaType data are also different. According to these data, the input data of the model is generated as working parameters. , the model can be trained to calculate W _nn1 (for example, in this example, the gradient descent method can be used to calculate W _nn1 ), so as to obtain the trained model for generating parameters.

It can be seen that, through the above steps 501 to 507, the training module for the generation model of the parameter in the training module for the prediction model of the parameter can realize the training of the generation model of the parameter according to the sample data. Subsequently, the work parameter correction model training module in the work parameter prediction model training module will perform related model training on the work parameter correction model through the following steps 508 to 512 .

Step 508: Obtain predicted operating parameters of multiple device antennas.

In some specific embodiments, after the trained working parameter generation model is obtained through step 507, the predicted working parameters of the device antenna can be obtained according to the sample data and the trained working parameter generation model, for example, by latitude and longitude prediction For example, the prediction result of the latitude and longitude (Latitude, Longtitude) of the device antenna can be obtained.

It can be understood that when the sample data of the training set includes the sample data of multiple device antennas, such as the sample data including the device antenna 1 (specifically including the configuration data of the device antenna 1, the working parameters of the device antenna 1, the terminal to the device antenna 1 MR data reported), the sample data of the device antenna 2 (specifically including the configuration data of the device antenna 2, the working parameters of the device antenna 2, the MR data reported by the terminal to the device antenna 2), etc., then according to the sample data of each device antenna and the trained work parameter generation model, the prediction results of the longitude and latitude of each device antenna can be obtained.

For example, the prediction results of the longitude and latitude of each device antenna obtained by the trained work parameter generation model are shown in Table 2:

Table 2

It should be noted that the above examples are only used to explain the embodiments of the present invention and not limit them.

Step 509: Calculate the top N antennas of the device antennas according to the list of common antenna cells of the device antennas and the low-dimensional MR data of the device antennas.

Among them, for the target device antenna, top N antennas represent the N device antennas most related to the target device antenna based on the preset rule among the surrounding device antennas of the target device antenna, and N is an integer greater than or equal to 1. For example, in some embodiments, the top N antennas are the N device antennas with the closest spatial distance from the target device antenna among the multiple device antennas around the target device antenna; in still other embodiments, the top N antennas are the target device antennas Among the multiple device antennas around the antenna, the N device antennas with the greatest degree of signal overlap with the target device antenna; in some other embodiments, the top N antennas are among the multiple device antennas around the target device antenna, and the number of times the terminal switches the most of N device antennas, and so on.

A method for determining the top N antennas of each device antenna is given below.

Through the above step 502, a list of common antenna cells of any device antenna among the multiple device antennas can be obtained. Through the above step 504, low-dimensional MR data of a common antenna cell of any device antenna among the multiple device antennas can be obtained. Then, the ID of the device antenna corresponding to other cells except the serving cell in the low-dimensional MR data of the target device antenna can be determined according to the common antenna cell list of any device antenna among the multiple device antennas. For example, for the low-dimensional MR data 1 of the device antenna 1 shown in the above-mentioned embodiment of FIG. 17 , it can be determined that the rate of cell 2 belongs to the device antenna 1 according to the common antenna cell list of any device antenna among the multiple device antennas, The cell 3 rate belongs to device antenna 2, the cell 4 rate belongs to device antenna 2, and so on. If there are only two cells in the low-dimensional MR data 1 that belong to the device antenna 2 (such as cell 3 and cell 4), then the device antenna 2 is said to appear twice in the low-dimensional MR data 1 of the device antenna 1, and other devices The same goes for the case of antennas (such as device antenna 3, etc.). Similarly, for other low-dimensional MR data of device antenna 1 (such as low-dimensional MR data 2, etc.), it is also possible to find out the IDs of device antennas corresponding to other cells except the serving cell, and then count the other The number of times the device antennas (such as device antenna 2, device antenna 3, etc., which can also be called adjacent antennas of device antenna 1) appear in other low-dimensional MR data of device antenna 1. In this way, the number of occurrences of each adjacent antenna in all low-dimensional MR data of the device antenna 1 can be counted.

It can be understood that, based on the above description, in all low-dimensional MR data of any device antenna of the multiple base station devices, the number of occurrences of each adjacent antenna of the device antenna can be further counted.

For example, the number of occurrences of each adjacent antenna of any device antenna of the plurality of device antennas in all low-dimensional MR data of the base station may be as shown in Table 3:

table 3

Then, the N adjacent antennas with the most occurrences of each device antenna are selected as the Top N antennas of the device antenna.

For example, in Table 3 above, the adjacent antennas of the device antenna have been sorted based on the number of occurrences. It can be seen that for the device antenna 1, the IDs of the adjacent antennas sorted according to the number of occurrences are Antenna-2, Antenna-13...Antenna-16. For device antenna 2, the IDs of adjacent antennas sorted according to the number of occurrences are Antenna-5, Antenna-1...Antenna-25, and so on.

Then, in a possible embodiment, the adjacent antennas of each device antenna may be as shown in Table 4 below:

Table 4

It should be noted that the above examples are only used to explain the technical solutions of the embodiments of the present invention, but are not limitations.

Step 510: Obtain the predicted operating parameters of each adjacent antenna in the top N antennas according to the predicted operating parameters of the multiple device antennas and the top N antennas of each device antenna.

It can be understood that through the above step 508, the predicted operating parameters of each device antenna (for example, Table 2) have been obtained. Therefore, based on the ID of any device antenna, the predicted operating parameters of any device antenna can be obtained; based on the topN of any device antenna The ID of each adjacent antenna in the antenna can obtain the predicted operating parameters of each adjacent antenna in the top N antennas of any device antenna. As shown in Table 5 below:

table 5

The above Table 5 shows the predicted parameters of any device antenna (or target device antenna) and its corresponding Top N antenna, then the corresponding Top N antenna of any device antenna can be collectively referred to as (1+top N) antenna group . Therefore, through this step 510, the predicted operating parameters of the (1+top N) antenna group of any device antenna can be obtained. For convenience of description, in this embodiment of the present invention, the predicted operating parameter of the (1+Top N) antenna group of the device antenna i may be recorded as "Feature _{basic_i} ", and the device antenna i is any device antenna among the plurality of device antennas.

Step 511: Obtain second sample feature data of the top N antennas according to the low-dimensional MR data of the base station equipment and the predicted operating parameters of each adjacent antenna in the top N antennas of the base station equipment.

The second sample feature data represents the measurement features (or joint antenna measurement features) presented by different device antennas in the same measurement of the UE (that is, in the same UE geographic location).

A method for obtaining the second sample characteristic data of the top N antennas is described below.

First, for the top N antennas of any device antenna, such as the top N antennas of device antenna A, a cell is determined according to the list of common antenna cells of each adjacent antenna of the device antenna A, and the cell is called a neighbor of device antenna A. cell, then the N adjacent antennas will respectively determine N adjacent cells, such N adjacent cells may be referred to as the top N adjacent cells of the device antenna A. For example, for the first adjacent antenna in the top N antennas of the device antenna A, if there are multiple cells in the common antenna cell of the first adjacent antenna, the first adjacent antenna at the low latitude of the first adjacent antenna may be selected exemplarily. In the MR data, the cell with the largest number of occurrences is used as the adjacent cell corresponding to the first adjacent antenna. By analogy, the neighboring cells corresponding to each neighboring antenna in the top N antennas, that is, the top N neighboring cells of the device antenna A can be determined respectively. That is to say, based on the above description, the top N neighboring cells of any device antenna can be determined.

Then, for any neighboring cell in the top N neighboring cells of any device antenna, for example, any neighboring cell in the top N neighboring cells of device antenna A, M can be selected from multiple low-dimensional MR data of device antenna A and associate the M low-dimensional MR data with the neighboring cell. Wherein, any one of the M low-dimensional MR data includes the measurement feature information of the neighboring cell (for example, the RSRP of the neighboring cell, the AOA of the neighboring cell, etc.), exemplarily, the M low-dimensional MR data The RSRP values of neighboring cells in the MR data of M may be different, and M is an integer greater than or equal to 1. In this way, the second sample feature data of the device antenna A can be extracted from the respective low-dimensional MR data of the M low-dimensional MR data, and each sample feature data can include the positioning information of the UE (such as the longitude where the UE is located). , the latitude where the UE is located, the altitude where the UE is located, etc.), two or more of the RSRP of the serving cell, the AOA of the serving cell, the RSRP of the neighboring cell, the AOA of the neighboring cell, and so on. That is, based on the above description, M low-dimensional MR data associated with any one of the top N neighboring cells can be determined, and M second sample feature data can be determined based on the M low-dimensional MR data. For the convenience of description, the M second sample feature data of the top N neighboring cells of the device antenna i may be recorded as "Feature _{join_i} ", and the device antenna i is any device antenna among the plurality of device antennas.

For example, the second sample characteristic data of the top N antenna of the device antenna i is shown in Table 6 below:

Table 6

It should be noted that the above examples are only used to explain the technical solutions of the embodiments of the present invention, but are not limited.

Step 512 , train an operating parameter correction model according to the engineering parameter set, the predicted operating parameters of the (1+top N) antenna group, and the second sample feature data of the top N antennas.

In a specific embodiment, the model training module (the work parameter correction model training module in the work parameter prediction model training module) can use a machine learning algorithm to construct a work parameter correction model (for example, a neural network algorithm model), according to the equipment in the training set. The working parameters of the antenna, the predicted working parameters of the (1+top N) antenna group obtained in step 510, and the second sample feature data of the top N antennas obtained in step 511 are used to perform model training on the working parameter correction model.

For example, the training process of the work parameter correction model can be expressed by the following formula:

(Latitude, Longtitude)=NN((Feature _{join_i} ,Feature _{basic_i} ),W _nn2 )

Among them, Latitude represents the latitude value in the industrial parameter of the device antenna i, Longtitude represents the longitude value in the industrial parameter of the device antenna i, NN represents the neural network algorithm, Feature _{join_i} represents the second sample feature data of the top N antenna of the device antenna, Feature _{basic_i} represents the predicted operating parameters of the (1+Top N) antenna group of the device antenna i, and W _nn2 represents the model parameters in the operating parameter correction model.

Therefore, the model training module performs model training based on a large number of sample data in the training set. For different device antennas, the corresponding data of Latitude, Longtitude, Feature _{join_i} and Feature _{basic_i} are also different. These data are used as the input of the model to correct the working parameters. Data, the model can be trained to calculate W _nn2 (for example, the gradient descent method can be used to calculate W _nn2 in this example), so as to obtain a trained model for correcting the working parameters.

It can be seen that in this embodiment, through steps 501 to 507, it is possible to train the engineering parameter generation model according to the sample data, and to calculate the model parameter W _nn1 of the engineering parameter generation model, and through steps 508 to 512, it can realize The parameter correction model is trained, and the model parameter W _nn2 of the work parameter correction model is calculated. It can be understood that the engineering parameter prediction model described in the embodiment of the present invention can be regarded as including an engineering parameter generation model and an engineering parameter correction model, and the model parameters of the engineering parameter prediction model can be regarded as including the model parameters of the engineering parameter generation model W _nn1 and The model parameter W _nn2 of the engineering parameter correction model, so based on the above steps 501 to 512, the training of the engineering parameter prediction model is completed.

It should also be noted that, in some possible embodiments, if only the work parameter generation model training module as shown in FIG. 4 is used to train the work parameter generation model (this situation can also be regarded as the work parameter prediction model training module) Including only the model training module for generating the working parameters), then the implementation process of the model training can also be referred to the description of the above steps 501-507. For the sake of brevity of the description, this article will not describe it in detail.

It should also be noted that, in some possible embodiments, if only the work parameter correction model training module as shown in FIG. 7 is used to train the work parameter correction model (this situation can also be regarded as the work parameter prediction model training module) only includes the model training module for working parameter correction), then the implementation process of the model training can also refer to the description of the above steps 508 to 512, the difference is (if there is a difference from the above step 508), the Feature in the training process _{basic_i} comes from the low-confidence engineering parameter set in the sample data (X) of the training set, that is, Feature _{basic_i} represents the engineering parameters of the (1+Top N) antenna group of the device antenna i in the low-confidence engineering parameter set; (Latitude, Longtitude) The high-confidence engineering parameter set in the sample data (Y) from the training set, that is, Latitude represents the latitude value in the engineering parameter of the device antenna i in the high-confidence engineering parameter set, and Longtitude represents The high-confidence engineering parameter set is the longitude value in the engineering parameter of the device antenna i. The engineering parameters in the low-credibility engineering parameter set are, for example, the engineering parameters obtained in a rougher manner (for example, the engineering parameters obtained through a manual measurement), and the engineering parameters in the high-confidence engineering parameter set are, for example. It is an industrial parameter obtained by a relatively accurate method (for example, an industrial parameter obtained by using multiple GPS measurements and multiple manual measurements). Then, based on the descriptions of the above steps 508 to 512, those skilled in the art will similarly understand the method for training the work parameter correction model by the work parameter correction model training module as shown in FIG. No further details.

It can be seen that in this embodiment of the present invention, a model for predicting the operating parameters of the device antenna can be constructed based on ready-made sample data (such as MR data, configuration data, and operating parameter data, etc.) by means of model training (such as in this embodiment). The engineering parameter prediction model), and the application of this model will realize the generation and correction of the engineering parameters of the equipment antenna, so as to obtain the predicted engineering parameters with high reliability. Therefore, implementing the technical solutions of the embodiments of the present invention can overcome the defects of the prior art, effectively reduce the acquisition cost of the equipment antenna engineering parameters, and improve the accuracy of the engineering parameters.

Based on the above-mentioned working parameter determination system 203, a method for predicting working parameters based on a working parameter prediction model provided by an embodiment of the present invention is described below. The working parameters include the longitude and latitude of the antenna as an example for description. Please refer to FIG. 19 . FIG. 19 is a schematic flowchart of a method for predicting an engineering parameter based on an engineering parameter prediction model provided by an embodiment of the present invention. The method may include but is not limited to the following steps:

Step 601: Acquire an MR dataset and a configuration dataset of a second geographic area.

Specifically, the model prediction module (for example, the engineering parameter generation model training module in the engineering parameter prediction model prediction module, or a separate engineering parameter generation model prediction module) can obtain the MR data set in the second geographical area from the data in the prediction training set And the configuration data set of the base station device.

Wherein, the second geographic area may be different from the first geographic area described in the foregoing step 501 . That is, the base station device located in the second geographic area and the base station device located in the first geographic area are not the same base station device.

The configuration data set includes configuration data of one or more base station devices, and the configuration data may include configuration information of network parameters of the base station devices, such as antenna types of device antennas, cell lists, and the like. The MR dataset includes a plurality of MR data. The multiple MR data may be reported by one or more UEs to the one or more base station devices, and each MR data may include, for example, location information of the UE (such as geographic information such as latitude and longitude information of the UE, altitude information, etc.) , at least two of the RSRP data of the serving cell detected by the UE, the RSRP data of the neighboring cell detected by the UE, the AOA of the serving cell, the AOA of the neighboring cell, the timing advance of the serving cell, the UE transmit power headroom, etc. one or more. For example, a certain UE may periodically (eg, every 10 seconds) upload MR data to the target device antenna in the second geographic area, then a plurality of MRs within a preset time period may be collected from the target device antenna Data is entered into this forecast set.

Step 602: Obtain a list of cells with a common antenna according to the configuration data. The specific implementation process can be similar to the description with reference to step 502 in the embodiment of FIG. 14 , and for the sake of brevity of the description, details are not repeated here.

Step 603: Obtain MR data corresponding to the cells with the same antenna according to the MR data set and the list of cells with the same antenna. The specific implementation process may be similar to the description with reference to step 503 in the embodiment of FIG. 14 , and for the sake of brevity of the description, details are not repeated here.

Step 604: Select feature information for the MR data corresponding to the common-antenna cell to obtain low-dimensional MR data of the common-antenna cell. The specific implementation process may be similar to the description with reference to step 504 in the embodiment of FIG. 14 , and for the sake of brevity of the description, details are not repeated here.

Step 605: Calculate the first sample feature data of the common-antenna cell according to the low-dimensional MR data of the common-antenna cell. The specific implementation process may be similar to the description with reference to step 505 in the embodiment of FIG. 14 , and for the sake of brevity of the description, details are not repeated here.

Step 606: Obtain the antenna type of the device antenna according to the configuration data. The specific implementation process may be similar to the description with reference to step 506 in the embodiment of FIG. 14 , and for the sake of brevity of the description, details are not repeated here.

Step 607: Input the first sample feature data and the antenna type into the trained working parameter generation model, thereby obtaining the predicted working parameters of the device antenna.

Wherein, the trained working parameter generation model is, for example, the working parameter generation model (eg, a neural network algorithm model) obtained through the training in the above-mentioned embodiment 501-step 507 of FIG. The feature data (Feature ₁₀₅₆ ) and the antenna type (AntennaType) are input into the trained working parameter generation model, and the predicted working parameters of the device antenna can be obtained. For example, taking the latitude and longitude prediction as an example, the prediction result of the latitude and longitude (Latitude, Longtitude) of the device antenna can be obtained.

It can be understood that when the input data of the training set includes sample data of multiple device antennas, such as sample data including device antenna 1 (specifically including configuration data of device antenna 1, MR data reported by the terminal to device antenna 1), device antenna 1 The sample data of the antenna 2 (specifically including the configuration data of the device antenna 2, the MR data reported by the terminal to the device antenna 2), etc., then, for different device antennas, the corresponding Feature ₁₀₅₆ and AntennaType data are also different. The Feature ₁₀₅₆ and AntennaType data corresponding to each device antenna are used as the input data of the work parameter generation model, and the predicted work parameters of each device antenna can be obtained.

It can be seen that through the above steps 601 to 607, the engineering parameter generation model prediction module in the engineering parameter prediction model prediction module can realize the predicted engineering parameters of the equipment antenna by using the engineering parameter generation model according to the sample data of the prediction set. Subsequently, the engineering parameter correction model prediction module in the engineering parameter prediction model prediction module will use the engineering parameter correction model to further process the predicted engineering parameters of the equipment antenna through the following steps 608 to 611, so as to obtain the reliability (accuracy). sex) higher predictive parameters.

Step 608: Calculate the top N antennas of the device antennas according to the list of common antenna cells of the device antennas and the low-dimensional MR data of the device antennas. The specific implementation process may be similar to the description with reference to step 509 in the embodiment of FIG. 14 , and for the sake of brevity of the description, details are not repeated here.

Step 609: Obtain the predicted operating parameters of each adjacent antenna in the top N antennas according to the predicted operating parameters of the multiple device antennas and the top N antennas of each device antenna. The specific implementation process may be similar to the description with reference to step 510 in the embodiment of FIG. 14 , and for the sake of brevity of the description, details are not repeated here.

Step 610: Obtain second sample feature data of the top N antennas according to the low-dimensional MR data of the base station equipment and the predicted operating parameters of each adjacent antenna in the top N antennas of the base station equipment. The specific implementation process may be similar to the description with reference to step 511 in the embodiment of FIG. 14 , and for the sake of brevity of the description, details are not repeated here.

Step 611: Input the predicted working parameters of the (1+top N) antenna group and the second sample feature data of the top N antennas into the trained working parameter correction model, thereby obtaining highly reliable predicted working parameters of the device antenna.

Wherein, the trained work parameter correction model is, for example, the work parameter correction model (such as a neural network algorithm model) obtained by training from steps 508 to 512 in the embodiment of FIG. 14 , a model prediction module (an work parameter prediction model prediction module The engineering parameter correction model prediction module in ) inputs the second sample feature data (Feature _{join_i} ) corresponding to the device antenna i and the predicted engineering parameters (Feature _{basic_i} ) of the (1+Top N) antenna group of the device antenna i into the training With a good working parameter correction model, a high-confidence predicted working parameter of the device antenna i can be obtained. For example, taking the latitude and longitude prediction as an example, a prediction result with high reliability of the latitude and longitude (Latitude, Longtitude) of the device antenna i can be obtained.

It can be understood that when the input data of the training set includes sample data of multiple device antennas, such as sample data including device antenna 1 (specifically including configuration data of device antenna 1, MR data reported by the terminal to device antenna 1), device antenna 1 The sample data of the antenna 2 (specifically including the configuration data of the device antenna 2, the MR data reported by the terminal to the device antenna 2), etc., then, for different device antennas, the corresponding second sample feature data, (1+Top N ) The predicted operating parameters of the antenna group are also different. Taking the second sample characteristic data corresponding to each device antenna and the (1+Top N) antenna group as the input data of the correction model for the operating parameters, the height of each device antenna can be obtained. Predictive parameters of reliability.

It can be seen that in this embodiment, through steps 601 to 607, the predicted operating parameters of the equipment antenna can be obtained by using the engineering parameter generation model according to the sample data of the prediction set, and through steps 608 to 611, the model can be corrected by using the engineering parameters. Further corrections are made to the predicted operating parameters of the equipment antenna to obtain high-confidence predicted operating parameters. It can be understood that the engineering parameter prediction model described in the embodiment of the present invention can be regarded as including an engineering parameter generation model and an engineering parameter correction model. Therefore, based on the above steps 601 to 611, the equipment antenna engineering parameters based on the engineering parameter prediction model are completed. Prediction.

It should also be noted that, in some possible embodiments, if only the working parameter generation model prediction module as shown in The engineering parameter prediction model prediction module only includes the engineering parameter generation model prediction module), then the implementation process of the model training can also refer to the description of the above steps 601-607. For the brevity of the description, this article will not describe it in detail.

It should also be noted that, in some possible embodiments, if only the working parameter correction model prediction module as shown in The engineering parameter prediction model prediction module only includes the engineering parameter correction model prediction module), then the implementation process of the model training can also refer to the description of the above steps 608-611, the difference is that in the prediction process (1+Top N) The predicted engineering parameters of the antenna group are not obtained through the engineering parameter generation model, but from the engineering parameter set in the sample data of the prediction set. The engineering parameters in the engineering parameter set can be regarded as the engineering parameters with low reliability. , for example, the ginseng obtained through rougher methods. Based on the above steps 608 to 611, those skilled in the art will similarly understand that the engineering parameter correction model prediction module as shown in FIG. The method, for the brevity of the description, will not be described in detail in this article.

It can be seen that in this embodiment of the present invention, a pre-trained model for predicting working parameters (such as the predicting model for working parameters in this embodiment) can be input based on ready-made sample data (such as MR data, configuration data, etc.) With this model, the working parameters of the equipment antenna can be generated and corrected, so as to obtain the predicted working parameters with high reliability. Therefore, implementing the technical solutions of the embodiments of the present invention can overcome the defects of the prior art, effectively reduce the acquisition cost of the equipment antenna engineering parameters, and improve the accuracy of the engineering parameters.

In order to better understand the technical solutions of the embodiments of the present invention, a method for correcting the original working parameters of the device antenna in some application scenarios is described below. Referring to Figure 20, the method includes but is not limited to the following steps:

Step 701, the method starts, and the variable i is set to 0.

Step 702: If i is equal to 0, the method jumps to step 703; otherwise, if i is not equal to 0, jumps to step 708.

Step 703, the method executes to select sample data.

In some embodiments, if the model prediction module in this method embodiment is an engineering parameter generation model prediction module, the sample data includes MR data (including UE positioning information) and configuration data of the base station device. The sample data has a corresponding relationship with the original working parameters of the device antenna in the original working parameter list.

In some embodiments, if the model prediction module in this embodiment of the method is an engineering parameter correction model, the sample data includes MR data of the base station equipment (including the positioning information of the UE), configuration data, and a list of original engineering parameters (which may be used herein). The original working parameters of the device antenna (that is, the to-be-corrected working parameters of the device antenna) in the original engineering parameter list for short.

In some embodiments, if the model prediction module in this method embodiment is an engineering parameter prediction model, the sample data includes MR data (including UE positioning information) and configuration data of the base station equipment. The sample data has a corresponding relationship with the original working parameters of the device antenna in the original working parameter list.

Wherein, the original working parameter table includes the original working parameters of one or more device antennas.

Step 704 , input the sample data into the model prediction module to perform engineering parameter prediction.

It can be understood that if the model prediction module in the method embodiment is an engineering parameter generation model prediction module, the engineering parameter generation model prediction module can use the engineering parameter generation model to realize the predicted engineering parameters of the equipment antenna according to the sample data. For the specific implementation process, reference may be made to the foregoing embodiment in FIG. 5 and the related descriptions of steps 601 to 607 in the embodiment of FIG. 19 , which are not repeated here for the sake of brevity of the description.

It can be understood that, if the model prediction module in this embodiment of the method is an engineering parameter correction model prediction module, the engineering parameter correction model prediction module can use the engineering parameter correction model to obtain the predicted engineering parameters of the equipment antenna according to the sample data. For the specific implementation process, reference may be made to the foregoing embodiment in FIG. 8 and the related descriptions of steps 608 to 611 in the embodiment of FIG. 19 , which are not repeated here for brevity of the description.

It can be understood that if the model prediction module in this method embodiment is an engineering parameter prediction model prediction module, then the engineering parameter prediction model prediction module can obtain the predicted engineering parameters of the equipment antenna by using the engineering parameter prediction model according to the sample data. For the specific implementation process, reference may be made to the foregoing embodiment in FIG. 11 and the related descriptions of steps 601 to 611 in the embodiment of FIG. 19 , which are not repeated here for brevity of the description.

Step 705: Set the variable i to i+=1.

Step 706, the method is executed to determine whether the average difference between the predicted working parameters of the equipment antenna and the original working parameters of the equipment antenna satisfies the preset condition, if the preset condition is not met, the method continues to step 707; if the preset condition is not met, Then jump to execution step 709.

For example, in some embodiments, there are multiple device antennas, and the predicted operating parameters of each device antenna are respectively denoted as (Px1, Py1), (Px2, Py2), . . . , (Pxn, Pyn), where, Arbitrary predicted operating parameters (Pxi, Pyi) represent the predicted operating parameters of the i-th device antenna (or antenna feeder).

The original working parameters of each device antenna are recorded as (Rx1, Ry1), (Rx2, Ry2), ..., (Rxn, Ryn), where any original working parameters (Rxi, Ryi) represent the original working parameters of the i-th device antenna. Ginseng.

Input the predicted operating parameters and original operating parameters of the above-mentioned equipment antennas into the following formulas:

If diff<preset value (for example, 0.0002, the present invention does not limit the preset value), it is considered that the accuracy of the working parameter meets the requirements. That is to say, if diff<preset value, it is said that the average difference between the predicted parameters of the device antenna and the original parameters of the device antenna satisfies the preset condition. If the difference is greater than or equal to the preset value, it is said that the average difference between the predicted parameters of the device antenna and the original parameters of the device antenna does not meet the preset conditions.

It should be noted that, in practical applications of this method, more preset conditions can be set, for example, the relationship between i and the maximum number of iterations (maxIterTimes) can be determined, for example, the maximum number of iterations is set to 5 (only an example, The present invention does not limit this). If i is greater than the maximum number of iterations, the method jumps to step 708 ; if i is less than or equal to the maximum number of iterations, the method continues to perform step 707 .

Step 707: The method executes updating the working parameters in the original working parameter list, and placing the original working parameters with a larger degree of difference into the abnormal working parameter list. Then jump to step 702.

Specifically, in some embodiments, an abnormal working parameter list is set for placing the original working parameter with a large difference between the predicted value of the working parameter and the original working parameter (eg, greater than or equal to the preset value described in step 706 ). Then, through step 706, if there are some equipment antennas with a large difference between the predicted operating parameters and the original operating parameters (for example, greater than or equal to the preset value described in step 706), the original operating parameters of such equipment antennas are put into List of exception parameters. If there are some equipment antennas whose predicted operating parameters differ from the original operating parameters (for example, smaller than the preset value described in step 706 ), the predicted operating parameters of such equipment antennas are replaced in the original operating parameter list. The original working parameters of the device antenna, so as to realize the update of the original working parameter list.

It should be noted that, in a possible implementation, only the ID of the device antenna corresponding to the abnormal operating parameter may be stored in the abnormal operating parameter list.

Step 708: Determine whether the abnormal work parameter list is empty. If the abnormal parameter list is empty, the method proceeds to step 709 ; if the abnormal parameter list is not empty, the method jumps to step 703 .

Step 709 , using the preset working parameters predicted through step 704 and satisfying the preset conditions in step 706 as the final corrected working parameters and outputting them. That is to say, by performing the above steps 701 to 709, the original operating parameters of each device antenna can be corrected, and more accurate predicted operating parameters of each device antenna can be obtained.

It can be seen that in this embodiment of the present invention, a pre-trained model for predicting working parameters (such as the predicting model for working parameters in this embodiment) can be input based on ready-made sample data (such as MR data, configuration data, etc.) This model can realize the generation and correction of the engineering parameters of the equipment antenna to obtain the predicted results of the engineering parameters. Based on the comparison of the predicted results of the engineering parameters with the original engineering parameters of the base station equipment, the original engineering parameters of the base station equipment can be updated to obtain the base station equipment. The working parameter value with high reliability of the device. Therefore, implementing the technical solutions of the embodiments of the present invention can overcome the defects of the prior art, effectively reduce the acquisition cost of the equipment antenna engineering parameters, and improve the accuracy of the engineering parameters.

The system architecture and method of the embodiments of the present invention are described in detail above. Based on the same inventive concept, the following continues to provide related devices of the embodiments of the present invention.

Referring to FIG. 21, FIG. 21 is a schematic structural diagram of a computing device 80 provided by an embodiment of the present invention. The computing device 80 includes: a data acquisition module 801 and a model training module 802, wherein:

A data acquisition module 801 is configured to acquire a first engineering parameter set, a first configuration data set in a first geographical area, and a first data set uploaded by a terminal to a first device antenna among a plurality of device antennas in the first geographical area. A measurement report data set; wherein the first engineering parameter set includes engineering parameters of the first device antenna, and the engineering parameters of the first device antenna include position data and attitude data of the first device antenna. at least one; the first configuration data set includes configuration data of the first device antenna, the configuration data of the first device antenna represents configuration information of network parameters of the first device antenna; the first measurement report The measurement report data in the data set includes the position data and signal received power data of the terminal; the first device antenna is any device antenna among the plurality of device antennas;

A model training module 802, configured to perform model training according to the engineering parameters of the first device antenna, the configuration data of the first device antenna, and the measurement report data set to obtain an antenna working parameter prediction model; the antenna working parameter The prediction model is used to output the second device antenna according to the second configuration data set in the second geographical area by the terminal and the second measurement report data set uploaded by the terminal to the second device antenna in the second geographical area engineering parameters.

In some embodiments, the antenna working parameter prediction model includes an antenna working parameter generation model; the model training module 802 is specifically configured to: obtain, according to the configuration data of the first device antenna and the measurement report data set, first sample feature data; the first sample feature data includes a plurality of received signal power data of a cell or a remote radio unit RRU belonging to the antenna of the first device, and each of the plurality of received signal power data. The position data of the terminal corresponding to the signal receiving power data; the antenna type of the first device antenna is obtained according to the configuration data of the first device antenna; the model is carried out according to the engineering parameters of the first device antenna and the first feature set training to obtain the antenna working parameter generation model; the first feature set includes the first sample feature data and the antenna type of the first device antenna, and the antenna working parameter generation model is used according to the input A feature set outputs engineering parameters.

In a possible embodiment, the model training module 802 is specifically configured to: determine, according to the configuration data of the first device antenna, the cell or RRU that belongs to the first device antenna; from the measurement report data set, determine Measurement report data corresponding to the cell or RRU; and feature extraction is performed according to the measurement report data corresponding to the cell or RRU to obtain the first sample feature data.

In some embodiments, the antenna working parameter prediction model includes both an antenna working parameter generation model and an antenna working parameter correction model; the engineering parameter set further includes engineering parameters of at least one other device antenna; the configuration data set It also includes configuration data of the at least one other device antenna; the at least one other device antenna represents a device antenna other than the first device antenna among the multiple device antennas; the model training module 802 is further configured to: Obtain second sample characteristic data according to the configuration data set and the measurement report data set, where the second sample characteristic data includes a plurality of received signal power data of a cell or an RRU of the antenna of the at least one other device, and all The position data of the terminal corresponding to each signal received power data in the plurality of signal received power data and the signal received power data of the cell or RRU of the antenna of the first device; according to the antenna working parameter generation model, the first device is obtained. The prediction result of the engineering parameters of the device antenna and the prediction result of the engineering parameters of the at least one other device antenna; model training is performed according to the engineering parameter set and the second feature set, and the antenna engineering parameter correction model is obtained; The second feature set includes the second sample feature data, the prediction result of the engineering parameter of the first device antenna, and the prediction result of the engineering parameter of the at least one other device antenna, and the antenna engineering parameter correction model is used for inputting The second feature set of output engineering parameters.

In a possible embodiment, the at least one other device antenna is top N device antennas surrounding the first device antenna, and the top N device antennas indicate that the first device antennas are the same as the first device antenna. The device antennas of the N most relevant device antennas, where N is an integer greater than or equal to 1.

In a possible embodiment, the model training module 802 is specifically configured to: determine, according to the configuration data set, a cell or RRU belonging to the first device antenna, and a cell or RRU belonging to the top N device antennas RRU; according to the first measurement report data set, set the cell of the top N device antennas or the cell of any device antenna in the RRU or the RRU to correspond to at least one measurement report data, and each of the at least one measurement report data The pieces of measurement report data include the received signal power data of the cell or RRU of any device antenna, the received signal power data of the cell or RRU of the first device antenna, and the location data of the terminal; according to the top N Feature extraction is performed on the cell of the device antenna or the measurement report data corresponding to the cell of each device antenna or the RRU in the RRU to obtain the second sample feature data.

The computing device 80 may be specifically used to implement the model training process described in the various embodiments of the present invention.

In a specific implementation, the program codes of the data acquisition module 801 and the model training module 802 may be stored in the memory 1022 described in the embodiment of FIG. 2 above, and may be called by the processor 1021 to execute the description in the embodiment of the present invention model training method.

In a specific implementation, the data acquisition module 801 and the model training module 802 can be jointly used to implement the function of the model training module for generating the work parameters described in the embodiment of FIG. The described function of the training module for the correction model of the engineering parameter may be jointly used to realize the function of the training module for the training module of the engineering parameter prediction model described in the above embodiment of FIG. 10 . For the specific implementation process, reference may be made to the relevant description of the embodiment in FIG. 12 or the embodiment in FIG. 14 , which is not repeated here for the sake of brevity of the description.

Referring to FIG. 22, FIG. 22 is a schematic structural diagram of a computing device 90 provided by an embodiment of the present invention. The computing device 90 includes: a data acquisition module 901 and an industrial parameter prediction module 902, wherein:

A data acquisition module 901, configured to acquire a second configuration data set in a second geographical area and a second measurement report data set uploaded by a terminal to a second device antenna in the second geographical area; wherein the second configuration The data set includes configuration data of the second device antenna, where the configuration data of the second device antenna represents configuration information of network parameters of the second device antenna; the measurement report data in the second measurement report data set includes all the location data and signal reception power data of the terminal; the second device antenna is any device antenna among multiple device antennas in the second geographic area;

An engineering parameter prediction module 902, configured to input the second configuration data set and the second measurement report data set into an antenna engineering parameter prediction model to obtain a prediction result of engineering parameters of the second device antenna; The antenna engineering parameter prediction model is based on the first engineering parameter set in the first geographical area, the first configuration data set, and the first device antenna uploaded by the terminal to the first device antenna among the plurality of device antennas in the first geographical area. A measurement report data set is obtained by training; the engineering parameters of the second device antenna include at least one of position data and attitude data of the first device antenna.

In some embodiments, the antenna working parameter prediction model includes an antenna working parameter generation model; the working parameter prediction module 902 is specifically configured to: according to the second measurement report data set and the configuration data of the second device antenna , obtain first sample feature data; the first sample feature data includes multiple signal received power data of the cell or remote radio unit RRU belonging to the second device antenna, and the multiple signal received power data The position data of the terminal corresponding to the received power data of each signal in the data; according to the configuration data of the second device antenna, the antenna type of the second device antenna is obtained; the first sample feature data and the second device The antenna type of the antenna is input into the antenna engineering parameter generation model, and a first prediction result of the engineering parameter of the second device antenna is obtained.

In a possible embodiment, the working parameter prediction module 902 is specifically configured to: determine a cell or RRU belonging to the second device antenna according to the configuration data of the second device antenna; from the second measurement report data set , determine the measurement report data corresponding to the cell or the RRU; perform feature extraction according to the measurement report data corresponding to the cell or the RRU to obtain the first sample feature data.

In some embodiments, the antenna working parameter prediction model includes both an antenna working parameter generation model and an antenna working parameter correction model; the second configuration data set further includes configuration data of at least one other device antenna; the at least one One other device antenna represents a device antenna other than the second device antenna among the plurality of device antennas; the operating parameter prediction module 902 is further configured to: according to the second configuration data set and the second measurement report A data set to obtain second sample feature data, where the second sample feature data includes multiple signal received power data of the cell or RRU of the at least one other device antenna, and each signal received in the multiple signal received power data The position data of the terminal corresponding to the power data and the signal received power data of the cell or RRU of the antenna of the second device; obtain the prediction result of the engineering parameters of the antenna of the at least one other device according to the generation model of the antenna engineering parameters; The second sample feature data, the first prediction result of the engineering parameter of the second device antenna, and the prediction result of the engineering parameter of the at least one other device antenna are input into the antenna engineering parameter correction model, and the first prediction result is obtained. Second prediction results of engineering parameters of two device antennas.

In a possible embodiment, the at least one other device antenna is topN device antennas surrounding the second device antenna, and the top N device antennas represent the second device antenna among the plurality of device antennas. The most relevant N device antennas, where N is an integer greater than or equal to 1.

In a possible embodiment, the working parameter prediction module 902 is specifically configured to: determine, according to the second configuration data set, the cells or RRUs that belong to the second device antenna, and the cells or RRUs that belong to the top N device antennas. cell or RRU; according to the second measurement report data set, the cell of the top N device antennas or the cell or RRU of any device antenna in the RRU is set to correspond to at least one measurement report data, and the at least one measurement report data Each measurement report data includes signal received power data of the cell of any device antenna, signal received power data of the cell of the second device antenna, and position data of the terminal; according to the top N device antennas Feature extraction is performed on the measurement report data corresponding to the cell or the RRU of each device antenna in the cell or the RRU to obtain the second sample feature data.

Wherein, the computing device 90 may be specifically used to implement the model-based prediction process of the operating parameters described in the various embodiments of the present invention.

In a specific implementation, the program codes of the data acquisition module 901 and the work parameter prediction module 902 may be stored in the memory 1022 described in the above-mentioned embodiment of FIG. A model-based approach to predicting engineering parameters is described.

In a specific implementation, the data acquisition module 901 and the work parameter prediction module 902 may be jointly used to implement the function of the work parameter generation model prediction module described in the above embodiment in FIG. 5 , or used together to implement the above embodiment in FIG. 8 The described functions of the engineering parameter correction model prediction module may be jointly used to implement the functions of the engineering parameter prediction model prediction module described in the embodiment of FIG. 11 . For the specific implementation process, reference may be made to the relevant description of the above-mentioned embodiment in FIG. 13 , the embodiment in FIG. 19 , or the embodiment in FIG. 20 , which is not repeated here for the sake of brevity of the description.

Based on the same inventive concept, an embodiment of the present invention provides an industrial parameter determination system, and the industrial parameter determination system may include the computing device 80 described in FIG. 21 and the computing device 90 described in FIG. 22 . In a specific implementation, the working parameter determination system is, for example, the working parameter determination system 201 described in the embodiment of FIG. 3 above, or the working parameter determination system 202 described in the embodiment of FIG. 6 , or the working parameter determination system described in the embodiment of FIG. 9 Refer to the determination system 203 .

In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware or any combination. When implemented in software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions, and when the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer may be a general purpose computer, special purpose computer, computer network, or other programmable device. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be downloaded from a network site, computer, server, or data center Transmission to another network site, computer, server, or data center by wire (eg, coaxial cable, optical fiber, digital subscriber line) or wireless (eg, infrared, microwave, etc.). The computer-readable storage medium may be any available medium that can be accessed by a computer, or may be a data storage device such as a server, a data center, or the like that includes one or more available media integrated. The usable media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes, etc.), optical media (eg, Digital Video Disc (DVD), etc.), or semiconductor media (eg, solid-state drives), and the like.

In the above-mentioned embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

Claims (18)

1. A method of predicting antenna engineering parameters, comprising:

acquiring a first engineering parameter set, a first configuration data set and a first measurement report data set uploaded to a first equipment antenna by a terminal; wherein the first set of engineering parameters comprises engineering parameters of the first device antenna including at least one of position data and attitude data of the first device antenna; the first set of configuration data comprises configuration data of the first device antenna, the configuration data of the first device antenna representing configuration information of network parameters of the first device antenna; measurement report data in the first measurement report data set comprises position data and signal received power data of the terminal;

performing model training according to the engineering parameters of the first equipment antenna, the configuration data of the first equipment antenna and the first measurement report data set to obtain an antenna engineering parameter prediction model; the antenna working parameter prediction model is used for outputting the engineering parameters of a second equipment antenna according to a second configuration data set and a second measurement report data set uploaded to the second equipment antenna by a terminal, wherein the second configuration data set comprises the configuration data of the second equipment antenna;

The antenna working parameter prediction model comprises an antenna working parameter generation model and an antenna working parameter correction model;

the performing model training according to the engineering parameters of the first device antenna, the configuration data of the first device antenna, and the first measurement report data set to obtain an antenna engineering parameter prediction model specifically includes:

performing model training according to the engineering parameters of the first equipment antenna, the configuration data of the first equipment antenna and the first measurement report data set to obtain an antenna working parameter generation model;

obtaining a prediction result of the engineering parameters of the first equipment antenna and a prediction result of the engineering parameters of the at least one other equipment antenna according to the antenna engineering parameter generation model;

performing model training according to the first engineering parameter set and the second characteristic set to obtain the antenna engineering parameter correction model; the second feature set comprises second sample feature data, a prediction result of the engineering parameters of the first device antenna and a prediction result of the engineering parameters of the at least one other device antenna, the antenna engineering parameter correction model is used for outputting the engineering parameters according to the input second feature set, and the first engineering parameter set further comprises the engineering parameters of the at least one other device antenna around the first device antenna; the second sample characteristic data includes multiple pieces of signal reception power data of the cell or the RRU of the at least one other device antenna, and position data of the terminal corresponding to each piece of signal reception power data in the multiple pieces of signal reception power data and signal reception power data of the cell or the RRU of the first device antenna.

2. The method of claim 1,

the performing model training according to the engineering parameters of the first device antenna, the configuration data of the first device antenna, and the first measurement report data set to obtain the antenna working parameter generation model includes:

obtaining first sample characteristic data of the first device antenna from the configuration data of the first device antenna and the first measurement report data set; the first sample characteristic data comprises a plurality of signal receiving power data of a cell or a remote radio unit RRU belonging to the first equipment antenna and position data of a terminal corresponding to each signal receiving power data in the plurality of signal receiving power data;

obtaining the antenna type of the first equipment antenna according to the configuration data of the first equipment antenna;

performing model training according to the engineering parameters and the first characteristic set of the first equipment antenna to obtain an antenna engineering parameter generation model; the first feature set comprises the first sample feature data and an antenna type of the first device antenna, and the antenna parameter generation model is used for outputting engineering parameters according to the input first feature set.

3. The method of claim 2, wherein obtaining first sample characterization data for the first device antenna based on the configuration data for the first device antenna and the first measurement report data set comprises:

determining a cell or RRU (radio remote unit) belonging to the first equipment antenna according to the configuration data of the first equipment antenna;

determining measurement report data corresponding to the cell or the RRU from the first measurement report data set;

and performing feature extraction according to the measurement report data corresponding to the cell or the RRU to obtain first sample feature data of the first equipment antenna.

4. A method according to claim 2 or 3, wherein the first set of engineering parameters further comprises engineering parameters of at least one other device antenna peripheral to the first device antenna; the first set of configuration data further comprises configuration data of the at least one other device antenna;

after the model training is performed according to the engineering parameters and the first feature set of the first device antenna to obtain the antenna engineering parameter generation model, the method further includes:

obtaining second sample characteristic data of the first device antenna from the first configuration data set and the first measurement report data set.

5. The method according to claim 4, wherein the at least one other device antenna is top N device antennas around the first device antenna, the top N device antennas representing N device antennas most correlated to the first device antenna among the plurality of device antennas, and N being an integer equal to or greater than 1.

6. The method of claim 5, wherein obtaining second sample characterization data for the first device antenna based on the first configuration data set and the first measurement report data set comprises:

determining a cell or RRU (radio remote unit) belonging to the first equipment antenna and a cell or RRU belonging to the top N equipment antennas according to the first configuration data set;

setting at least one measurement report data corresponding to a cell of the top N device antennas or a cell of any one of RRUs according to the first measurement report data set, wherein each measurement report data in the at least one measurement report data comprises signal receiving power data of the cell of any one of the device antennas or the RRUs, signal receiving power data of the cell of the first device antenna or the RRU and position data of the terminal;

And performing feature extraction according to the cell of the top N device antennas or the cell of each device antenna in the RRUs or the measurement report data corresponding to the RRU, and obtaining second sample feature data of the first device antenna.

7. A method of predicting antenna engineering parameters, comprising:

acquiring a second configuration data set and a second measurement report data set uploaded to a second equipment antenna by the terminal; wherein the second configuration data set comprises configuration data of the second device antenna, the configuration data of the second device antenna representing configuration information of network parameters of the second device antenna; measurement report data in the second measurement report data set comprises position data and signal received power data of the terminal;

inputting the second configuration data set and the second measurement report data set into an antenna engineering parameter prediction model to obtain a prediction result of the engineering parameters of the second equipment antenna; the antenna engineering parameter prediction model is obtained by training according to a first engineering parameter set, a first configuration data set and a first measurement report data set uploaded to a first equipment antenna in a first antenna set by a terminal; the engineering parameters of the second device antenna comprise at least one of position data and attitude data of the first device antenna;

The antenna working parameter prediction model comprises an antenna working parameter generation model and an antenna working parameter correction model;

inputting the second configuration data set and the second measurement report data set into an antenna engineering parameter prediction model to obtain a prediction result of the engineering parameters of the second equipment antenna, including:

obtaining a first prediction result of the engineering parameters of the first equipment antenna according to the antenna engineering parameter generation model;

obtaining a prediction result of the engineering parameters of at least one other equipment antenna according to the antenna engineering parameter generation model;

inputting second sample characteristic data of the second equipment antenna, a first prediction result of the engineering parameters of the second equipment antenna and a prediction result of the engineering parameters of the at least one other equipment antenna into the antenna engineering parameter correction model to obtain a second prediction result of the engineering parameters of the second equipment antenna; the second sample characteristic data of the second device antenna includes multiple pieces of signal reception power data of the cell or the RRU of the at least one other device antenna, and location data of a terminal corresponding to each piece of signal reception power data in the multiple pieces of signal reception power data and the signal reception power data of the cell or the RRU of the second device antenna.

8. The method of claim 7, wherein the first set of engineering parameters comprises engineering parameters of the first device antenna including at least one of position data and attitude data of the first device antenna; the first set of configuration data comprises configuration data of the first device antenna, the configuration data of the first device antenna representing configuration information of network parameters of the first device antenna; the measurement report data in the first measurement report data set comprises position data and signal received power data of the terminal.

9. The method according to claim 7 or 8, wherein obtaining a first prediction of the engineering parameters of the first device antenna from the antenna engineering parameter generation model comprises:

obtaining first sample characteristic data of the second device antenna according to the second measurement report data set and the configuration data of the second device antenna; the first sample characteristic data of the second device antenna comprises a plurality of signal receiving power data of a cell or a Remote Radio Unit (RRU) which is subordinate to the second device antenna and position data of a terminal corresponding to each signal receiving power data in the plurality of signal receiving power data;

Obtaining the antenna type of the second equipment antenna according to the configuration data of the second equipment antenna;

and inputting the first sample characteristic data of the second equipment antenna and the antenna type of the second equipment antenna into the antenna engineering parameter generation model to obtain a first prediction result of the engineering parameters of the second equipment antenna.

10. The method of claim 9, wherein the antenna parameter generation model is obtained by model training according to engineering parameters and a first feature set of the first device antenna; the first feature set includes first sample feature data of the first device antenna and an antenna type of the first device antenna, where the first sample feature data of the first device antenna includes multiple signal reception power data of a cell or an RRU that belongs to the first device antenna, and location data of a terminal corresponding to each signal reception power data in the multiple signal reception power data.

11. The method of claim 9, wherein obtaining the first sample characteristic data of the second device antenna according to the second measurement report data set and the configuration data of the second device antenna comprises:

Determining a cell or RRU (radio remote unit) belonging to the second equipment antenna according to the configuration data of the second equipment antenna;

determining measurement report data corresponding to the cell or the RRU from the second measurement report data set;

and performing feature extraction according to the measurement report data corresponding to the cell or the RRU to obtain first sample feature data of the second equipment antenna.

12. The method according to any one of claims 8, 10 and 11, wherein the second configuration data set further comprises configuration data of at least one other device antenna peripheral to the second device antenna;

after obtaining the first prediction result of the engineering parameter of the second device antenna, the method further includes:

obtaining second sample characteristic data of the second device antenna from the second configuration data set and the second measurement report data set.

13. The method of claim 12, wherein the antenna engineering parameter correction model is obtained by model training according to the first engineering parameter set and the second feature set; wherein the first engineering parameter set comprises engineering parameters of the first equipment antenna and engineering parameters of at least one other equipment antenna around the first equipment antenna; the second feature set comprises second sample feature data of the first device antenna, a prediction result of the engineering parameter of the first device antenna, and a prediction result of the engineering parameter of at least one other device antenna around the first device antenna; the second sample characteristic data includes multiple pieces of signal reception power data of a cell of at least one other device antenna or an RRU around the first device antenna, and position data of a terminal corresponding to each piece of signal reception power data in the multiple pieces of signal reception power data and signal reception power data of the cell or the RRU of the first device antenna; the prediction result of the engineering parameter of the first equipment antenna and the prediction result of the engineering parameter of at least one other equipment antenna around the first equipment antenna are obtained according to the antenna engineering parameter generation model.

14. The method according to claim 12, wherein the at least one other device antenna around the second device antenna is top N device antennas around the second device antenna, the top N device antennas representing N device antennas most correlated to the second device antenna among the plurality of device antennas, where N is an integer greater than or equal to 1.

15. The method of claim 14, wherein obtaining second sample characterization data for the second device antenna based on the second configuration data set and the second measurement report data set comprises:

determining a cell or RRU (radio remote unit) belonging to the second equipment antenna and a cell or RRU belonging to the top N equipment antennas according to the second configuration data set;

setting at least one measurement report data corresponding to the cell of the top N device antennas or the cell of any one of the RRUs according to the second measurement report data set, wherein each measurement report data in the at least one measurement report data comprises signal receiving power data of the cell of any one device antenna, signal receiving power data of the cell of the second device antenna and position data of the terminal;

And performing feature extraction according to the cell of the top N device antennas or the cell of each device antenna in the RRUs or the measurement report data corresponding to the RRU, and obtaining second sample feature data of the second device antenna.

16. A computing device that predicts antenna engineering parameters, the computing device comprising a processor and a memory, wherein:

the memory is used for storing a program code, a first engineering parameter set, a first configuration data set and a first measurement report data set uploaded to a first equipment antenna by a terminal; wherein the first set of engineering parameters comprises engineering parameters of the first device antenna including at least one of position data and attitude data of the first device antenna; the first set of configuration data comprises configuration data of the first device antenna, the configuration data of the first device antenna representing configuration information of network parameters of the first device antenna; measurement report data in the first measurement report data set comprises position data and signal received power data of the terminal;

the processor is configured to execute the program code in the memory to implement the method of any of claims 1-6.

17. A computing device that predicts antenna engineering parameters, the computing device comprising a processor and a memory, wherein:

the memory is used for storing a program code, a second configuration data set and a second measurement report data set uploaded to a second equipment antenna by the terminal; wherein the second configuration data set comprises configuration data of the second device antenna, the configuration data of the second device antenna representing configuration information of network parameters of the second device antenna; measurement report data in the second measurement report data set comprises position data and signal received power data of the terminal;

the processor is configured to execute the program code in the memory to implement the method of any of claims 7-15.

18. A system for predicting antenna engineering parameters, the system comprising the computing device for training an antenna parameters prediction model of claim 16 and the computing device for predicting parameters based on the antenna parameters prediction model of claim 17, wherein:

the computing device for training the antenna engineering parameter prediction model is specifically configured to obtain a first engineering parameter set, a first configuration data set, and a first measurement report data set uploaded to the first device antenna by a terminal; performing model training according to the engineering parameters of the first equipment antenna, the configuration data of the first equipment antenna and the first measurement report data set to obtain an antenna engineering parameter prediction model; inputting the antenna parameters prediction model to the computing device for predicting the parameters based on the antenna parameters prediction model;

The computing device for predicting the power parameters based on the antenna power parameter prediction model is specifically configured to obtain a second configuration data set and a second measurement report data set uploaded to a second device antenna by the terminal; inputting the second configuration data set and the second measurement report data set into an antenna engineering parameter prediction model to obtain a prediction result of the engineering parameters of the second equipment antenna; the second set of configuration data includes configuration data for the second device antenna.

Images