Base station work parameter consistency checking method and device and medium

Abstract

The invention provides a base station work parameter consistency checking method and device and a medium, and relates to the technical field of networks. The method comprises the following steps: acquiring various base station work parameter data and various rule information from various platforms or systems; according to the plurality of base station work parameter data and the plurality of rule information, obtaining a plurality of target numerical value information and a plurality of target text information with a uniform target checking content format; and according to the various target numerical value information and the various target text information, completing consistency check on the target check content in the various base station work parameter data. According to the method, the numerical information and the text information are combined, and the records of the target checking content in various platforms and systems are compared and analyzed, so that the consistency checking of the target checking content in various base station work parameter data is realized, and the accuracy of the work parameter data recording and the consistency of the same content in multiple platforms and systems are improved.

Description

Technical Field

[0001] The present application relates at least to the field of network technology, and in particular to a method, device and medium for verifying the consistency of base station working parameters. Background Art

[0002] There is inconsistency in the key information of base station engineering parameters of different standards and platforms of operators. Base station engineering parameters are very important to the network quality of mobile communications. Accurate and complete base station engineering parameter data is one of the key factors in building a high-quality mobile communication network.

[0003] Therefore, it is necessary to conduct consistency verification on the operating parameters of base stations of different standards and platforms of operators to improve the accuracy and consistency of the operating parameter data, provide a solid foundation for daily basic maintenance of base stations and network optimization, and promote the efficiency and intelligence level of operators' decision-making. Summary of the Invention

[0004] In response to the above-mentioned deficiencies, this application provides a base station working parameter consistency verification method, device and medium to solve the following technical problems: how to effectively integrate multi-standard and multi-platform base station working parameter data to improve the accuracy and consistency of working parameter data.

[0005] In a first aspect, the present application provides a method for checking consistency of working parameters of a base station, the method comprising:

[0006] Obtain various base station working parameter data and various rule information from various platforms or standards;

[0007] Based on various base station working parameter data and various rule information, multiple target numerical information and multiple target text information with unified target verification content format are obtained;

[0008] Based on multiple target numerical information and multiple target text information, the target verification content is checked for consistency in multiple base station engineering parameter data.

[0009] Furthermore, based on various base station working parameter data and various rule information, various target numerical information and various target text information with a unified target verification content format are obtained, specifically including:

[0010] Divide various base station engineering parameter data to obtain all original value information, original name information and original number information of various base station engineering parameter data, as well as various value rule information and number rule information of various platforms or standards;

[0011] Inputting the original numerical value information, original name information, original number information, numerical value rule information, and number rule information into the pre-trained target verification content recognition model respectively;

[0012] The target verification content recognition model outputs a variety of target numerical information in a unified format based on the original numerical information and numerical rule information, and outputs a variety of target text information in a unified format based on the original name information, original numbering information and numbering rule information.

[0013] Furthermore, the method further includes pre-training a target verification content recognition model, specifically including:

[0014] Construct training set samples and test set samples, which include sample original value information, sample original name information, sample original number information, sample value rule information, sample number rule information, and labeled target value information and labeled target text information of sample base station working parameter data from multiple platforms and formats;

[0015] Input the training set samples and the test set samples into the target verification content recognition model, and the target verification content recognition model outputs the training target numerical information according to the sample original numerical information and the sample numerical rule information, and outputs the training target text information according to the sample original name information, the sample original number information and the sample number rule information;

[0016] According to the difference between the training target numerical information and the labeled target numerical information, as well as the difference between the training target text information and the labeled target text information, the parameters of the target verification content recognition model are adjusted until the preset model training end conditions are met.

[0017] Furthermore, multiple base station operating parameter data and multiple rule information from multiple platforms or standards are obtained, including:

[0018] Acquire working parameter data of a first base station from a first platform or a first standard and working parameter data of a second base station from a second platform or a second standard;

[0019] Obtaining first numbering rule information of a first platform or a first standard and second numbering rule information of a second platform or a second standard, where the numbering rule information includes distance and direction information between the base station sequence number and the reference base station;

[0020] First numerical rule information of a first platform or a first format and second numerical rule information of a second platform or a second format are obtained, where the numerical rule information includes measurement method and unit information of each original numerical information.

[0021] Furthermore, the target verification content recognition model outputs a variety of target numerical information in a unified format based on the original numerical information and the numerical rule information, and outputs a variety of target text information in a unified format based on the original name information, the original number information, and the number rule information, specifically including:

[0022] In response to the target verification content including the geographical location of the base station,

[0023] The target verification content recognition model outputs target numerical information based on the original base station geographic location numerical information and the measurement method of the original base station geographic location numerical information, and obtains the first base station geographic location numerical information and the second base station geographic location numerical information including the longitude information, latitude information and altitude information based on the unified benchmark,

[0024] The target verification content recognition model outputs target text information based on the original base station's own name information, the original base station's own number information, the original base station's neighboring area name information, the original base station's neighboring area number information, and the distance and direction information between the base station serial number and the reference base station, and obtains the first base station geographic location text information and the second base station geographic location text information including the base station's own administrative area information and the distance and direction information between the base station itself and the reference base station, as well as the base station's neighboring area administrative area information and the distance and direction information between the base station's neighboring area and the reference base station.

[0025] Furthermore, based on the multiple target numerical information and multiple target text information, the target verification content is checked for consistency in the multiple base station engineering parameter data, specifically including:

[0026] Obtaining a first similarity between the numerical information of the geographic location of the first base station and the numerical information of the geographic location of the second base station;

[0027] Obtain a second similarity between the geographic location text information of the first base station and the geographic location text information of the second base station;

[0028] In response to the first similarity being greater than a first threshold and the second similarity being greater than a second threshold, acquiring the first base station engineering parameter data and the second base station engineering parameter data having the same base station geographic location;

[0029] The base station geographic location numerical information in the first base station working parameter data and the second base station working parameter data having the same base station geographic location is uniformly modified to the average value of the first base station geographic location numerical information and the second base station geographic location numerical information.

[0030] Furthermore, the target verification content recognition model outputs a variety of target numerical information in a unified format based on the original numerical information and the numerical rule information, and outputs a variety of target text information in a unified format based on the original name information, the original number information, and the number rule information, specifically including:

[0031] In response to the target verification content including base station radio frequency parameters,

[0032] The target verification content recognition model outputs target text information according to the original base station radio frequency parameter name information, and obtains base station radio frequency parameter text information that uniformly names the first original base station radio frequency parameter name information and the second original base station radio frequency parameter name information.

[0033] The target verification content recognition model obtains the corresponding original base station RF parameter numerical information and its unit information based on the uniformly named base station RF parameter text information, outputs the target numerical information based on the corresponding original base station RF parameter numerical information and its unit information, and obtains the first base station RF parameter numerical information and the second base station RF parameter numerical information using a unified unit.

[0034] Furthermore, based on the multiple target numerical information and multiple target text information, the target verification content is verified in the multiple base station working parameter data, specifically including:

[0035] Merging the working parameter data of the first base station and the working parameter data of the second base station having the same base station geographical location;

[0036] Modify the original base station radio frequency parameter name information in the first base station working parameter data and the second base station working parameter data into uniformly named base station radio frequency parameter text information;

[0037] The original base station radio frequency parameter value information in the first base station working parameter data and the second base station working parameter data is modified to the first base station radio frequency parameter value information and the second base station radio frequency parameter value information using a unified unit.

[0038] In a second aspect, the present application provides a base station working parameter consistency verification device, the device comprising:

[0039] Data acquisition module, used to obtain various base station working parameter data and various rule information from various platforms or standards;

[0040] The unified processing module is connected to the data acquisition module and is used to obtain multiple target numerical information and multiple target text information with a unified target verification content format based on multiple base station working parameter data and multiple rule information;

[0041] The consistency verification module is connected to the unified processing module and is used to complete consistency verification of target verification content in multiple base station working parameter data based on multiple target numerical information and multiple target text information.

[0042] In a third aspect, the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the base station working parameter consistency verification method as described above is implemented.

[0043] The present application provides a method, device and medium for checking the consistency of base station working parameters. By combining numerical information and text information, the records of target verification content in multiple platforms and standards are compared and analyzed, thereby realizing consistency verification of target verification content in multiple base station working parameter data, thereby improving the accuracy of working parameter data records and the consistency of the same content in multiple platforms and multiple standards. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flow chart of a method for checking consistency of base station working parameters according to an embodiment of the present application;

[0045] Figure 2 This is a schematic structural diagram of a base station working parameter consistency verification device according to an embodiment of the present application;

[0046] Figure 3 This is a flow chart of another method for checking consistency of base station working parameters according to an embodiment of the present application;

[0047] Figure 4 This is a flow chart of a method for checking consistency of base station location parameters according to an embodiment of the present application. DETAILED DESCRIPTION

[0048] In order to enable those skilled in the art to better understand the technical solution of the present application, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0049] It should be understood that the specific embodiments and drawings described herein are only used to explain the present application, rather than to limit the present application.

[0050] It can be understood that, in the absence of conflict, the various embodiments and features in the embodiments of the present application can be combined with each other.

[0051] It will be understood that, for the sake of ease of description, the drawings of this application only show the parts related to this application, while the parts not related to this application are not shown in the drawings.

[0052] It can be understood that each module and unit involved in the embodiments of the present application may correspond to only one physical structure, or may be composed of multiple physical structures, or multiple modules and units may be integrated into one physical structure.

[0053] It is understood that, in the absence of conflict, the functions and steps marked in the flowcharts and block diagrams of the present application may occur in an order different from that marked in the drawings.

[0054] It is understood that the flowcharts and block diagrams of the present application illustrate the possible architectures, functions, and operations of the systems, devices, equipment, and methods according to the various embodiments of the present application. Each box in the flowchart or block diagram may represent a module, unit, program segment, or code, which contains executable instructions for implementing the specified functions. Moreover, each box or combination of boxes in the block diagram and flowchart may be implemented by a hardware-based device that implements the specified functions, or by a combination of hardware and computer instructions.

[0055] It can be understood that the modules and units involved in the embodiments of the present application can be implemented by software or hardware, for example, the modules and units can be located in a processor.

[0056] Example 1:

[0057] like Figure 1 As shown, the present application provides a method for checking consistency of base station working parameters, the method comprising:

[0058] S1. Acquire multiple base station working parameter data and multiple rule information from multiple platforms or standards;

[0059] S2. Acquire multiple target numerical information and multiple target text information in a unified target verification content format based on multiple base station working parameter data and multiple rule information;

[0060] S3. Based on multiple target numerical information and multiple target text information, the target verification content is verified in multiple base station working parameter data.

[0061] In this embodiment, the method combines numerical information and text information to compare and analyze the records of target verification content in multiple platforms and formats, thereby achieving consistency verification of target verification content in multiple base station engineering parameter data, thereby improving the accuracy of engineering parameter data records and the consistency of the same content in multiple platforms and formats. Figure 1 The method shown is applied to Figure 2 The device shown.

[0062] More specifically, this embodiment provides a method and apparatus for verifying the consistency of working parameters of base stations of different platforms or standards using multimodal technology. Accurate and complete base station working parameter data is one of the key factors in building a high-quality mobile communication network. There is a lack of uniformity in the names, numbers and other key information of base station working parameters of different standards and platforms of operators, which causes difficulties in network maintenance and optimization, reduced operation and maintenance efficiency and other problems. This embodiment collects working parameter data of base stations from multiple platforms or different standards, identifies and uniformly marks the common fields in the working parameters, and uses GIS (Geographic Information System) geographic algorithms, semantic recognition, outlier detection and other algorithms to improve the accuracy and timeliness of the consistency verification of working parameter data, providing a solid foundation for daily basic maintenance of base stations and network optimization, and promoting the efficiency and intelligence level of operators' decision-making.

[0063] Base station engineering parameters, or base station engineering parameters, are a set of data describing various characteristics and operating conditions of a base station. These primarily include the base station's geographic location (latitude and longitude), altitude, antenna height, antenna azimuth, downtilt angle, cell coverage radius, transmit power, frequency, and scrambling code. Base station engineering parameters are crucial to mobile communication network quality. They serve as essential foundational data for the construction, maintenance, and optimization of mobile communication networks, playing a crucial role in ensuring network coverage, controlling interference, and improving network performance and service quality. On the one hand, base station engineering parameters are a crucial data foundation for network performance evaluation. By analyzing the relationship between these parameters and network performance indicators (such as signal strength, signal-to-noise ratio, and bit error rate), network problems can be promptly identified. For example, if signal strength in a cell is generally low, combining these parameters with the data can determine whether the cause is an antenna fault or improper transmit power settings. On the other hand, when a network fault occurs, base station engineering parameters can help quickly locate the fault and its cause. For example, if a signal outage occurs in a certain area, checking the engineering parameters of the base stations in that area to check for changes in antenna azimuth and downtilt angles and whether the equipment's transmit power is normal helps quickly identify the fault and shortens repair time. In short, accurate and complete base station engineering parameter data is one of the key factors in building a high-quality mobile communication network.

[0064] However, there is inconsistency in key information such as the names and numbers of base station engineering parameters across different standards and platforms across operators. The main reasons are: 1) With the continuous advancement of communication technology from 2G (second-generation mobile communication technology) to 5G (fifth-generation mobile communication technology), each generation of standards has experienced differences in base station identification and positioning. Early communication network construction lacked a unified long-term plan. With technological development and network upgrades, the coexistence of new and old systems has further complicated the inconsistency in base station names, locations, and numbers. For example, some early base stations may be difficult to adjust to their ideal locations during subsequent network optimization due to site selection constraints, making it difficult to align their numbers with new base stations. 2) Base station equipment from different manufacturers and suppliers has differences in internal coding and parameter settings, resulting in inconsistent base station numbering across different platforms. 3) During network construction and planning, each operator will plan and number base stations based on their business needs, maintenance strategies, and resource availability at different times. This results in different planning approaches and methods for each phase, making it difficult to standardize base station locations and numbers.

[0065] In response to the above shortcomings, this embodiment uses geographic information system technology, machine learning, data retrieval and other means to compare and analyze the relationship between the working parameters of multi-source base stations on different platforms or standards of operators, and completes the unified naming of working parameter fields, consistency verification of public field information in working parameters, and information consistency update, so that operators can effectively integrate multi-standard and multi-platform base station working parameter data, identify potential problems and optimize network resource allocation, while improving the accuracy and timeliness of working parameter data processing, providing a solid foundation for daily basic maintenance of base stations and network optimization, and promoting the efficiency and intelligence level of operators' decision-making.

[0066] like Figure 3As shown, the technical solution of this embodiment specifically includes the following aspects: 1) collecting base station working parameter data from multiple platforms (such as 2G, 3G, 4G, and 5G network management systems) or different standards (such as GSM (Global System for Mobile Communications), LTE (Long Term Evolution), and NR (New Radio)); 2) extracting common fields in the working parameters (the number that uniquely identifies the base station, the base station name, the location / latitude and longitude, the city / region, the antenna height, the supporting information of the computer room, etc.) through an automated recognition method, and uniformly identifying them according to the rules; 3) using GIS geographic algorithms, semantic recognition, and other methods to identify working parameters belonging to base stations in the same geographical location, and uniformly merging the geographic related information in the working parameters; 4) identifying and revising inconsistent information in the working parameters. The specific technical advantages of the technical solution of this embodiment include: 1) Multimodal data fusion: combining numerical information and text information, making full use of the complementarity of two different types of data. For example, for the geographic location information of the base station, the data is preliminarily divided through geographic clustering, and then the data within the same geographic cluster is secondary clustered using semantic similarity, thereby improving the accuracy and effectiveness of clustering; 2) Custom clustering process: Based on the standard clustering algorithm, a unique two-step clustering strategy is designed, such as first performing geographic spatial clustering, and then performing semantic clustering for each geographic cluster. This hierarchical clustering method can better adapt to complex data set structures; 3) Flexible parameter setting: By setting parameters such as geo_eps (geographic neighborhood radius), geo_min_samples (geographic minimum number of samples) and semantic_threshold (semantic similarity threshold), users can flexibly adjust the clustering granularity and similarity threshold according to the characteristics of different data sets, thereby improving the versatility and adaptability of the algorithm.

[0067] In one embodiment, S2, based on multiple base station working parameter data and multiple rule information, obtains multiple target numerical information and multiple target text information in a unified target verification content format, specifically including:

[0068] Divide various base station engineering parameter data to obtain all original value information, original name information and original number information of various base station engineering parameter data, as well as various value rule information and number rule information of various platforms or standards;

[0069] Inputting the original numerical value information, original name information, original number information, numerical value rule information, and number rule information into the pre-trained target verification content recognition model respectively;

[0070] The target verification content recognition model outputs a variety of target numerical information in a unified format based on the original numerical information and numerical rule information, and outputs a variety of target text information in a unified format based on the original name information, original numbering information and numbering rule information.

[0071] In this embodiment, the base station working parameter data is classified and processed, mainly including dividing the base station working parameter data into numerical, name and number categories. Usually, the first text information corresponding to the target verification content can be extracted from the name data, and the second text information of the target verification content can be learned from the number data in combination with the numbering rules. The numerical data may have different benchmarks or different units according to different numerical acquisition rules, so they need to be checked for consistency separately. Due to the complexity of the content of the base station working parameter data, a pre-trained machine learning algorithm (target verification content recognition model) is used to obtain a variety of target numerical information and a variety of target text information with a unified format for the target verification content. The specific machine learning algorithm can adopt an algorithm model known in the art, such as a logistic regression classification model, a decision tree model, an artificial neural network, etc. By pre-training the machine learning model for one or some target verification contents, the machine learning model obtains the target output information according to the specified input information. The specified input information includes the original numerical information, original name information, original number information and corresponding rule information obtained by dividing the base station working parameter data.

[0072] In one embodiment, the method further includes pre-training a target verification content recognition model, specifically including:

[0073] Construct training set samples and test set samples, which include sample original value information, sample original name information, sample original number information, sample value rule information, sample number rule information, and labeled target value information and labeled target text information of sample base station working parameter data from multiple platforms and formats;

[0074] Input the training set samples and the test set samples into the target verification content recognition model, and the target verification content recognition model outputs the training target numerical information according to the sample original numerical information and the sample numerical rule information, and outputs the training target text information according to the sample original name information, the sample original number information and the sample number rule information;

[0075] According to the difference between the training target numerical information and the labeled target numerical information, as well as the difference between the training target text information and the labeled target text information, the parameters of the target verification content recognition model are adjusted until the preset model training end conditions are met.

[0076] In this embodiment, the process of training the machine learning model mainly includes constructing samples for training, training the model based on the samples, evaluating the effect of the model training based on the sample annotations, and finally obtaining a target verification content recognition model that achieves a preset output effect, that is, using this model can obtain an ideal output effect for target numerical information and target text information.

[0077] Specifically, the following methods and processes are used to handle inconsistencies in base station engineering parameters across different systems:

[0078] 1) Collect all relevant database tables that can be collected, including table structure (field name, data type, etc.) and some sample data. According to the meaning and use of each field in the base station working parameters, mark the fields representing the same attribute. For example, for ['Cell Network Management Name', 'Cell Chinese Name', 'Sector Chinese Name', 'Direction Angle', 'Antenna Angle'], mark them as [0,0,0,1,1]. Clean the Chinese field names to remove punctuation and special characters, convert full-width characters to half-width characters, and unify uppercase and lowercase characters. The pseudo code is as follows:

[0079]

[0080]

[0081] Use the jieba library (a widely used third-party Chinese word segmentation library in Python) to segment the field names. The pseudo code is as follows:

[0082] import jieba

[0083] def chinese_tokenizer(text):

[0084] return jieba.lcut(text)

[0085] We use TF-IDF (Term Frequency-Inverse Document Frequency) vectorization code to extract text features. We use TfidfVectorizer in the sklearn library (a free software machine learning library for the Python programming language) to calculate the term frequency-inverse document frequency (TF-IDF) of each word in the field name and convert the field name into a numeric vector. This vector representation can highlight important words in the field name and reflect the semantic characteristics of the field name. The pseudo code is as follows:

[0086] from sklearn.feature_extraction.text import TfidfVectorizer

[0087] import pandas as pd

[0088] corpus = ['Cell Network Management Name', 'Cell Chinese Name', 'Sector Chinese Name', 'Direction Angle', 'Antenna Angle', 'Downtilt Angle', 'Tilt Angle', 'Antenna Tilt Angle']

[0089] #Assume that fields is a list of all collected field names

[0090] fields=[clean_text(field)for field in fields]

[0091] tokenized_fields=[chinese_tokenizer(field)for field in fields]

[0092] vectorizer=TfidfVectorizer(tokenizer=lambda x:x)

[0093] tfidf_matrix=vectorizer.fit_transform(tokenized_fields)

[0094] #Assume y is the label

[0095] y=[0,0,0,1,1,2,2,2]

[0096] 2) Use the labeled field pairs as samples and randomly divide them into training and test set samples. Select the logistic regression model and classification model for training. The training set is used for model training, and the test set is used to adjust model parameters and evaluate the model's generalization ability. Model performance is evaluated using metrics such as recall rate and F1 value. Recall rate measures the model's ability to identify positive samples, while F1 value comprehensively considers precision and recall rate. Based on the evaluation results of the test set, adjust the model parameters and optimize model performance. The pseudo code is as follows:

[0097] # Divide into training set and test set

[0098] X_train,X_test,y_train,y_test=train_test_split(tfidf_matrix,y,test_size=0.2,random_state=42)

[0099] #Training logistic regression model

[0100] model = LogisticRegression()

[0101] model.fit(X_train,y_train)

[0102] #Model evaluation

[0103] y_pred=model.predict(X_test)

[0104] accuracy=accuracy_score(y_test,y_pred)

[0105] #Calculate the macro average recall

[0106] recall=recall_score(y_test,y_pred,average='macro')

[0107] #Calculate the macro average F1 value

[0108] f1=f1_score(y_test,y_pred,average='macro')

[0109] print(f"Accuracy:{accuracy}")

[0110] print(f"Recall rate:{recall}")

[0111] print(f"F1 value:{f1}")

[0112] A variety of target verification content recognition models can be trained for a variety of target verification contents respectively. The main improvement of the target verification content recognition model is that it can learn from the input data the output target numerical information and target text information that are consistent with the labeled target numerical information and target text information format, thereby serving as the basis for checking whether the base station working parameter data from different platforms or standards are consistent or whether consistency modification is required.

[0113] In one embodiment, S1, acquiring multiple base station working parameter data and multiple rule information from multiple platforms or standards, specifically includes:

[0114] Acquire working parameter data of a first base station from a first platform or a first standard and working parameter data of a second base station from a second platform or a second standard;

[0115] Obtaining first numbering rule information of a first platform or a first standard and second numbering rule information of a second platform or a second standard, where the numbering rule information includes distance and direction information between the base station sequence number and the reference base station;

[0116] First numerical rule information of a first platform or a first format and second numerical rule information of a second platform or a second format are obtained, where the numerical rule information includes measurement method and unit information of each original numerical information.

[0117] In this embodiment, the information involved in the working parameter data of base stations of different platforms or standards is very extensive and can be mainly divided into the following categories, each of which contains multiple specific parameters: 1) Basic information of the base station includes the site name, site code / base station ID (different platforms may have different coding rules, and the base station number is usually composed of the following parts: city code, network code, manufacturer code, indoor or private network code (optional) and base station serial number), detailed address, longitude / latitude / altitude, administrative region (province / city / district / county / street), property information (owned / leased), room type, base station type, standard (GSM (2G) , CDMA (2G), UMTS (3G), LTE (4G), NR (5G)), equipment vendor, equipment model, activation time; 2) RF parameters including frequency band information (900MHz, 1800MHz, 2100MHz, 2.6GHz, 3.5GHz, etc.), transmit power, antenna type (such as plate antenna, cylindrical antenna), antenna gain, antenna height, antenna azimuth, mechanical downtilt, electrical downtilt, antenna polarization, channel bandwidth (such as 5MHz, 10MHz, 20MHz), cell identification parameters under each standard (PCI (Physical Cell 1) Physical cell identifier (Physical cell identifier) / PSC (Primary synchronization code) / LAC (Location Area Code) / TAC (Tracking Area Code) / Cell ID (Cell identification code); 2) Transmission parameters include transmission mode (fiber, microwave, satellite), transmission bandwidth, base station IP (Internet Protocol) address, etc.; 3) Network configuration parameters include neighboring cell relations, switching parameters, QoS (Quality of Service) parameters, TAC (Tracking Area Code), PLMN (Public Land Mobile Network) number; 4) Supporting and environmental information includes lightning protection measures, power supply conditions, and computer room conditions; 5) Maintenance information includes maintenance records and contact information.

[0118] Due to different standards and platforms, information differences and inconsistencies occur, and some public information that should be consistent will be different: 1) Information that originally had common attributes is inconsistent: for example, base station name, latitude and longitude, machine room information, property information, etc.; 2) Because the data formats exported by different platforms may be different, such as using different file formats such as CSV, Excel, XML, JSON, etc., it is difficult to share information between data; 3) Different equipment vendors or network planning tools may use different parameter names, and even if the meaning is the same, there will be different parameter names and definitions; 4) The platforms have inconsistent requirements for data storage accuracy; 5) The data update frequency of different platforms may be different. Therefore, when processing base station working parameter data from multiple platforms or different standards, data standardization and conversion are required, and then unified analysis and application are required, which requires the use of corresponding algorithms or scripts.

[0119] In one embodiment, the target verification content recognition model outputs multiple target numerical information in a unified format based on the original numerical information and the numerical rule information, and outputs multiple target text information in a unified format based on the original name information, the original number information, and the number rule information, specifically including:

[0120] In response to the target verification content including the geographical location of the base station,

[0121] The target verification content recognition model outputs target numerical information based on the original base station geographic location numerical information and the measurement method of the original base station geographic location numerical information, and obtains the first base station geographic location numerical information and the second base station geographic location numerical information including the longitude information, latitude information and altitude information based on the unified benchmark,

[0122] The target verification content recognition model outputs target text information based on the original base station's own name information, the original base station's own number information, the original base station's neighboring area name information, the original base station's neighboring area number information, and the distance and direction information between the base station serial number and the reference base station, and obtains the first base station geographic location text information and the second base station geographic location text information including the base station's own administrative area information and the distance and direction information between the base station itself and the reference base station, as well as the base station's neighboring area administrative area information and the distance and direction information between the base station's neighboring area and the reference base station.

[0123] In this embodiment, for the base station geographic location information in the base station engineering parameter data, there are problems such as inaccurate recording of longitude and latitude information and different coordinate systems used by different platforms, resulting in the inability to accurately identify base stations with the same geographic location based on longitude and latitude information. Although the name information of the base station usually includes information such as the administrative division of the base station location, it is not accurate enough. In this embodiment, the base station location text information is obtained by assisting the base station number information and the base station neighboring area information. The text information has the same design format as the geographic location text information manually annotated during model training, such as the design of its own province / city / district / county / street-the distance between itself and the benchmark base station-the neighboring area province / city / district / county / street-the distance between the neighboring area and the benchmark base station. More base station location information can be obtained, thereby improving the accuracy of identifying co-located base stations. The benchmark base station can be a part of the base station for which accurate location information has been measured.

[0124] Specifically, if Figure 4 As shown in the figure, the unified format conversion and storage of data use GIS (geographic information system) technology to achieve unified format conversion and geographic storage of base station data. The step-by-step implementation process is as follows: 1) First, data in different formats are converted into a unified GeoJSON (a format for encoding various geographic data structures) format for data parsing and exchange. Use Python (for example, using GeoPandas and Shapely libraries) or other suitable programming languages ​​to write scripts to read base station data in various raw data formats (such as CSV, JSON, XML) and convert them into the target format. During the conversion process, data cleaning is performed, such as processing missing values ​​and outliers, and data verification is performed to ensure the accuracy and completeness of the converted data; 2) In the PostGIS database, create a table for storing base station data. The table should contain base station ID, longitude, latitude and other related attribute fields, and use PostGIS geometry types (such as geometry (Point, 4326)) to store the data in the postgtresql database; 3) In order to improve the efficiency of geographic spatial queries, use the CREATE INDEX statement to create a spatial index on the field containing geographic location information. The pseudo code is as follows:

[0125] Python

[0126] import geopandas as gpd from sqlalchemy import create_engine

[0127] #1. Read data (e.g., from a CSV file)

[0128] gdf = gpd.read_file("Base Station Engineering Parameter Data.csv", geometry = "geometry", crs = "EPSG:4326")

[0129] #2. Connect to the PostGIS database

[0130] engine=create_engine('postgresql: / / user:password@host:port / database')

[0131] #3. Write GeoDataFrame to PostGIS

[0132] gdf.to_postgis("Base Station Engineering Parameter Data s",engine,if_exists="append",index=False)

[0133] 4. Create a spatial index (execute SQL directly in PostGIS) #CREATE INDEX base_stations_geom_idx ON base_stations USING GIST(geom)

[0134] Figure 4 Among them, wgs84 (World Geodetic System-1984 Coordinate System is an internationally adopted geocentric coordinate system); the proj library is an open source library mainly used for conversion between geographic coordinate systems and map projections; the Shaply library is a Python library for interpreting machine learning model predictions, which calculates the contribution of each feature to model predictions based on the Shapley value in game theory; GeoPandas is a library for simplifying geospatial data processing in Python; PostGIS is an extension of PostgreSQL, specifically for storing, querying and analyzing geospatial data; SQLAlchemy is an open source software under the Python programming language; ORM (Object-Relational Mapping) is a programming technology used to implement data conversion between different types of systems in object-oriented programming languages.

[0135] In one embodiment, S3, based on the multiple target numerical information and the multiple target text information, performs consistency verification on the target verification content in the multiple base station working parameter data, specifically including:

[0136] Obtaining a first similarity between the numerical information of the geographic location of the first base station and the numerical information of the geographic location of the second base station;

[0137] Obtain a second similarity between the first base station geographic location text information and the second base station geographic location text information;

[0138] In response to the first similarity being greater than a first threshold and the second similarity being greater than a second threshold, acquiring the first base station engineering parameter data and the second base station engineering parameter data having the same base station geographic location;

[0139] The base station geographic location numerical information in the first base station working parameter data and the second base station working parameter data having the same base station geographic location is uniformly modified to the average value of the first base station geographic location numerical information and the second base station geographic location numerical information.

[0140] In this embodiment, geographic information of base stations at the same geographic location is identified and corrected. When unifying the engineering parameter data of base stations on multiple platforms or with different standards, it is particularly important to merge the information of base stations at the same geographic location, because the geographic location is related to the allocation of resources and the configuration of network optimization parameters. The traditional method for identifying geographic information of base stations at the same geographic location is based on longitude and latitude clustering. If the longitude and latitude deviations of the data reported by the engineering parameter are large, inaccurate clustering is likely to occur. This embodiment combines geographic information with text information describing the base station name, fully utilizing the complementarity of the two different types of data. The data is initially divided through geographic clustering, and the data within the same geographic cluster is secondary clustered using semantic similarity, thereby improving the accuracy and effectiveness of clustering.

[0141] The calculation methods include: using pandas from the Python library list for data processing, geopy for calculating geographic distance, sklearn's DBSCAN (Density-Based Spatial Clustering of Applications with Noise, a density-based clustering algorithm) for geographic clustering, SentenceTransformer for text semantic encoding, cosine_similarity for calculating semantic similarity, and numpy for numerical calculations, specifically including:

[0142] 1) Extract the latitude and longitude columns from the base station's engineering parameters to form a coordinate array called coordinates. Define a function called geodesic_distance to calculate geographic distance. Use geopy's geodesic function to calculate the distance between two points. Use the DBSCAN algorithm to cluster the coordinate data. Set eps to the geographic neighborhood radius, min_samples to the minimum number of samples, and metric to the custom geographic distance calculation function. Add the resulting geographic clustering labels, geo_labels, to the data frame df to form a new column called geo_cluster. The pseudo code is as follows:

[0143] coordinates=df[['latitude','longitude']].values

[0144] def geodesic_distance(x,y):

[0145] return geodesic(x,y).meters

[0146] geo_db=DBSCAN(eps=geo_eps,min_samples=geo_min_samples,metric=geodesic_distance)

[0147] geo_labels=geo_db.fit_predict(coordinates)

[0148] df['geo_cluster']=geo_labels

[0149] 2) Initialize a SentenceTransformer model, use the all-MiniLM-L6-v2 pre-trained model, encode the text column text_column in the data frame into a semantic vector, and obtain semantic_embeddings. The pseudo code is as follows:

[0150] model=SentenceTransformer('all-MiniLM-L6-v2')

[0151] semantic_embeddings=model.encode(df['text_column'])

[0152] SentenceTransformer, based on the Transformer architecture, converts sentences into fixed-length semantic vectors to measure semantic similarity between sentences. all-MiniLM-L6-v2 is a pre-trained model within the SentenceTransformer framework. It aims to significantly reduce the number of model parameters and improve inference speed while preserving model performance. It utilizes knowledge distillation to transfer knowledge from a large model (teacher model) to a smaller model (student model), enabling self-supervised learning on large text corpora. Common self-supervised tasks include the Masked Language Model (MLM), which requires the model to predict the original vocabulary based on masked words in a sentence. For example, in word prediction, the teacher model predicts the probability distribution of each word, and the student model learns to mimic this probability distribution, rather than simply learning the correct answer, thereby acquiring richer semantic knowledge. SentenceTransformer uses a dual-tower architecture, with two identical MiniLM models encoding two sentences respectively. In the fine-tuning stage, the model takes sentence pairs as input and minimizes the vector distance of semantically similar sentence pairs and maximizes the vector distance of semantically dissimilar sentence pairs through contrastive learning.

[0153] 3) Merge geographic clustering and semantic similarity, initialize an empty list final_clusters to store the final clustering results, traverse each geographic cluster label geo_cluster, and extract the data subframe cluster_df belonging to the current geographic cluster. If the number of data rows in the subframe is greater than 1: ① Extract the subframe index cluster_indices and the corresponding semantic vector cluster_embeddings; ② Calculate the cosine similarity matrix similarity_matrix between the semantic vectors; ③ Initialize the semantic cluster label list semantic_labels; ④ By traversing the similarity matrix, perform semantic clustering according to the semantic similarity threshold semantic_threshold to generate semantic cluster labels; ⑤ Combine the geographic cluster label and the semantic cluster label and add them to final_clusters. If there is only one row of data in the subframe, directly combine the geographic cluster label and the default semantic cluster label 0 and add them to final_clusters. The pseudo code is as follows:

[0154]

[0155] In one embodiment, the target verification content recognition model outputs multiple target numerical information in a unified format based on the original numerical information and the numerical rule information, and outputs multiple target text information in a unified format based on the original name information, the original number information, and the number rule information, specifically including:

[0156] In response to the target verification content including base station radio frequency parameters,

[0157] The target verification content recognition model outputs target text information according to the original base station radio frequency parameter name information, and obtains base station radio frequency parameter text information that uniformly names the first original base station radio frequency parameter name information and the second original base station radio frequency parameter name information.

[0158] The target verification content recognition model obtains the corresponding original base station RF parameter numerical information and its unit information based on the uniformly named base station RF parameter text information, outputs the target numerical information based on the corresponding original base station RF parameter numerical information and its unit information, and obtains the first base station RF parameter numerical information and the second base station RF parameter numerical information using a unified unit.

[0159] In this embodiment, in order to achieve common field identification and unified identification, some specific field names may be different for different network standards and different equipment manufacturers, but their meanings and functions are similar. Ensuring the consistency of the identification of these common fields in various systems can avoid data conflicts, improve work efficiency, and enhance network performance. If the data is inconsistent, it will lead to network optimization analysis errors, neighboring area configuration errors, inaccurate positioning and other problems. These common fields are usually stored in multiple systems, such as: a network management system (NMS, Network Management Subsystem) for monitoring and managing the entire network, a network optimization system for network optimization and performance analysis, a GIS system for geographic information display and analysis, and an OSS (Operation Support Systems) system for operation support and management.

[0160] The following are some common common fields of base station working parameters, classified by type: 1) Identification category: Base station ID (Cell ID, eNodeB ID, etc.), the unique identifier of the base station; Cell Identity (CI), the unique identifier of the cell; Physical Cell ID (PCI) is used to distinguish different cells under the same base station; Cell name, the easy-to-read name of the base station or cell, which is convenient for identification and management; Tracking Area Code (TAC) is used for location management; LMNID (Public Land Mobile Network ID), the operator network identifier; 2) Frequency category: Frequency point (ARFCN, EARFCN), the frequency used by the cell; Bandwidth, the spectrum bandwidth occupied by the cell; 3) Location category: Longitude, the geographical longitude of the base station; Latitude, the geographical latitude of the base station; Altitude, the altitude of the base station; Antenna height, the height of the antenna relative to the ground; Azimuth, the direction the antenna points; Downtilt angle, the angle at which the antenna is tilted downward; Power category: Transmit power, the transmit power of the base station; Reference Signal Power (RSPower), used by UE (User Equipment) for RSRP (Reference Signal Power) Receiving Power (RSRP), the reference signal power measured; Network configuration category: Neighbor cell list, a list of cells adjacent to this cell; Switching parameters, parameters that control UE switching between different cells; Equipment information category: Equipment model, the model of the base station equipment; Software version, the software version of the base station equipment; Others: Coverage type, coverage range, such as macro station, micro station, pico station, etc.; Region, the administrative region to which the base station belongs; Status: the operating status of the base station, such as normal, under maintenance, faulty, etc.

[0161] The names of common fields in base station engineering parameters may be inconsistent in different systems. For example, "cell name" may be labeled "physical site cell name," "CellName," "sector Chinese name," etc. To unify the naming and management of these common fields in base station engineering parameters, a trained model is used to classify and predict all field names, assigning each field name a category label. A unified Chinese name is determined for each category, such as unifying level 0 as "cell name," level 1 as "direction angle," and level 2 as "inclination angle." Then, all field names under the same category are corrected to this unified name. The pseudo code is as follows:

[0162]

[0163] After a unified name is identified, a unified representation method is used for the parameters under the name, such as a unified unit.

[0164] In one embodiment, S3, based on the multiple target numerical information and the multiple target text information, performs consistency verification on the target verification content in the multiple base station working parameter data, specifically including:

[0165] Merging the working parameter data of the first base station and the working parameter data of the second base station having the same base station geographical location;

[0166] Modify the original base station radio frequency parameter name information in the first base station working parameter data and the second base station working parameter data into uniformly named base station radio frequency parameter text information;

[0167] The original base station radio frequency parameter value information in the first base station working parameter data and the second base station working parameter data is modified to the first base station radio frequency parameter value information and the second base station radio frequency parameter value information using a unified unit.

[0168] In this example, completing the consistency check primarily involves merging data frames and records within the same cluster, converting final_clusters to a data frame, and naming the columns. For example, for base station geographic locations, the merge method is used to merge the final clustering results into the original data frame df. The data frame df is grouped by final_cluster, and an aggregation operation is performed on each group to calculate the average latitude and longitude. The text columns are concatenated, and the merged result data frame, merged_df, is returned. The pseudo code is as follows:

[0169]

[0170] This embodiment uses a method and device for multi-modal consistency verification of the working parameters of multiple-source base stations on different platforms or standards of operators. It utilizes geographic information system (GIS) technology, machine learning, data retrieval, semantic recognition, and other means to compare and analyze the relationships between the working parameters of multiple-source base stations on different platforms or standards of operators, complete unified working parameter field naming, consistency verification of common field information in working parameters, and information consistency update. This enables operators to effectively integrate working parameter data of base stations on multiple standards and platforms, identify potential problems, and optimize network resource allocation. At the same time, it improves the accuracy and timeliness of working parameter data processing, increasing efficiency by more than 50%, providing a solid foundation for daily basic base station maintenance and network optimization, and promoting the efficiency and intelligence level of operator decision-making.

[0171] Example 2:

[0172] like Figure 2 As shown, the present application provides a base station working parameter consistency verification device, the device comprising:

[0173] Data acquisition module 1, used to obtain various base station working parameter data and various rule information from various platforms or standards;

[0174] The unified processing module 2 is connected to the data acquisition module 1 and is used to obtain multiple target numerical information and multiple target text information in a unified target verification content format based on multiple base station working parameter data and multiple rule information;

[0175] The consistency verification module 3 is connected to the unified processing module 2 and is used to complete consistency verification of target verification content in multiple base station working parameter data based on multiple target numerical information and multiple target text information.

[0176] In one embodiment, the unified processing module 2 specifically includes:

[0177] A data partitioning unit is used to partition various base station working parameter data to obtain all original numerical value information, original name information, and original number information of various base station working parameter data, as well as various numerical value rule information and number rule information of various platforms or standards;

[0178] A data input unit, connected to the data division unit, is used to input the original numerical information, original name information, original number information, and numerical rule information and number rule information into the pre-trained target verification content recognition model respectively;

[0179] The model output unit is connected to the data input unit and is used for the target verification content recognition model to output a variety of target numerical information in a unified format based on the original numerical information and numerical rule information, and to output a variety of target text information in a unified format based on the original name information, original numbering information and numbering rule information.

[0180] In one embodiment, the device further includes a model training module for pre-training a target verification content recognition model, specifically including:

[0181] The sample unit is used to construct training set samples and test set samples. The training set samples and test set samples include sample original value information, sample original name information, sample original number information, sample value rule information, sample number rule information, and annotated target value information and annotated target text information of sample base station working parameter data from various platforms and formats;

[0182] a training unit connected to the sample unit, and configured to input the training set samples and the test set samples into the target verification content recognition model, wherein the target verification content recognition model outputs the training target numerical information based on the sample original numerical information and the sample numerical rule information, and outputs the training target text information based on the sample original name information, the sample original number information, and the sample number rule information;

[0183] The parameter adjustment unit is connected to the training unit and is used to adjust the parameters of the target verification content recognition model according to the difference between the training target numerical information and the labeled target numerical information, as well as the difference between the training target text information and the labeled target text information, until the preset model training end conditions are met.

[0184] In one embodiment, the data acquisition module 1 specifically includes:

[0185] An operation parameter data acquisition unit, configured to acquire operation parameter data of a first base station from a first platform or a first standard and operation parameter data of a second base station from a second platform or a second standard;

[0186] A numbering rule acquisition unit, configured to acquire first numbering rule information of a first platform or a first standard and second numbering rule information of a second platform or a second standard, wherein the numbering rule information includes distance and direction information between the base station number and the reference base station;

[0187] The numerical rule acquisition unit is used to acquire first numerical rule information of a first platform or a first format and second numerical rule information of a second platform or a second format. The numerical rule information includes measurement methods and unit information of each original numerical information.

[0188] In one embodiment, the model output unit is specifically configured to:

[0189] In response to the target verification content including the geographical location of the base station,

[0190] The target verification content recognition model outputs target numerical information based on the original base station geographic location numerical information and the measurement method of the original base station geographic location numerical information, and obtains the first base station geographic location numerical information and the second base station geographic location numerical information including the longitude information, latitude information and altitude information based on the unified benchmark,

[0191] The target verification content recognition model outputs target text information based on the original base station's own name information, the original base station's own number information, the original base station's neighboring area name information, the original base station's neighboring area number information, and the distance and direction information between the base station serial number and the reference base station, and obtains the first base station geographic location text information and the second base station geographic location text information including the base station's own administrative area information and the distance and direction information between the base station itself and the reference base station, as well as the base station's neighboring area administrative area information and the distance and direction information between the base station's neighboring area and the reference base station.

[0192] In one embodiment, the consistency checking module 3 specifically includes:

[0193] A first similarity unit is configured to obtain a first similarity between the numerical information of the geographic location of the first base station and the numerical information of the geographic location of the second base station;

[0194] A second similarity unit is configured to obtain a second similarity between the geographic location text information of the first base station and the geographic location text information of the second base station;

[0195] a similarity judgment unit connected to the first similarity unit and the second similarity unit, configured to obtain the first base station working parameter data and the second base station working parameter data having the same base station geographical location in response to the first similarity being greater than a first threshold and the second similarity being greater than a second threshold;

[0196] The position calibration unit is connected to the similarity judgment unit and is used to uniformly modify the base station geographic location numerical information in the first base station working parameter data and the second base station working parameter data having the same base station geographic location to the average value of the first base station geographic location numerical information and the second base station geographic location numerical information.

[0197] In one embodiment, the model output unit is further configured to:

[0198] In response to the target verification content including base station radio frequency parameters,

[0199] The target verification content recognition model outputs target text information according to the original base station radio frequency parameter name information, and obtains base station radio frequency parameter text information that uniformly names the first original base station radio frequency parameter name information and the second original base station radio frequency parameter name information.

[0200] The target verification content recognition model obtains the corresponding original base station RF parameter numerical information and its unit information based on the uniformly named base station RF parameter text information, outputs the target numerical information based on the corresponding original base station RF parameter numerical information and its unit information, and obtains the first base station RF parameter numerical information and the second base station RF parameter numerical information using a unified unit.

[0201] In one embodiment, the consistency checking module 3 specifically includes:

[0202] A data merging unit, configured to merge the working parameter data of a first base station and the working parameter data of a second base station having the same base station geographical location;

[0203] A name unification unit, configured to modify original base station radio frequency parameter name information in the first base station working parameter data and the second base station working parameter data into uniformly named base station radio frequency parameter text information;

[0204] The value unification unit is used to modify the original base station radio frequency parameter value information in the first base station working parameter data and the second base station working parameter data into the first base station radio frequency parameter value information and the second base station radio frequency parameter value information using a unified unit.

[0205] Example 3:

[0206] Embodiment 3 of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the base station working parameter consistency verification method as described in embodiment 1 is implemented, or the base station working parameter consistency verification device as described in embodiment 2 is implemented.

[0207] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, computer program elements or other data). Computer-readable storage media include, but are not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable read only memory), flash memory or other memory technology, CD-ROM (Compact Disc Read-Only Memory), digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0208] In addition, the present application may also provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the base station working parameter consistency verification method as described in Example 1. The computer device may be the base station working parameter consistency verification device as described in Example 2.

[0209] The memory is connected to the processor, the memory may be a flash memory, a read-only memory or other memory, and the processor may be a central processing unit or a single-chip microcomputer.

[0210] Embodiments 1-3 of the present application provide a method, device, and medium for checking the consistency of base station working parameters. By combining numerical information and text information, comparative analysis is performed on the records of target verification content in multiple platforms and standards, thereby realizing consistency verification of the target verification content in multiple base station working parameter data, thereby improving the accuracy of working parameter data records and the consistency of the same content in multiple platforms and multiple standards.

[0211] It is understood that the above embodiments are merely exemplary embodiments for illustrating the principles of the present application, and the present application is not limited thereto. Those skilled in the art may make various modifications and improvements without departing from the spirit and substance of the present application, and such modifications and improvements are also considered to be within the scope of protection of the present application.

Claims

1. A method for checking consistency of base station working parameters, characterized in that: The method comprises: Obtain various base station working parameter data and various rule information from various platforms or standards; Based on various base station working parameter data and various rule information, multiple target numerical information and multiple target text information with unified target verification content format are obtained; Based on multiple target numerical information and multiple target text information, the target verification content is checked for consistency in multiple base station engineering parameter data.

2. The method according to claim 1, characterized in that Based on various base station engineering parameter data and various rule information, multiple target numerical information and multiple target text information with unified target verification content format are obtained, including: Divide the various base station working parameter data respectively to obtain all original value information, original name information and original number information of the various base station working parameter data, as well as various value rule information and number rule information of various platforms or standards; Inputting the original numerical value information, original name information, original number information, numerical value rule information, and number rule information into the pre-trained target verification content recognition model respectively; The target verification content recognition model outputs a variety of target numerical information in a unified format based on the original numerical information and numerical rule information, and outputs a variety of target text information in a unified format based on the original name information, original numbering information and numbering rule information.

3. The method according to claim 2, characterized in that The method further includes pre-training a target verification content recognition model, specifically including: Construct training set samples and test set samples, which include sample original value information, sample original name information, sample original number information, sample value rule information, sample number rule information, and labeled target value information and labeled target text information of sample base station working parameter data from multiple platforms and formats; Input the training set samples and the test set samples into the target verification content recognition model, and the target verification content recognition model outputs the training target numerical information according to the sample original numerical information and the sample numerical rule information, and outputs the training target text information according to the sample original name information, the sample original number information and the sample number rule information; According to the difference between the training target numerical information and the labeled target numerical information, as well as the difference between the training target text information and the labeled target text information, the parameters of the target verification content recognition model are adjusted until the preset model training end conditions are met.

4. The method according to claim 2 or 3, characterized in that Obtain various base station operating parameter data and various rule information from various platforms or standards, including: Acquire working parameter data of a first base station from a first platform or a first standard and working parameter data of a second base station from a second platform or a second standard; Obtaining first numbering rule information of a first platform or a first standard and second numbering rule information of a second platform or a second standard, where the numbering rule information includes distance and direction information between the base station sequence number and the reference base station; First numerical rule information of a first platform or a first format and second numerical rule information of a second platform or a second format are obtained, where the numerical rule information includes measurement method and unit information of each original numerical information.

5. The method according to claim 4, characterized in that The target verification content identification model outputs multiple target numerical information in a unified format based on the original numerical information and numerical rule information, and outputs multiple target text information in a unified format based on the original name information, original number information, and number rule information, specifically including: In response to the target verification content including the geographical location of the base station, The target verification content recognition model outputs target numerical information based on the original base station geographic location numerical information and the measurement method of the original base station geographic location numerical information, and obtains the first base station geographic location numerical information and the second base station geographic location numerical information including the longitude information, latitude information and altitude information based on the unified benchmark, The target verification content recognition model outputs target text information based on the original base station's own name information, the original base station's own number information, the original base station's neighboring area name information, the original base station's neighboring area number information, and the distance and direction information between the base station serial number and the reference base station, and obtains the first base station geographic location text information and the second base station geographic location text information including the base station's own administrative area information and the distance and direction information between the base station itself and the reference base station, as well as the base station's neighboring area administrative area information and the distance and direction information between the base station's neighboring area and the reference base station.

6. The method according to claim 5, characterized in that Based on multiple target numerical information and multiple target text information, the target verification content is verified in multiple base station engineering parameter data, including: Obtaining a first similarity between the numerical information of the geographic location of the first base station and the numerical information of the geographic location of the second base station; Obtain a second similarity between the geographic location text information of the first base station and the geographic location text information of the second base station; In response to the first similarity being greater than a first threshold and the second similarity being greater than a second threshold, acquiring the first base station engineering parameter data and the second base station engineering parameter data having the same base station geographic location; The base station geographic location numerical information in the first base station working parameter data and the second base station working parameter data having the same base station geographic location is uniformly modified to the average value of the first base station geographic location numerical information and the second base station geographic location numerical information.

7. The method according to claim 6, characterized in that The target verification content identification model outputs multiple target numerical information in a unified format based on the original numerical information and numerical rule information, and outputs multiple target text information in a unified format based on the original name information, original number information, and number rule information, specifically including: In response to the target verification content including base station radio frequency parameters, The target verification content recognition model outputs target text information according to the original base station radio frequency parameter name information, and obtains base station radio frequency parameter text information that uniformly names the first original base station radio frequency parameter name information and the second original base station radio frequency parameter name information. The target verification content recognition model obtains the corresponding original base station RF parameter numerical information and its unit information based on the uniformly named base station RF parameter text information, outputs the target numerical information based on the corresponding original base station RF parameter numerical information and its unit information, and obtains the first base station RF parameter numerical information and the second base station RF parameter numerical information using a unified unit.

8. The method according to claim 7, characterized in that Based on multiple target numerical information and multiple target text information, the target verification content is verified in multiple base station engineering parameter data, including: Merging the working parameter data of the first base station and the working parameter data of the second base station having the same base station geographical location; Modify the original base station radio frequency parameter name information in the first base station working parameter data and the second base station working parameter data into uniformly named base station radio frequency parameter text information; The original base station radio frequency parameter value information in the first base station working parameter data and the second base station working parameter data is modified to the first base station radio frequency parameter value information and the second base station radio frequency parameter value information using a unified unit.

9. A base station working parameter consistency verification device, characterized in that: The device comprises: Data acquisition module, used to obtain various base station working parameter data and various rule information from various platforms or standards; The unified processing module is connected to the data acquisition module and is used to obtain multiple target numerical information and multiple target text information with a unified target verification content format based on multiple base station working parameter data and multiple rule information; The consistency verification module is connected to the unified processing module and is used to complete consistency verification of target verification content in multiple base station working parameter data based on multiple target numerical information and multiple target text information.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the base station working parameter consistency verification method according to any one of claims 1 to 8 is implemented.

Images