CN114339855B - Wireless network coverage rate evaluation method and device and computing equipment - Google Patents
Wireless network coverage rate evaluation method and device and computing equipment Download PDFInfo
- Publication number
- CN114339855B CN114339855B CN202011032719.5A CN202011032719A CN114339855B CN 114339855 B CN114339855 B CN 114339855B CN 202011032719 A CN202011032719 A CN 202011032719A CN 114339855 B CN114339855 B CN 114339855B
- Authority
- CN
- China
- Prior art keywords
- wireless network
- coverage rate
- network coverage
- characteristic
- feature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Abstract
The embodiment of the invention relates to the technical field of artificial intelligence and discloses a wireless network coverage rate assessment method, a wireless network coverage rate assessment device and computing equipment. The method comprises the following steps: acquiring actual measurement data related to coverage rate of the wireless network in a plurality of areas; selecting a training set and a verification set from the measured data; determining characteristic items for a wireless network coverage rate evaluation model based on data of a training set, and establishing an algorithm model to be tested; inputting the characteristic value corresponding to the characteristic item in the data of the verification set into an algorithm model to be tested to obtain the coverage rate of the wireless network to be tested, and determining the algorithm model to be tested as a wireless network coverage rate evaluation model if the difference value between the coverage rate of the wireless network to be tested and the real wireless network coverage rate meets the preset requirement; and inputting the characteristic value corresponding to the characteristic item of the region to be evaluated into a wireless network coverage rate evaluation model to obtain the wireless network coverage rate of the region to be evaluated. Through the mode, the embodiment of the invention realizes intelligent evaluation of the coverage rate of the wireless network.
Description
Technical Field
The embodiment of the invention relates to the technical field of artificial intelligence, in particular to a wireless network coverage rate assessment method, a wireless network coverage rate assessment device and computing equipment.
Background
With the development of wireless network technology, users have increasingly demanded more and more wireless network technology. To accommodate technical advances and user needs, adjustments to wireless network coverage are needed.
In carrying out embodiments of the present invention, the inventors found that: in the related art, the wireless network coverage rate adjustment is performed on a specific area mainly by means of adjusting power, adjusting an antenna azimuth angle, creating a station, and the like. In order to obtain the actual effect of the wireless network coverage rate adjustment, a person needs to be arranged to go to the field to perform actual test, data related to the wireless network coverage rate adjustment are obtained, the person is arranged to analyze and process the related data, and finally the actual value of the wireless network coverage rate is obtained. It can be seen that the related art has a relatively high cost and low efficiency for evaluating the coverage rate of the wireless network.
Disclosure of Invention
In view of the above problems, embodiments of the present invention provide a wireless network coverage rate evaluation method, device, and computing device, which are used to solve the problems in the prior art that the cost of evaluating the wireless network coverage rate is relatively high and the efficiency is relatively low.
According to an aspect of the embodiment of the present invention, there is provided a wireless network coverage rate evaluation method, including:
acquiring actual measurement data related to coverage rate of the wireless network in a plurality of areas;
selecting a training set and a verification set from the measured data according to a preset distribution proportion;
determining characteristic items for a wireless network coverage rate evaluation model based on the data of the training set, and establishing an algorithm model to be tested;
inputting the characteristic value corresponding to the characteristic item in the data of the verification set into the algorithm model to be tested to obtain the coverage rate of the wireless network to be tested, and determining the algorithm model to be tested as a wireless network coverage rate evaluation model if the difference value between the coverage rate of the wireless network to be tested and the real wireless network coverage rate meets the preset requirement;
and inputting the characteristic values corresponding to the characteristic items of the region to be evaluated into the wireless network coverage rate evaluation model to obtain the wireless network coverage rate of the region to be evaluated.
In an alternative manner, the measured data includes the bottom test data of the plurality of areas before the wireless network optimization and the retest data after the wireless network optimization.
In an optional manner, after the training set and the verification set are selected from the measured data according to the preset allocation proportion, the method further includes:
and performing super-parameter tuning by using K-fold cross validation in the training set.
In an alternative manner, the determining feature items for a wireless network coverage assessment model based on the data of the training set further includes:
screening out a first characteristic item set affecting the coverage rate of the wireless network based on the training set;
removing feature items with the association degree lower than a preset threshold value from the first feature item set according to a preset association degree analysis mode to obtain a second feature item set;
performing distribution conversion on the characteristic items in the second characteristic item set according to a preset conversion mode, and removing at least one characteristic item in the characteristic items which are related to each other and are related to the wireless network coverage rate from the converted second characteristic item set to obtain a third characteristic item set;
and performing feature screening of a machine learning model by using recursive feature elimination and cross verification from the third feature item set to obtain a fourth feature item set, and determining the feature items in the fourth feature item set as feature items for a wireless network coverage rate evaluation model.
In an alternative, the method further comprises:
performing data processing on the characteristic items in the first characteristic item set, wherein the data processing mode comprises the following steps: log operation, square operation, and polynomial addition and subtraction operation between characteristic items.
In an alternative manner, the algorithm model to be tested includes: one or more of a random forest regression tree algorithm, a gradient lifting regression tree algorithm, and a ridge regression algorithm.
In an optional manner, before inputting the feature value corresponding to the feature item in the data of the verification set into the algorithm model to be tested, the method further includes:
and optimizing the algorithm model to be tested by using a grid search technology, a learning curve and a verification curve.
According to another aspect of the embodiment of the present invention, there is provided a wireless network coverage rate evaluation apparatus including:
the measured data acquisition module is used for acquiring measured data related to coverage rate of the wireless network in a plurality of areas;
the data selection module is used for selecting a training set and a verification set from the actual measurement data according to a preset distribution proportion;
the algorithm model building module is used for determining characteristic items for the wireless network coverage rate evaluation model based on the data of the training set and building an algorithm model to be tested;
the evaluation model determining module is used for inputting the characteristic value corresponding to the characteristic item in the data of the verification set into the to-be-detected algorithm model to obtain the to-be-detected wireless network coverage rate, and determining the to-be-detected algorithm model as a wireless network coverage rate evaluation model if the difference value between the to-be-detected wireless network coverage rate and the real wireless network coverage rate meets the preset requirement;
the coverage rate obtaining module is used for inputting the characteristic value corresponding to the characteristic item of the region to be evaluated into the wireless network coverage rate evaluating model to obtain the wireless network coverage rate of the region to be evaluated.
According to another aspect of an embodiment of the present invention, there is provided a computing device including: the device comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus;
the memory is configured to store at least one executable instruction that causes the processor to perform operations of the wireless network coverage assessment method as described above.
According to still another aspect of the embodiments of the present invention, there is provided a computer-readable storage medium having stored therein at least one executable instruction that, when run on a wireless network coverage assessment device, causes the wireless network coverage assessment device to perform operations of the wireless network coverage assessment method as described above.
According to the embodiment of the invention, the training set and the verification set are selected from the measured data according to the preset distribution proportion by acquiring the measured data of the multiple areas related to the wireless network coverage rate, the characteristic item for the wireless network coverage rate assessment model is determined based on the data of the training set, the algorithm model to be tested is established, the wireless network coverage rate assessment model can be further determined by inputting the data of the verification set into the algorithm model to be tested, and the wireless network coverage rate of the area to be assessed can be obtained according to the characteristic value corresponding to the characteristic item of the area to be assessed by the wireless network coverage rate assessment model. According to the embodiment of the invention, the on-site drive test can be carried out without manual work, so that the evaluation cost of the wireless network coverage rate is reduced and the efficiency is high.
The foregoing description is only an overview of the technical solutions of the embodiments of the present invention, and may be implemented according to the content of the specification, so that the technical means of the embodiments of the present invention can be more clearly understood, and the following specific embodiments of the present invention are given for clarity and understanding.
Drawings
The drawings are only for purposes of illustrating embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:
fig. 1 is a schematic flow chart of a wireless network coverage rate evaluation method according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a wireless network coverage rate evaluation method according to another embodiment of the present invention;
fig. 3 is a schematic structural diagram of a wireless network coverage rate evaluation device according to an embodiment of the present invention;
FIG. 4 illustrates a schematic diagram of a computing device provided by an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention may be embodied in various forms and should not be limited to the embodiments set forth herein.
Fig. 1 shows a flowchart of an embodiment of a wireless network coverage assessment method of the present invention, which is performed by a computing device. In an embodiment of the present invention, executable instructions are stored in a memory space of a computing device, where the executable instructions may cause a processor to perform a wireless network coverage assessment method. As shown in fig. 1, the method comprises the steps of:
step 120: and acquiring actual measurement data of a plurality of areas related to the coverage rate of the wireless network.
The plurality of areas are areas subjected to wireless network coverage optimization. The wireless network optimization can be performed on a plurality of areas respectively by means of power adjustment, antenna azimuth adjustment, station establishment and the like. The wireless network may be a wireless network of various frequency bands, such as the FDD900 band.
In an optional manner of the embodiment of the present invention, the actually measured data is manual test data, which includes the model test data of the plurality of areas before the wireless network optimization and the retest data after the wireless network optimization. The touchdown test data may include data affecting regional wireless network coverage prior to wireless network optimization and resulting regional wireless network coverage. The retest data may include data that affects regional wireless network coverage after wireless network optimization and resulting regional wireless network coverage.
Step 140: and selecting a training set and a verification set from the measured data according to a preset distribution proportion.
The predetermined dispensing ratio may be, for example, 80%/20%, 70%/30%, or other ratios. For example, 80% of the measured data may be selected for the training set and 20% of the measured data may be selected for the validation set. Generally, training set data can be selected as much as possible, and the accuracy of the algorithm is improved.
Step 160: and determining characteristic items for the wireless network coverage rate evaluation model based on the data of the training set, and establishing a to-be-tested algorithm model.
In an alternative manner of the embodiment of the present invention, step 160 may further include the steps of:
step 161: and screening out a first characteristic item set affecting the coverage rate of the wireless network based on the training set.
Wherein the area may be divided into several regularly distributed grids. The feature items may include, for example, area comprehensive coverage, grid properties (road or natural village), number of grids covered (RSRP > = -110dbm & SINR > = -3 dB), total number of grids, average RSRP (Reference Signal Receiving Power, reference signal received power), average SINR (Signal to Interference plus Noise Ratio ), RSRP > = -110 sample points, SINR > = -3 sample points, nearest site distance, azimuth of the site at grid center point, number of 1Km up power cells, 1Km up power amplitude (summation), 1Km up power amplitude (average), number of 2Km up power cells, 2Km up power amplitude (summation), 2Km up power amplitude (average), number of RSRP > = -105 sample points, RSRP > = -100 sample points, RSRP > = -90 sample points, number of SINR > = 0 sample points.
Step 162: and removing the characteristic items with the association degree lower than a preset threshold value from the first characteristic item set according to a preset association degree analysis mode to obtain a second characteristic item set.
The preset association degree analysis mode can be one or more of analysis modes such as distribution analysis, box graph analysis, joint probability density analysis and the like. In one embodiment, before eliminating, from the first feature item set, feature items having a degree of association with the wireless network coverage rate lower than a preset threshold according to a preset degree of association analysis, abnormal data in the first feature item set may be first removed. The anomaly data in the first feature term set may be identified based on empirical analysis.
Step 163: and carrying out distribution conversion on the characteristic items in the second characteristic item set according to a preset conversion mode, and removing at least one characteristic item in the characteristic items which are related to each other and the coverage rate of the wireless network from the converted second characteristic item set to obtain a third characteristic item set.
The preset conversion mode may be, for example, conversion by using a box-cox technique, and conversion by using a log1p (x) function formula. The correlation between the feature item and the wireless network coverage and the feature item may be analyzed using a Pearson correlation coefficient method (Pearson Correlation Coefficient ) and a Spearman correlation coefficient method (Spearman's rank correlation coefficient, spearman rank correlation coefficient) in statistics before eliminating at least one of the feature items that are each associated with each other and the wireless network coverage from the converted second feature item set. The weighting of the feature items under the linear model can be more accurate by eliminating at least one of the feature items which are associated with each other and each associated with the wireless network coverage rate from the converted second feature item set.
Step 164: and performing feature screening of a machine learning model by using recursive feature elimination and cross verification from the third feature item set to obtain a fourth feature item set, and determining the feature items in the fourth feature item set as feature items for a wireless network coverage rate evaluation model.
Wherein the recursive feature elimination (Recursive feature elimination) is to select the best feature item by iteratively modeling, and then continue to select the best feature item over the remaining feature items until all feature items are traversed. Cross Validation (Cross Validation) is performed by first dividing the raw data into training data and Validation data, then training the classifier with the training data, and testing the training with the Validation data to obtain a model.
In an alternative manner of the embodiment of the present invention, the data processing may be performed on the feature items in the first feature item set, and then step 162 is performed. The data processing mode comprises the following steps: log operation, square operation, and polynomial addition and subtraction operation between characteristic items. By performing data processing on the feature items in the first feature item set, consistency and robustness of the feature items in the first feature item set can be improved. The data processing may employ, for example, the following standardized processing formula:
wherein X is original data, X std Standard deviation data of X, X mean Mean value data of X, X norm And the data is normalized by X. The accuracy of the data in the first feature item set can be improved by processing the data in the first feature item set by using a standardized processing formula.
The method comprises the steps of establishing a to-be-tested algorithm model according to data of a training set and determined characteristic items for a wireless network coverage rate evaluation model, wherein the to-be-tested algorithm model is used for outputting the wireless network coverage rate of an area to be evaluated according to the characteristic items of the area to be evaluated.
The algorithm model to be tested can be a machine learning algorithm model and can be selected according to the data scale, for example, in the embodiment of the invention, one or more of a random forest regression tree algorithm, a gradient lifting regression tree algorithm and a ridge regression algorithm with better performance and accuracy under the small data scale are selected due to the smaller data scale of the training set and the verification set. It can be appreciated that the random forest regression tree algorithm, the gradient lifting regression tree algorithm and the ridge regression algorithm are all common machine learning algorithms, and will not be described here.
Step 180: inputting the characteristic value corresponding to the characteristic item in the data of the verification set into the algorithm model to be tested to obtain the coverage rate of the wireless network to be tested, and determining the algorithm model to be tested as a wireless network coverage rate evaluation model if the difference value between the coverage rate of the wireless network to be tested and the real wireless network coverage rate meets the preset requirement.
The feature values corresponding to the feature items in the data of the verification set can be respectively input into a plurality of machine learning algorithm models, so that a plurality of wireless network coverage rates to be detected are obtained. And determining a to-be-detected algorithm model corresponding to the to-be-detected wireless network coverage rate closest to the real wireless network coverage rate as a wireless network coverage rate evaluation model.
Step 200: and inputting the characteristic values corresponding to the characteristic items of the region to be evaluated into the wireless network coverage rate evaluation model to obtain the wireless network coverage rate of the region to be evaluated.
The feature items of the region to be evaluated and the corresponding feature values thereof can be obtained according to the result of the fuzzing test of the region to be evaluated. The feature values corresponding to the feature items of the region to be evaluated and related parameters for performing wireless network optimization on the region to be evaluated can be input into a wireless network coverage rate evaluation model together, so that the wireless network coverage rate of the region to be evaluated, which is output by the model, is obtained, and the actual effect of performing wireless network optimization on the region to be evaluated is analyzed.
In the embodiment of the invention, the training set and the verification set are selected from the actually measured data according to the preset distribution proportion by acquiring the actually measured data of a plurality of areas related to the wireless network coverage rate, the characteristic item for the wireless network coverage rate evaluation model is determined based on the data of the training set, the algorithm model to be tested is established, the wireless network coverage rate evaluation model can be further determined by inputting the data of the verification set into the algorithm model to be tested, and the wireless network coverage rate of the area to be evaluated can be obtained according to the characteristic value corresponding to the characteristic item of the area to be evaluated by the wireless network coverage rate evaluation model. According to the embodiment of the invention, the on-site drive test can be carried out without manual work, so that the evaluation cost of the wireless network coverage rate is reduced and the efficiency is high.
Fig. 2 shows a flowchart of another embodiment of the wireless network coverage assessment method of the present invention, which is performed by a computing device. In an embodiment of the present invention, executable instructions are stored in a memory space of a computing device, where the executable instructions may cause a processor to perform a wireless network coverage assessment method. As shown in fig. 2, this embodiment is different from the above embodiment in that, before inputting, in the step 180, the feature value corresponding to the feature item in the data of the verification set into the to-be-detected algorithm model to obtain the to-be-detected wireless network coverage rate, the wireless network coverage rate evaluation method further includes:
step 171: and performing super-parameter tuning by using K-fold cross validation in the training set.
The K-fold cross verification may be, for example, cross verification, where the training set is divided into ten parts by cross verification, 9 parts of data in the training set are alternately used as training data, and 1 part is used as test data. The under fitting and the over fitting of the algorithm model to be tested can be reduced through K-fold cross validation, so that the hyper-parameters of the algorithm model to be tested are more optimized.
Step 172: and optimizing the algorithm model to be tested by using a grid search technology, a learning curve and a verification curve.
The grid search technology, the learning curve and the verification curve are all conventional means for optimizing an algorithm model in the machine learning field, and are not described herein. In an optional manner of the embodiment of the present invention, the optimizing the model to be measured may further include pre-pruning and post-pruning the random forest regression tree algorithm, pre-pruning and post-pruning the gradient lifting regression tree algorithm, regularizing penalty on the ridge regression algorithm, and optimizing the model to be measured with new retested data.
In the embodiment of the invention, the hyper-parameters are optimized by using K-fold cross validation in the training set, and the algorithm model to be tested is optimized by using a grid search technology, a learning curve and a validation curve, so that the algorithm model to be tested can be more accurate, the under-fitting and the over-fitting are reduced, and the optimal algorithm model can be selected as a wireless network coverage rate evaluation model through the actual test of the algorithm model to be tested.
Fig. 3 is a schematic structural diagram of an embodiment of the wireless network coverage rate evaluation device of the present invention. As shown in fig. 3, the apparatus 300 includes: the system comprises a measured data acquisition module 310, a data selection module 320, an algorithm model establishment module 330, an evaluation model determination module 340 and a coverage rate acquisition module 350.
A measured data obtaining module 310, configured to obtain measured data related to coverage of the wireless network in a plurality of areas;
the data selecting module 320 is configured to select a training set and a verification set from the measured data according to a preset allocation proportion;
an algorithm model building module 330, configured to determine feature items for a wireless network coverage rate evaluation model based on the data of the training set, and build an algorithm model to be tested;
the evaluation model determining module 340 is configured to input a feature value corresponding to the feature item in the data of the verification set into the to-be-detected algorithm model to obtain a to-be-detected wireless network coverage rate, and determine the to-be-detected algorithm model as a wireless network coverage rate evaluation model if a difference value between the to-be-detected wireless network coverage rate and a real wireless network coverage rate meets a preset requirement;
the coverage rate obtaining module 350 is configured to input a feature value corresponding to the feature item of the region to be evaluated into the wireless network coverage rate evaluation model, so as to obtain the wireless network coverage rate of the region to be evaluated.
In an alternative manner, the measured data acquired by the measured data acquiring module 310 includes the bottom test data of the plurality of areas before the wireless network optimization and the retest data after the wireless network optimization.
In an alternative manner, the algorithm model building module 330 is further configured to:
screening out a first characteristic item set affecting the coverage rate of the wireless network based on the training set;
removing feature items with the association degree lower than a preset threshold value from the first feature item set according to a preset association degree analysis mode to obtain a second feature item set;
performing distribution conversion on the characteristic items in the second characteristic item set according to a preset conversion mode, and removing at least one characteristic item in the characteristic items which are related to each other and are related to the wireless network coverage rate from the converted second characteristic item set to obtain a third characteristic item set;
and performing feature screening of a machine learning model by using recursive feature elimination and cross verification from the third feature item set to obtain a fourth feature item set, and determining the feature items in the fourth feature item set as feature items for a wireless network coverage rate evaluation model.
In an alternative manner, the algorithm model building module 330 is further configured to:
performing data processing on the characteristic items in the first characteristic item set, wherein the data processing mode comprises the following steps: log operation, square operation, and polynomial addition and subtraction operation between characteristic items.
In an alternative manner, the algorithm model to be tested established by the algorithm model establishing module 330 includes: one or more of a random forest regression tree algorithm, a gradient lifting regression tree algorithm, and a ridge regression algorithm.
In an alternative, the apparatus 300 further comprises a model optimization module 360 for super-parameter tuning within the training set using K-fold cross-validation.
In an alternative manner, the model optimization module 360 is further configured to optimize the algorithm model to be tested using a grid search technique, a learning curve, and a verification curve.
In the embodiment of the invention, the measured data of a plurality of areas related to the wireless network coverage rate can be acquired through the measured data acquisition module, the data selection module can select a training set and a verification set from the measured data according to the preset distribution proportion, the algorithm model establishment module can determine the characteristic items for the wireless network coverage rate assessment model based on the data of the training set and establish the algorithm model to be tested, the assessment model determination module can determine the wireless network coverage rate assessment model for the wireless network coverage rate assessment of the area to be assessed, and the coverage rate acquisition module can input the characteristic values corresponding to the characteristic items of the area to be assessed into the wireless network coverage rate assessment model and obtain the wireless network coverage rate of the area to be assessed. It can be seen that the wireless network coverage rate evaluation device of the embodiment of the invention can establish a wireless network coverage rate evaluation model, and output the wireless network coverage rate of the area to be evaluated according to the established wireless network coverage rate evaluation model, and compared with the manual field drive test, the wireless network coverage rate evaluation device can improve the evaluation efficiency of the wireless network coverage rate and reduce the evaluation cost.
FIG. 4 illustrates a schematic diagram of an embodiment of a computing device of the present invention, and the embodiments of the present invention are not limited to a particular implementation of the computing device.
As shown in fig. 4, the computing device may include: a processor 402, a communication interface (Communications Interface) 404, a memory 406, and a communication bus 408.
Wherein: processor 402, communication interface 404, and memory 406 communicate with each other via communication bus 408. A communication interface 404 for communicating with network elements of other devices, such as clients or other servers. The processor 402 is configured to execute the program 410, and may specifically perform the relevant steps in the embodiment of the wireless network coverage assessment method described above.
In particular, program 410 may include program code including computer-executable instructions.
The processor 402 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of the present invention. The one or more processors included by the computing device may be the same type of processor, such as one or more CPUs; but may also be different types of processors such as one or more CPUs and one or more ASICs.
Memory 406 for storing programs 410. Memory 406 may comprise high-speed RAM memory or may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
Program 410 may be specifically invoked by processor 402 to cause a computing device to:
acquiring actual measurement data related to coverage rate of the wireless network in a plurality of areas;
selecting a training set and a verification set from the measured data according to a preset distribution proportion;
determining characteristic items for a wireless network coverage rate evaluation model based on the data of the training set, and establishing an algorithm model to be tested;
inputting the characteristic value corresponding to the characteristic item in the data of the verification set into the algorithm model to be tested to obtain the coverage rate of the wireless network to be tested, and determining the algorithm model to be tested as a wireless network coverage rate evaluation model if the difference value between the coverage rate of the wireless network to be tested and the real wireless network coverage rate meets the preset requirement;
and inputting the characteristic values corresponding to the characteristic items of the region to be evaluated into the wireless network coverage rate evaluation model to obtain the wireless network coverage rate of the region to be evaluated.
In an alternative manner, the measured data includes the bottom test data of the plurality of areas before the wireless network optimization and the retest data after the wireless network optimization.
In an alternative manner, the determining feature items for a wireless network coverage assessment model based on the data of the training set further includes:
screening out a first characteristic item set affecting the coverage rate of the wireless network based on the training set;
removing feature items with the association degree lower than a preset threshold value from the first feature item set according to a preset association degree analysis mode to obtain a second feature item set;
performing distribution conversion on the characteristic items in the second characteristic item set according to a preset conversion mode, and removing at least one characteristic item in the characteristic items which are related to each other and are related to the wireless network coverage rate from the converted second characteristic item set to obtain a third characteristic item set;
and performing feature screening of a machine learning model by using recursive feature elimination and cross verification from the third feature item set to obtain a fourth feature item set, and determining the feature items in the fourth feature item set as feature items for a wireless network coverage rate evaluation model.
In an optional manner, after the training set and the verification set are selected from the measured data according to the preset allocation proportion, the method further includes:
and performing super-parameter tuning by using K-fold cross validation in the training set.
In an alternative, the program 410 is invoked by the processor 402 to cause the computing device to:
performing data processing on the characteristic items in the first characteristic item set, wherein the data processing mode comprises the following steps: log operation, square operation, and polynomial addition and subtraction operation between characteristic items.
In an alternative manner, the algorithm model to be tested includes: one or more of a random forest regression tree algorithm, a gradient lifting regression tree algorithm, and a ridge regression algorithm.
In an alternative, the program 410 is invoked by the processor 402 to cause the computing device to:
before inputting the characteristic values corresponding to the characteristic items in the data of the verification set into the algorithm model to be tested, optimizing the algorithm model to be tested by using a grid search technology, a learning curve and a verification curve.
In the embodiment of the invention, the program on the computing device can be called by the processor to enable the computing device to execute the wireless network coverage rate assessment method in any method embodiment, so that the network coverage rate of the area to be assessed can be automatically assessed, the accuracy of the network coverage rate assessment is improved, and the cost of the network coverage rate assessment is reduced.
Embodiments of the present invention provide a computer readable storage medium storing at least one executable instruction that, when executed on a processor, causes the processor to perform the wireless network coverage assessment method of any of the method embodiments described above.
The executable instructions may be particularly useful for causing a processor to:
acquiring actual measurement data related to coverage rate of the wireless network in a plurality of areas;
selecting a training set and a verification set from the measured data according to a preset distribution proportion;
determining characteristic items for a wireless network coverage rate evaluation model based on the data of the training set, and establishing an algorithm model to be tested;
inputting the characteristic value corresponding to the characteristic item in the data of the verification set into the algorithm model to be tested to obtain the coverage rate of the wireless network to be tested, and determining the algorithm model to be tested as a wireless network coverage rate evaluation model if the difference value between the coverage rate of the wireless network to be tested and the real wireless network coverage rate meets the preset requirement;
and inputting the characteristic values corresponding to the characteristic items of the region to be evaluated into the wireless network coverage rate evaluation model to obtain the wireless network coverage rate of the region to be evaluated.
In an alternative manner, the measured data includes the bottom test data of the plurality of areas before the wireless network optimization and the retest data after the wireless network optimization.
In an optional manner, after the training set and the verification set are selected from the measured data according to the preset allocation proportion, the method further includes:
and performing super-parameter tuning by using K-fold cross validation in the training set.
In an alternative manner, the determining feature items for a wireless network coverage assessment model based on the data of the training set further includes:
screening out a first characteristic item set affecting the coverage rate of the wireless network based on the training set;
removing feature items with the association degree lower than a preset threshold value from the first feature item set according to a preset association degree analysis mode to obtain a second feature item set;
performing distribution conversion on the characteristic items in the second characteristic item set according to a preset conversion mode, and removing at least one characteristic item in the characteristic items which are related to each other and are related to the wireless network coverage rate from the converted second characteristic item set to obtain a third characteristic item set;
and performing feature screening of a machine learning model by using recursive feature elimination and cross verification from the third feature item set to obtain a fourth feature item set, and determining the feature items in the fourth feature item set as feature items for a wireless network coverage rate evaluation model.
In an alternative manner, the algorithm model to be tested includes: one or more of a random forest regression tree algorithm, a gradient lifting regression tree algorithm, and a ridge regression algorithm.
In one alternative, the executable instructions cause the computing device to: performing data processing on the characteristic items in the first characteristic item set, wherein the data processing mode comprises the following steps: log operation, square operation, and polynomial addition and subtraction operation between characteristic items.
In one alternative, the executable instructions cause the computing device to:
before the characteristic values corresponding to the characteristic items in the data of the verification set are input into the algorithm model to be tested, the algorithm model to be tested is optimized by utilizing a grid search technology, a learning curve and a verification curve.
In the embodiment of the invention, at least one executable instruction is stored in the computer readable storage medium, and when the executable instruction runs on the processor, the processor executes the wireless network coverage rate assessment method in any of the method embodiments, so that the network coverage rate of the area to be assessed can be automatically assessed, the accuracy of network coverage rate assessment is improved, and the cost of network coverage rate assessment is reduced.
The algorithms or displays presented herein are not inherently related to any particular computer, virtual system, or other apparatus. Various general-purpose systems may also be used with the teachings herein. The required structure for a construction of such a system is apparent from the description above. In addition, embodiments of the present invention are not directed to any particular programming language. It will be appreciated that the teachings of the present invention described herein may be implemented in a variety of programming languages, and the above description of specific languages is provided for disclosure of enablement and best mode of the present invention.
In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the above description of exemplary embodiments of the invention, various features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be construed as reflecting the intention that: i.e., the claimed invention requires more features than are expressly recited in each claim.
Those skilled in the art will appreciate that the modules in the apparatus of the embodiments may be adaptively changed and disposed in one or more apparatuses different from the embodiments. The modules or units or components of the embodiments may be combined into one module or unit or component, and they may be divided into a plurality of sub-modules or sub-units or sub-components. Any combination of all features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or units of any method or apparatus so disclosed, may be used in combination, except insofar as at least some of such features and/or processes or units are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings), may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, third, etc. do not denote any order. These words may be interpreted as names. The steps in the above embodiments should not be construed as limiting the order of execution unless specifically stated.
Claims (9)
1. A wireless network coverage assessment method, the method comprising:
acquiring actual measurement data related to coverage rate of the wireless network in a plurality of areas;
selecting a training set and a verification set from the measured data according to a preset distribution proportion;
determining characteristic items for a wireless network coverage rate evaluation model based on the data of the training set, and establishing an algorithm model to be tested, wherein the method comprises the following steps: screening out a first characteristic item set affecting the coverage rate of the wireless network based on the training set; removing feature items with the association degree lower than a preset threshold value from the first feature item set according to a preset association degree analysis mode to obtain a second feature item set; performing distribution conversion on the characteristic items in the second characteristic item set according to a preset conversion mode, and removing at least one characteristic item in the characteristic items which are related to each other and are related to the wireless network coverage rate from the converted second characteristic item set to obtain a third characteristic item set; performing feature screening of a machine learning model by using recursive feature elimination and cross verification from the third feature item set to obtain a fourth feature item set, and determining the feature items in the fourth feature item set as feature items for a wireless network coverage rate evaluation model;
inputting the characteristic value corresponding to the characteristic item in the data of the verification set into the algorithm model to be tested to obtain the coverage rate of the wireless network to be tested, and determining the algorithm model to be tested as a wireless network coverage rate evaluation model if the difference value between the coverage rate of the wireless network to be tested and the real wireless network coverage rate meets the preset requirement;
and inputting the characteristic values corresponding to the characteristic items of the region to be evaluated into the wireless network coverage rate evaluation model to obtain the wireless network coverage rate of the region to be evaluated.
2. The method of claim 1, wherein the measured data comprises a background test data of the plurality of regions prior to wireless network optimization and a retest data after wireless network optimization.
3. The method according to claim 1 or 2, wherein after selecting the training set and the verification set from the measured data according to a preset allocation ratio, the method further comprises:
and performing super-parameter tuning by using K-fold cross validation in the training set.
4. The method according to claim 1, wherein the method further comprises:
performing data processing on the characteristic items in the first characteristic item set, wherein the data processing mode comprises the following steps: log operation, square operation, and polynomial addition and subtraction operation between characteristic items.
5. The method of claim 1, wherein the algorithm model to be tested comprises: one or more of a random forest regression tree algorithm, a gradient lifting regression tree algorithm, and a ridge regression algorithm.
6. The method according to claim 1, wherein before inputting the feature value corresponding to the feature item in the data of the verification set into the algorithm model to be tested, the method further comprises:
and optimizing the algorithm model to be tested by using a grid search technology, a learning curve and a verification curve.
7. A wireless network coverage assessment apparatus, the apparatus comprising:
the measured data acquisition module is used for acquiring measured data related to coverage rate of the wireless network in a plurality of areas;
the data selection module is used for selecting a training set and a verification set from the actual measurement data according to a preset distribution proportion;
the algorithm model building module is used for determining characteristic items for the wireless network coverage rate evaluation model based on the data of the training set and building an algorithm model to be tested, and comprises the following steps: screening out a first characteristic item set affecting the coverage rate of the wireless network based on the training set; removing feature items with the association degree lower than a preset threshold value from the first feature item set according to a preset association degree analysis mode to obtain a second feature item set; performing distribution conversion on the characteristic items in the second characteristic item set according to a preset conversion mode, and removing at least one characteristic item in the characteristic items which are related to each other and are related to the wireless network coverage rate from the converted second characteristic item set to obtain a third characteristic item set; performing feature screening of a machine learning model by using recursive feature elimination and cross verification from the third feature item set to obtain a fourth feature item set, and determining the feature items in the fourth feature item set as feature items for a wireless network coverage rate evaluation model;
the evaluation model determining module is used for inputting the characteristic value corresponding to the characteristic item in the data of the verification set into the to-be-detected algorithm model to obtain the to-be-detected wireless network coverage rate, and determining the to-be-detected algorithm model as a wireless network coverage rate evaluation model if the difference value between the to-be-detected wireless network coverage rate and the real wireless network coverage rate meets the preset requirement;
the coverage rate obtaining module is used for inputting the characteristic value corresponding to the characteristic item of the region to be evaluated into the wireless network coverage rate evaluating model to obtain the wireless network coverage rate of the region to be evaluated.
8. A computing device, comprising: the device comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus;
the memory is configured to store at least one executable instruction that causes the processor to perform the operations of the wireless network coverage assessment method according to any one of claims 1-6.
9. A computer storage medium having stored therein at least one executable instruction that, when run on a wireless network coverage assessment device, causes the wireless network coverage assessment device to perform the operations of the wireless network coverage assessment method according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011032719.5A CN114339855B (en) | 2020-09-27 | 2020-09-27 | Wireless network coverage rate evaluation method and device and computing equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011032719.5A CN114339855B (en) | 2020-09-27 | 2020-09-27 | Wireless network coverage rate evaluation method and device and computing equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114339855A CN114339855A (en) | 2022-04-12 |
| CN114339855B true CN114339855B (en) | 2023-08-15 |
Family
ID=81011728
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011032719.5A Active CN114339855B (en) | 2020-09-27 | 2020-09-27 | Wireless network coverage rate evaluation method and device and computing equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114339855B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109495898A (en) * | 2017-09-12 | 2019-03-19 | 中国移动通信集团设计院有限公司 | A kind of the index quantification prediction technique and equipment of wireless network covering |
| CN109587721A (en) * | 2017-09-28 | 2019-04-05 | 中国移动通信集团浙江有限公司 | A kind of subzone network coverage evaluating method and device |
| CN110418354A (en) * | 2019-08-06 | 2019-11-05 | 北京邮电大学 | It is a kind of that propagation model wireless network planning method is exempted from based on machine learning |
| US10701566B1 (en) * | 2019-06-28 | 2020-06-30 | T-Mobile Usa, Inc. | Multidimensional analysis and network response |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9441975B2 (en) * | 2012-06-05 | 2016-09-13 | Apple Inc. | System and method for generating signal coverage information from client metrics |
| EP3366057B1 (en) * | 2015-10-23 | 2019-08-07 | Telefonaktiebolaget LM Ericsson (publ) | Cell operation in a wireless communications network |
-
2020
- 2020-09-27 CN CN202011032719.5A patent/CN114339855B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109495898A (en) * | 2017-09-12 | 2019-03-19 | 中国移动通信集团设计院有限公司 | A kind of the index quantification prediction technique and equipment of wireless network covering |
| CN109587721A (en) * | 2017-09-28 | 2019-04-05 | 中国移动通信集团浙江有限公司 | A kind of subzone network coverage evaluating method and device |
| US10701566B1 (en) * | 2019-06-28 | 2020-06-30 | T-Mobile Usa, Inc. | Multidimensional analysis and network response |
| CN110418354A (en) * | 2019-08-06 | 2019-11-05 | 北京邮电大学 | It is a kind of that propagation model wireless network planning method is exempted from based on machine learning |
Non-Patent Citations (1)
| Title |
|---|
| 图像识别技术在5G网络覆盖率评估中的应用探讨;吕非彼 等;《邮电设计技术》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114339855A (en) | 2022-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102938071B (en) | Fuzzy clustering analysis method for detecting synthetic aperture radar (SAR) image changes based on non-local means | |
| CN108918815B (en) | Method for predicting heavy metal risk of soil | |
| CN111860982A (en) | Wind power plant short-term wind power prediction method based on VMD-FCM-GRU | |
| CN103308463B (en) | Characteristic spectrum area selection method for near infrared spectrum | |
| CN113329437B (en) | Wireless network signal propagation path loss prediction method and electronic equipment | |
| CN108375363B (en) | Antenna azimuth deflection checking method, device, equipment and medium | |
| CN105653450A (en) | Software defect data feature selection method based on combination of modified genetic algorithm and Adaboost | |
| CN112735097A (en) | Regional landslide early warning method and system | |
| CN111598457B (en) | Method and device for determining quality of power wireless network | |
| Jafarzadeh et al. | Examination of various feature selection approaches for daily precipitation downscaling in different climates | |
| CN105930900A (en) | Method and system for predicting hybrid wind power generation | |
| CN113222210A (en) | Typhoon short-term prediction method and device based on multi-mode generation confrontation | |
| CN114339855B (en) | Wireless network coverage rate evaluation method and device and computing equipment | |
| CN116223962B (en) | Method, device, equipment and medium for predicting electromagnetic compatibility of wire harness | |
| CN113569345A (en) | Numerical control system reliability modeling method and device based on multi-source information fusion | |
| CN113420722A (en) | Emergency linkage method and system for airport security management platform | |
| Kovalchuk et al. | Towards ensemble simulation of complex systems | |
| CN116719714A (en) | Training method and corresponding device for screening model of test case | |
| Nagao et al. | Geographical clustering of path loss modeling for wireless emulation in various environments | |
| CN113890833B (en) | Network coverage prediction method, device, equipment and storage medium | |
| CN115459868A (en) | Millimeter wave communication performance evaluation method and system in complex environment | |
| CN116131979A (en) | Wireless channel prediction method and system under remote interference, storage medium and terminal | |
| CN110489797B (en) | Electromagnetic environment identification method suitable for external radiation source radar | |
| CN114462127A (en) | Structural damage identification method and system based on deep extreme learning machine | |
| CN111459838B (en) | Software defect prediction method and system based on manifold alignment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |