Whole-network common-coverage cell identification method and device, electronic equipment and storage medium

Abstract

The invention discloses a whole network common coverage cell identification method and device, electronic equipment and a storage medium. The method comprises the following steps: determining a field intensity correlation index based on 4 / 5G inter-system MR measurement data; the field intensity correlation index is an index between adjacent intervals of the 4 / 5G inter-system; generating a first AI training sample set based on the 4 / 5G cell work parameters; the training sample set comprises a common-coverage sample set and a non-common-coverage sample set; based on a field intensity correlation index, performing data cleaning on the first AI training sample set to obtain a second AI training sample set; and carrying out AI model training based on the second AI training sample set, and carrying out whole network cell co-coverage identification through the trained AI model. By adopting the technical scheme of the invention, the whole network cell co-coverage identification is carried out through the trained AI model, the workload of manual field verification is reduced, the human resources are saved, the optimization cost is saved, and the network awareness is improved at the same time.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a method, apparatus, electronic device, and storage medium for identifying cells with full network coverage. Background Technology

[0002] In the optimization of high-load 4G / 5G wireless cells, load balancing is often performed, including intra-system load balancing and inter-system load balancing. Cells that can be offloaded need to meet the co-coverage relationship so that network optimization personnel can formulate offloading strategies.

[0003] Existing technical solutions based on Pearson correlation coefficient to identify co-coverage relationships and specific feature data are not applicable to co-coverage relationships between different systems (4G / 5G). They also have high requirements for the accuracy of site latitude and longitude, cell sector number and azimuth in the engineering parameters. However, in practice, it is difficult to guarantee the real-time update of the engineering parameters due to daily optimization work. Therefore, there is an urgent need for a method that can accurately identify whether cells of different wireless systems have co-coverage. Summary of the Invention

[0004] This invention provides a method, apparatus, electronic device, and storage medium for identifying shared coverage cells across a network, in order to solve the problem of low identification efficiency for shared coverage cells.

[0005] According to one aspect of the present invention, a method for identifying cells with full network coverage is provided, the method comprising:

[0006] Field strength correlation index is determined based on 4 / 5G inter-system MR measurement data; the field strength correlation index is the index of the 4 / 5G inter-system neighbor interval;

[0007] The first AI training sample set is generated based on 4 / 5G cell parameters; the training sample set includes two types of sample sets: shared coverage and non-shared coverage.

[0008] Based on the field strength correlation index, the first AI training sample set is cleaned to obtain the second AI training sample set.

[0009] The AI ​​model is trained based on the second AI training sample set, and the trained AI model is used to identify the full network coverage of cells.

[0010] According to another aspect of the present invention, a full-network coverage cell identification device is provided, the device comprising:

[0011] The index determination module is used to determine the field strength correlation index based on 4 / 5G inter-system MR measurement data; the field strength correlation index is the index of the 4 / 5G inter-system neighbor interval;

[0012] The sample construction module is used to generate the first AI training sample set based on 4 / 5G cell parameters; the training sample set includes two types of sample sets: shared coverage and non-shared coverage.

[0013] The sample cleaning module is used to clean the first AI training sample set based on the field strength correlation index to obtain the second AI training sample set.

[0014] The co-coverage identification module is used to train the AI ​​model based on the second AI training sample set, and to identify the co-coverage of cells across the entire network using the trained AI model.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] At least one processor; and

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program that can be executed by the at least one processor, which is then executed by the at least one processor to enable the at least one processor to perform the full network coverage cell identification method according to any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the whole-network co-coverage cell identification method according to any embodiment of the present invention.

[0020] The technical solution of this invention determines the field strength correlation index based on 4 / 5G inter-system MR measurement data; the field strength correlation index is an index between neighboring cells of the 4 / 5G inter-system; a first AI training sample set is generated based on 4 / 5G cell engineering parameters; the training sample set includes two types of sample sets: co-coverage and non-co-coverage; based on the field strength correlation index, the first AI training sample set is cleaned to obtain a second AI training sample set; an AI model is trained based on the second AI training sample set, and the trained AI model is used to identify co-coverage of cells across the entire network, reducing the workload of manual on-site verification, saving human resources, saving optimization costs, and improving network awareness.

[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart of a method for identifying cells with full network coverage according to Embodiment 1 of the present invention;

[0024] Figure 2 This is a flowchart of another method for identifying cells with full network coverage provided in Embodiment 2 of the present invention;

[0025] Figure 3 This is a schematic diagram of the generation of co-coverage cell pairs applicable to embodiments of the present invention;

[0026] Figure 4 This is a schematic diagram of a full-network coverage cell identification device according to Embodiment 3 of the present invention;

[0027] Figure 5 This is a schematic diagram of the structure of an electronic device that implements the full-network coverage cell identification method of this invention. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] Example 1

[0031] Figure 1 This is a flowchart of a method for identifying cells with full network coverage provided in Embodiment 1 of the present invention. This embodiment is applicable to the identification of cells with full network coverage. This method can be executed by a cell identification device with full network coverage, which can be implemented in hardware and / or software. This cell identification device can be configured in an electronic device with data processing capabilities. Figure 1 As shown, the method includes:

[0032] S110. Determine the field strength correlation index based on 4 / 5G inter-system MR measurement data; the field strength correlation index is the index of the 4 / 5G inter-system neighbor interval.

[0033] Based on the Pearson correlation coefficient method, the proportion of effective inter-system measurement sampling points, field strength correlation coefficient, and field strength difference standard deviation index of 4G / 5G cells are calculated. These indicators all characterize the possibility of co-coverage between 4G / 5G cells to a certain extent.

[0034] S120. Generate the first AI training sample set based on 4 / 5G cell parameters; the training sample set includes two types of sample sets: shared coverage and non-shared coverage.

[0035] During AI model training, two types of samples, one with shared coverage and the other with non-shared coverage, are needed as the training set. One method can determine the accurate shared coverage relationship: if 4G / 5G inter-system cells share an RRU, then they definitely share coverage. However, cell pairs with shared RRUs between 4G / 5G inter-system cells only account for about 15% of the total number of 5G cells in the entire network, and most of them are combinations of 4G 3DMIMO and 5G 2.6G frequency bands.

[0036] Therefore, based on work experience and the 4 / 5G cell parameters of each cell pair, this application selects some cell pairs as non-co-coverage cell pairs and co-coverage cell pairs.

[0037] S130. Based on the field strength correlation index, the first AI training sample set is cleaned to obtain the second AI training sample set.

[0038] The field strength correlation index includes the percentage of effective sampling points, the field strength correlation coefficient, and the standard deviation of the field strength difference.

[0039] Specifically, based on the field strength correlation index, the first AI training sample set is cleaned to obtain the second AI training sample set, which includes:

[0040] For each field strength correlation index, the first AI training sample set is cleaned based on the normal distribution theory to obtain the second AI training sample set.

[0041] The first AI training sample set was cleaned.

[0042] Because the accuracy of engineering parameters cannot be guaranteed—for example, the azimuth angle may be incorrect due to a lack of timely updates by optimization personnel—it can lead to errors in the determination of co-coverage relationships. Therefore, data cleaning is necessary.

[0043] To address this, a statistical method based on normal distribution was employed, utilizing three indicators—field strength correlation between 4G and 5G inter-system measurements, standard deviation of field strength difference, and percentage of effective sampling points—to improve the accuracy of co-coverage classification. The average value of a given dataset was then calculated. and standard deviation It can remove abnormal sampling points at the edges.

[0044] Based on prior knowledge, a certain threshold T is also set for the total number of sampling points. Regions with too few sampling points are considered to have abnormal MR data and are removed from the training set.

[0045] Considering that the larger the proportion of effective heterogeneous system measurement sampling points in the first and second regions, the larger the field strength correlation coefficient and the smaller the standard deviation of the field strength difference, the more obvious the co-coverage characteristic. However, the accurate proportion of engineering parameters cannot be determined, so a cyclical experimental method is needed to set the data cleaning threshold.

[0046] In the first experiment, a 2σ standard (selecting the optimal 97.5% of samples for each indicator) was selected. For each indicator, μ-2σ and μ+2σ were used as standards for indicator cleaning and filtering, retaining 97.5% of the sample data. Therefore, data filtering was performed based on the calculated standard deviation of each indicator to ensure accurate labeling of the co-covered samples.

[0047] in, ;

[0048] ;

[0049] ;

[0050] In the formula, and represents the mean and standard deviation, respectively; r represents the field strength correlation coefficient; v represents the standard deviation of the field strength difference; This represents the percentage of valid sampling points. The method for calculating the percentage of valid sampling points includes:

[0051] Determine the total number of sampling points and the number of valid sampling points, wherein the valid sampling points are sampling points that meet preset conditions;

[0052] The percentage of valid sampling points is obtained by dividing the number of valid sampling points by the total number of sampling points.

[0053] A sampling point can be the location where MR data is reported during MR measurement of the first cell. MR measurement refers to Measurement Report, which is the wireless environment measurement data actively or passively reported by the terminal to the base station in a mobile communication network. A sampling point refers to the time and location of the mobile terminal during MR measurement.

[0054] When conducting mobile communication network measurements, mobile terminals periodically report MR measurement data. The time and location of each mobile terminal during MR measurement are the sampling points.

[0055] by Let represent the proportion of valid sampling points, which is also the first parameter. Then:

[0056] ;

[0057] Ne represents the number of valid sampling points; This indicates the total number of sampling points.

[0058] Since the technical problem to be solved by this application is to determine whether the first cell and the second cell are 4G / 5G co-covered cells, the networks of the first cell and the second cell must be different. That is, when the first cell is a 4G cell, the second cell needs to be a 5G cell, and correspondingly, when the second cell is a 4G cell, the first cell needs to be a 5G cell.

[0059] In response, 4G users report the MR sampling points measured by 4G to the 5G cell, and 5G users report the MR sampling points measured by 5G to the 4G cell. Each user reports one sampling point at preset intervals to measure the field strength (RSRP, Reference Signal Receiving Power) of the first and second cells. Each cell's MR data contains measurement results from different users at different times and locations in the second cell; therefore, a cross-system forest pair will have multiple rows of sampling point data. If no cross-system cell information is detected in a measurement, a sampling point will still be generated, but the neighbor cell information will be blank.

[0060] Referring to Table 1, the inter-system MR data received by the 4G cell is the 4G measurement of 5G inter-system MR data, which includes the field strength, frequency, PCI, etc. of the 4 / 5G neighboring cells.

[0061] Table 1

[0062]

[0063] See Table 2. The 5G measurement of 4G inter-system MR data includes the field strength, frequency, PCI, etc. of 5G cells and 4G cells.

[0064] 4G measurement of 5G inter-system MR data This indicates that 5G is used to measure MR data from 4G different systems. express.

[0065] Table 2

[0066]

[0067] To accurately determine the field strength correlation coefficient, it is calculated using the Pearson correlation coefficient formula, as follows:

[0068] ;

[0069] The covariance of the random variable of field strength in the service cell and neighboring cells of different systems;

[0070] ;

[0071] , These are the field strength variances for the serving cell and neighboring cells from different systems, respectively.

[0072] ;

[0073] ;

[0074] , respectively, are the average field strengths of the serving cell and the neighboring cells of the different systems, where i represents the i-th MR sampling point; N represents the total number of sampling points.

[0075] In one alternative approach, the expression for the standard deviation of the field strength difference is:

[0076] ;

[0077] In the formula, This represents the standard deviation of the electric field strength difference. Indicates the total number of sampling points; The expression is: ; This represents the difference in field strength between the serving cell and the neighboring cell of the different system at the i-th sampling point.

[0078] ;

[0079] Among them, LTEScRSRP represents the serving cell field strength, and NRNcRSRP represents the neighboring cell field strength of the different system.

[0080] S140. Train the AI ​​model based on the second AI training sample set, and use the trained AI model to identify the full network cell coverage.

[0081] After obtaining the second AI training sample set, it can be used to...

[0082] Optionally, an AI model is trained based on a second AI training sample set, and the trained AI model is used to identify the full network cell coverage, including:

[0083] At the current stage, if the accuracy of the second AI training sample set corresponding to the field strength correlation index is greater than the accuracy of the AI ​​model's recognition result of the second AI training sample set, then the cleaning threshold of the normal distribution theory will be increased.

[0084] The first AI training sample set is cleaned and a second AI training sample set is generated. The AI ​​model is then retrained based on the regenerated second AI training sample set. The trained AI model is then used to identify the full network cell coverage again until the accuracy of the second AI training sample set corresponding to the field strength correlation index is less than the accuracy of the AI ​​model's identification result based on the second AI training sample set.

[0085] The basic training process includes: training library construction, feature extraction, hyperparameter tuning, learning process - obtaining the model (75% of the training library data), and testing the accuracy of the model (25% of the training library data).

[0086] Training library construction: A shared-coverage library is built based on a shared-coverage sample set, and a non-shared-coverage library is built based on a non-shared-coverage sample set;

[0087] Feature extraction: Key indicators are extracted from the shared coverage database and the non-shared coverage database, including: correlation of field strength between primary and neighboring cells, standard deviation of field strength difference between primary and neighboring cells, and proportion of shared coverage sampling points. These fields are used as training input source data.

[0088] Hyperparameter tuning: Optimize hyperparameters such as the number of decision tree layers, the minimum number of samples required for internal nodes to split, and the minimum number of samples required for leaf nodes;

[0089] Learning process - obtaining the model (75% training library data): By using 75% of the sample set data and based on the configured hyperparameters, AI training is performed to obtain the AI ​​model;

[0090] Accuracy testing of the model (25% training data): The prediction results of the AI ​​model are tested and verified using the remaining 25% of the sample set data. Hyperparameters and the AI ​​model are continuously optimized. When the final AI model accuracy reaches 95%, the AI ​​model result is output. The AI ​​model output is a prediction of the co-coverage category based on different primary and secondary cell field strength correlations, primary and secondary cell field strength difference standard deviations, and the proportion of shared coverage sampling points. The output categories are 0 and 1, where 0 represents non-co-coverage and 1 represents co-coverage.

[0091] The co-coverage identification results based on this proposal are compared and verified with the co-coverage identification results based on the engineering parameter method.

[0092] For example, the proposed method identified a total of 17,958 pairs of 4 / 5G inter-system co-coverage cell groups. Based on the engineering parameter method, 30,350 groups were identified. After screening, the number of sampling points was greater than T and had inter-system sampling data, but without data cleaning, 12,135 groups remained. There are 10,421 pairs of co-coverage groups that overlap between the proposed method and the engineering parameter method.

[0093] Based on the identification results, the accuracy of identifying co-coverage using the engineering parameter method is approximately [percentage missing].

[0094] 10421 / 12135 = 85.87%;

[0095] Compared with data filtering conditions based on engineering parameters, the lowest accuracy rate was [missing value].

[0096] ;

[0097] How can the data filtered by the three filtering criteria overlap to achieve the highest accuracy rate?

[0098]

[0099] Based on the results of the third step, the accuracy of the initial selection of the 2σ standard (selecting 97.5% of the best samples for each indicator) (between 92.5% and 97.5%) was higher than the accuracy of the final result (85.87%). Therefore, the screening threshold was lowered to σ.

[0100] Returning to the first step of sample set cleaning and fusion, the cleaning threshold for each indicator is reduced to σ, retaining 84% of the optimal samples for each indicator. See the formula below for details, where... The meanings are shown in Table 7. and These represent the mean and standard deviation, respectively.

[0101] ;

[0102] The training samples were regenerated, trained, and identified, resulting in the identification of 16,226 pairs of 4 / 5G inter-system co-coverage cell groups. Based on the engineering parameter method, 30,350 groups were identified. After screening for sampling points with a number greater than T and having inter-system sampling data, but without data cleaning, 12,135 groups remained. There are 10,135 pairs of co-coverage groups that overlap between the proposed method and the engineering parameter method.

[0103] Based on the identification results, the accuracy of identifying co-coverage using the engineering parameter method is approximately:

[0104] 10135 / 12135 = 83.52%;

[0105] Compared with data filtering conditions based on engineering parameters, the lowest accuracy was achieved when the three filtering conditions did not overlap at all.

[0106] ;

[0107] The highest accuracy rate is achieved when all three screening criteria completely overlap.

[0108] ;

[0109] The initial accuracy of the σ standard (selecting 97.5% of the best samples for each indicator) was between 52% and 84%, and the final accuracy of 83.82% was reasonable, so the cycle could be terminated.

[0110] Considering the different coverage areas of 4G / 5G cells across different frequency bands and the varying characteristic values ​​of the field strength correlation index between pairs of cells, it is necessary to construct AI modeling training sample sets for each frequency band. AI algorithms are then used to train models on the training data of each frequency band. This proposal uses the decision tree AI algorithm for modeling training, but other algorithms, such as random forests, can also be used.

[0111] The technical solution of this application determines the field strength correlation index based on 4 / 5G inter-system MR measurement data; the field strength correlation index is the index between neighboring cells of the 4 / 5G inter-system; a first AI training sample set is generated based on 4 / 5G cell engineering parameters; the training sample set includes two types of sample sets: co-coverage and non-co-coverage; based on the field strength correlation index, the first AI training sample set is cleaned to obtain a second AI training sample set; an AI model is trained based on the second AI training sample set, and the trained AI model is used to identify co-coverage of cells across the entire network, reducing the workload of manual on-site verification, saving human resources, saving optimization costs, and improving network awareness.

[0112] Example 2

[0113] Figure 2 This invention provides a flowchart of another method for identifying cells with full network coverage. This embodiment further optimizes the process of generating the first AI training sample set based on 4 / 5G cell parameters in the aforementioned embodiments, building upon the above embodiments. This embodiment can be combined with various optional schemes in one or more of the above embodiments. Figure 2 As shown, the method for identifying cells with full network coverage in this embodiment may include the following steps:

[0114] S210. Determine the field strength correlation index based on 4 / 5G inter-system MR measurement data; the field strength correlation index is the index of the 4 / 5G inter-system neighbor interval.

[0115] S220. Based on a preset first azimuth difference, determine a pair of cells with shared coverage from all cells with the same site and sector; the azimuth difference of the pair of cells with shared coverage is less than the first azimuth difference.

[0116] S230. Based on a preset second azimuth difference value, determine non-co-coverage cell pairs from various cells that share the same site but do not share the same sector; the azimuth difference value of the non-co-coverage cell pairs is greater than the second azimuth difference value; the first azimuth difference value is less than the second azimuth difference value.

[0117] See Figure 3 Based on cell pairs with co-located sites and the same sector, and setting a certain threshold for the difference in azimuth angle, such as a difference of no more than 30 degrees, a suspected co-coverage cell pair is generated across the entire network. For example, at a certain site, cell 1 in the 900M frequency band and cell 1 in the 2.6G frequency band generate a co-coverage cell pair.

[0118] Based on cell pairs located in different sectors at the same site, and with a certain threshold set for the azimuth difference (e.g., a difference of no less than 60 degrees), suspected non-co-coverage cell pairs are generated across the entire network. For example, at a certain site, cell 1 in the 700MHz band and cell 2 in the 2.6GHz band are considered a non-co-coverage cell pair.

[0119] S240. Based on the field strength correlation index, perform data cleaning on the first AI training sample set to obtain the second AI training sample set.

[0120] S250: Trains an AI model based on the second AI training sample set, and uses the trained AI model to identify the full network cell coverage.

[0121] Using the technical solution of this application, based on a preset first azimuth difference value, co-coverage cell pairs are determined from various cells that share the same site and sector; the azimuth difference value of the co-coverage cell pairs is less than the first azimuth difference value; based on a preset second azimuth difference value, non-co-coverage cell pairs are determined from various cells that share the same site but do not share the same sector; the azimuth difference value of the non-co-coverage cell pairs is greater than the second azimuth difference value; the first azimuth difference value being less than the second azimuth difference value can improve the comparison between co-coverage cell pairs and non-coverage cell pairs, thereby achieving rapid generation of the sample set and ensuring the accuracy of the generation.

[0122] Example 3

[0123] Figure 4This invention provides a structural block diagram of a full-network shared coverage cell identification device, applicable to the identification of full-network shared coverage cell pairs. This full-network shared coverage cell identification device can be implemented in hardware and / or software, and can be configured in an electronic device with data processing capabilities. For example... Figure 4 As shown, the full-network co-coverage cell identification device of this embodiment may include: an index determination module 310, a sample construction module 320, a sample cleaning module 330, and a co-coverage identification module 340. Wherein:

[0124] The index determination module 310 is used to determine the field strength correlation index based on 4 / 5G inter-system MR measurement data; the field strength correlation index is the index of the 4 / 5G inter-system neighbor interval;

[0125] The sample construction module 320 is used to generate the first AI training sample set based on 4 / 5G cell parameters; the training sample set includes two types of sample sets: shared coverage and non-shared coverage.

[0126] The sample cleaning module 330 is used to clean the first AI training sample set based on the field strength correlation index to obtain the second AI training sample set.

[0127] The coverage identification module 340 is used to train the AI ​​model based on the second AI training sample set, and to identify the coverage of cells across the entire network through the trained AI model.

[0128] Based on the above embodiments, optionally, the field strength correlation index includes the proportion of effective sampling points, the field strength correlation coefficient, and the standard deviation of the field strength difference;

[0129] The sample cleaning module 330 includes:

[0130] For each field strength correlation index, the first AI training sample set is cleaned based on the normal distribution theory to obtain the second AI training sample set.

[0131] Based on the above embodiments, optionally, the sample construction module 320 includes:

[0132] Based on a preset first azimuth difference, a pair of cells with shared coverage is determined from all cells with shared site and shared sector; the azimuth difference of the pair of cells with shared coverage is less than the first azimuth difference.

[0133] Based on the above embodiments, optionally, the sample construction module 320 includes:

[0134] Based on a preset second azimuth difference, non-co-coverage cell pairs are determined from all cells that share the same site but do not share the same sector; the azimuth difference of the non-co-coverage cell pairs is greater than the second azimuth difference; the first azimuth difference is less than the second azimuth difference.

[0135] Based on the above embodiments, optionally, the total coverage identification module 340 includes:

[0136] At the current stage, if the accuracy of the second AI training sample set corresponding to the field strength correlation index is greater than the accuracy of the AI ​​model's recognition result of the second AI training sample set, then the cleaning threshold of the normal distribution theory will be increased.

[0137] The first AI training sample set is cleaned and a second AI training sample set is generated. The AI ​​model is then retrained based on the regenerated second AI training sample set. The trained AI model is then used to identify the full network cell coverage again until the accuracy of the second AI training sample set corresponding to the field strength correlation index is less than the accuracy of the AI ​​model's identification result based on the second AI training sample set.

[0138] Based on the above embodiments, optionally, the method for calculating the proportion of effective sampling points includes:

[0139] Determine the total number of sampling points and the number of valid sampling points, wherein the valid sampling points are sampling points that meet preset conditions;

[0140] The percentage of valid sampling points is obtained by dividing the number of valid sampling points by the total number of sampling points.

[0141] The full-network coverage cell identification device provided in this embodiment of the invention can execute the full-network coverage cell identification method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0142] Example 4

[0143] Figure 5 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0144] like Figure 5 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0145] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0146] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the whole-network coverage cell identification method.

[0147] In some embodiments, the full-network shared coverage cell identification method can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the full-network shared coverage cell identification method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the full-network shared coverage cell identification method by any other suitable means (e.g., by means of firmware).

[0148] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0149] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0150] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0151] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0152] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0153] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0154] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0155] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for identifying cells with full network coverage, characterized in that, include: Field strength correlation index is determined based on 4 / 5G inter-system MR measurement data; the field strength correlation index is the index of the 4 / 5G inter-system neighbor interval; The first AI training sample set is generated based on 4 / 5G cell parameters; the training sample set includes two types of sample sets: shared coverage and non-shared coverage. Based on the field strength correlation index, the first AI training sample set is cleaned to obtain the second AI training sample set. The AI ​​model is trained based on the second AI training sample set, and the trained AI model is used to identify the full network coverage of cells.

2. The method according to claim 1, characterized in that, The field strength correlation indicators include the percentage of effective sampling points, the field strength correlation coefficient, and the standard deviation of the field strength difference. Specifically, based on the field strength correlation index, the first AI training sample set is cleaned to obtain the second AI training sample set, which includes: For each field strength correlation index, the first AI training sample set is cleaned based on the normal distribution theory to obtain the second AI training sample set.

3. The method according to claim 1, characterized in that, The first AI training sample set was generated based on 4 / 5G cell parameters, including: Based on a preset first azimuth difference, a pair of cells with shared coverage is determined from all cells with shared site and shared sector; the azimuth difference of the pair of cells with shared coverage is less than the first azimuth difference.

4. The method according to claim 3, characterized in that, The first AI training sample set was generated based on 4 / 5G cell parameters, and also includes: Based on a preset second azimuth difference, non-co-coverage cell pairs are determined from all cells that share the same site but do not share the same sector; the azimuth difference of the non-co-coverage cell pairs is greater than the second azimuth difference; the first azimuth difference is less than the second azimuth difference.

5. The method according to claim 1, characterized in that, The AI ​​model is trained based on the second AI training sample set, and the trained AI model is used to identify the full network cell coverage, including: At the current stage, if the accuracy of the second AI training sample set corresponding to the field strength correlation index is greater than the accuracy of the AI ​​model's recognition result of the second AI training sample set, then the cleaning threshold of the normal distribution theory will be increased. The first AI training sample set is cleaned and a second AI training sample set is generated. The AI ​​model is then retrained based on the regenerated second AI training sample set. The trained AI model is then used to identify the full network cell coverage again until the accuracy of the second AI training sample set corresponding to the field strength correlation index is less than the accuracy of the AI ​​model's identification result based on the second AI training sample set.

6. The method according to claim 2, characterized in that, The method for calculating the proportion of effective sampling points includes: Determine the total number of sampling points and the number of valid sampling points, wherein the valid sampling points are sampling points that meet preset conditions; The percentage of valid sampling points is obtained by dividing the number of valid sampling points by the total number of sampling points.

7. A cell identification device with full network coverage, characterized in that, include: The index determination module is used to determine the field strength correlation index based on 4 / 5G inter-system MR measurement data; the field strength correlation index is the index of the 4 / 5G inter-system neighbor interval; The sample construction module is used to generate the first AI training sample set based on 4 / 5G cell parameters; the training sample set includes two types of sample sets: shared coverage and non-shared coverage. The sample cleaning module is used to clean the first AI training sample set based on the field strength correlation index to obtain the second AI training sample set. The co-coverage identification module is used to train the AI ​​model based on the second AI training sample set, and to identify the co-coverage of cells across the entire network using the trained AI model.

8. The apparatus according to claim 7, characterized in that, The field strength correlation indicators include the percentage of effective sampling points, the field strength correlation coefficient, and the standard deviation of the field strength difference. The sample cleaning module includes: The sample set filtering module is used to clean the first AI training sample set based on the normal distribution theory for each field strength correlation index, so as to obtain the second AI training sample set.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the full network coverage cell identification method according to any one of claims 1-6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are used to cause a processor to execute the whole-network co-coverage cell identification method according to any one of claims 1-6.

Images