A base station installation optimization method and system for signal twin model analysis

Abstract

The application provides a base station installation optimization method and system based on signal twin model analysis, and belongs to the field of base station layout optimization. The method comprises the following steps: randomly configuring a base station distribution position list and matching a base station cell list; determining a theoretical signal receiving power distribution and a theoretical signal to noise ratio distribution of a target area; correcting the theoretical signal receiving power and the theoretical signal to noise ratio to determine a corrected signal receiving power and a corrected signal to noise ratio; partitioning the target area to obtain an abnormal signal area; and when the abnormal signal area is equal to 0, setting the base station distribution position list as a base station installation target scheme. The application solves the technical problem that the determination of the base station installation position in the prior art only depends on theoretical signal analysis, which leads to a deviation between the actual signal coverage effect after the base station is deployed and the theoretical expectation, and achieves the technical effect that the theoretical signal parameters are corrected by combining the actual environment, the base station installation layout is optimized, and the actual deployment effect meets the theoretical expectation.

Description

Technical Field

[0001] This invention relates to the field of base station layout optimization, and in particular to a base station installation optimization method and system based on signal twin model analysis. Background Technology

[0002] With the development of mobile communication technology and the increasing demands of users for communication quality, the rational deployment of base stations has become a key factor in ensuring the coverage quality and service level of communication networks. The determination of base station installation locations directly affects signal coverage and communication quality.

[0003] Currently, the determination of base station installation locations mainly relies on theoretical signal analysis methods. For example, based on radio wave propagation models, the signal coverage range of the base station is obtained by calculating the theoretical received signal power and the theoretical signal-to-interference-plus-noise ratio, and the installation location of the base station is determined accordingly. However, since theoretical signal analysis is often based on idealized propagation environment assumptions, the actual signal coverage in practical applications deviates significantly from theoretical expectations. When base stations are deployed according to the theoretical analysis results, the actual signal coverage effect often does not match the theoretical expectations, affecting the overall performance of the communication network and the user experience. Summary of the Invention

[0004] This invention addresses the technical problem in existing technologies where the determination of base station installation locations relies solely on theoretical signal analysis, leading to discrepancies between actual signal coverage and theoretical expectations after base station deployment. It provides a base station installation optimization method and system based on signal twin model analysis to solve this problem.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0006] In a first aspect, the present invention provides a base station installation optimization method based on signal twin model analysis, comprising: randomly configuring a base station distribution location list; matching a base station cell list from a base station cell database based on the base station model, wherein a base station cell represents the service range of a pre-stored base station providing wireless signals; processing the base station distribution location list and the base station cell list using a signal twin model to determine the theoretical signal received power distribution and the theoretical signal interference-to-noise ratio distribution of the target area; traversing the base station distribution location list and the base station cell list according to the building distribution topology of the target area, correcting the theoretical signal received power and the theoretical signal interference-to-noise ratio, and determining the corrected signal received power and the corrected signal interference-to-noise ratio; partitioning the target area based on the corrected signal received power and the corrected signal interference-to-noise ratio to obtain abnormal signal areas; and setting the base station distribution location list as the base station installation target scheme when the abnormal signal area is equal to 0.

[0007] Secondly, the present invention provides a base station installation optimization system based on signal twin model analysis, comprising: a base station configuration module, used to randomly configure a list of base station distribution locations and match a list of base station cells from a base station cell database based on the base station model, wherein a base station cell represents the service range of a pre-stored base station providing wireless signals; a signal modeling module, used to process the list of base station distribution locations and the list of base station cells using a signal twin model to determine the theoretical signal received power distribution and the theoretical signal interference-to-noise ratio distribution of a target area; a signal correction module, used to traverse the list of base station distribution locations and the list of base station cells according to the building distribution topology of the target area, correct the theoretical signal received power and the theoretical signal interference-to-noise ratio, and determine the corrected signal received power and the corrected signal interference-to-noise ratio; an anomaly detection module, used to partition the target area based on the corrected signal received power and the corrected signal interference-to-noise ratio to obtain anomaly signal areas; and a scheme determination module, used to set the list of base station distribution locations as the target scheme for base station installation when the number of anomaly signal areas is equal to 0.

[0008] The beneficial effects of this invention are:

[0009] A randomized list of base station locations is configured, and the base station cell list is matched from the base station cell database based on the base station model. This provides fundamental base station deployment configuration data for subsequent signal analysis, ensuring accurate matching between base station models and their corresponding service areas during the analysis process. Using a signal twin model, the list of base station locations and the list of base station cells are processed to determine the theoretical signal received power distribution and theoretical signal-to-noise ratio (SNR) distribution in the target area. This yields the signal distribution under theoretical conditions, providing a reliable theoretical benchmark for subsequent environmental corrections. Based on the building topology of the target area, the list of base station locations and the list of base station cells are traversed to correct the theoretical signal received power and theoretical SNR, determining the corrected signal received power and corrected SNR. The impact of buildings on signal propagation in the actual geographical environment is considered, making the theoretical calculation results closer to actual environmental conditions and effectively reducing the deviation between theory and reality. Based on the corrected signal received power and corrected SNR, the target area is divided into zones to obtain abnormal signal areas. Through zone analysis, areas with poor signal coverage are identified, providing clear improvement targets and basis for base station layout optimization. When the abnormal signal area is equal to 0, the base station installation target scheme is set according to the base station distribution location list, which ensures that the final base station installation scheme can achieve full coverage and no abnormal signal area, thereby ensuring that the actual deployment effect reaches the theoretical expectation.

[0010] The above technical solution achieves the technical effect of correcting theoretical signal parameters based on the actual environment, optimizing the base station installation layout, and ensuring that the actual deployment effect meets the theoretical expectations. Attached Figure Description

[0011] Figure 1 A flowchart illustrating a base station installation optimization method for signal twin model analysis provided by the present invention;

[0012] Figure 2 This is a schematic diagram of the structure of a base station installation optimization system for signal twin model analysis provided by the present invention.

[0013] In the attached diagram, the components represented by each number are as follows:

[0014] Base station configuration module 11, signal modeling module 12, signal correction module 13, anomaly detection module 14, and scheme determination module 15. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 are within the scope of protection of the present invention.

[0016] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0017] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0018] Example 1, as Figure 1 As shown, this embodiment of the invention provides a base station installation optimization method based on signal twin model analysis, including:

[0019] S1. Randomly configure a list of base station distribution locations, and match the list of base station cells from the base station cell database based on the base station model. Here, the base station cell represents the service range of the wireless signal provided by the pre-stored base station.

[0020] Specifically, firstly, within the target area, a list of base station distribution locations is generated using a random configuration method. This random configuration method refers to randomly generating several candidate base station installation locations within a preset geographical area, excluding locations where base stations can be installed.

[0021] Next, the base station model uniformly used in this deployment plan is obtained, and the base station cell list is matched against the base station cell database based on this model. The base station cell database is a pre-built database that stores data categorized by base station model, covering the service range of different base stations providing wireless signals. Base station models can include classification information such as equipment manufacturer identification, power level, and frequency band support capabilities, for example, specific models like Huawei AAU5613 and ZTE AAUV2. The base station cell, as the basic coverage unit corresponding to the base station model, represents the effective range of wireless signal service provided by a single base station in an ideal unobstructed propagation environment. This effective range of wireless signal service is determined by parameters such as transmit power, antenna gain, operating frequency band, maximum coverage radius, vertical beamwidth, and horizontal beamwidth, forming a specific geographical coverage area, which serves as the service range of the base station providing wireless signals.

[0022] Specifically, based on a unified base station model, the corresponding base station cell is retrieved from the base station cell database. This cell is then applied to each base station location in the base station distribution location list, configuring the same wireless signal service range for each location, thus generating a base station cell list. This list contains two core elements: the base station's installation location and the service range of the wireless signal provided by the base station, providing foundational data for subsequent signal propagation analysis.

[0023] By configuring the base station distribution location list and matching the base station cell list, the spatial distribution information of the base stations and the corresponding coverage characteristic parameters were obtained, laying a data foundation for subsequent signal twin model analysis.

[0024] S2. Using a signal twin model, process the list of base station distribution locations and the list of base station cells to determine the theoretical signal received power distribution and the theoretical signal interference-to-noise ratio distribution of the target area.

[0025] Specifically, after obtaining the list of base station locations and the list of base station cells, a pre-trained signal twin model is used to process and analyze these lists. The signal twin model is a deep learning-based signal propagation prediction model. By learning the mapping relationship between base station configuration and signal distribution in a large number of areas without signal obstructions, this model can accurately predict signal coverage under ideal propagation conditions.

[0026] The signal twin model takes a list of base station locations and a list of base station cells as input data. The base station location list provides the geographical coordinates of each base station, and the base station cell list provides the service range of each base station's wireless signal coverage. Through its internal neural network structure, the signal twin model comprehensively analyzes the spatial distribution characteristics of base stations, the service range of wireless signals, and the mutual influence relationships between base stations. It calculates the theoretical signal propagation characteristics of each base station location within the target area, including the theoretical received power distribution and the theoretical signal-to-interference-plus-noise ratio (SINR) distribution. Specifically, the theoretical received power distribution characterizes the distribution of base station signal strength received at each base station location within the target area; the theoretical SINR distribution characterizes the signal quality distribution at each base station location within the target area. The theoretical received power distribution and theoretical SINR distribution reflect the signal coverage effect and signal quality distribution of the base station location list under an ideal, unobstructed propagation environment.

[0027] By determining the theoretical signal received power distribution and theoretical signal interference-to-noise ratio distribution in the target area, a theoretical reference benchmark is provided for subsequent signal correction analysis combined with actual environmental factors.

[0028] S3. Based on the building distribution topology of the target area, traverse the base station distribution location list and the base station cell list, correct the theoretical signal received power and the theoretical signal interference-to-noise ratio, and determine the corrected signal received power and the corrected signal interference-to-noise ratio.

[0029] Specifically, after obtaining the theoretical signal received power distribution and theoretical signal-to-interference-plus-noise ratio (SINR) distribution of the target area, environmental corrections are made to these distributions based on the building topology of the target area. The building topology of the target area refers to the spatial distribution structure of buildings within the target area, including information such as the building's location coordinates, height, shape, and material properties. These factors can affect the propagation path of wireless signals through factors such as obstruction, reflection, and attenuation.

[0030] First, we iterate through the base station distribution location list and the base station cell list for each base station configuration. For each base station's coverage area, we analyze the impact of the building distribution topology in the target area on the base station's signal propagation. Specifically, we extract the base station's geographical location and coverage range parameters, then identify the building distribution within the base station's coverage area, and analyze the obstruction and attenuation effects of buildings on the signal propagation path.

[0031] Subsequently, based on the building distribution topology of the target area, the signal propagation loss correction for each base station at each location within its coverage area is calculated. For the theoretical signal received power, a power attenuation correction value is calculated based on the obstruction and attenuation effects of buildings. The theoretical signal received power is then superimposed with the power attenuation correction value to obtain the corrected signal received power. For the theoretical signal-to-interference-plus-noise ratio (SINR), considering the multipath effect and reflection interference caused by buildings, an SINR correction value is calculated. The theoretical SINR is then superimposed with the SINR correction value to obtain the corrected SINR.

[0032] Through the above correction process, corrected signal received power and corrected signal interference-to-noise ratio were obtained, taking into account the influence of the actual building environment. The corrected signal parameters can more accurately reflect the signal coverage effect and signal quality distribution of the base station in the actual deployment environment, providing a more realistic data basis for subsequent base station optimization analysis.

[0033] S4. Based on the corrected signal received power and the corrected signal interference-to-noise ratio, the target area is divided into regions to obtain abnormal signal regions.

[0034] Specifically, after obtaining the corrected signal received power and corrected signal interference-to-noise ratio (JNR), signal quality zoning analysis is performed on the target area based on the corrected signal received power and corrected JNR to identify and classify abnormal signal areas. Abnormal signal areas refer to areas within the target area where the signal coverage quality does not meet the preset standards, including three types: weak coverage areas, over-coverage areas, and interference areas.

[0035] First, establish rules for identifying abnormal signal areas. The rule for weak coverage areas is: when the corrected signal received power at a location is below a preset minimum coverage threshold and the corrected signal interference-to-noise ratio (IRR) does not meet communication quality requirements, that location is identified as a weak coverage area. The rule for over-coverage areas is: when the corrected signal received power at a location is too high and there is overlap of strong signals from multiple base stations causing a decrease in the corrected signal IRR, that location is identified as an over-coverage area. The rule for interference areas is: when the corrected signal received power at a location is at a moderate level but the corrected signal IRR is severely degraded, that location is identified as an interference area.

[0036] Then, the process of identifying abnormal signal areas is performed by traversing each geographical location within the target area. Specifically, for each geographical location, the corrected signal received power and corrected signal interference-to-noise ratio (EMIR) are extracted and compared one by one according to preset weak coverage determination rules, over-coverage determination rules, and interference area determination rules. When the corrected signal received power and corrected EMIR of a certain location meet any one of the above three determination rules, the location is identified as an abnormal signal area.

[0037] Through the above-described partitioning analysis process, the signal quality assessment of all locations within the target area was completed, and abnormal signal areas, including weak coverage areas, over-coverage areas, and interference areas, were obtained. This provides a clear basis for problem identification and improvement for subsequent base station deployment optimization decisions.

[0038] S5. When the abnormal signal area is equal to 0, set the base station distribution location list as the base station installation target scheme.

[0039] Specifically, after identifying abnormal signal areas in the target area, the number of abnormal signal areas is statistically evaluated. When the number of abnormal signal areas is 0, it indicates that the base station deployment scheme corresponding to the current base station distribution location list does not have weak coverage areas, over-coverage areas, or interference areas in the target area, meaning that the signal coverage quality of all areas meets the preset standard requirements.

[0040] In this context, an abnormal signal area of ​​0 indicates that the corrected signal received power and corrected signal interference-to-noise ratio at each location within the target area meet communication quality requirements, and the signal coverage configuration among base stations has reached an ideal balance. At this point, the current list of base station locations already provides effective coverage of the target area, requiring no further optimization. Therefore, the current list of base station locations is selected as the target base station installation plan. This target plan includes verified base station installation location information, which, considering the impact of the actual building environment, can provide wireless signal coverage services that meet communication quality requirements for the target area.

[0041] Through the above-mentioned determination process, the base station deployment plan was confirmed, and a target base station installation plan that can be directly used for actual engineering deployment was obtained, achieving precise optimization of the base station installation location.

[0042] Furthermore, embodiments of this application also include:

[0043] S6. When the abnormal signal area is not equal to 0, update the base station distribution location list and then execute the loop;

[0044] S7. If the number of cycles is greater than or equal to the preset number of cycles and no target base station installation scheme appears, the fitness of several historical base station distribution location lists is evaluated based on the fitness evaluation rules to obtain the fitness of several historical base station distribution schemes.

[0045] S8. Based on the adaptability of the aforementioned historical base station distribution schemes, perform guided updates on the base station distribution locations to obtain expanded base station distribution locations, and then execute the loop.

[0046] In one feasible implementation, when the abnormal signal area is not equal to 0, it indicates that the deployment scheme corresponding to the current base station distribution location list still has weak coverage areas, over-coverage areas, or interference areas, failing to meet the signal coverage quality requirements. In this case, the base station distribution location list is updated and optimized. Specifically, firstly, the spatial distribution characteristics and specific types of abnormal signal areas are analyzed. For weak coverage areas, optimization can be achieved by adding base station locations within or around the area, or by adjusting existing base station locations closer to the weak coverage area; for over-coverage areas, optimization can be achieved by deleting redundant base stations or adjusting base station locations away from the over-coverage area; for interference areas, optimization can be achieved by adjusting the relative positional relationships of related base stations or replacing some base station locations. The update operation includes various adjustment strategies such as precise movement of base station locations, configuration of new base station locations, and deletion of redundant base station locations. After the update is completed, the complete processing flow of steps S2 to S4 is re-executed, i.e., the theoretical signal received power distribution and theoretical signal interference-to-noise ratio distribution of the target area are recalculated using the signal twin model, environmental correction analysis is performed based on the building distribution topology to determine the corrected signal parameters, and abnormal signal areas are re-identified and re-divided to verify whether the updated base station distribution scheme achieves the expected optimization effect.

[0047] When no suitable base station installation plan is found after the number of iterations exceeds or equals a preset threshold, it indicates that the current optimization strategy may be trapped in a local optimum, making it difficult to find a globally optimal base station deployment plan. At this point, a fitness evaluation mechanism based on historical data is activated to comprehensively evaluate the performance of several historical base station distribution location lists accumulated during the iteration process. Specifically, firstly, several historical base station distribution location lists generated in historical iterations and their corresponding abnormal signal area analysis results are collected. Then, based on preset fitness evaluation rules, a quantitative score is calculated for each historical base station distribution location list to obtain the fitness of several historical base station distribution plans. First, the area sizes of weak coverage areas, over-coverage areas, and interference areas in each historical base station distribution location list are statistically analyzed. The areas of each type of abnormal area are multiplied by their corresponding weights, and then the three weighted results are summed to obtain the fitness of the historical base station distribution plan for that historical base station distribution location list. By calculating the fitness of several historical base station distribution plans one by one, the fitness of several historical base station distribution plans is obtained, reflecting the overall performance differences between different historical base station distribution plans.

[0048] Subsequently, based on the fitness of several historical base station distribution schemes, historical base station distribution schemes with high fitness values ​​were identified, and effective configuration features and optimization experiences from these schemes were extracted to guide subsequent base station distribution location guidance and update operations. First, historical base station distribution schemes with high fitness values ​​were analyzed as high-quality historical base station distribution schemes, and key features such as the spatial distribution patterns of base station locations, the distance configuration characteristics between base stations, and their positional relationships relative to abnormal signal areas were extracted. Then, based on these extracted key features, guidance and update operations were performed on the base station distribution locations. According to the base station distribution patterns of high-quality historical base station distribution schemes, new candidate base station locations were added to the current base station distribution location list, forming an expanded base station location search space; existing base station locations were adjusted to move closer to efficient location areas in high-quality historical base station distribution schemes; and the configuration advantages of multiple high-quality historical schemes were integrated to generate a more optimal base station distribution scheme.

[0049] Through the aforementioned guided update process, the expanded base station distribution locations are obtained. These expanded base station distribution locations not only retain valid information from the current base station distribution location list but also incorporate configuration experience from high-quality historical base station distribution schemes. Subsequently, using the expanded base station distribution locations as new input data, the complete cyclical processing flow of steps S2 to S4 is re-executed to continue searching for the optimal base station installation target scheme.

[0050] Furthermore, the fitness evaluation rules include:

[0051] S71. Configure a first weight for weak coverage areas, configure a second weight for over-coverage areas, and configure a third weight for interference areas, wherein the first weight is less than the second weight, and the second weight is less than the third weight;

[0052] S72. Based on the first weight, the second weight, and the third weight, the weak coverage area, the over-coverage area, and the interference area are weighted by area and then summed to obtain the fitness of the base station distribution location list.

[0053] In a preferred embodiment, in order to accurately quantify the impact of different types of abnormal signal regions on the overall performance of the base station deployment scheme, corresponding weight coefficients are configured for each type of abnormal signal region.

[0054] Specifically, a first weight is assigned to weak coverage areas. This weight is relatively small because weak coverage areas mainly affect signal coverage range. Although they may cause localized areas to lose signal, the impact is relatively limited and can be effectively improved by adding base stations or adjusting their locations. A second weight is assigned to over-coverage areas. This weight is at a medium level because over-coverage areas waste base station resources and may cause signal interference between neighboring cells, affecting overall network efficiency, but not directly interrupting communication services. A third weight is assigned to interfering areas. This weight is the largest because interfering areas severely degrade signal quality, leading to unstable communication connections, increased data transmission error rates, and even potential communication outages, having the most serious impact on user experience and network performance. The weighting order is set as follows: first weight < second weight, second weight < third weight, i.e., weak coverage area weight < over-coverage area weight < interfering area weight, ensuring that the fitness assessment prioritizes and addresses the most severe interference issues.

[0055] Then, based on the configured first, second, and third weights, area weighted calculations are performed on weak coverage areas, over-coverage areas, and interference areas, and the fitness value of the base station distribution location list is obtained by summing the results. First, the total area of ​​weak coverage areas, over-coverage areas, and interference areas within the target area are calculated separately, with the area unit being square meters. Then, the total area of ​​weak coverage areas is multiplied by the first weight to obtain the weighted area value of weak coverage areas; the total area of ​​over-coverage areas is multiplied by the second weight to obtain the weighted area value of over-coverage areas; and the total area of ​​interference areas is multiplied by the third weight to obtain the weighted area value of interference areas. Finally, the weighted area values ​​of the three types of abnormal signal areas are summed, i.e.: Base station distribution location list fitness = Total area of ​​weak coverage areas × First weight + Total area of ​​over-coverage areas × Second weight + Total area of ​​interference areas × Third weight.

[0056] By using fitness evaluation rules, the fitness of several historical base station distribution schemes in a list of historical base station locations can be evaluated, reflecting the overall performance level of the corresponding base station deployment scheme, and providing a quantitative evaluation basis for subsequent guidance updates and scheme selection.

[0057] Furthermore, using a signal twin model, the list of base station distribution locations and the list of base station cells are processed to determine the theoretical signal received power distribution and theoretical signal-to-interference-plus-noise ratio distribution of the target area, including:

[0058] S21. Collect multiple sets of data in areas without signal obstructions, wherein any set of the multiple sets of data includes base station distribution location list record data, base station cell list record data, and a label indicating the distribution of signal received power, as well as a label indicating the distribution of signal interference-to-noise ratio.

[0059] The process for determining the tag that identifies the distribution of received signal power is as follows:

[0060] S211. Collect data on the distribution location list and base station cell list of areas without signal obstructions.

[0061] S212. Using the base station distribution location list recorded data and the base station cell list recorded data as constraints, collect several signal reception power distribution recorded data in areas without signal obstructions.

[0062] S213. Perform mode analysis on the recorded data of the plurality of signal received power distributions to obtain labels that identify the signal received power distributions, and determine the labels that identify the signal interference-to-noise ratio distributions in the same manner.

[0063] S22. Using tags that identify the distribution of received signal power and tags that identify the distribution of signal interference-to-noise ratio as supervision, and using the base station distribution location list record data and the base station cell list record data as input, retrieve the multiple sets of data to train the signal twin model.

[0064] In a preferred embodiment, to construct a signal twin model capable of accurately predicting theoretical signal distribution, multiple sets of data need to be collected from signal-obstruction-free areas. These signal-obstruction-free areas refer to ideal propagation environments free from buildings, terrain obstacles, or other signal propagation barriers. Signal data collected in such environments accurately reflects the theoretical signal propagation characteristics of the base station. Specifically, different base station configurations are deployed in multiple different signal-obstruction-free test areas, and corresponding base station distribution location lists and base station cell lists are recorded. Simultaneously, comprehensive signal strength and quality measurements are performed within the test areas using professional signal measurement equipment, obtaining labels identifying signal received power distribution and signal interference-to-noise ratio (SNR) distribution. Each set of data contains four core elements: base station distribution location list data, recording the geographical coordinates and installation parameters of the base station; base station cell list data, recording the coverage area and technical parameters of the base station; labels identifying signal received power distribution, recording the actual signal received power values ​​at each location within the test area; and labels identifying SNR distribution, recording the actual SNR values ​​at each location within the test area.

[0065] The process for determining the tags that identify the distribution of received signal power is as follows:

[0066] First, within a selected area free of signal obstructions, base station location list data and base station cell list data were collected. The base station location list data contained the precise geographic coordinates of all base stations in this test scenario. The base station cell list data contained technical parameters such as the corresponding base station's transmit power, antenna gain, operating frequency band, and coverage area. Then, using the collected base station location list data and base station cell list data as constraints, signal measurements were performed within the unobstructed area. These constraints ensured the consistency and accuracy of the signal measurement process with the base station configuration parameters. By setting up dense measurement points within the test area and using professional signal analyzers and other equipment, several signal received power distribution records were collected. These data reflected the actual signal received power values ​​at each measurement point under a specific base station configuration. Subsequently, mode analysis was performed on the collected signal received power distribution records. Mode analysis refers to statistically analyzing multiple measurement results at the same geographical location and extracting the most frequently occurring value as the representative signal received power value for that location. Mode analysis can effectively eliminate the influence of random errors and outliers in the measurement process, obtain more stable and reliable signal received power distribution data, and form a label to identify the signal received power distribution.

[0067] The same data acquisition and mode analysis methods were used to determine the tags that indicate the signal-to-interference-plus-noise ratio (SINR) distribution. Specifically, by measuring the SINR values ​​at various locations within the same test area and performing repeated measurements and mode analysis, stable and reliable tags that indicate the SINR distribution were obtained.

[0068] Subsequently, based on multiple sets of collected data, a supervised learning method was used to train the signal twin model. Specifically, the training process used labels identifying the distribution of received signal power and the distribution of signal interference-to-noise ratio (SNR) as supervision signals, and data recorded in the base station location list and the base station cell list as input features, retrieving multiple sets of data to construct a complete training dataset. During training, the signal twin model learned the complex mapping relationship between base station configuration parameters and signal distribution through deep learning algorithms, while extracting key information from input features such as base station location and coverage parameters, and establishing a nonlinear mapping relationship between the received signal power distribution and SNR distribution in the target area. Through repeated forward propagation, error calculation, and backpropagation processes, the parameters in the signal twin model were continuously optimized and adjusted until the error between the output of the signal twin model and the true labels reached the preset convergence condition. After training, the signal twin model has the ability to accurately predict the theoretical signal distribution of a corresponding area based on any base station configuration parameters, providing a reliable theoretical signal calculation tool for subsequent base station deployment optimization analysis.

[0069] Furthermore, based on the building distribution topology of the target area, the list of base station locations and the list of base station cells are traversed to correct the theoretical signal received power and the theoretical signal-to-interference-plus-noise ratio (SINR), determining the corrected signal received power and corrected SINR, including:

[0070] S31. Extract the first base station distribution location from the base station distribution location list, and extract the first base station cell from the base station cell list;

[0071] S32. Based on the distribution location of the first base station and the first base station cell, and in conjunction with the building distribution topology of the target area, configure the first signal received power deviation vector distribution and the first signal interference-to-noise ratio deviation vector distribution of the first base station cell distribution area.

[0072] S33. When all base stations are analyzed, sum the signal received power deviation vector distributions of all signal received power deviation vector distributions in the distribution area of ​​the first base station cell to obtain the received power deviation vector distribution of the first region, and add it to the received power deviation vector distributions of several regions.

[0073] S34. When all base stations are analyzed, sum the signal interference ratio (SNR) deviation vector distributions of all cells in the distribution area of ​​the first base station cell at the same location to obtain the signal interference ratio (SNR) deviation vector distribution of the first region, and add it to the signal interference ratio (SNR) deviation vector distributions of several regions.

[0074] S35. Sum the received power deviation vector distribution of the several regions with the theoretical signal received power at the same position to obtain the corrected signal received power;

[0075] S36. The signal interference-to-noise ratio deviation vector distribution of the several regions is summed with the theoretical signal interference-to-noise ratio at the same position to obtain the corrected signal interference-to-noise ratio.

[0076] In a preferred embodiment, to perform accurate environmental correction analysis on the signal propagation of each base station, it is necessary to extract the configuration information of each base station individually. Specifically, the location of a first base station is extracted from the list of base station locations, and this location information includes the geographic coordinate parameters of the first base station. Simultaneously, the first base station cell corresponding to the location of the first base station is extracted from the list of base station cells, and this first base station cell includes the service range of the wireless signal provided by the first base station.

[0077] Then, based on the extracted locations of the first base station and the first base station cell, an environmental impact analysis is performed in conjunction with the building distribution topology of the target area, and signal correction parameters for the distribution area of ​​the first base station cell are configured. The building distribution topology of the target area includes detailed parameters such as the location coordinates, height information, planar shape, and building materials of the buildings. These factors will have complex effects on the propagation path of wireless signals, such as obstruction, reflection, diffraction, and absorption. Specifically, the theoretical coverage area of ​​the first base station cell is first determined, and then the distribution of all buildings within this coverage area is identified. By calculating the relative positional relationship between the buildings and the first base station, the degree of obstruction of the signal propagation path by the buildings, and the signal reflection characteristics of the building surfaces, the specific impact of the buildings on signal propagation is determined. Based on the above environmental impact analysis, the first signal received power deviation vector distribution and the first signal interference-to-noise ratio deviation vector distribution are configured for the distribution area of ​​the first base station cell. The first signal received power deviation vector distribution refers to the distribution of the correction amount of the signal parameters deviating from the theoretical value due to the influence of the building environment at each location point within the coverage area of ​​the first base station. The first signal received power deviation vector distribution reflects the degree of influence of building obstruction and attenuation on signal strength, and the first signal interference-to-noise ratio deviation vector distribution reflects the degree of influence of multipath effects and reflection interference caused by buildings on signal quality.

[0078] After completing the environmental impact analysis of all base stations in the base station distribution location list and base station cell list, the signal deviations generated by each base station are spatially aggregated. Specifically, the signal received power deviation vector distributions within the distribution area of ​​the first base station cell are summed at the same location. Summing at the same location means accumulating the signal received power deviation values ​​from different base stations at each geographical location point within the target area to obtain the overall signal received power deviation at that location point. Through this process, the first area received power deviation vector distribution is obtained, reflecting the overall signal received power correction amount caused by the building environment at each location point within the distribution area of ​​the first base station cell. Subsequently, the first area received power deviation vector distribution is added to the received power deviation vector distributions of several areas. Using the same processing method as in step S33, the signal interference-to-noise ratio (SNR) deviation vector distributions within the distribution area of ​​the first base station cell are summed at the same location. By accumulating the SNR deviation values ​​from different base stations at each geographical location point, the first area SNR deviation vector distribution is obtained. The first region's signal-to-interference-plus-noise ratio (SNR) deviation vector distribution reflects the overall SNR correction caused by the building environment at each location point within the distribution area of ​​the first base station cell, including the combined effects of multipath interference, reflection noise, and signal quality degradation. After calculation, the first region's SNR deviation vector distribution is added to the SNR deviation vector distributions of several regions.

[0079] Subsequently, based on the aggregated received power deviation vector distributions of several regions, environmental correction processing is performed on the theoretical signal received power. Specifically, the received power deviation vector distributions of several regions are summed with the theoretical signal received power at the same location. The theoretical signal received power value at each geographical location within the target area is added to the received power deviation value at the corresponding location to obtain the actual signal received power value at that location after considering the influence of the building environment. By performing point-by-point calculations across the entire target area, the corrected signal received power is obtained. The corrected signal received power accurately reflects the signal coverage strength distribution of the base station in the actual building environment, providing a more realistic data basis for subsequent coverage quality assessment. Using the same calculation method as in step S35, the signal interference-to-noise ratio (SNR) deviation vector distributions of several regions are summed with the theoretical SNR at the same location. By adding the theoretical SNR value at each geographical location to the corresponding SNR deviation value, the corrected SNR after considering the influence of the building environment is obtained. The corrected SNR accurately reflects the signal quality distribution of the base station in the actual building environment, comprehensively considering the impact of various environmental factors such as multipath interference, reflection noise, and signal attenuation caused by buildings on communication quality.

[0080] Through the above environmental correction process, the corrected signal received power and corrected signal interference-to-noise ratio are obtained, which can more accurately reflect the signal coverage performance and communication quality level of the base station deployment scheme in the actual environment, and provide a reliable technical basis for optimizing the decision of base station installation location.

[0081] Furthermore, based on the distribution location of the first base station and the first base station cell, and in conjunction with the building distribution topology of the target area, a first signal received power deviation vector distribution and a first signal interference-to-noise ratio deviation vector distribution are configured for the distribution area of ​​the first base station cell, including:

[0082] S321. Extract the first building distribution topology of the first base station cell from the building distribution topology of the target area;

[0083] S322. Using the distribution topology of the first building and the distribution location of the first base station as constraints, retrieve the signal received power distribution sample set and the signal interference-to-noise ratio distribution sample set;

[0084] S323. Using the signal twin model, process the distribution location of the first base station and the first base station cell to obtain the first theoretical signal received power distribution and the first theoretical signal interference-to-noise ratio distribution.

[0085] S324. Traverse the signal received power distribution sample set, calculate the deviation with the first theoretical signal received power distribution, obtain the signal received power deviation vector distribution set, perform LOF outlier factor minimum extraction at the same location, and obtain the first signal received power deviation vector distribution of the first base station cell distribution area.

[0086] S325. Traverse the signal interference-to-noise ratio (SNR) distribution sample set, calculate the deviation from the first theoretical SNR distribution, obtain the SNR deviation vector distribution set, perform LOF outlier minimum extraction at the same location, and obtain the first SNR deviation vector distribution in the distribution area of ​​the first base station cell.

[0087] In a preferred embodiment, to accurately analyze the specific impact of the building environment on the signal propagation of the first base station, building information related to the first base station cell is extracted from the building distribution topology of the target area. Specifically, based on the coverage area of ​​the first base station cell, a corresponding spatial region is delineated in the building distribution topology of the target area, and the first building distribution topology within that region is extracted. The first building distribution topology contains detailed information on all buildings within the coverage area of ​​the first base station cell, including key parameters such as the building's geographical coordinates, planar shape outline, building height, number of floors, building material type, and relative positional relationships between buildings. These building parameters directly affect the propagation path, propagation loss, and propagation quality of the wireless signal within that region.

[0088] Then, based on the extracted topology of the first building distribution and the location of the first base station as constraints, a matching signal distribution sample set is retrieved from a pre-built historical database. The constraints ensure that the retrieved sample data has similar environmental characteristics and base station configuration features to the current analysis scenario. Specifically, the retrieval process first establishes environmental similarity matching rules, which comprehensively consider the similarity across multiple dimensions, including building density, building height distribution, the relative positional relationship between the base station and the building, and base station configuration parameters. Through a multi-dimensional matching algorithm, historical test scenarios highly similar to the current topology of the first building distribution and the location of the first base station are selected from the historical database, and the signal received power distribution and signal interference-to-noise ratio (SNR) distribution are extracted from them, forming a signal received power distribution sample set and a SNR distribution sample set. The signal received power distribution sample set contains actual measured signal received power distribution data under multiple similar environments, and the SNR distribution sample set contains the corresponding SNR distribution data. These samples provide a true reference benchmark for subsequent deviation calculations and environmental corrections.

[0089] Subsequently, a pre-trained signal twin model is used to process the location and configuration parameters of the first base station cell to calculate signal propagation under ideal unobstructed conditions. Specifically, the signal twin model takes the geographical coordinates of the first base station location and the technical parameters of the first base station cell as input, and calculates the signal propagation under ideal unobstructed conditions using a deep neural network. The outputs are a first theoretical signal received power distribution and a first theoretical signal-to-noise ratio (SNR) distribution. The first theoretical signal received power distribution reflects the signal strength distribution of the first base station at various locations under ideal propagation conditions, and the first theoretical SNR distribution reflects the corresponding signal quality distribution.

[0090] To determine the specific impact of the building environment on signal reception power, a comparative analysis of signal distribution samples under actual conditions and theoretical signal distributions was conducted. Specifically, each sample data in the signal reception power distribution sample set was traversed, and the deviation was calculated point-by-point with the first theoretical signal reception power distribution. The numerical values ​​of each signal reception power distribution sample in the sample set were subtracted from the first theoretical signal reception power distribution at the same geographical location to obtain the signal reception power deviation vector distribution for that sample. By calculating the deviation for all samples in the sample set, a signal reception power deviation vector distribution set was obtained, containing deviation distribution data from multiple different test scenarios. To extract the most representative and reliable deviation data from multiple deviation samples, the Local Outlier Factor (LOF) minimum extraction method was adopted. LOF is an anomaly detection algorithm based on local density, capable of identifying outliers and abnormal points in a dataset. Specifically, for each geographical location, the LOF outlier factor value of each sample data in the signal reception power deviation vector distribution set for that location was calculated, and the sample data with the smallest LOF outlier factor was selected as the representative deviation value for that location. By extracting the minimum LOF value of all locations within the distribution area of ​​the first base station cell, the first signal received power deviation vector distribution of the first base station cell distribution area is obtained. This deviation vector distribution can accurately reflect the typical influence pattern of the building environment on the signal received power of the first base station, eliminating interference from abnormal measurement values ​​and random errors.

[0091] Simultaneously, using the same technical method as in step S324, deviation analysis and optimization extraction of the signal interference-to-noise ratio (SNR) are performed. Specifically, each sample data in the SNR distribution sample set is traversed, and the deviation is calculated point-by-point with the first theoretical SNR distribution. The SNR deviation vector distribution of each sample is obtained through numerical subtraction, forming a SNR deviation vector distribution set. Subsequently, the minimum LOF outlier value extraction method at the same location is used to optimize the data of the SNR deviation vector distribution set. By calculating the LOF outlier value of each location point in each sample data, the sample data with the smallest outlier value is selected as the representative SNR deviation value for that location point, ensuring that the extracted deviation data has the highest reliability and representativeness.

[0092] Through the above processing, the first signal-to-interference-plus-noise ratio (SIR) deviation vector distribution in the distribution area of ​​the first base station cell is obtained. This deviation vector distribution accurately reflects the typical influence characteristics of the building environment on the SIR of the first base station, including the combined effects of multipath interference, reflection noise, and signal quality degradation.

[0093] Through the complete processing flow of steps S321 to S325, the first signal received power deviation vector distribution and the first signal interference-to-noise ratio deviation vector distribution of the first base station cell distribution area were successfully configured. These deviation vector distribution data provide accurate and reliable correction parameters for subsequent signal environment correction calculations.

[0094] Furthermore, based on the corrected signal received power and the corrected signal interference-to-noise ratio, the target region is divided into regions to obtain abnormal signal regions, including:

[0095] S41. Based on the distribution range of base station cells, obtain the number of overlapping cells in the first location of the target area;

[0096] S42. Extract the first position correction signal received power and the first position correction signal interference ratio from the correction signal received power and the correction signal interference ratio.

[0097] S43. Configure weak coverage determination rules, over-coverage determination rules, and interference area determination rules through the user terminal;

[0098] S44. When the number of overlapping cells at the first location, the received power of the corrected signal at the first location, and the interference-to-noise ratio of the corrected signal at the first location satisfy any one of the weak coverage determination rule, the over-coverage determination rule, and the interference area determination rule, the signal is added to the abnormal signal area.

[0099] In a preferred embodiment, firstly, based on the distribution range of base station cells, a cell coverage overlap analysis is performed on the target area to obtain the number of overlapping cells at each location point within the target area. The number of overlapping cells refers to the number of base station cells that can receive a valid signal at a specific geographical location; this parameter reflects the signal coverage richness and base station deployment density at that location. The calculation process for the number of overlapping cells is as follows: traversing each geographical location point within the target area, and determining whether the location point is within the effective coverage range of each base station cell based on its coverage range parameters. When the signal received power at a location point exceeds the minimum coverage threshold of a base station cell, that base station cell is included in the overlapping cell statistics for that location point. Through cumulative statistics, the number of overlapping cells at the first location and the distribution of overlapping cell counts across all locations within the target area are obtained.

[0100] To identify abnormal signal areas at specific locations, key signal parameters for those locations are extracted. Specifically, a first-location corrected signal received power is extracted from the corrected signal received power, and a first-location corrected signal interference-to-noise ratio (IRR) is extracted from the corrected signal received power. The first-location corrected signal received power reflects the actual signal strength level at that location after considering the influence of the building environment, while the first-location corrected signal IRR reflects the actual signal quality level at that location.

[0101] Subsequently, the system receives user-configured rules for weak coverage, over-coverage, and interference areas from the user terminal, establishing a standardized identification system for abnormal signal areas. The configuration of these rules is based on mobile communication network technical standards and practical engineering experience, ensuring the accuracy and practicality of abnormal area identification.

[0102] The following are exemplary rules for determining weak coverage, over-coverage, and interference areas:

[0103] Table 1 Weak Coverage Judgment Rules

[0104]

[0105] As shown in Table 1 above, the weak coverage determination rule is as follows: when the corrected signal received power at a location is lower than the minimum coverage threshold and the corrected signal interference-to-noise ratio cannot meet the basic communication quality requirements, and the number of overlapping cells is small, the location is determined to be a weak coverage area. Specific technical indicators include: for 4G LTE networks, the RSRP (Reference Signal Received Power) threshold is set to less than or equal to -110dBm, the SINR (Signal-to-Interference-plus-Noise Ratio) threshold is set to less than or equal to 0dB, and the overlapping cell number threshold is set to less than or equal to 1; for 5G NR networks, the RSRP threshold is set to less than or equal to -120dBm, the SINR threshold is set to less than or equal to -3dB, and the overlapping cell number threshold is set to less than or equal to 1.

[0106] Table 2 Over-coverage Decision Rules

[0107]

[0108] As shown in Table 2 above, the over-coverage determination rule is as follows: when the corrected signal received power at a location is too high and there is strong signal overlap from multiple base stations causing a decrease in the corrected signal interference-to-noise ratio, the location is determined to be an over-coverage area. Specific technical indicators include: for 4G LTE networks, the RSRP threshold is set to greater than or equal to -95dBm, the SINR threshold is set to less than 10dB, and the overlapping cell number threshold is set to greater than or equal to 3; for 5G NR networks, the RSRP threshold is set to greater than or equal to -100dBm, the SINR threshold is set to less than 5dB, and the overlapping cell number threshold is set to greater than or equal to 3.

[0109] Table 3 Rules for Determining Interference Areas

[0110]

[0111] As shown in Table 3 above, the interference area determination rule is as follows: when the corrected signal received power at a location is at a moderate level but the corrected signal interference-to-noise ratio (IRR) is severely degraded, that location is determined to be an interference area. Specific technical indicators include: for 4G LTE networks, the RSRP threshold is set to a moderately strong range of -95 to -80 dBm, the SINR threshold is set to less than 5 dB, and the overlapping cell number threshold is set to greater than or equal to 2; for 5G NR networks, the RSRP threshold is set to a range of -100 to -85 dBm, the SINR threshold is set to less than 0 dB, and the overlapping cell number threshold is set to greater than or equal to 2.

[0112] Subsequently, based on the configured judgment rules, abnormal signal areas are determined for each location point within the target area. Specifically, the number of overlapping cells at the first location, the corrected signal received power at the first location, and the corrected signal interference-to-noise ratio at the first location are compared one by one with preset weak coverage judgment rules, over-coverage judgment rules, and interference area judgment rules. The judgment process uses a logical OR operation, meaning that when the signal parameters of a location point meet any one of the three judgment rules, that location point is identified as an abnormal signal area. The specific judgment process is as follows: First, check whether the location point meets all the conditions of the weak coverage judgment rule; if it does, it is marked as a weak coverage area. If it does not meet the weak coverage conditions, then check whether it meets the over-coverage judgment rule; if it does, it is marked as an over-coverage area. If neither of the first two conditions is met, then check whether it meets the interference area judgment rule; if it does, it is marked as an interference area.

[0113] Through the above-described determination process, location points that meet the abnormal conditions are added to the classification markers of abnormal signal areas, and the areas are classified accordingly based on the specific determination rule type met. Ultimately, abnormal signal areas containing weak coverage areas, over-coverage areas, and interference areas are obtained, providing precise problem localization and improvement targets for subsequent base station deployment optimization.

[0114] Example 2, as Figure 2 As shown, based on the same inventive concept as the base station installation optimization method for signal twin model analysis provided in Embodiment 1, this embodiment of the invention also provides a base station installation optimization system for signal twin model analysis, comprising:

[0115] Base station configuration module 11 is used to randomly configure a list of base station distribution locations and match a list of base station cells from a base station cell database based on the base station model. Here, a base station cell represents the service range of the wireless signal provided by a pre-stored base station.

[0116] Signal modeling module 12 is used to process the base station distribution location list and the base station cell list using a signal twin model to determine the theoretical signal received power distribution and theoretical signal interference-to-noise ratio distribution of the target area;

[0117] The signal correction module 13 is used to traverse the base station distribution location list and the base station cell list according to the building distribution topology of the target area, correct the theoretical signal received power and the theoretical signal interference-to-noise ratio, and determine the corrected signal received power and the corrected signal interference-to-noise ratio.

[0118] Anomaly detection module 14 is used to partition the target area based on the corrected signal received power and the corrected signal interference-to-noise ratio to obtain anomaly signal areas;

[0119] The scheme determination module 15 is used to set the base station distribution location list as the base station installation target scheme when the abnormal signal area is equal to 0.

[0120] Furthermore, embodiments of this application also include an iterative optimization module, the execution steps of which include:

[0121] When the abnormal signal area is not equal to 0, update the base station distribution location list and then execute the loop;

[0122] If the target base station installation scheme is not found after the number of iterations is greater than or equal to the preset number of iterations, the fitness of several historical base station distribution locations is evaluated based on the fitness evaluation rules to obtain the fitness of several historical base station distribution schemes.

[0123] Based on the adaptability of the aforementioned historical base station distribution schemes, a guided update is performed on the base station distribution locations to obtain expanded base station distribution locations, and the process is repeated.

[0124] Furthermore, the fitness evaluation rules include:

[0125] A first weight is assigned to weak coverage areas, a second weight is assigned to over-coverage areas, and a third weight is assigned to interfering areas, wherein the first weight is less than the second weight, and the second weight is less than the third weight;

[0126] Based on the first weight, the second weight, and the third weight, the weak coverage area, the over-coverage area, and the interference area are weighted by area and then summed to obtain the fitness of the base station distribution location list.

[0127] Furthermore, the execution steps of the signal modeling module 12 include:

[0128] Multiple sets of data are collected from areas without signal obstructions. Each set of data includes a list of base station distribution locations, a list of base station cells, a label indicating the distribution of received signal power, and a label indicating the distribution of signal interference-to-noise ratio.

[0129] The process for determining the tag that identifies the distribution of received signal power is as follows:

[0130] Collect data on the distribution of base stations and the list of base station cells in areas without signal obstructions.

[0131] Using the base station distribution location list and the base station cell list as constraints, collect several signal reception power distribution records in areas without signal obstructions;

[0132] Perform a mode analysis at the same location on the recorded data of the signal received power distribution to obtain a label that identifies the signal received power distribution, and determine a label that identifies the signal interference-to-noise ratio distribution in the same way;

[0133] Using labels identifying the distribution of received signal power and labels identifying the distribution of signal interference-to-noise ratio as supervision, and taking the data recorded in the base station distribution location list and the data recorded in the base station cell list as input, the signal twin model is trained by retrieving the multiple sets of data.

[0134] Furthermore, the execution steps of the signal correction module 13 include:

[0135] Extract the first base station distribution location from the base station distribution location list, and extract the first base station cell from the base station cell list;

[0136] Based on the distribution location of the first base station and the first base station cell, and in combination with the building distribution topology of the target area, configure the first signal received power deviation vector distribution and the first signal interference-to-noise ratio deviation vector distribution of the first base station cell distribution area;

[0137] Once all base stations have been analyzed, the signal received power deviation vector distributions of all cells in the distribution area of ​​the first base station cell are summed at the same location to obtain the received power deviation vector distribution of the first region, which is then added to the received power deviation vector distributions of several regions.

[0138] Once all base stations have been analyzed, the signal interference ratio (SIR) deviation vector distributions of all cells in the distribution area of ​​the first base station cell are summed at the same location to obtain the signal interference ratio (SIR) deviation vector distribution of the first region, which is then added to the signal interference ratio (SIR) deviation vector distributions of several regions.

[0139] The corrected signal received power is obtained by summing the received power deviation vector distribution of the several regions with the theoretical signal received power at the same position.

[0140] The corrected signal interference ratio (SNR) is obtained by summing the distribution of the signal interference ratio deviation vector in the several regions with the theoretical signal interference ratio at the same position.

[0141] Furthermore, the execution steps of the signal correction module 13 also include:

[0142] Extract the first building distribution topology of the first base station cell from the building distribution topology of the target area;

[0143] Using the topology of the first building distribution and the location of the first base station distribution as constraints, retrieve the signal received power distribution sample set and the signal interference-to-noise ratio distribution sample set;

[0144] By processing the distribution location of the first base station and the first base station cell using the signal twin model, a first theoretical signal received power distribution and a first theoretical signal interference-to-noise ratio distribution are obtained;

[0145] Traverse the signal received power distribution sample set, calculate the deviation from the first theoretical signal received power distribution, obtain the signal received power deviation vector distribution set, perform LOF outlier minimum extraction at the same location, and obtain the first signal received power deviation vector distribution of the first base station cell distribution area.

[0146] Traverse the signal interference-to-noise ratio (SNR) distribution sample set, calculate the deviation from the first theoretical SNR distribution, obtain the SNR deviation vector distribution set, perform LOF outlier minimum extraction at the same location, and obtain the first SNR deviation vector distribution in the distribution area of ​​the first base station cell.

[0147] Furthermore, the execution steps of the anomaly detection module 14 include:

[0148] Based on the distribution range of base station cells, the number of overlapping cells in the first location of the target area is obtained;

[0149] The first position corrected signal received power and the first position corrected signal interference ratio are extracted from the corrected signal received power and the corrected signal interference ratio.

[0150] Configure weak coverage determination rules, over-coverage determination rules, and interference area determination rules through the user terminal;

[0151] When the number of overlapping cells at the first location, the received power of the corrected signal at the first location, and the interference-to-noise ratio of the corrected signal at the first location satisfy any one of the weak coverage determination rule, the over-coverage determination rule, and the interference area determination rule, it is added to the abnormal signal area.

[0152] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0153] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0154] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0155] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0156] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0157] Although preferred embodiments of the invention have been described, those skilled in the art, once they have learned the basic inventive concept, can make other changes and modifications to these embodiments.

[0158] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A base station installation optimization method based on signal twin model analysis, characterized in that, include: A randomized list of base station distribution locations is configured, and a list of base station cells is matched from a base station cell database based on the base station model. The base station cell represents the service range of the wireless signal provided by the pre-stored base station. Using a signal twin model, the list of base station locations and the list of base station cells are processed to determine the theoretical signal received power distribution and the theoretical signal-to-noise ratio distribution of the target area; Based on the building distribution topology of the target area, the list of base station distribution locations and the list of base station cells are traversed to correct the theoretical signal received power and the theoretical signal interference-to-noise ratio, and the corrected signal received power and the corrected signal interference-to-noise ratio are determined. Based on the corrected signal received power and the corrected signal interference-to-noise ratio, the target area is divided into regions to obtain abnormal signal regions. When the abnormal signal area is equal to 0, the list of base station distribution locations is set as the base station installation target scheme; Using a signal twin model, the list of base station locations and the list of base station cells are processed to determine the theoretical signal received power distribution and the theoretical signal-to-interference-plus-noise ratio distribution of the target area, including: Multiple sets of data are collected from areas without signal obstructions. Each set of data includes a list of base station distribution locations, a list of base station cells, a label indicating the distribution of received signal power, and a label indicating the distribution of signal interference-to-noise ratio. The process for determining the tag that identifies the distribution of received signal power is as follows: Collect data on the distribution of base stations and the list of base station cells in areas without signal obstructions. Using the base station distribution location list and the base station cell list as constraints, collect several signal reception power distribution records in areas without signal obstructions; Perform mode analysis at the same location on the recorded data of the signal received power distribution to obtain the label identifying the signal received power distribution, and determine the label identifying the signal interference-to-noise ratio distribution in the same way; Using labels that identify the distribution of received signal power and labels that identify the distribution of signal interference-to-noise ratio as supervision, and using the data recorded in the base station distribution location list and the data recorded in the base station cell list as input, the signal twin model is trained by retrieving the multiple sets of data. Based on the building distribution topology of the target area, the list of base station locations and the list of base station cells are traversed to correct the theoretical signal received power and the theoretical signal-to-interference-plus-noise ratio (SINR), determining the corrected signal received power and corrected SINR, including: Extract the first base station distribution location from the base station distribution location list, and extract the first base station cell from the base station cell list; Based on the distribution location of the first base station and the first base station cell, and in combination with the building distribution topology of the target area, configure the first signal received power deviation vector distribution and the first signal interference-to-noise ratio deviation vector distribution of the first base station cell distribution area; Once all base stations have been analyzed, the signal received power deviation vector distributions of all cells in the distribution area of ​​the first base station cell are summed at the same location to obtain the received power deviation vector distribution of the first region, which is then added to the received power deviation vector distributions of several regions. Once all base stations have been analyzed, the signal interference ratio (SIR) deviation vector distributions of all cells in the distribution area of ​​the first base station cell are summed at the same location to obtain the signal interference ratio (SIR) deviation vector distribution of the first region, which is then added to the signal interference ratio (SIR) deviation vector distributions of several regions. The corrected signal received power is obtained by summing the received power deviation vector distribution of the several regions with the theoretical signal received power at the same position. The corrected signal interference ratio is obtained by summing the signal interference ratio deviation vector distribution of the several regions with the theoretical signal interference ratio at the same position. Based on the distribution location of the first base station and the first base station cell, and in conjunction with the building distribution topology of the target area, a first signal received power deviation vector distribution and a first signal interference-to-noise ratio deviation vector distribution are configured for the distribution area of ​​the first base station cell, including: Extract the first building distribution topology of the first base station cell from the building distribution topology of the target area; Using the topology of the first building distribution and the location of the first base station distribution as constraints, retrieve the signal received power distribution sample set and the signal interference-to-noise ratio distribution sample set; By processing the distribution location of the first base station and the first base station cell using the signal twin model, a first theoretical signal received power distribution and a first theoretical signal interference-to-noise ratio distribution are obtained; Traverse the signal received power distribution sample set, calculate the deviation from the first theoretical signal received power distribution, obtain the signal received power deviation vector distribution set, perform LOF outlier minimum extraction at the same location, and obtain the first signal received power deviation vector distribution of the first base station cell distribution area. Traverse the signal interference-to-noise ratio (SNR) distribution sample set, calculate the deviation from the first theoretical SNR distribution, obtain the SNR deviation vector distribution set, perform LOF outlier factor minimum extraction at the same location, and obtain the first SNR deviation vector distribution of the first base station cell distribution area. Based on the corrected signal received power and the corrected signal interference-to-noise ratio, the target area is divided into regions to obtain abnormal signal regions, including: Based on the distribution range of base station cells, the number of overlapping cells in the first location of the target area is obtained; The first position corrected signal received power and the first position corrected signal interference ratio are extracted from the corrected signal received power and the corrected signal interference ratio. Configure weak coverage determination rules, over-coverage determination rules, and interference area determination rules through the user terminal; When the number of overlapping cells at the first location, the received power of the corrected signal at the first location, and the interference-to-noise ratio of the corrected signal at the first location satisfy any one of the weak coverage determination rule, the over-coverage determination rule, and the interference area determination rule, it is added to the abnormal signal area.

2. The method as described in claim 1, characterized in that, Also includes: When the abnormal signal area is not equal to 0, update the base station distribution location list and then execute the loop; If the target base station installation scheme is not found after the number of iterations is greater than or equal to the preset number of iterations, the fitness of several historical base station distribution locations is evaluated based on the fitness evaluation rules to obtain the fitness of several historical base station distribution schemes. Based on the adaptability of the aforementioned historical base station distribution schemes, a guided update is performed on the base station distribution locations to obtain expanded base station distribution locations, and the process is repeated.

3. The method as described in claim 2, characterized in that, The fitness evaluation rules include: A first weight is assigned to weak coverage areas, a second weight is assigned to over-coverage areas, and a third weight is assigned to interfering areas, wherein the first weight is less than the second weight and the third weight. Based on the first weight, the second weight, and the third weight, the weak coverage area, the over-coverage area, and the interference area are weighted by area and then summed to obtain the fitness of the base station distribution location list.

4. A base station installation optimization system based on signal twin model analysis, characterized in that, For implementing the method as described in any one of claims 1 to 3, comprising: The base station configuration module is used to randomly configure the list of base station distribution locations and match the base station cell list from the base station cell database based on the base station model. The base station cell represents the service range of the wireless signal provided by the pre-stored base station. The signal modeling module is used to process the list of base station distribution locations and the list of base station cells using a signal twin model to determine the theoretical signal received power distribution and the theoretical signal interference-to-noise ratio distribution of the target area. The signal correction module is used to traverse the base station distribution location list and the base station cell list according to the building distribution topology of the target area, correct the theoretical signal received power and the theoretical signal interference-to-noise ratio, and determine the corrected signal received power and the corrected signal interference-to-noise ratio. Anomaly detection module is used to partition the target area based on the corrected signal received power and the corrected signal interference-to-noise ratio to obtain anomaly signal areas; The scheme determination module is used to set the list of base station distribution locations as the base station installation target scheme when the abnormal signal area is equal to 0.

Images