Communication signal coverage optimization method and system based on artificial intelligence

Abstract

The invention discloses a communication signal coverage optimization method based on artificial intelligence, and relates to the field of artificial intelligence. Selecting a plurality of grid areas to collect signal intensity characteristic data; judging whether each grid region is a weak signal region and a signal blind region or not according to the preprocessed signal intensity characteristic data of the grid region; obtaining multiple groups of regional prediction training data; each group of region prediction training data comprises signal intensity characteristic data of each grid region and an identifier indicating whether the grid region is a weak signal region or a signal blind region; performing machine learning training on the multiple groups of region prediction training data to obtain a region prediction training model; the area prediction training model is used for judging whether the to-be-tested area is a weak signal area or a signal blind area; when the signal area is the weak signal area, dynamically adjusting the base station transmitting power of the to-be-tested area; and when the blind area is the signal blind area, dynamically adjusting the base station transmitting power, the antenna azimuth angle, the antenna downward inclination angle and / or the base station position of the base station in the to-be-tested area.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to a communication signal coverage optimization method and system based on artificial intelligence. Background Art

[0002] Due to the problem of uneven signal coverage in the existing communication network, the communication quality in some areas has declined, affecting the user experience. Currently, the signal coverage and communication quality are improved mainly by manually adjusting the base station parameters. This method is time-consuming and inefficient, and it is difficult to adapt to the rapidly changing network environment. In addition, with the continuous growth of network traffic, how to dynamically allocate resources to optimize network performance has become a major challenge facing the communication field. Therefore, there is a need for a communication signal coverage optimization method and system based on artificial intelligence, which can automatically optimize base station parameters and signal allocation, thereby improving the signal coverage quality and efficiency of the communication network, reducing labor costs, and enhancing network adaptability and robustness. Summary of the invention

[0003] The technical problem to be solved by the present invention is how to dynamically allocate communication signal resources to optimize network performance and reduce labor costs. The purpose is to provide a communication signal coverage optimization method and system based on artificial intelligence, which automatically optimizes base station parameters and resource allocation through artificial intelligence algorithms, thereby improving the signal coverage quality and efficiency of the communication network, enhancing the adaptability and robustness of network signals, and reducing labor costs.

[0004] The present invention is achieved through the following technical solutions:

[0005] A communication signal coverage optimization method based on artificial intelligence, comprising:

[0006] Divide the region into grid areas;

[0007] Selecting a plurality of the grid areas according to the requirements of the communication network coverage, and collecting signal strength characteristic data of each of the grid areas;

[0008] Preprocessing the signal strength characteristic data of each of the grid areas, and judging whether each of the grid areas is a weak signal area or a signal blind area according to the preprocessed signal strength characteristic data of each of the grid areas;

[0009] Acquire multiple groups of regional prediction training data; each group of the regional prediction training data includes the signal strength characteristic data of each grid area, and an identification of whether the grid area is a weak signal area or a signal blind area; the multiple groups of regional prediction training data are trained by machine learning to obtain a regional prediction training model; the regional prediction training model is used to determine whether the area to be tested is a weak signal area or a signal blind area;

[0010] When the area to be tested is a weak signal area, the base station transmission power of the area to be tested is dynamically adjusted to improve the communication quality of the area to be tested; when the area to be tested is a signal blind area, the base station transmission power, antenna azimuth, antenna downtilt angle, and / or base station position of the base station in the area to be tested are dynamically adjusted to expand the signal coverage range of the area to be tested.

[0011] The dynamically adjusting the base station transmission power of the area to be tested to improve the communication quality of the area to be tested includes:

[0012] When dynamically adjusting the base station transmission power of the area to be tested, obtain the real-time communication quality, energy consumption and signal coverage of the area to be tested, and select the base station transmission power that meets the preset communication quality standard, the lowest energy consumption and the maximum signal coverage.

[0013] When the base station transmission power of the area to be tested is dynamically adjusted, the updated signal strength characteristic data of the area to be tested is simultaneously input into the regional prediction training model to determine whether the area to be tested is a weak signal area; when the area to be tested is not a weak signal area, the real-time communication quality and energy consumption of the area to be tested are obtained according to the corresponding base station transmission power.

[0014] The method for optimizing communication signal coverage based on artificial intelligence also includes: collecting multiple groups of signal quality optimization training data; each group of the signal quality optimization training data includes the base station transmission power before adjustment of the weak signal area, and the base station transmission power after adjustment that meets the preset communication quality standard and has the lowest energy consumption; through training, multiple groups of signal quality optimization training data are used to obtain a base station power optimization model for adjusting the base station transmission power in the weak signal area.

[0015] The method of dynamically adjusting the antenna azimuth angle, antenna downtilt angle, base station transmission power, and / or base station position of the base station in the area to be tested to expand the signal coverage of the area to be tested includes:

[0016] When dynamically adjusting the antenna azimuth and / or antenna downtilt angle of the base station corresponding to the area to be tested, obtain the real-time signal coverage and communication quality of the area to be tested; select the antenna azimuth and / or antenna downtilt angle that meets the preset communication quality standard and has the maximum signal coverage.

[0017] When the antenna azimuth angle and / or antenna downtilt angle are dynamically adjusted, the updated signal strength characteristic data of the area to be tested is simultaneously input into the regional prediction training model to determine whether the area to be tested is a signal blind spot; when the area to be tested is not a signal blind spot, the real-time signal coverage and communication quality of the area to be tested are obtained according to the corresponding base station position.

[0018] The method of dynamically adjusting the antenna azimuth angle, antenna downtilt angle, base station transmission power, and / or base station position of the base station in the area to be tested to expand the signal coverage of the area to be tested includes:

[0019] Dynamically adjust the base station position of the area to be tested, obtain the real-time signal coverage of the area to be tested, and the signal interference between the corresponding base station and other base stations; select the base station position that satisfies the minimum signal interference between the corresponding base station and other base stations and the maximum signal coverage;

[0020] When dynamically adjusting the base station transmission power of the area to be tested, obtain the real-time communication quality, energy consumption and signal coverage of the area to be tested, and select the base station transmission power that meets the preset communication quality standard, the lowest energy consumption and the maximum signal coverage.

[0021] When the base station position of the area to be tested is dynamically adjusted, the updated signal strength characteristic data of the area to be tested is simultaneously input into the regional prediction training model to determine whether the area to be tested is a signal blind spot; when the area to be tested is not a signal blind spot, the real-time signal coverage range of the area to be tested and the signal interference between the corresponding base station and other base stations are obtained according to the corresponding base station position.

[0022] The method for optimizing communication signal coverage based on artificial intelligence also includes: collecting multiple groups of signal range optimization training data; each group of the signal range optimization training data includes the antenna azimuth, antenna downtilt, and / or base station position before the signal blind spot is adjusted, and the antenna azimuth and / or antenna downtilt that meet the preset communication quality standard and the maximum signal coverage range after adjustment, and / or the base station position that meets the minimum signal interference between the corresponding base station and other base stations and the maximum signal coverage range; through training, multiple groups of signal quality optimization training data are used to obtain a base station power optimization model for adjusting the antenna azimuth, the antenna downtilt, and / or the base station position of the signal blind spot.

[0023] A communication signal coverage optimization system based on artificial intelligence, comprising:

[0024] Grid area module: used to divide the region into multiple grid areas;

[0025] Feature collection module: used to select a plurality of the grid areas according to the requirements of the communication network coverage, and collect signal strength feature data of each of the grid areas;

[0026] Area judgment module: used for preprocessing the signal strength characteristic data of each grid area, and judging whether each grid area is a weak signal area and a signal blind area according to the preprocessed signal strength characteristic data of each grid area;

[0027] Regional training module: used to obtain multiple groups of regional prediction training data; each group of regional prediction training data includes the signal strength characteristic data of each grid area, and an identification of whether the grid area is a weak signal area or a signal blind area; multiple groups of regional prediction training data are trained by machine learning to obtain a regional prediction training model; the regional prediction training model is used to determine whether the area to be tested is a weak signal area or a signal blind area;

[0028] Base station adjustment module: used to improve the communication quality of the area to be tested by dynamically adjusting the base station transmission power of the area to be tested when the area to be tested is a weak signal area; when the area to be tested is a signal blind area, dynamically adjust the base station transmission power, antenna azimuth, antenna downtilt angle, and / or base station position of the base station in the area to be tested to expand the signal coverage of the area to be tested.

[0029] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0030] The area is gridded, so that the corresponding communication network data can be collected according to different grid areas, which is convenient for analyzing the network signal strength characteristics of the corresponding area according to the network usage needs within the established communication range, and improving the efficiency of optimizing the overall network quality of the area; according to the signal strength characteristic data, it is identified whether the grid area is a weak signal area and a signal blind area. When the corresponding area is identified, the weak signal area improves the communication quality in the designated area by adjusting the base station transmission power, and the signal blind area expands the signal coverage by adjusting the antenna azimuth and downtilt angle, base station transmission power and base station location, thereby automatically optimizing base station parameters and resource allocation, improving the signal coverage quality and efficiency of the communication network, enhancing the adaptability and robustness of the network signal, and reducing labor costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0032] In the figure:

[0033] Figure 1 This is a flowchart of a communication signal coverage optimization method based on artificial intelligence according to an embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention.

[0035] Example

[0036] like Figure 1 As shown, the embodiment of the present application provides a communication signal coverage optimization method based on artificial intelligence, including:

[0037] Divide the region into grid areas;

[0038] Selecting a plurality of the grid areas according to the requirements of the communication network coverage, and collecting signal strength characteristic data of each of the grid areas;

[0039] Preprocessing the signal strength characteristic data of each of the grid areas, and judging whether each of the grid areas is a weak signal area or a signal blind area according to the preprocessed signal strength characteristic data of each of the grid areas;

[0040] Acquire multiple groups of regional prediction training data; each group of the regional prediction training data includes the signal strength characteristic data of each grid area, and an identification of whether the grid area is a weak signal area or a signal blind area; the multiple groups of regional prediction training data are trained by machine learning to obtain a regional prediction training model; the regional prediction training model is used to determine whether the area to be tested is a weak signal area or a signal blind area;

[0041] When the area to be tested is a weak signal area, the base station transmission power of the area to be tested is dynamically adjusted to improve the communication quality of the area to be tested; when the area to be tested is a signal blind area, the base station transmission power, antenna azimuth, antenna downtilt angle, base station transmission power, and / or base station position of the base station in the area to be tested are dynamically adjusted to expand the signal coverage range of the area to be tested.

[0042] Among them, when dividing the region into multiple grid areas, the existing gridding operation is adopted.

[0043] According to the grid where the communication network coverage is located, select the corresponding grid area to collect signal strength characteristic data. Signal strength characteristics: such as average signal strength, signal strength fluctuation, etc. The above content can be obtained through existing technologies. The above signal strength is detected by sensors, existing data is collected through the network, and it is collected by detection equipment, or the average value of the signal strength collected over a period of time is calculated, and the signal strength fluctuation curve is drawn according to the changes.

[0044] Optionally, the signal strength characteristic data of each grid area can be pre-processed to clean the data or normalize the data. For example, the data is cleaned using existing technologies such as outlier detection, missing value processing, and data deduplication. Determine whether it is a weak signal or a blind area based on the various signal strength characteristics. For example, when the average signal strength is lower than a preset threshold or the signal strength fluctuation exceeds a preset range, the area is judged to be a weak signal area; when the average signal strength is close to 0, or there is no signal strength fluctuation, the area is judged to be a signal blind area.

[0045] Optionally, after the signal strength feature data is cleaned, normalization processing can be performed on the same features, using the Z-score standardization method to make each feature have the same scale, thereby improving the model training efficiency.

[0046] Normalization formula:

[0047]

[0048] Here, z is the normalized value, x is the original value, μ is the mean of all values, and σ is the standard deviation.

[0049] Optionally, the regional prediction training model uses batch gradient descent (BGD) or its variants (such as Adam optimizer) to train the model. By setting the appropriate batch size, learning rate and number of iterations, the best model performance can be obtained. The trained model is used to predict the signal strength feature data set of the test area to identify whether it is a weak signal area or a signal blind area. According to the prediction results, the transmit power P, antenna direction (antenna azimuth angle θ, antenna downtilt angle θ) are dynamically adjusted. ) and location d multiple base station parameters to optimize signal quality and signal coverage.

[0050] Optional, for multiple base station parameters The results of optimizing signal quality and signal coverage can be optimized using the gradient descent algorithm to obtain the minimum value of the following objective function f:

[0051]

[0052] in, Indicates the communication quality. Indicates energy consumption, represents the interference between base stations, H(P,θ,φ,d) represents the signal coverage, and α, β, γ, and δ are weight coefficients. In the objective function f of the transmit power P, β and γ are both 0; in the objective function f of the antenna direction (antenna azimuth angle θ and / or antenna downtilt angle φ), α, β, and γ are all 0; in the objective function f of the position d, α, β, γ, and δ are all 0.

[0053] Optionally, dynamically adjust various base station parameters When optimizing the objective function f, an iterative optimization algorithm such as gradient descent or conjugate gradient method is used. For each parameter update step, the gradient of the objective function is calculated and the parameter value is adjusted according to the learning rate. For example, adjusting the antenna azimuth angle can be expressed as:

[0054]

[0055] Among them, θ old Indicates the antenna azimuth before adjustment; θ new represents the adjusted antenna azimuth; η is the learning rate; represents the partial derivative value of f(θ) with respect to θ; f(θ) represents the objective function f obtained by adjusting θ.

[0056] The collected signal strength feature data is preprocessed to clean the data, including removing invalid values, outliers, duplicate records and other noise data. The cleaning process can be carried out in the following steps: Outlier detection: using statistical methods and machine learning algorithms, such as box plots and cluster analysis, to identify outliers in the data; missing value processing: taking methods such as deletion, filling or interpolation for missing data to ensure the integrity of the data set; data deduplication: removing duplicate records through data unique identifiers, such as timestamps and location information.

[0057] Optionally, the neural network structure of the regional prediction training model adopts a deep learning network with a multi-layer perceptron (MLP) structure to adapt to the high dimensionality and nonlinear characteristics of the signal coverage data.

[0058] Among them, a deep learning network with 3 hidden layers can be used, each hidden layer contains 100 neurons, and the ReLU activation function is used to improve the nonlinear modeling ability of the network and reduce the gradient disappearance problem.

[0059] By adjusting the learning rate to 0.001, the batch size to 64, and iterative training for 100 epochs, a model that can more accurately predict signal coverage and quality was obtained, and the signal resource allocation in each area was adjusted in advance to ensure the user's communication experience. At the same time, by adjusting the relevant base station parameters, the signal quality and coverage area were significantly improved.

[0060] Loss function of regional prediction training model: Mean square error (MSE) is used as the loss function To minimize the difference between the predicted value and the actual value.

[0061] The corresponding mathematical formula is as follows:

[0062]

[0063] Among them, y represents the actual value, n represents the number of samples, Represents the predicted value, y i represents the ith actual value, represents the i-th predicted value.

[0064] The dynamically adjusting the base station transmission power of the area to be tested to improve the communication quality of the area to be tested includes:

[0065] When dynamically adjusting the base station transmission power of the area to be tested, obtain the real-time communication quality, energy consumption and signal coverage of the area to be tested, and select the base station transmission power that meets the preset communication quality standard, the lowest energy consumption and the maximum signal coverage.

[0066] Among them, by adjusting the base station transmission power of the area to be tested multiple times, select the parameters that meet the quality standards, and combine the energy consumption and signal coverage to obtain the final base station transmission power. The communication quality, energy consumption and signal coverage are the data recorded after the current adjustment of the base station transmission power. The evaluation of communication quality can be carried out in any existing way, such as using the value of any one or more network performance indicators in delay, bandwidth, packet loss rate, stability, and security to determine whether the threshold corresponding to the communication quality standard is reached, so as to evaluate the communication quality according to the corresponding quality requirements. The signal coverage can be obtained by real-time detection and uploading, so as to record the base station transmission power when the quality standard is met, and at the same time combine the energy consumption and the size of the signal coverage to obtain the final base station transmission power. Among them, the energy consumption of the area to be tested can be obtained in any existing way, such as using existing data, or calculated using the base station transmission power. Generally speaking, the greater the power, the lower the energy consumption, so the minimum power value is selected as the value with the lowest energy consumption when the communication quality standard is met.

[0067] When the base station transmission power of the area to be tested is dynamically adjusted, the updated signal strength characteristic data of the area to be tested is simultaneously input into the regional prediction training model to determine whether the area to be tested is a weak signal area; when the area to be tested is not a weak signal area, the real-time communication quality and energy consumption of the area to be tested are obtained according to the corresponding base station transmission power.

[0068] The method for optimizing communication signal coverage based on artificial intelligence also includes: collecting multiple groups of signal quality optimization training data; each group of the signal quality optimization training data includes the base station transmission power before adjustment of the weak signal area, and the base station transmission power after adjustment that meets the preset communication quality standard and has the lowest energy consumption; through training, multiple groups of signal quality optimization training data are used to obtain a base station power optimization model for adjusting the base station transmission power in the weak signal area.

[0069] The method of dynamically adjusting the antenna azimuth angle, antenna downtilt angle, base station transmission power, and / or base station position of the base station in the area to be tested to expand the signal coverage of the area to be tested includes:

[0070] When dynamically adjusting the antenna azimuth and / or antenna downtilt angle of the base station corresponding to the area to be tested, obtain the real-time signal coverage and communication quality of the area to be tested; select the antenna azimuth and / or antenna downtilt angle that meets the preset communication quality standard and has the maximum signal coverage.

[0071] When the antenna azimuth angle and / or antenna downtilt angle are dynamically adjusted, the updated signal strength characteristic data of the area to be tested is simultaneously input into the regional prediction training model to determine whether the area to be tested is a signal blind spot; when the area to be tested is not a signal blind spot, the real-time signal coverage and communication quality of the area to be tested are obtained according to the corresponding base station position.

[0072] After adjusting the base station parameters of the area to be tested several times, the signal strength characteristic data of the grid area, such as signal strength and bit error rate, are collected again, and the regional prediction training model is used to make another judgment until the area is no longer a weak signal area or a signal blind area.

[0073] The method of dynamically adjusting the antenna azimuth angle, antenna downtilt angle, base station transmission power, and / or base station position of the base station in the area to be tested to expand the signal coverage of the area to be tested includes:

[0074] Dynamically adjust the base station position of the area to be tested, obtain the real-time signal coverage of the area to be tested, and the signal interference between the corresponding base station and other base stations; select the base station position that satisfies the minimum signal interference between the corresponding base station and other base stations and the maximum signal coverage;

[0075] When dynamically adjusting the base station transmission power of the area to be tested, obtain the real-time communication quality, energy consumption and signal coverage of the area to be tested, and select the base station transmission power that meets the preset communication quality standard, the lowest energy consumption and the maximum signal coverage.

[0076] When the base station position of the area to be tested is dynamically adjusted, the updated signal strength characteristic data of the area to be tested is simultaneously input into the regional prediction training model to determine whether the area to be tested is a signal blind spot; when the area to be tested is not a signal blind spot, the real-time signal coverage range of the area to be tested and the signal interference between the corresponding base station and other base stations are obtained according to the corresponding base station position.

[0077] The method for optimizing communication signal coverage based on artificial intelligence also includes: collecting multiple groups of signal range optimization training data; each group of the signal range optimization training data includes the antenna azimuth, antenna downtilt, and / or base station position before the signal blind spot is adjusted, and the antenna azimuth and / or antenna downtilt that meet the preset communication quality standard and the maximum signal coverage range after adjustment, and / or the base station position that meets the minimum signal interference between the corresponding base station and other base stations and the maximum signal coverage range; through training, multiple groups of signal quality optimization training data are used to obtain a base station power optimization model for adjusting the antenna azimuth, the antenna downtilt, and / or the base station position of the signal blind spot.

[0078] In summary, the embodiments of the present application provide a communication signal coverage optimization method and system based on artificial intelligence:

[0079] The area is gridded, so that the corresponding communication network data can be collected according to different grid areas, which is convenient for analyzing the network signal strength characteristics of the corresponding area according to the network usage needs within the established communication range, and improving the efficiency of optimizing the overall network quality of the area; according to the signal strength characteristic data, it is identified whether the grid area is a weak signal area and a signal blind area. When the corresponding area is identified, the weak signal area improves the communication quality in the designated area by adjusting the base station transmission power, and the signal blind area expands the signal coverage by adjusting the antenna azimuth and downtilt angle, base station transmission power and base station location, thereby automatically optimizing base station parameters and resource allocation, improving the signal coverage quality and efficiency of the communication network, enhancing the adaptability and robustness of the network signal, and reducing labor costs.

[0080] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A communication signal coverage optimization method based on artificial intelligence, characterized in that: include: Divide the region into grid areas; Selecting a plurality of the grid areas according to the requirements of the communication network coverage, and collecting signal strength characteristic data of each of the grid areas; Preprocessing the signal strength characteristic data of each of the grid areas, and judging whether each of the grid areas is a weak signal area or a signal blind area according to the preprocessed signal strength characteristic data of each of the grid areas; Acquire multiple groups of regional prediction training data; each group of the regional prediction training data includes the signal strength characteristic data of each grid area, and an identification of whether the grid area is a weak signal area or a signal blind area; the multiple groups of regional prediction training data are trained by machine learning to obtain a regional prediction training model; the regional prediction training model is used to determine whether the area to be tested is a weak signal area or a signal blind area; When the area to be tested is a weak signal area, the base station transmission power of the area to be tested is dynamically adjusted to improve the communication quality of the area to be tested; when the area to be tested is a signal blind area, the base station transmission power, antenna azimuth, antenna downtilt angle, base station transmission power, and / or base station position of the base station in the area to be tested are dynamically adjusted to expand the signal coverage range of the area to be tested.

2. The communication signal coverage optimization method based on artificial intelligence according to claim 1 is characterized in that: The dynamically adjusting the base station transmission power of the area to be tested to improve the communication quality of the area to be tested includes: When dynamically adjusting the base station transmission power of the area to be tested, obtain the real-time communication quality, energy consumption and signal coverage of the area to be tested, and select the base station transmission power that meets the preset communication quality standard, the lowest energy consumption and the maximum signal coverage.

3. The communication signal coverage optimization method based on artificial intelligence according to claim 2 is characterized in that: When the base station transmission power of the area to be tested is dynamically adjusted, the updated signal strength characteristic data of the area to be tested is simultaneously input into the regional prediction training model to determine whether the area to be tested is a weak signal area; when the area to be tested is not a weak signal area, the real-time communication quality and energy consumption of the area to be tested are obtained according to the corresponding base station transmission power.

4. The communication signal coverage optimization method based on artificial intelligence according to claim 2 is characterized in that: Also includes: Collect multiple groups of signal quality optimization training data; each group of the signal quality optimization training data includes the base station transmission power before adjustment in the weak signal area, and the base station transmission power after adjustment that meets the preset communication quality standard and has the lowest energy consumption; through training, multiple groups of signal quality optimization training data are used to obtain a base station power optimization model for adjusting the base station transmission power in the weak signal area.

5. The communication signal coverage optimization method based on artificial intelligence according to claim 1 is characterized in that: The method of dynamically adjusting the antenna azimuth angle, antenna downtilt angle, base station transmission power, and / or base station position of the base station in the area to be tested to expand the signal coverage of the area to be tested includes: When dynamically adjusting the antenna azimuth and / or antenna downtilt angle of the base station corresponding to the area to be tested, obtain the real-time signal coverage and communication quality of the area to be tested; select the antenna azimuth and / or antenna downtilt angle that meets the preset communication quality standard and has the maximum signal coverage.

6. The communication signal coverage optimization method based on artificial intelligence according to claim 5 is characterized in that: When the antenna azimuth angle and / or antenna downtilt angle are dynamically adjusted, the updated signal strength characteristic data of the area to be tested is simultaneously input into the regional prediction training model to determine whether the area to be tested is a signal blind spot; when the area to be tested is not a signal blind spot, the real-time signal coverage and communication quality of the area to be tested are obtained according to the corresponding base station position.

7. The method for optimizing communication signal coverage based on artificial intelligence according to claim 5, characterized in that: The method of dynamically adjusting the antenna azimuth angle, antenna downtilt angle, base station transmission power, and / or base station position of the base station in the area to be tested to expand the signal coverage of the area to be tested includes: Dynamically adjust the base station position of the area to be tested, obtain the real-time signal coverage of the area to be tested, and the signal interference between the corresponding base station and other base stations; select the base station position that satisfies the minimum signal interference between the corresponding base station and other base stations and the maximum signal coverage; When dynamically adjusting the base station transmission power of the area to be tested, obtain the real-time communication quality, energy consumption and signal coverage of the area to be tested, and select the base station transmission power that meets the preset communication quality standard, the lowest energy consumption and the maximum signal coverage.

8. The method for optimizing communication signal coverage based on artificial intelligence according to claim 7, characterized in that: When the base station position of the area to be tested is dynamically adjusted, the updated signal strength characteristic data of the area to be tested is simultaneously input into the regional prediction training model to determine whether the area to be tested is a signal blind spot; when the area to be tested is not a signal blind spot, the real-time signal coverage range of the area to be tested and the signal interference between the corresponding base station and other base stations are obtained according to the corresponding base station position.

9. The communication signal coverage optimization method based on artificial intelligence according to claim 7 is characterized in that: Also includes: Collect multiple groups of signal range optimization training data; each group of the signal range optimization training data includes the antenna azimuth, antenna downtilt, and / or base station position before the signal blind spot is adjusted, the antenna azimuth and / or antenna downtilt that meet the preset communication quality standard and the maximum signal coverage range after adjustment, and / or the base station position that meets the minimum signal interference between the corresponding base station and other base stations and the maximum signal coverage range; through training, multiple groups of signal quality optimization training data obtain a base station power optimization model for adjusting the antenna azimuth, the antenna downtilt, and / or the base station position of the signal blind spot.

10. A communication signal coverage optimization system based on artificial intelligence, characterized in that: include: Grid area module: used to divide the region into multiple grid areas; Communication collection module: used to select multiple grid areas to collect communication data according to the requirements of communication network coverage; Feature extraction module: used for cleaning and feature extraction of the communication data of each grid area to obtain feature data; Area judgment module: used for preprocessing the signal strength characteristic data of each grid area, and judging whether each grid area is a weak signal area and a signal blind area according to the preprocessed signal strength characteristic data of each grid area; Regional training module: used to obtain multiple groups of regional prediction training data; each group of regional prediction training data includes the signal strength characteristic data of each grid area, and an identification of whether the grid area is a weak signal area or a signal blind area; multiple groups of regional prediction training data are trained by machine learning to obtain a regional prediction training model; the regional prediction training model is used to determine whether the area to be tested is a weak signal area or a signal blind area; Base station adjustment module: used to improve the communication quality of the area to be tested by dynamically adjusting the base station transmission power of the area to be tested when the area to be tested is a weak signal area; when the area to be tested is a signal blind area, dynamically adjust the base station transmission power, antenna azimuth, antenna downtilt angle, and / or base station position of the base station in the area to be tested to expand the signal coverage of the area to be tested.

Images