A mobile communication device deployment method and apparatus

Abstract

The application provides a mobile communication device deployment method and device, taking a deployment control device as an execution subject, the mobile communication device being a base station, and the method comprises the following steps: S1: the deployment control device acquires basic data of a vertical industry target scene, wherein the basic data comprises scene layout data, terminal device data and signal transmission environment data; S2: the deployment control device pre-processes the basic data, removes abnormal data through a 3σ criterion, and adopts a min-max standardization method to standardize the basic data after removing the abnormal data, so as to obtain standardized data. The application realizes customized deployment of the mobile communication device in the vertical industry scene, breaks through the technical blank of the prior art that focuses on the public network scene and is not adapted to the differentiated needs of the vertical industry, solves the low-latency and high-reliability needs of the industrial scene and the signal stability needs of the vehicle-road cooperation scene in a targeted manner, and fills the technical barriers of the customized deployment of the vertical industry.

Description

Technical Field

[0001] This invention relates to the field of mobile communication technology, and in particular to a method and apparatus for deploying mobile communication devices. Background Technology

[0002] With the widespread adoption of 5G and 5G-A technologies, mobile communication devices have been extensively used in various vertical industries, including the Industrial Internet, vehicle-to-everything (V2X) communication, and smart healthcare. These vertical industry scenarios have different and stringent requirements for mobile communication latency, reliability, and signal stability, which differ significantly from the general deployment requirements of public network scenarios.

[0003] However, existing technologies mostly focus on base station deployment in public network scenarios, without designing customized deployment solutions for the differentiated needs of vertical industries. For example, in the industrial internet scenario, industrial workshops are densely populated with equipment and large equipment causes severe obstruction. Existing deployment methods do not take into account the distribution of workshop equipment and signal obstruction, and cannot meet the low latency (<5ms) and high reliability (99.999%) requirements of industrial equipment communication, resulting in delays and packet loss in industrial control signal transmission, which affects production efficiency. In the vehicle-road cooperative scenario, vehicles move at high speeds and road layouts are complex. Existing deployment methods do not dynamically adjust the base station coverage based on vehicle movement trajectories and road layouts, resulting in poor signal stability of high-speed mobile terminals, which are prone to handover interruptions and signal attenuation, affecting the real-time transmission of vehicle-road cooperative data. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the defects of the existing technology. The present invention proposes a method and apparatus for deploying mobile communication devices.

[0005] To address the issue that existing technologies primarily focus on base station deployment in public network scenarios and fail to design customized deployment solutions for the differentiated needs of vertical industries—for example, in industrial internet scenarios, where industrial workshops are densely populated with large equipment and suffer from severe signal obstruction—existing deployment methods do not incorporate customized planning based on equipment distribution and signal obstruction, failing to meet the low-latency communication requirements of industrial equipment. This results in delays and packet loss in industrial control signal transmission, impacting production efficiency. In vehicle-road cooperative scenarios, where vehicles move at high speeds and road layouts are complex, existing deployment methods do not dynamically adjust base station coverage based on vehicle movement trajectories and road layouts, leading to poor signal stability for high-speed mobile terminals and issues such as handover interruptions and signal attenuation, thus affecting the real-time transmission of vehicle-road cooperative data—the technical solution adopted in this invention is: A method and apparatus for deploying a mobile communication device, with a deployment control device as the executing entity, wherein the mobile communication device is a base station, and the method includes the following steps: S1: Deploy control equipment to acquire basic data of target scenarios in vertical industries. The basic data includes scenario layout data, terminal device data, and signal transmission environment data. S2: Deploy control equipment to preprocess the basic data, remove abnormal data using the 3σ criterion, and standardize the basic data after removing abnormal data using the min-max standardization method to obtain standardized data; S3: Deploy control equipment to extract features from the standardized data, extract occlusion feature parameters through convolutional neural network, extract terminal demand parameters through linear regression, and extract movement trajectory feature parameters through adaptive Kalman filter algorithm to obtain scene feature parameters; S4: The deployment control device determines whether the scene feature parameters meet the preset deployment threshold conditions; the preset deployment threshold conditions are: in industrial scenarios, the latency parameter in the terminal requirement parameters is <5ms and the reliability parameter is ≥99.999%; in vehicle-road cooperative scenarios, the signal stability parameter corresponding to the movement trajectory feature parameters is ≥99.5%; S5: If the scene feature parameters meet the preset deployment threshold conditions, the deployment control device calculates the deployment parameters of the mobile communication device based on the scene feature parameters using a reinforcement learning algorithm. The deployment parameters include base station location coordinates, base station beam angle, and base station transmit power. If not, the deployment control device returns to step S1 to reacquire basic data and optimizes the abnormal data removal threshold in the preprocessing process. S6: The deployment control device outputs the deployment parameters and controls the mobile communication device to complete the deployment according to the deployment parameters; at the same time, the deployment control device collects the communication status data after deployment in real time, uses the communication status data as new basic data, and executes steps S3-S5 in a loop to realize dynamic optimization of the deployment parameters.

[0006] Preferably, the scene layout data in step S1 includes the coordinates of industrial workshop equipment and the coordinates of vehicle-road cooperative road nodes; the terminal equipment data includes the location of industrial control terminals and the speed and location of vehicle-road cooperative mobile terminals; the signal transmission environment data includes the size of obstructions, the location of obstructions, and signal propagation loss data.

[0007] Preferably, the calculation formula for the min-max normalization method in step S2 is as follows: in, For standardized basic data, As the original basic data, The minimum value in the original basic data. This is the maximum value in the original base data.

[0008] 4. The mobile communication device deployment method according to claim 1, wherein the calculation formula for the occlusion feature parameter in step S3 is: in, The occlusion coefficient is... For the height of the obstruction, Preset the height for mobile communication devices. The horizontal distance between the obstruction and the base station's preset location. Preset the angle for the base station beam; when When, it is determined that there is no obvious obstruction; when When, it is judged as moderate occlusion; when At that time, it was determined to be severe occlusion.

[0009] Preferably, the movement trajectory feature parameters in step S3 are extracted using an adaptive Kalman filter algorithm, which includes a state prediction equation and a state update equation; a state vector is defined. ,in , for Moment mobile terminal coordinate, , for Mobile terminals at all times velocity in the direction; State prediction equation: State update equation: in, for Predicted state at any given time Here is the state transition matrix. for The optimal estimated state at time t. For the control matrix, for The amount of control at any given moment for The covariance matrix of the predicted state at each time step. The process noise covariance matrix is ​​obtained by... Factors are dynamically adjusted. H is the Kalman gain, and H is the observation matrix. To observe the noise covariance matrix, the β factor is dynamically adjusted. Let k be the observation value at time k. It is an identity matrix; the α factor is calculated using the motion state consistency index. The factor is calculated using communication quality indicators.

[0010] Preferably, the comprehensive reward function of the reinforcement learning algorithm in step S5 is: in, For the overall reward value, , , These are the weighting coefficients for coverage performance, communication quality, and deployment cost, respectively. ; To cover performance bonus values, This is a communication quality bonus value. The deployment cost reward value is used to determine the optimal deployment parameters for the deployment control equipment by maximizing the overall reward value.

[0011] Preferably, the communication status data in step S6 includes actual latency data, actual reliability data, and actual signal stability data; the deployment control device compares the actual latency data with a preset latency threshold, the actual reliability data with a preset reliability threshold, and the actual signal stability data with a preset signal stability threshold, calculates the deviation value, and adjusts the deployment parameters based on the deviation value.

[0012] A mobile communication device deployment apparatus, integrated into a deployment control device, includes a data acquisition module, a data preprocessing module, a feature extraction module, a threshold judgment module, a parameter calculation module, a deployment control module, and a dynamic optimization module; The data acquisition module is used to acquire basic data of the target scenario in the vertical industry, including scenario layout data, terminal device data and signal transmission environment data; it is also used to collect communication status data after deployment in real time. The data preprocessing module is used to... The criteria remove outliers from the base data, using... The standardization method standardizes the base data after removing outliers to obtain standardized data; The feature extraction module is used to extract occlusion feature parameters through a convolutional neural network, extract terminal demand parameters through linear regression, and extract movement trajectory feature parameters through an adaptive Kalman filter algorithm to obtain scene feature parameters. The threshold judgment module is used to determine whether the scene feature parameters meet the preset deployment threshold conditions. The preset deployment threshold conditions are: in industrial scenarios, latency parameter < 5ms and reliability parameter ≥ 99.999%; in vehicle-road cooperative scenarios, signal stability parameter ≥ 99.5%. The parameter calculation module is used to calculate the deployment parameters of the mobile communication device based on the scene feature parameters and through a reinforcement learning algorithm when the scene feature parameters meet the preset deployment threshold conditions. The deployment parameters include base station location coordinates, base station beam angle, and base station transmit power. When the conditions are not met, the data acquisition module is triggered to reacquire the basic data and optimize the abnormal data removal threshold. The deployment control module is used to output the deployment parameters and control the mobile communication device to complete the deployment according to the deployment parameters; The dynamic optimization module is used to take the communication status data collected by the data acquisition module as new basic data, trigger the feature extraction module, threshold judgment module and parameter calculation module to work in a loop, so as to realize the dynamic optimization of deployment parameters.

[0013] Preferably, the data preprocessing module uses The standardization method, the occlusion feature parameter calculation formula used in the feature extraction module, the adaptive Kalman filter algorithm, and the reinforcement learning algorithm used in the parameter calculation module are all consistent with the formulas and algorithms corresponding to claims 3-6.

[0014] Compared with the prior art, the beneficial effects of the present invention are: This invention enables customized deployment of mobile communication devices in vertical industry scenarios, breaking through the technological gap of existing technologies that focus on public network scenarios and fail to adapt to the differentiated needs of vertical industries. It specifically addresses the low latency and high reliability requirements of industrial scenarios and the signal stability requirements of vehicle-road cooperative scenarios, filling the technical barriers to customized deployment in vertical industries.

[0015] This invention employs a comprehensive workflow design encompassing data preprocessing, feature extraction, threshold determination, and dynamic optimization, combined with... Guidelines Algorithms such as standardization, adaptive Kalman filtering, and reinforcement learning enable precise calculation and dynamic adjustment of deployment parameters. In industrial scenarios, it can stably meet the requirements of latency <5ms and reliability ≥99.999%. In vehicle-road cooperative scenarios, it can make the signal stability of high-speed mobile terminals ≥99.5%, significantly improving the quality of mobile communication in vertical industries and solving the core pain points of existing deployment methods.

[0016] The method and apparatus of this invention can be directly applied to vertical industries such as industrial internet and vehicle-road collaboration. The deployment process is highly automated, requiring minimal manual intervention, thus reducing deployment and maintenance costs. Furthermore, the deployment parameters can be dynamically optimized according to changes in the scenario, adapting to the differentiated scenario needs of different vertical industries. It is highly practical and scalable. Attached Figure Description

[0017] The disclosure of this invention is illustrated with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. In the drawings, the same reference numerals are used to refer to the same parts. Wherein: Figure 1 This is a flowchart of the mobile communication device deployment method of the present invention; Figure 2 A block diagram illustrating the deployment apparatus for the mobile communication device of the present invention. Detailed Implementation

[0018] The present invention will be further described in detail below with reference to specific embodiments. It should be noted that these embodiments are only used to explain the present invention and are not intended to limit the scope of protection of the present invention.

[0019] Please see Figures 1-2 The mobile communication device deployment method of the present invention takes a deployment control device as the execution subject, and the specific implementation process is as follows: The deployment control device first acquires the basic data of the target scenario in the vertical industry, preprocesses the data to ensure data accuracy, extracts the scenario feature parameters and determines whether they meet the preset threshold. If they do, the deployment parameters are calculated and the device is deployed. Finally, by collecting communication status data in real time, the deployment parameters are iteratively optimized to ensure that the deployment effect meets the needs of the vertical industry.

[0020] Detailed implementation of each step Step 1: Deploy the control equipment. Through its built-in data acquisition interface, connect to various data acquisition devices in vertical industry scenarios (such as IoT sensors in industrial workshops, roadside units for vehicle-to-everything (V2X) communication, GPS positioning devices, etc.) to obtain basic data for the target scenario, specifically including: In industrial scenarios, data such as the length, width, and height of the industrial workshop, the three-dimensional coordinates (x, y, z) of various production equipment within the workshop, and the spacing between equipment are collected. In vehicle-road cooperative scenarios, data such as the length, width, coordinates of road nodes, intersection locations, and curve radii of roads are collected.

[0021] In industrial scenarios, the system collects the installation coordinates of industrial control terminals and sensor terminals, as well as the terminals' communication requirements (latency requirements, data transmission rate requirements). In vehicle-road cooperative scenarios, the system collects the real-time location coordinates (x, y), driving speed v, and driving direction θ of mobile terminals (vehicles), as well as the terminals' communication bandwidth requirements.

[0022] Collect the three-dimensional dimensions (length, width, height) and position coordinates of obstructions (industrial equipment, buildings, trees, etc.) in the scene, path loss data during signal propagation, and environmental parameters that affect signal transmission, such as ambient temperature and humidity.

[0023] In this embodiment, the data acquisition frequency is set to 10Hz to ensure the real-time nature of the data. The acquired data is stored in a structured format in the local database of the deployed control device for easy subsequent processing.

[0024] Step Two: Deploy control equipment to preprocess the collected basic data, which consists of two steps: outlier removal and data standardization. The 3σ criterion is used to remove outliers. The specific logic is as follows: calculate the mean of each type of basic data. and standard deviation If a data point x satisfies If a data point exceeds the mean ± 3 times the standard deviation, it is considered an abnormal data point and is removed. For example, in industrial scenarios, if a data point in the latency data exceeds the mean ± 3 times the standard deviation, it is considered an abnormal data point and is removed to avoid affecting subsequent parameter calculations.

[0025] The min-max standardization method is used to map the base data after removing outliers to the [0,1] interval, eliminating the influence of different dimensions of data. The standardization calculation formula is as follows: in, Here, x represents the standardized base data, and x represents the original base data. The minimum value in the original basic data. This represents the maximum value in the original base data. For example, if the x-coordinate of equipment in an industrial workshop ranges from [0, 100m], and the x-coordinate of a certain piece of equipment is 50m, then the standardized data is (50-0) / (100-0) = 0.5.

[0026] After preprocessing, standardized data is obtained and stored in the cache of the deployment control device for subsequent feature extraction.

[0027] Step 3: Deploy control equipment to extract features from standardized data, specifically extracting occlusion feature parameters, terminal requirement parameters, and movement trajectory feature parameters. The specific implementation is as follows: A convolutional neural network (CNN) is used to extract features from the standardized occlusion data (size, position), and the occlusion coefficient k is output as the occlusion feature parameter. The calculation formula is as follows: Where k is the obstruction coefficient, h is the height of the obstruction (the actual height is obtained by inverse normalization of the standardized data), H is the preset height of the mobile communication equipment (base station) (set to 10m in this embodiment), d is the horizontal distance between the obstruction and the preset position of the base station (the actual distance is obtained by inverse normalization of the standardized data), and θ is the preset angle of the base station beam (set to 30° in this embodiment). When k < 0.3, it is determined to be no obvious obstruction; when 0.3 ≤ k < 0.7, it is determined to be moderate obstruction; when k ≥ 0.7, it is determined to be severe obstruction. For example, if the height of an obstruction is h = 5m, the horizontal distance between it and the preset position of the base station is d = 10m, and θ = 30°, then k = 5 / (10 - 10 × tan30°) ≈ 5 / (10 - 5.77) = 1.18 ≥ 0.7, it is determined to be severe obstruction, and the position of the base station needs to be adjusted to avoid the obstruction during subsequent deployment.

[0028] Simultaneously, combining the wireless propagation model (optimized based on the COST231-Walfisch-Ikegami model, adapted to vertical industry scenarios), the path loss in non-line-of-sight transmission is calculated, and occlusion feature parameters are supplemented. The formula for calculating non-line-of-sight path loss is as follows: in, For free space loss, This refers to the fading loss caused by diffraction at the lower edge of the roof. This refers to the fading loss caused by multiple diffraction along the roof. The calculation formula is: Where ω is the width of the street where the receiver is located (vehicle-road cooperative scenario) or the width of the workshop passage (industrial scenario). The average height of the building. The height of the receiving antenna is used; this optimization formula is adapted to the signal propagation characteristics of vertical industry scenarios, improving the accuracy of path loss calculation.

[0029] A linear regression algorithm is used to fit the standardized terminal device data to extract terminal requirement parameters, including latency requirement parameter t and reliability requirement parameter R for industrial scenarios, and signal stability requirement parameter S for vehicle-to-everything (V2X) scenarios. In this embodiment, the latency requirement t < 5ms and the reliability requirement R ≥ 99.999% for industrial scenarios are specified; the signal stability requirement S ≥ 99.5% for V2X scenarios is specified. The logic for extracting terminal requirement parameters is as follows: by fitting the relationship between terminal location and communication requirements through linear regression, the requirement priority of terminals at different locations is obtained, providing a basis for subsequent deployment parameter calculations.

[0030] The reliability parameter R is calculated using the polar code-optimized bit error rate model, and the formula is as follows: in, To reduce the bit error rate, this embodiment optimizes the process using polar coding techniques. Ensure that R ≥ 99.999%.

[0031] An adaptive Kalman filter algorithm is used to process the location data of mobile terminals (vehicles) in a vehicle-road cooperative scenario, extracting movement trajectory feature parameters (predicted position, movement speed, and trajectory offset). The specific implementation of the algorithm is as follows: First, define the state vector. ,in , Let x and y coordinates be the vehicle at time k. , Let be the vehicle's velocity in the x and y directions at time k; State transition matrix , where T is the sampling period (T=0.1s in this embodiment); Control matrix B=0 (no external control input), observation matrix ; Process noise covariance matrix Through dynamic adjustment of the α factor, the calculation logic of the α factor is as follows: First, calculate the motion state consistency index. ,in Let be the vehicle acceleration at time t. Let t be the angular velocity of the vehicle. , These are the weights for acceleration and angular velocity, respectively, with values ​​ranging from [0.1, 0.9]. Then calculate the α factor. ,in ; Last Update , The initial process noise covariance matrix (in this embodiment) ); Observation noise covariance matrix Through dynamic adjustment of the β factor, the β factor is based on communication quality indicators. calculate, ,in To receive signal strength, For packet error rate, For communication delay, Maximum allowable delay (5ms); β factor ,renew , The initial observation noise covariance matrix (in this embodiment) ); The optimal estimated state (i.e., trajectory characteristic parameters) at time k is calculated according to the following state prediction equation and state update equation: State prediction equation: State update equation: in, Let I be the observation value (actual position of the vehicle) at time k, and let I be a 4th-order identity matrix. Through the above algorithm, the position of the vehicle at the next time moment can be accurately predicted, providing a basis for adjusting the coverage of the base station.

[0032] Step 4: The deployment control device compares the extracted scene feature parameters with the preset deployment threshold conditions to determine whether the deployment requirements are met. The preset deployment threshold conditions are as follows: The terminal requirements parameters include latency (t < 5ms) and reliability (R ≥ 99.999%); among the occlusion characteristics, the proportion of heavily occluded areas is ≤ 10%. The signal stability parameter S corresponding to the movement trajectory characteristic parameters is ≥99.5%; among the occlusion characteristic parameters, there is no severe occlusion in the core area of ​​the road (lane, intersection).

[0033] The deployment control device compares the scene feature parameters with the above thresholds one by one. If all parameters meet the threshold requirements, it is determined that the preset deployment threshold conditions are met, and step five is executed. If any parameter does not meet the threshold requirements, it is determined that it is not met, and the process returns to step one to reacquire basic data and adjusts the abnormal data removal threshold of the 3σ criterion (adjusting 3σ to 2.5σ to improve the accuracy of data collection), and preprocessing is performed again.

[0034] Step 5: When the scene feature parameters meet the preset deployment threshold conditions, the deployment control device calculates the deployment parameters of the mobile communication device based on the scene feature parameters using a reinforcement learning algorithm. The deployment parameters include the base station location coordinates (x, y, z), the base station beam angle θ, and the base station transmit power P. The specific implementation is as follows: The base station deployment process is modeled as a reinforcement learning decision-making process. The state space consists of scene feature parameters (occlusion coefficient k, latency t, reliability R, predicted location of mobile trajectory, etc.), the action space consists of the adjustment amount of deployment parameters (base station position offset, beam angle adjustment, transmit power adjustment), and the reward function is the comprehensive reward value R, which is used to evaluate the deployment effect.

[0035] Calculation of the comprehensive reward function: in, , , (Set according to the demand weight of the vertical industry); , , The values ​​of all values ​​are in the range of [0,1], ensuring the rationality of the comprehensive reward value R.

[0036] To cover performance bonus values, The area of ​​the uncovered region is calculated using occlusion characteristic parameters and the base station coverage range. The formula for calculating the base station coverage radius is: in, The basic coverage radius is set to 50m in this embodiment. It is the beam factor (positively correlated with the beam angle θ). This is the downslope penalty factor (value range [0.8, 1.0]). The communication quality bonus value, in industrial scenarios Vehicle-road cooperative scenarios ; As a reward value for deployment costs, The actual deployment cost is related to the number of base stations and their locations.

[0037] Reinforcement learning agents iterate continuously, selecting actions (deployment parameter adjustments) that maximize the overall reward value R, ultimately obtaining the optimal deployment parameters. For example, in industrial scenarios, for heavily obstructed areas, the agent adjusts the base station's location coordinates to increase the horizontal distance d between the base station and the obstruction, while simultaneously adjusting the beam angle θ to reduce the obstruction coefficient k, ensuring latency t < 5ms and reliability R ≥ 99.999%. In vehicle-road cooperative scenarios, the agent predicts the location based on vehicle trajectories and adjusts the base station's coverage radius and beam angle to ensure signal stability S ≥ 99.5%.

[0038] In this embodiment, the base station transmit power P is calculated using the reference link budget model, and the calculation formula is as follows: Where L is the path loss (including free space loss and occlusion loss). To meet the power requirements for receiving, For the transmit antenna gain, For receiving antenna gain, Losses due to feeder and human body obstruction (in this embodiment) ).

[0039] Step Six: Deployment control equipment outputs the calculated deployment parameters to the mobile communication equipment (base station) through the communication interface, and controls the base station to complete the installation and debugging according to the deployment parameters, ensuring that the base station location, beam angle and transmission power meet the requirements.

[0040] The deployment control equipment collects real-time communication status data after deployment, including actual latency data. Actual reliability data Actual signal stability data The sampling frequency is 5Hz; these communication status data are used as new basic data, and steps three to five are executed cyclically to dynamically adjust the deployment parameters.

[0041] Calculate the deviation between the actual parameters and a preset threshold. If the deviation is ≤5%, maintain the current deployment parameters; if the deviation is >5%, adjust the deployment parameters using a reinforcement learning algorithm. For example, in an industrial scenario, if the actual latency... If the deviation from the preset threshold of 5ms is 4%, the current deployment parameters will be maintained; if If the deviation is 10%, the base station transmit power P is adjusted to reduce path loss and latency.

[0042] Industrial Internet Scenario Examples This embodiment is applied to a smart manufacturing workshop with dimensions of 100m×80m×10m. There are 50 industrial control devices in the workshop, distributed in different areas. Some large devices (8m high) have signal obstruction. The industrial control terminals are required to have a communication latency of <5ms and a reliability of ≥99.999%.

[0043] The specific implementation process of the mobile communication device deployment method and apparatus of the present invention is as follows: The deployment of control equipment uses IoT sensors in the workshop to acquire basic data: workshop layout data (100m×80m×10m, 3D coordinates of 50 industrial control devices), terminal device data (location of industrial control terminals, latency requirement <5ms, reliability requirement ≥99.999%), and signal transmission environment data (size and location of large equipment, signal path loss data). The deployment control equipment uses the 3σ criterion to remove abnormal data (such as abnormal path loss data collected by individual sensors), and the data is standardized by the min-max standardization method to obtain standardized data; Extract scene feature parameters: occlusion coefficient k is calculated using a convolutional neural network. For large devices, k=0.8 (severe occlusion), and for small devices, k=0.2 (no obvious occlusion). Terminal requirement parameters (latency t=4.2ms, reliability R=99.9995%) are extracted using linear regression. Threshold judgment: latency t=4.2ms<5ms, reliability R=99.9995%≥99.999%, heavy obstruction area ratio 8%≤10%, meeting the preset deployment threshold conditions; Deployment parameters were calculated: using a reinforcement learning algorithm, the deployment parameters for the three base stations were obtained: Base station 1 coordinates (25m, 40m, 10m), beam angle 30°, transmit power 43dBm; Base station 2 coordinates (75m, 40m, 10m), beam angle 30°, transmit power 43dBm; Base station 3 coordinates (50m, 80m, 10m), beam angle 45°, transmit power 45dBm (avoiding areas obstructed by large equipment). Deployment Execution and Dynamic Optimization: The system controlled three base stations to complete deployment according to the deployment parameters, and collected communication status data in real time. The actual latency t=4.1ms and the actual reliability R=99.9996%, which met the requirements. After one month of operation, due to the adjustment of the equipment layout in the workshop, the obstruction coefficient k in a certain area increased to 0.9, and the actual latency t=5.3ms. The deployment control equipment automatically adjusted the beam angle of base station 3 to 50° and the transmit power to 46dBm. After the adjustment, the actual latency t=4.3ms, which returned to the threshold range.

[0044] In this embodiment, the mobile communication latency in the industrial workshop is stabilized between 4.0-4.5ms and the reliability is stabilized above 99.999% through the method and apparatus of the present invention. This solves the pain point that the existing deployment methods cannot meet the low latency and high reliability requirements of industrial scenarios and improves production efficiency.

[0045] Vehicle-Road Cooperative Scenarios Example This embodiment is applied to a section of an urban expressway, which is 2km long and 30m wide, includes 2 intersections, and the vehicle speed range is 60-120km / h. The vehicle-road cooperative mobile terminal is required to have a signal stability of ≥99.5%.

[0046] The specific implementation process of the mobile communication device deployment method and apparatus of the present invention is as follows: The deployment of control equipment uses roadside units and GPS positioning devices to acquire basic data: road layout data (2km×30m, intersection location, curve radius), terminal equipment data (real-time vehicle location, driving speed, communication bandwidth requirements), and signal transmission environment data (size and location of trees and buildings on both sides of the road, signal path loss data). The deployment control equipment uses the 3σ criterion to remove abnormal data (such as abnormal location data from GPS positioning), and standardizes the data using the min-max standardization method to obtain standardized data; Scene feature parameters were extracted: occlusion coefficient k was calculated using a convolutional neural network, with k=0.4 for trees on both sides of the road (moderate occlusion) and k=0.6 for buildings (moderate occlusion); vehicle trajectory feature parameters (predicted position and speed) were extracted using an adaptive Kalman filter algorithm; and signal stability requirement parameter S=99.6% was extracted using linear regression. Threshold judgment: The signal stability parameter S=99.6%≥99.5%, there is no severe obstruction in the core area of ​​the road, and the preset deployment threshold conditions are met; Deployment parameters were calculated: Using a reinforcement learning algorithm, the deployment parameters for four base stations were obtained: Base station 1 coordinates (500m, 15m, 12m), beam angle 25°, transmit power 44dBm; Base station 2 coordinates (1000m, 15m, 12m), beam angle 25°, transmit power 44dBm; Base station 3 coordinates (1500m, 15m, 12m), beam angle 25°, transmit power 44dBm; Base station 4 coordinates (2000m, 15m, 12m), beam angle 25°, transmit power 44dBm. Simultaneously, based on vehicle trajectory prediction, the dynamic adjustment period for the base station beams was set to 0.1s. Deployment Execution and Dynamic Optimization: The system controls four base stations to complete deployment according to the deployment parameters and collects communication status data in real time. When the vehicle travels at a speed of 120 km / h, the actual signal stability S=99.7%, which meets the requirements. When the vehicle enters the intersection and its movement trajectory deviates, the deployment control equipment predicts the vehicle's position through an adaptive Kalman filter algorithm and dynamically adjusts the base station beam angle to ensure that the signal stability is always ≥99.5%, thus solving the pain point of poor signal stability in existing deployment methods.

[0047] In this embodiment, using the method and apparatus of the present invention, the signal stability of vehicles in a vehicle-road cooperative scenario remains stable at over 99.5% across different driving speeds (60-120 km / h), with signal stability maintained at 99.7%-99.8% at high speeds (100-120 km / h). The base station handover interruption rate is reduced to below 0.05%, representing a significant improvement over existing deployment methods (handover interruption rate 1.2%, signal stability 98.3%). Simultaneously, the base station beam dynamic adjustment response time is ≤0.1s, enabling precise tracking of vehicle movement trajectories and avoiding signal attenuation and interruption caused by high-speed vehicle movement, road curves, and obstructions. This effectively ensures the real-time and stable transmission of vehicle-road cooperative data (such as vehicle location, driving status, and road condition warnings).

[0048] Furthermore, the deployment process in this embodiment is fully automated, from basic data acquisition and deployment parameter calculation to base station deployment and dynamic optimization, requiring no manual intervention. Deployment efficiency is improved by more than 60% compared to traditional manual deployment methods, and the deployment cost per kilometer is reduced by 30%, significantly lowering the deployment and maintenance costs of vehicle-road cooperative scenarios. This embodiment fully verifies the feasibility and practicality of the method and device of the present invention in vehicle-road cooperative scenarios, effectively addressing the pain points of existing deployment methods such as poor adaptability, insufficient signal stability, and low deployment efficiency, providing reliable technical support for the large-scale development of the vehicle-road cooperative industry.

[0049] The technical scope of this invention is not limited to the content described above. Those skilled in the art can make various modifications and variations to the above embodiments without departing from the technical concept of this invention, and all such modifications and variations should fall within the protection scope of this invention.

Claims

1. A method for deploying mobile communication devices, characterized in that, The method, which uses a deployment control device as the executing entity and the mobile communication device as a base station, includes the following steps: S1: Deploy control equipment to acquire basic data of target scenarios in vertical industries. The basic data includes scenario layout data, terminal device data, and signal transmission environment data. S2: Deploy control equipment to preprocess the basic data, remove abnormal data using the 3σ criterion, and standardize the basic data after removing abnormal data using the min-max standardization method to obtain standardized data; S3: Deploy control equipment to extract features from the standardized data, extract occlusion feature parameters through convolutional neural network, extract terminal demand parameters through linear regression, and extract movement trajectory feature parameters through adaptive Kalman filter algorithm to obtain scene feature parameters; S4: The deployment control device determines whether the scene feature parameters meet the preset deployment threshold conditions; the preset deployment threshold conditions are: in industrial scenarios, the latency parameter in the terminal requirement parameters is <5ms and the reliability parameter is ≥99.999%; in vehicle-road cooperative scenarios, the signal stability parameter corresponding to the movement trajectory feature parameters is ≥99.5%; S5: If the scene feature parameters meet the preset deployment threshold conditions, the deployment control device calculates the deployment parameters of the mobile communication device based on the scene feature parameters using a reinforcement learning algorithm. The deployment parameters include base station location coordinates, base station beam angle, and base station transmit power. If not, the deployment control device returns to step S1 to reacquire basic data and optimizes the abnormal data removal threshold in the preprocessing process. S6: The deployment control device outputs the deployment parameters and controls the mobile communication device to complete the deployment according to the deployment parameters; at the same time, the deployment control device collects the communication status data after deployment in real time, uses the communication status data as new basic data, and executes steps S3-S5 in a loop to realize dynamic optimization of the deployment parameters.

2. The mobile communication device deployment method according to claim 1, characterized in that, The scene layout data in step S1 includes the coordinates of industrial workshop equipment and the coordinates of vehicle-road cooperative road nodes; the terminal equipment data includes the location of industrial control terminals and the speed and location of vehicle-road cooperative mobile terminals; the signal transmission environment data includes the size of obstructions, the location of obstructions, and signal propagation loss data.

3. The mobile communication device deployment method according to claim 1, characterized in that, The calculation formula for the min-max normalization method described in step S2 is as follows: in, For standardized basic data, As the original basic data, The minimum value in the original basic data. This is the maximum value in the original base data.

4. The mobile communication device deployment method according to claim 1, characterized in that, The formula for calculating the occlusion feature parameters in step S3 is as follows: in, The occlusion coefficient is... For the height of the obstruction, Preset the height for mobile communication devices. The horizontal distance between the obstruction and the base station's preset location. Preset the angle for the base station beam; when When, it is determined that there is no obvious obstruction; when When, it is judged as moderate occlusion; when At that time, it was determined to be severe occlusion.

5. The mobile communication device deployment method according to claim 1, characterized in that, The movement trajectory feature parameters mentioned in step S3 are extracted using an adaptive Kalman filter algorithm, which includes a state prediction equation and a state update equation; a state vector is defined. ,in , for Moment mobile terminal coordinate, , for Mobile terminals at all times velocity in the direction; State prediction equation: State update equation: in, for Predicted state at any given time Here is the state transition matrix. for The optimal estimated state at time t. For the control matrix, for The amount of control at any given moment for The covariance matrix of the predicted state at each time step. The process noise covariance matrix is ​​obtained by... Factors are dynamically adjusted. H is the Kalman gain, and H is the observation matrix. To observe the noise covariance matrix, the β factor is dynamically adjusted. Let k be the observation value at time k. It is an identity matrix; the α factor is calculated using the motion state consistency index. The factor is calculated using communication quality indicators.

6. The mobile communication device deployment method according to claim 1, characterized in that, The comprehensive reward function of the reinforcement learning algorithm described in step S5 is: in, For the overall reward value, , , These are the weighting coefficients for coverage performance, communication quality, and deployment cost, respectively. ; To cover performance bonus values, This is a communication quality bonus value. The deployment cost reward value is used to determine the optimal deployment parameters for the deployment control equipment by maximizing the overall reward value.

7. The mobile communication device deployment method according to claim 1, characterized in that, The communication status data mentioned in step S6 includes actual latency data, actual reliability data, and actual signal stability data; The deployment control device compares the actual latency data with preset latency thresholds, the actual reliability data with preset reliability thresholds, and the actual signal stability data with preset signal stability thresholds, calculates the deviation value, and adjusts the deployment parameters based on the deviation value.

8. A mobile communication device deployment apparatus, characterized in that, Integrated into deployment control equipment, it includes a data acquisition module, a data preprocessing module, a feature extraction module, a threshold judgment module, a parameter calculation module, a deployment control module, and a dynamic optimization module; The data acquisition module is used to acquire basic data of the target scenario in the vertical industry, including scenario layout data, terminal device data and signal transmission environment data; it is also used to collect communication status data after deployment in real time. The data preprocessing module is used to... The criteria remove outliers from the base data, using... The standardization method standardizes the base data after removing outliers to obtain standardized data; The feature extraction module is used to extract occlusion feature parameters through a convolutional neural network, extract terminal demand parameters through linear regression, and extract movement trajectory feature parameters through an adaptive Kalman filter algorithm to obtain scene feature parameters. The threshold judgment module is used to determine whether the scene feature parameters meet the preset deployment threshold conditions. The preset deployment threshold conditions are: in industrial scenarios, the latency parameter is <5ms and the reliability parameter is ≥99.999%. In vehicle-road cooperative scenarios, signal stability parameters are ≥99.5%; The parameter calculation module is used to calculate the deployment parameters of the mobile communication device based on the scene feature parameters and through a reinforcement learning algorithm when the scene feature parameters meet the preset deployment threshold conditions. The deployment parameters include base station location coordinates, base station beam angle, and base station transmit power. When the conditions are not met, the data acquisition module is triggered to reacquire the basic data and optimize the abnormal data removal threshold. The deployment control module is used to output the deployment parameters and control the mobile communication device to complete the deployment according to the deployment parameters; The dynamic optimization module is used to take the communication status data collected by the data acquisition module as new basic data, trigger the feature extraction module, threshold judgment module and parameter calculation module to work in a loop, so as to realize the dynamic optimization of deployment parameters.

9. The mobile communication device deployment apparatus according to claim 8, characterized in that, The data preprocessing module uses The standardization method, the occlusion feature parameter calculation formula used in the feature extraction module, the adaptive Kalman filter algorithm, and the reinforcement learning algorithm used in the parameter calculation module are all consistent with the formulas and algorithms corresponding to claims 3-6.

Images