5G signal blind area AI prediction compensation method and system

Abstract

The invention relates to the technical field of signal compensation, in particular to a 5G signal blind area AI prediction compensation method and system, and the method comprises the following steps: based on an urban region, fusing remote sensing image surface reflectance and micro base station signal intensity, recognizing a signal attenuation risk region, analyzing the signal change trend and fluctuation characteristics, and combining the user connection frequency and density to obtain a 5G signal blind area AI prediction compensation result. And judging trend deviation, screening relay nodes, determining a switching sequence, adjusting coverage parameters, and generating a node switching compensation instruction set. Through fusion application of multi-source environment information and signal dynamic data, pre-recognition of a risk area is carried out, synchronous analysis of a signal fluctuation time sequence rule and a user behavior mode is achieved, on the basis of multi-parameter linkage judgment, relay node state switching is automatically driven, a coverage blind area is compensated, and real-time matching of signals and interference characteristics is combined, so that the reliability of the system is improved. Signal coverage continuity is improved, instant distribution of compensation response is realized, and the duration of a signal unstable area is effectively shortened.

Description

Technical Field

[0001] This invention relates to the field of signal compensation technology, and in particular to a 5G signal blind spot AI prediction compensation method and system. Background Technology

[0002] Signal compensation primarily involves effectively compensating for and optimizing areas with insufficient signal coverage or signal attenuation in wireless communication networks. This includes monitoring and analyzing the distribution of wireless signals, identifying network coverage blind spots and weak signal areas, and improving signal coverage through methods such as adjusting base station configurations, adding compensation equipment, optimizing antenna deployment, and improving transmission parameters. Traditional 5G signal blind spot compensation refers to addressing signal blind spots in 5G communication networks caused by the easily obstructed propagation characteristics of high-frequency signals. This is achieved by detecting blind spots based on historical coverage data or user terminal feedback, and then compensating for coverage in the detected blind spots using methods such as adding small base stations, distributed antennas, power adjustment, or directional antenna adjustment.

[0003] Existing technologies mostly rely on static historical information and user feedback to identify network blind spots. Their ability to perceive changes in areas in real time is limited, making it difficult to cope with dynamic signal fluctuations caused by rapid changes in the signal environment or high-density populations. There are response delays in adjusting base stations and compensation facilities, making it difficult to cover short-term weak areas caused by weather disturbances or building layouts. This results in intermittent communication services in some areas. In complex urban scenarios or densely populated areas, the risk of signal attenuation is easily overlooked, and the ability to continuously compensate and coordinate with signals is limited. Summary of the Invention

[0004] To address the technical problems existing in the prior art, this invention provides a 5G signal blind spot AI prediction and compensation method and system. The technical solution is as follows: On the one hand, a 5G signal blind spot AI prediction and compensation method is provided, including the following steps: S1: Based on urban areas, analyze surface reflectivity, collect signal strength, calculate the difference between the two, and combine the influence of environmental factors to identify areas at risk of signal attenuation under reflection interference, and obtain the regional shading distribution. S2: Based on the regional masking distribution, calculate the changing trend of signal intensity in the target area, compare the signal intensity change amplitude at each time node, determine the signal fluctuation pattern, analyze the abnormal change trend of periodic signal, and obtain the sequence fluctuation characteristics; S3: Based on the sequence fluctuation characteristics, analyze the changes in user connection frequency in the high-reflection occlusion area, calculate the influence of user activity density on the signal slope, set the signal change slope criterion, determine the continuous period slope deviation trend, and obtain the trend offset signal. S4: Based on the trend offset signal, analyze the signal strength change in the adjacent area, calculate the correspondence between the interference signal waveform and the signal change in the target area, determine the relay handover priority order, and obtain the handover coordination parameters; S5: Based on the handover coordination parameters, analyze the spatial relationship between the current relay node signal coverage area and the target attenuation zone, determine whether the existing compensation behavior meets the requirements, adjust the coverage parameters and issue a handover command to obtain the node handover compensation command set.

[0005] On the other hand, the regional shading distribution includes shading area identification number, risk level classification, and spatial mapping information; the sequence fluctuation characteristics include periodic fluctuation segments, change trend identifiers, and feature extraction codes; the trend offset signal includes slope change labels, user density characteristics, and abnormal trend identifiers; the handover coordination parameters include node handover sequence number, handover priority category, and feature matching results; and the node handover compensation instruction set includes handover control type and compensation scheduling number.

[0006] On the other hand, the specific steps for obtaining the regional shading distribution are as follows: S101: Based on urban areas, analyze the surface reflection data frames acquired from L-band remote sensing images, identify regional pixels with signal reflection, compare the pixel reflectivity with the signal strength data collected by micro base stations, integrate the two types of data according to spatial coordinates, and obtain the reflectivity intensity distribution characteristics. S102: Based on the reflectivity intensity distribution characteristics, calculate the difference between reflectivity and signal strength of each coordinate unit, analyze wind speed, precipitation and building distribution parameters, optimize the collaborative processing flow between differences and environmental factors, determine the spatial location with reflection interference characteristics, and obtain the interference risk grid identifier. S103: Based on the interference risk grid identifier, analyze the geospatial coordinates corresponding to the grid within the urban area, compare the risk level with the pixel connectivity features in the remote sensing image, encode the combination information of space and interference level, and obtain the regional occlusion distribution.

[0007] On the other hand, the steps for obtaining the sequence fluctuation characteristics are as follows: S201: Based on the regional shading distribution, analyze the spatial mapping information and shading area number, calculate the signal intensity change of the corresponding spatial unit in the continuous time series, mine the increase or decrease state of the signal at each time point through sequence difference operation, optimize and compare the change amplitude between each time point, and obtain the signal trend characteristics. S202: Based on the signal trend characteristics, compare the differences in the characteristics of each change segment, optimize the timing synchronization parameters of the signal acquisition process, identify the intervals of signal intensity transitions between consecutive time nodes, and aggregate the transition states and spatial distributions within each change interval to obtain the fluctuation transition distribution index. S203: Based on the fluctuation turning point distribution index, analyze the continuity and trend stability of the signal change trajectory, screen the abnormal variation features in the periodic sequence, and cross-mark the abnormal segment and the normal segment to obtain the sequence fluctuation features.

[0008] On the other hand, the steps for obtaining the trend offset signal are as follows: S301: Based on the sequence fluctuation characteristics, analyze the user connection records in the high-reflectivity building obstruction area, calculate the frequency data of user access to the network in different time periods in each obstruction area, and determine the fluctuation type of user connection frequency in each interval to obtain the connection frequency change sequence. S302: Based on the connection frequency variation sequence, analyze the correlation parameters between user activity density and signal strength change in each spatial unit, calculate the influence factor of user activity density on the signal strength change trend over time, and optimize the signal change slope criterion set based on user distribution. S303: Call the signal change slope criterion, filter the signal slope deviation parameter within the continuous period, determine whether the signal change trend meets the abnormal change judgment condition, assign a trend offset label to the fluctuation segment that meets the condition, and obtain the trend offset signal.

[0009] On the other hand, the specific steps for obtaining the switching coordination parameters are as follows: S401: Based on the trend offset signal, analyze the spatial location of the marker, filter the mobile edge communication unit covering the location, compare the signal strength change amplitude of each communication unit in the target area and its adjacent areas, determine the correlation characteristics of signal fluctuations in adjacent areas, and obtain the neighboring area interference characteristics. S402: Based on the neighboring cell interference characteristics, calculate the changing trend of the interference signal waveform in the neighboring area, analyze the correspondence between the trend and the change in the signal strength of the target area, determine whether the interference signal trajectory and the signal fluctuation trajectory of the target area meet the relay switching conditions, and obtain the relay switching factor. S403: Filter the edge communication units with handover adaptation characteristics among the relay handover factors, analyze the order of the communication units in the relay node sequence, determine the handover priority relationship between the communication units and the target area, and obtain the handover coordination parameters.

[0010] On the other hand, the steps for obtaining the node switching compensation instruction set are as follows: S501: Based on the switching coordination parameters, analyze the relationship between the node switching sequence number and the spatial correspondence, determine the overlap between the current relay node signal coverage boundary and the geographical mapping area of ​​the signal attenuation zone, compare the interaction positions of the two in the spatial structure, and obtain the spatial coverage mapping relationship. S502: Based on the spatial coverage mapping relationship, determine whether the signal compensation behavior of the relay node meets the communication requirements in the signal attenuation zone, compare the change rate of the node signal strength with the fluctuation amplitude of the interference signal in the neighboring cell, analyze the adaptation status of the current compensation configuration, and obtain the compensation adaptation identifier. S503: Based on the compensation adaptation identifier, adjust the parameters of relay nodes with inconsistent configurations, optimize the coverage angle and transmission direction of the corresponding nodes, and complete the scheduling association between the target area number and the handover instruction according to the spatial matching result to obtain the node handover compensation instruction set.

[0011] On the other hand, the surface reflectivity represents the ability of various areas on the ground to reflect wireless signals. It is the proportion of energy reflected back after a radar or wireless signal encounters a surface object to the incident energy. The environmental factors refer to natural or artificial conditions that affect the propagation of wireless signals. The signal attenuation risk area refers to the geographical grid or area where signal attenuation occurs.

[0012] On the other hand, the trend of signal strength refers to the increase or decrease of signal strength values ​​in the same area over a period of time, and the signal fluctuation pattern refers to the analysis of the amplitude of change to uncover the characteristics of signal strength changes in the time series.

[0013] On the other hand, a 5G signal blind zone AI prediction and compensation system is provided. This system is applied to the 5G signal blind zone AI prediction and compensation method, including: The occlusion identification module is based on urban areas, analyzes surface reflectivity, collects signal strength, calculates the difference between the two, and combines the influence of environmental factors to identify areas at risk of signal attenuation under reflection interference, thus obtaining the regional occlusion distribution. Based on the regional masking distribution, the fluctuation feature extraction module calculates the changing trend of signal intensity in the target area, compares the signal intensity change amplitude at each time node, judges the signal fluctuation pattern, analyzes the abnormal change trend of periodic signal, and obtains the sequence fluctuation features. Based on the sequence fluctuation characteristics, the user slope analysis module analyzes the changes in user connection frequency in high-reflection occlusion areas, calculates the impact of user activity density on signal slope, sets signal change slope criteria, judges the continuous period slope deviation trend, and obtains the trend offset signal. Based on the trend offset signal, the handover coordination module analyzes the signal strength changes in adjacent areas, calculates the correspondence between the interference signal waveform and the signal changes in the target area, determines the relay handover priority order, and obtains the handover coordination parameters. Based on the handover coordination parameters, the compensation control module analyzes the spatial relationship between the current relay node signal coverage area and the target attenuation zone, determines whether the existing compensation behavior meets the requirements, adjusts the coverage parameters, and issues a handover command to obtain a node handover compensation command set.

[0014] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: By integrating multi-source environmental information and dynamic signal data, the system proactively identifies risk areas based on signal attenuation trends, enabling synchronous analysis of signal fluctuation patterns and user behavior patterns. Based on multi-parameter linkage judgment, it automatically drives relay node status switching to compensate for coverage blind spots. Combined with real-time matching of signal and interference characteristics, it improves signal coverage continuity, achieves immediate allocation of compensation responses, effectively shortens the duration of unstable signal areas, and enhances the continuous connectivity of 5G communication networks and the efficiency of compensation coordination in high-density scenarios. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of the main steps of the present invention; Figure 2 This is a flowchart of steps S1 of the present invention; Figure 3 This is a flowchart of steps S2 of the present invention; Figure 4 This is a flowchart of steps S3 of the present invention; Figure 5 This is a flowchart of step S4 of the present invention; Figure 6 This is a flowchart of steps S5 of the present invention; Figure 7 This is a system block diagram of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0018] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0019] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0020] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0021] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0022] This invention provides an AI prediction and compensation method for 5G signal blind spots, such as... Figure 1 As shown, it includes the following steps: S1: Based on urban areas, analyze the surface reflectance obtained from L-band remote sensing images, combine it with the signal strength collected by ground micro base stations, calculate the difference between reflectance and signal strength, determine the interference effect of environmental factors, identify the signal attenuation risk areas under the reflection interference standard, and obtain the regional shading distribution. S2: Based on the regional masking distribution, calculate the changing trend of regional signal strength, optimize the timing synchronization during signal acquisition, compare the change amplitude of continuous signal strength at each time node, determine the signal fluctuation pattern, and analyze whether there are any abnormalities in the signal change trend of each period to obtain the sequence fluctuation characteristics. S3: Based on the sequence fluctuation characteristics, analyze the changes in user connection frequency in the high-reflectivity building obstruction area, calculate the impact of user activity density on signal slope changes, optimize the signal change slope criterion set based on user distribution, determine whether the continuous period signal slope deviation has an abnormal trend, and obtain the trend offset signal. S4: Based on the trend offset signal, filter the mobile edge communication units in the target area, analyze the signal strength change amplitude of their adjacent areas, calculate the correspondence between the interference signal waveform and the signal strength change in the target area, determine whether the joint characteristics of the two meet the relay handover standard, determine the priority order of relay handover, and obtain the handover coordination parameters. S5: Based on the handover coordination parameters, analyze the spatial relationship between the current relay node signal coverage area and the target signal attenuation area, determine whether the existing relay node signal compensation behavior can meet the regional communication needs, compare the signal strength change rate with the fluctuation of interference signals in neighboring cells, adjust the coverage parameters of the relay node, and issue an early handover command to obtain the node handover compensation command set.

[0023] The regional shading distribution includes shading area identification number, risk level classification, and spatial mapping information; the sequence fluctuation characteristics include periodic fluctuation segments, change trend indicators, and feature extraction codes; the trend offset signal includes slope change labels, user density characteristics, and abnormal trend indicators; the handover coordination parameters include node handover sequence number, handover priority category, and feature matching results; and the node handover compensation instruction set includes handover control type and compensation scheduling number.

[0024] In S1, L-band remote sensing imagery refers to remote sensing data obtained by imaging the urban surface using L-band (wavelength approximately 1-2 GHz) radar or satellites to acquire information reflecting the type, structure, and distribution of ground features; surface reflectivity represents the ability of different areas of the ground to reflect wireless signals, which is the proportion of energy reflected back after a radar or wireless signal encounters a surface object to the incident energy; signal strength refers to the power or amplitude of a 5G wireless signal collected by a micro base station or receiving device at a specific geographical location, used to assess the communication quality of that location; environmental factors refer to natural or artificial conditions that affect the propagation of wireless signals, such as wind speed, precipitation, temperature, humidity, and building distribution; reflection interference standard refers to a standard set based on the difference between surface reflectivity and signal strength, used to determine whether the signal attenuation risk caused by reflection in a certain area reaches a level that requires handling; signal attenuation risk area refers to the geographical grid or area identified according to the standard where signal attenuation (i.e., a significant decrease in signal strength) occurs.

[0025] In S2, the signal strength trend refers to the increase or decrease of signal strength values ​​within the same area over a period of time, used to analyze signal stability and potential risks; time synchronization refers to aligning data according to a unified time reference when collecting multi-source data, in order to accurately analyze the signal's characteristics over time; each time node refers to the different time points when signal strength data is collected, which are the basic units constituting the time series data sequence; the change amplitude refers to the magnitude of the change in signal strength between two consecutive time nodes, that is, the increase or decrease of the signal over time; the signal fluctuation pattern refers to revealing the characteristics of signal strength changes over time series by analyzing a series of change amplitudes, such as stability, oscillation, or abrupt change.

[0026] In S3, user connection frequency variation refers to the change in the number of times a user accesses the 5G network in a certain area at different times, which is used to reflect the dynamics of user flow and distribution; signal slope variation refers to the rate of change of signal strength over time, which reflects the speed of signal attenuation or recovery by calculating the slope of the signal strength curve; signal change slope criterion refers to the reference standard set based on actual user distribution and historical statistics to judge whether the signal change trend is abnormal; signal slope deviation refers to the difference between the actual measured signal strength slope and the criterion standard, and a significant deviation can be considered as an abnormality in the signal environment.

[0027] In S4, the target area refers to the specific spatial location or geographical range that has been identified as having a risk of signal attenuation and requires key attention and compensation; the mobile edge communication unit refers to a 5G relay node or small base station with flexible deployment capabilities and dynamically adjustable coverage, used to enhance signals in blind spots or weak coverage areas; the interference signal waveform refers to the time-series characteristic data of interference signals from surrounding communication units, used to analyze the impact of interference on signal quality; the correspondence refers to the correlation established between the signal strength change in the target area and the surrounding interference waveform through data comparison, used to further determine influencing factors; the joint feature refers to the comprehensive analysis of the signal strength change and interference waveform characteristics of the target area, used as a criterion for relay handover conditions; and the relay handover standard refers to the specific standards or rules set based on the analysis of multiple parameters to determine whether relay nodes need to be switched in advance.

[0028] In S5, spatial relationship refers to the spatial distribution and mutual influence between the current relay node signal coverage area and the signal attenuation risk area; signal compensation behavior refers to the specific operation process by which the relay node dynamically adjusts its own parameters to enhance or cover the signal in the attenuation area; area communication demand refers to the information requirements such as the actual network connection demand and bandwidth demand of users in the signal attenuation area; and handover command refers to the control command issued by the system to the relay node in advance to switch or adjust the signal coverage parameters based on the data analysis results.

[0029] like Figure 2 As shown, the specific steps for obtaining the regional occlusion distribution are as follows: S101: Based on urban areas, analyze the surface reflection data frames acquired from L-band remote sensing images, identify regional pixels with signal reflection, compare the pixel reflectivity with the signal strength data collected by micro base stations, integrate the two types of data according to spatial coordinates, and obtain the reflectivity intensity distribution characteristics. High-resolution images covering the city are acquired using L-band remote sensing equipment. The reflectance value corresponding to each surface pixel is extracted from the data. This value is obtained by processing image echo information. The pixels are then arranged into reflectance data frames according to their latitude and longitude in the image. For example, pixels with higher reflectance values ​​in residential areas typically indicate reinforced concrete structures, corresponding to reflectance values ​​higher than 0.65. In green areas such as parks or grasslands, the reflectance is lower, often below 0.35. Subsequently, signal strength data collected by micro base stations is extracted. Data records that are close to the time of acquisition of the remote sensing images and coincide with their coordinate locations are selected. The signal data is then aligned and matched with image pixels according to their latitude and longitude positions. Within a positional deviation of less than 5 meters, the two... Each dataset is integrated, and each integrated unit contains the reflectance of a pixel and the signal strength value at the corresponding location. Then, the relative difference between the reflectance and signal strength values ​​is compared to form a contrast value, and a distribution layer is constructed. Each spatial unit is assigned a combined value of reflectance-intensity pair. If the reflectance of a high-rise building pixel is 0.82 and the signal strength value at its corresponding location is low (-95dBm), then this combination is marked as a significant contrast area. On the other hand, the reflectance of a pixel in a grassland is 0.25 and the signal strength is high (-75dBm), so its combination has low contrast. This type of data combination is performed on all coordinate units and uniformly labeled to form a reflectance and signal strength comparison distribution map, thereby obtaining the reflectance intensity distribution characteristics.

[0030] S102: Based on the reflectivity intensity distribution characteristics, calculate the difference between reflectivity and signal intensity of each coordinate unit, analyze wind speed, precipitation and building distribution parameters, optimize the collaborative processing flow between differences and environmental factors, determine the spatial location with reflective interference characteristics, and obtain interference risk grid identification. First, an interpolation operation is performed on each location coordinate unit to compare the difference between the reflectance value and the normalized signal strength value at that location. For areas with large differences, i.e., units with high reflectance and weak signal strength, their corresponding spatial coordinates are extracted and temporarily marked as suspected risk points. Then, environmental factor data, including wind speed, precipitation, and building density, are introduced. Wind speed and precipitation records for the same time period are extracted from historical meteorological datasets. The proportion of building area is extracted as a building density parameter using urban 3D modeling or remote sensing layers. For wind speeds exceeding 6 m / s, precipitation exceeding 15 mm, and building density exceeding 60%, [further analysis is needed]. All areas are considered high-impact factors. Then, the difference of each coordinate point is combined with the scores of the corresponding three environmental factors to form a collaborative vector. The vector is then weighted and comprehensively scored. If the total score exceeds a certain threshold, the coordinate point is considered to not only have a prominent difference in reflectivity and signal strength, but also to have environmental factors further aggravating signal interference. If the difference in a certain area is large, and the wind speed is 8 meters per second, the precipitation is 20 millimeters, and the building density is 75%, its comprehensive score is significantly high. Such coordinate points are formally marked as reflection interference risk units and assigned a risk level. An interference risk grid layer is constructed for subsequent shading analysis.

[0031] S103: Based on the interference risk grid identification, analyze the geospatial coordinates of the grids within the urban area, compare the risk level with the pixel connectivity features in the remote sensing image, encode the combination information of space and interference level, and obtain the regional occlusion distribution. Extract all spatial grid numbers marked as high-risk and retrieve their geographical locations in the original remote sensing image. Obtain the latitude and longitude coordinates of each grid using a coordinate lookup method. Then, screen the corresponding pixel regions in the remote sensing image. Analyze the spatial connectivity between adjacent high-reflectivity pixels using the 8-neighbor pixel connectivity rule. If adjacent pixels are found to have consecutive reflectance values ​​higher than a specified level, and the total area of ​​the connected region exceeds approximately 30 square meters, the region is determined to have an effective shading effect and is marked as a connected shading block. Then, based on the size of the shading block and its corresponding risk level... The risk level is combined and coded, and the shading area is divided into several level intervals. For example, less than 20 square meters is classified as low intensity, 20 to 40 square meters as medium intensity, and more than 40 square meters as high intensity. The risk level is divided into 7 to 10 levels according to the total score interval calculated in the previous step. A spatial coding identifier with a structure such as "level-shading intensity" is constructed. If the risk level of a certain area is 9 and the shading intensity is medium, it is marked as "9-2". This coding information is bound to the corresponding spatial grid to form an attribute field and reprojected into the urban geographic information system to generate a complete regional shading distribution layer.

[0032] like Figure 3 As shown, the specific steps for obtaining sequence fluctuation characteristics are as follows: S201: Based on the regional shading distribution, analyze the spatial mapping information and shading area number, calculate the signal intensity change of the corresponding spatial unit in the continuous time series, mine the increase or decrease state of the signal at each time point through sequence difference operation, optimize and compare the change amplitude between each time point, and obtain the signal trend characteristics. First, spatial mapping information is extracted. The geospatial unit corresponding to each masked area number is correlated with the signal observation grid. Each pixel in the masked area is mapped to the latitude and longitude coordinates of the actual city. A continuous time series data record is established for each spatial unit. The 5G signal strength observation values ​​for each unit over a period of time are retrieved. The signal strength data comes from real-time reports from micro base stations or terminal devices. Specifically, the signal strength of a numbered unit within a certain masked area was -89dBm at 08:00 on May 1, 2024, -91dBm at 08:10, and -88dBm at 08:20. The signal values ​​corresponding to each time point are arranged along the time axis to form a continuous time series data record. Continuing the time-series signal strength sequence, a sequence difference operation is performed on the signal strength at each time point. The values ​​of two consecutive time points are compared point by point to obtain the increase or decrease status at each time point. A positive value indicates signal enhancement, while a negative value indicates signal attenuation. If the absolute value of the difference is greater than 3dBm, it is marked as significant fluctuation. The amplitude of change between each time point is compared, and the amplitude of signal change between all time points is summarized. The amplitude of fluctuation is quantified by using the maximum, minimum, and average values ​​of the change interval to determine the fluctuation range of the signal in the unit within the observation period. Combined with the spatial number of the shading area, all unit data are processed in batches to summarize the overall trend characteristics of signal strength change of each spatial unit under the time series.

[0033] S202: Based on signal trend characteristics, compare the differences in characteristics of each change segment, optimize the timing synchronization parameters of the signal acquisition process, identify the intervals of signal intensity transitions between consecutive time nodes, and aggregate the transition states and spatial distributions within each change interval to obtain the fluctuation transition distribution index. Data from each variation segment is extracted, and the signal strength differences within each time period are compared one by one. The states of signal enhancement and attenuation within continuous intervals are distinguished. The average difference amplitude of each variation segment is calculated. A difference amplitude below 2dBm is classified as a stable segment, 2 to 5dBm as a general fluctuation segment, and above 5dBm as a severe fluctuation segment. All segments are categorized according to the above intervals. The timing synchronization parameters of the signal acquisition process are optimized, and timestamps are uniformly aligned to acquisition points based on a 5-minute timeframe to avoid timing misalignment caused by acquisition delays between different data sources. [Further details are needed for the next step.] For each consecutive time point, the signal strength sequence identifies the transition intervals where the signal value changes from enhancement to attenuation or from attenuation to enhancement. The time and spatial location of the transition are recorded, and all transition points are aggregated into their respective change intervals. The number and distribution of transition points within each change interval are counted. If a certain obscured area experiences a transition by first enhancing and then suddenly decreasing the signal strength within a 10-minute acquisition period, this segment is marked as having a transition state. The transition intervals are then aggregated on the spatial layer, and the transition points and spatial distribution of all intervals are summarized, outputting a fluctuation transition distribution index.

[0034] S203: Based on the fluctuation turning point distribution index, analyze the continuity and trend stability of the signal change trajectory, screen the abnormal variation characteristics in the periodic sequence, and cross-mark the abnormal and normal segments to obtain the sequence fluctuation characteristics. A continuous analysis of the signal variation trajectory of each spatial unit is performed to check whether the signal intensity changes continuously with time, exhibiting monotonically increasing or decreasing or frequently fluctuating phenomena. The stability of the fluctuation segments of each spatial unit is tested. Segments in which the signal intensity remains within 5 dBm within the same period are marked as stable sequences. If the signal intensity changes drastically by more than 5 dBm within a short period within the same period, it is marked as an abnormal fluctuation segment. All abnormal segments are screened out and cross-labeled with normal segments. The degree of overlap between abnormal and normal segments in time and space is compared. A corresponding fluctuation feature label is assigned to the continuous signal sequence of each spatial unit. The ratio of the occurrence of abnormal segments to normal segments in the entire observation period is calculated, and the complete sequence fluctuation feature is output.

[0035] like Figure 4 As shown, the specific steps for obtaining the trend offset signal are as follows: S301: Based on the sequence fluctuation characteristics, analyze the user connection records in the high-reflectivity building obstruction area, calculate the frequency data of user access to the network in different time periods in each obstruction area, and determine the fluctuation type of user connection frequency in each interval to obtain the connection frequency change sequence. Select spatial areas identified as high-reflectivity building obstructions. Extract user connection records for each obstruction area. Summarize the user access logs of all base stations or access points within the designated area to create a statistical table of user connection counts for each obstruction area in different time periods. For example, in the obstruction area below an office building, the number of user connections from 08:00 to 09:00 on May 1, 2024, is 28; from 09:00 to 10:00, it is 32; from 10:00 to 11:00, it is 16, and so on. Arrange the total number of user accesses in each time period to form a time series, categorize them by obstruction area number and correspond them one-to-one with the aforementioned spatial numbers, and then... The user connection frequency data for each time period is compared within the interval. The increase or decrease in the number of connections between two consecutive time periods is calculated. If the increase or decrease is greater than or equal to 10 times, it is marked as a sharp change. If it is between 3 and 10 times, it is marked as a normal change. If it is less than 3 times, it is a stable segment. All interval change types are compared and each change interval type is attached to the corresponding time series. The synchronization with the signal strength fluctuation period is also compared. If the abnormal signal strength fluctuation period is found to overlap with the user connection frequency change segment, the segment is numbered as a correlated fluctuation. The detailed statistics and change type determination of user connection frequency in all obstructed areas under different time periods are completed, and a connection frequency change sequence is generated.

[0036] S302: Based on the connection frequency variation sequence, analyze the correlation parameters between user activity density and signal strength changes in each spatial unit, calculate the influence factor of user activity density on the signal strength change trend over time, and optimize the signal change slope criterion set based on user distribution. The user activity density within each spatial unit for each time period is retrieved. The user activity density is calculated by dividing the number of user connections in the same area per unit time by the spatial area of ​​that area. Specifically, if an obstructed area has an area of ​​300 square meters and there are 24 user connections between 08:00 and 09:00, the activity density is 8 connections per 100 square meters. Then, the signal strength change values ​​for each time period are summarized, and user activity density and signal strength changes over time are paired item by item. The trend of signal strength change within each activity density interval is statistically analyzed. Signal strength is calculated separately for three groups: low activity density (less than 5 connections per 100 square meters), medium activity density (5 to 10 connections per 100 square meters), and high activity density (more than 10 connections per 100 square meters). The average slope over time is used to determine the signal strength. If the signal strength decreases as activity density increases, and the absolute value of the average slope is greater than 3 dBm / hour, the area is considered a high-sensitivity zone. Conversely, if the absolute value of the slope is less than 1 dBm / hour, the area is considered a low-sensitivity zone. These statistical results are used to set threshold values ​​for the signal slope. For example, the slope criterion for high-sensitivity zones is set to -3 dBm / hour, for medium-sensitivity zones to -2 dBm / hour, and for low-sensitivity zones to -1 dBm / hour. Sample statistical data is used. For example, in actual samples, the activity density in a high-sensitivity zone is found to be 12 times / 100 square meters, and the signal strength slope is -3.5 dBm / hour. Therefore, -3 dBm / hour is used as the criterion for this zone, forming a signal slope criterion set based on user distribution.

[0037] S303: Call the signal change slope criterion, filter the signal slope deviation parameter within the continuous period, determine whether the signal change trend meets the abnormal change judgment condition, assign a trend offset label to the fluctuation segment that meets the condition, and obtain the trend offset signal. The signal strength sequence of each occluded area within a continuous sampling period is traversed. The slope value of the signal strength is calculated for each adjacent time period. The slope value of each period is compared with the criterion of the corresponding area. If the actual slope value of a certain interval is lower than the criterion threshold, for example, the criterion is -3dBm / hour and the actual slope is -4dBm / hour, then the interval is marked as an abnormal deviation. The slope deviation values ​​of all intervals are arranged in chronological order. All sequence segments that do not meet the criterion intervals are filtered out and it is determined whether they appear continuously. If there is an abnormal slope deviation in three consecutive periods in an occluded area, a trend offset label is assigned to the occluded area. All labeled spatial unit numbers are collected and archived according to the occluded area number to form the spatial distribution result of the trend offset signal.

[0038] like Figure 5 As shown, the specific steps for obtaining the switching coordination parameters are as follows: S401: Based on the trend offset signal, analyze the spatial location of the marker, filter the mobile edge communication units covering the location, compare the signal strength change amplitude of each communication unit in the target area and its adjacent areas, determine the correlation characteristics of signal fluctuations in adjacent areas, and obtain the neighboring area interference characteristics. Retrieve spatial location data of all labeled trend offsets. For each trend offset signal's spatial coordinates, pinpoint its actual geographical location within the urban network. Compare the deployment of surrounding communication units at that location, and filter all mobile edge communication units that can cover that spatial point. Compare each unit number with the spatial location mapping table to determine the list of relay nodes that can participate in coverage. Subsequently, retrieve the signal strength change records for the communication units in each target area and its adjacent areas over the most recent period, extracting the maximum, minimum, and average fluctuation amplitude of the signal strength within the same time period. For communication units in the target area, if the signal strength fluctuation amplitude exceeds... A value of 10 dBm is marked as severe fluctuation, 5 to 10 dBm as moderate fluctuation, and below 5 dBm as stable fluctuation. The signal strength variation amplitude of each communication unit in the adjacent area is classified, the fluctuation type of each communication unit in the same time period is compared, the number of time periods in which the fluctuations are synchronized or opposite is counted, and the correlation characteristics of the signal fluctuations in adjacent areas are determined. If a neighboring area and the target area show synchronized fluctuations at least 80% of the time points, it is classified as a strongly correlated neighboring area. If it is less than 50%, it is classified as a weakly correlated neighboring area. The data of all spatial units are combined, the fluctuation correlation distribution of all adjacent areas is statistically recorded, and the neighboring area interference characteristics corresponding to each target area are output.

[0039] S402: Based on the characteristics of neighboring cell interference, calculate the changing trend of the interference signal waveform in the neighboring area, analyze the correspondence between the trend and the change of the signal strength in the target area, determine whether the interference signal trajectory and the signal fluctuation trajectory in the target area meet the relay switching conditions, and obtain the relay switching factor. The trend of interference signal intensity changing with time for each communication unit in adjacent areas during the sampling period is calculated one by one. The signal intensity fluctuation of each sampling point in the adjacent area is paired with the signal intensity change of the target area in the same time period. The correspondence between the two sets of signal intensity curves in terms of rate of change, peak occurrence time, and abrupt change interval is analyzed. The consistency of the signal intensity change direction of the two at key moments is statistically analyzed. If the signal intensity of the two sets of curves decreases or increases synchronously in most time periods, it is determined that the interference signal trajectory and the signal fluctuation trajectory of the target area have a high coupling relationship. Cases with a coupling degree exceeding 70% are recorded as strong interference. The impact is considered as follows: if the percentage is less than 40%, it is considered as weak interference. The signal fluctuations of the target area and neighboring areas are compared to see if they are synchronized. Combined with the aforementioned parameters such as fluctuation amplitude, direction of change, and number of synchronizations, it is determined whether the relay handover conditions are met. The baseline value of the handover conditions is set as the proportion of signal synchronization fluctuation periods being higher than 70% and the signal strength of the target area being lower than -90dBm within a continuous period. If the above conditions are met, the time period is recorded as the handover adaptation period. The parameters of all time periods in the handover adaptation period are statistically analyzed to generate relay handover factors. All relay handover factors that meet the handover conditions are summarized and output according to the spatial unit and sampling time period archive number.

[0040] S403: Filter edge communication units with handover adaptation characteristics from the relay handover factors, analyze the order of communication units in the relay node sequence, determine the handover priority relationship between the communication units and the target area, and obtain the handover coordination parameters. Edge communication units with handover adaptation characteristics are selected from all relay handover factors. The spatial attributes of the corresponding relay nodes are retrieved sequentially according to the handover factor number of each unit. All communication units are sorted in the relay node sequence by priority. The priority is ranked by a comprehensive ranking of multiple parameters such as signal strength compensation amplitude, fluctuation synchronization degree, and spatial overlap degree in the handover factors. Units with signal strength compensation amplitude greater than 15dBm, fluctuation synchronization degree greater than 85%, and spatial overlap area greater than 80% are ranked first, and so on down. The handover priority relationship between each communication unit and the target area is judged in turn according to the above ranking. If there are multiple switchable units in the same target area, the unit ranked first is selected as the first choice for handover. The handover sequence number, priority category, feature matching and other data are recorded as handover coordination parameters. The set of handover coordination parameters between all target areas and relay nodes is output.

[0041] like Figure 6 As shown, the specific steps for obtaining the node switching compensation instruction set are as follows: S501: Based on the handover coordination parameters, analyze the relationship between the node handover sequence number and the spatial correspondence, determine the overlap between the current relay node signal coverage boundary and the geographical mapping area of ​​the signal attenuation zone, compare the interaction positions of the two in the spatial structure, and obtain the spatial coverage mapping relationship. The switching sequence number and spatial location information of each relay node are retrieved. The latitude and longitude coordinates of the node in the urban space are matched one-to-one with the geographic mapping area coordinates of the target signal attenuation zone. The vector data of the signal coverage boundary of each node are called to determine whether there is an intersection between the coverage boundary and the outer contour of the signal attenuation zone. The spatial overlap area of ​​the intersection area is calculated. If the overlap area accounts for more than 50% of the total area of ​​the signal attenuation zone, it is marked as effective coverage; otherwise, it is considered insufficient coverage. The interaction relationship of the overlapping area in spatial structure is further analyzed. The azimuth angles of the coverage boundary passing through the signal attenuation zone are counted. The distance between the center point of the node coverage boundary and the geometric center of the signal attenuation zone is compared. If it is less than 30 meters, it is considered as high spatial overlap; otherwise, it is considered as general overlap. Combined with the node switching sequence number, the spatial coverage of all nodes is sorted by number and the coverage status flag is recorded to form a spatial structure mapping table corresponding to the coverage boundary and the signal attenuation zone, thus obtaining the spatial coverage mapping relationship.

[0042] S502: Based on the spatial coverage mapping relationship, determine whether the signal compensation behavior of the relay node meets the communication requirements in the signal attenuation zone, compare the change rate of the node signal strength with the fluctuation amplitude of the interference signal in the neighboring cell, analyze the adaptation status of the current compensation configuration, and obtain the compensation adaptation identifier. The signal compensation behavior of each relay node is analyzed one by one. The rate of change of the compensated signal strength of the node in the recent period is extracted and compared with the actual detected increase rate of user signal strength in the signal attenuation zone. If the rate of change of the node's signal strength is greater than 3dBm / min and the rate of increase of the signal strength in the signal attenuation zone is less than 2dBm / min, it is determined that the current compensation configuration fails to meet the communication requirements. In addition, the fluctuation amplitude data of the interference signal in the neighboring cell is retrieved, and the relationship between the current node's signal change rate and the fluctuation amplitude of the interference signal in the neighboring cell is statistically analyzed. If the node's signal rate is lower than the fluctuation amplitude of the interference in the neighboring cell, it indicates that the compensation capability is limited, and the compensation behavior of the node is marked as unsuitable. Conversely, if the node's signal change rate is higher than the fluctuation amplitude of the interference in the neighboring cell, and the average signal strength in the signal attenuation zone is above -85dBm, it is considered as suitable for compensation. The compensation suitability flag is set to 1, and the unsuitability flag is set to 0. A list of compensation suitability flags is output according to the spatial number and compensation status of all nodes.

[0043] S503: Based on the compensation adaptation identifier, adjust the parameters of relay nodes with inconsistent configurations, optimize the coverage angle and transmission direction of the corresponding nodes, and complete the scheduling association between the target area number and the handover command according to the spatial matching result to obtain the node handover compensation command set. All relay nodes identified as incompatible with compensation are retrieved, and their current configuration parameters are adjusted. Priority is given to adjusting the node's coverage angle and rotating the antenna direction to bring the coverage boundary closer to the main direction of the signal attenuation zone. At the same time, the node's transmission direction is optimized, aligning the main lobe with the attenuation area with the lowest signal strength. The node's transmission power and coverage radius are then reset. During the adjustment process, the spatial overlap area between the node's coverage area and the signal attenuation zone needs to be recalculated after each parameter adjustment. If the overlap area increases by more than 20% after adjustment, the current parameters are retained; otherwise, fine-tuning continues until the set improvement threshold or the maximum number of adjustments is reached. Subsequently, all adjusted nodes are matched with the target area number, and each adjustment action and parameter is included in the handover instruction list, archived in order of spatial location, handover type, and scheduling number to form a node handover compensation instruction set.

[0044] like Figure 7 As shown, a 5G signal blind spot AI prediction and compensation system includes: The occlusion identification module is based on urban areas, analyzes surface reflectivity, collects signal strength, calculates the difference between the two, and combines the influence of environmental factors to identify areas at risk of signal attenuation under reflection interference, thus obtaining the regional occlusion distribution. The fluctuation feature extraction module calculates the changing trend of signal intensity in the target area based on the regional masking distribution, compares the signal intensity change amplitude at each time node, judges the signal fluctuation pattern, analyzes the abnormal change trend of periodic signal, and obtains the sequence fluctuation features. The user slope analysis module analyzes the changes in user connection frequency in high-reflectivity occlusion areas based on sequence fluctuation characteristics, calculates the impact of user activity density on signal slope, sets signal change slope criteria, judges the continuous period slope deviation trend, and obtains the trend offset signal. The handover coordination module analyzes the signal strength changes in adjacent areas based on the trend offset signal, calculates the correspondence between the interference signal waveform and the signal changes in the target area, determines the relay handover priority order, and obtains the handover coordination parameters. Based on the handover coordination parameters, the compensation control module analyzes the spatial relationship between the current relay node signal coverage area and the target attenuation zone, determines whether the existing compensation behavior meets the requirements, adjusts the coverage parameters, and issues a handover command to obtain the node handover compensation command set.

[0045] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1.A 5G signal blind area AI prediction compensation method, characterized in that, The method comprises: S1: based on the urban area, analyzing the surface reflectivity, collecting the signal strength, calculating the difference between the two, combining the influence of environmental factors, identifying the signal attenuation risk area under the reflection interference, and obtaining the regional shielding distribution; S2: based on the regional shielding distribution, calculate the change trend of the signal strength of the target area, compare the signal strength change amplitude of each time node, judge the signal fluctuation law, analyze the periodic signal change trend anomaly, and obtain the sequence fluctuation characteristics; S3: based on the sequence fluctuation characteristics, analyze the user connection frequency change in the high reflection shielding area, calculate the influence of user activity density on signal slope, set signal change slope criterion, judge continuous period slope deviation trend, and obtain trend offset signal; S4: based on the trend offset signal, analyze the signal strength change of the adjacent area, calculate the corresponding relationship between the interference signal waveform and the target area signal change, determine the relay switching priority, and obtain the switching coordination parameter; S5: based on the switching coordination parameter, analyze the spatial relationship between the current relay node signal coverage range and the target attenuation area, judge whether the existing compensation behavior meets the demand, adjust the coverage parameter and issue the switching instruction, and obtain the node switching compensation instruction set. 2.The 5G signal blind area AI prediction compensation method of claim 1, wherein, The regional shielding distribution includes shielding area identification number, risk level classification, spatial mapping information, the sequence fluctuation characteristics include periodic fluctuation section, change trend identification, feature extraction code, the trend offset signal includes slope change label, user intensive characteristics, abnormal trend identification, the switching coordination parameter includes node switching serial number, switching priority category, feature matching result, and the node switching compensation instruction set includes switching control type, compensation scheduling number. 3.The 5G signal blind area AI prediction compensation method of claim 1, wherein, The acquisition step of the regional shielding distribution is specifically: S101: based on the urban area, analyze the surface reflection data frame obtained from the L-band remote sensing image, identify the area pixel with signal reflection condition, compare the pixel reflectivity with the signal strength data collected by the micro base station, and correspondingly integrate the two types of data according to the spatial coordinates to obtain the reflectivity intensity distribution characteristics; S102: based on the reflectivity intensity distribution characteristics, calculate the difference between the reflectivity and signal strength of each coordinate unit, analyze the wind speed, precipitation and building distribution parameters, optimize the collaborative processing process between the difference and environmental factors, judge the spatial position with reflection interference characteristics, and obtain the interference risk grid identification; S103: based on the interference risk grid identification, analyze the geographical spatial coordinates corresponding to the grid in the urban area, compare the risk level with the pixel connectivity characteristics in the remote sensing image, encode the combination information of space and interference level, and obtain the regional shielding distribution. 4.The 5G signal blind area AI prediction compensation method of claim 1, wherein, The acquisition step of the sequence fluctuation characteristics is specifically: S201: based on the regional shielding distribution, analyze the spatial mapping information and shielding area number, calculate the signal strength change of the corresponding spatial unit in the continuous time sequence, mine the increase and decrease state of the signal at each time through sequence difference operation, optimize the comparison of the change amplitude between each time point, and obtain the signal trend characteristics; S202: Based on the signal trend characteristics, compare the difference characteristics of each change section, optimize the timing synchronization parameters of the signal acquisition process, identify the interval of the signal strength transition between the continuous time nodes, and aggregate the transition state and spatial distribution in each change interval to obtain the fluctuation transition distribution index; S203: Based on the fluctuation transition distribution index, analyze the continuity and trend stability of the signal change trajectory, filter the abnormal fluctuation characteristics in the cycle sequence, and cross-label the abnormal section and the normal section to obtain the sequence fluctuation characteristics. 5.The 5G signal blind area AI prediction compensation method of claim 1, wherein, The acquisition step of the trend deviation signal is specifically: S301: Based on the sequence fluctuation characteristics, analyze the user connection records of the high-reflective building shielding area, calculate the frequency data of user access to the network in each shielding area, and judge the fluctuation type of user connection frequency in each interval to obtain the connection frequency change sequence; S302: Based on the connection frequency change sequence, analyze the correlation parameters of user activity density and signal strength change in each spatial unit, calculate the influence factor of user activity density on the time-varying trend of signal strength, and optimize the signal change slope criterion set based on user distribution; S303: Call the signal change slope criterion to filter the signal slope deviation parameters in the continuous period, judge whether the signal change trend meets the abnormal change judgment condition, and assign a trend deviation label to the fluctuation section that meets the condition to obtain the trend deviation signal. 6.The 5G signal blind area AI prediction compensation method of claim 1, wherein, The acquisition step of the switching coordination parameter is specifically: S401: Based on the trend deviation signal, analyze the identified spatial position, filter the mobile edge communication unit covering the position, compare the signal strength change amplitude of each communication unit in the target area and its adjacent area, judge the correlation characteristics of adjacent area signal fluctuation, and obtain the adjacent area interference characteristics; S402: Based on the adjacent area interference characteristics, calculate the change trend of the interference signal waveform of the adjacent area, analyze the corresponding relationship between the trend and the signal strength change of the target area, judge whether the interference signal trajectory and the signal fluctuation trajectory of the target area meet the relay switching condition, and obtain the relay switching factor; S403: Filter the edge communication unit with switching adaptation characteristics in the relay switching factor, analyze the ordering of the communication unit in the relay node sequence, judge the switching priority relationship between the communication unit and the target area, and obtain the switching coordination parameter. 7.The 5G signal blind area AI prediction compensation method of claim 1, wherein, The acquisition step of the node switching compensation instruction set is specifically: S501: Based on the switching coordination parameter, analyze the spatial correspondence between the node switching sequence number and the spatial correspondence, judge the overlap state of the current relay node signal coverage boundary and the geographical mapping area of the signal attenuation area, compare the interaction position of the two in the spatial structure, and obtain the spatial coverage mapping relationship; S502: Based on the spatial coverage mapping relationship, judge whether the signal compensation behavior of the relay node meets the communication demand in the signal attenuation area, compare the node signal strength change rate and the fluctuation amplitude of the adjacent area interference signal, analyze the adaptation of the current compensation configuration, and obtain the compensation adaptation identifier; S503: Based on the compensation adaptation identifier, adjust the configuration inconsistent relay node parameters, optimize the coverage angle and transmission direction of the corresponding node, complete the scheduling association of target area numbering and switching instruction according to the spatial matching result, and obtain the node switching compensation instruction set. 8.The 5G signal blind area AI prediction compensation method of claim 1, wherein, The ground reflectivity represents the reflection ability of each region of the ground to wireless signals, which is the proportion of the reflected energy of the radar or wireless signal after encountering the ground object to the incident energy, the environmental factor refers to the natural or artificial conditions that will affect the propagation of wireless signals, and the signal attenuation risk area refers to the identified geographical grid or area where signal attenuation occurs. 9.The 5G signal blind area AI prediction compensation method of claim 1, wherein, The signal strength change trend refers to the increase and decrease change law of the signal strength value in the same region within a period of time, and the signal fluctuation law refers to the change characteristics of the signal strength in the time sequence by analyzing the change amplitude. 10.A 5G signal blind area AI prediction compensation system, the system is used to implement the 5G signal blind area AI prediction compensation method according to any one of claims 1-9, characterized in that, The system comprises: The shadow identification module analyzes the ground reflectivity based on the urban area, collects the signal strength, calculates the difference between the two, combines the influence of environmental factors, identifies the signal attenuation risk area under the reflection interference, and obtains the regional shadow distribution; The fluctuation feature extraction module calculates the change trend of the signal strength of the target area based on the regional shadow distribution, compares the signal strength change amplitude of each time node, judges the signal fluctuation law, analyzes the periodic signal change trend anomaly, and obtains the sequence fluctuation feature; The user slope analysis module analyzes the user connection frequency change in the high reflection shielding area based on the sequence fluctuation feature, calculates the influence of user activity density on signal slope, sets signal change slope criterion, judges the continuous period slope deviation trend, and obtains the trend offset signal; The switching coordination module analyzes the signal strength change of the adjacent area based on the trend offset signal, calculates the corresponding relationship between the interference signal waveform and the target area signal change, determines the relay switching priority, and obtains the switching coordination parameter; The compensation control module analyzes the spatial relationship between the current relay node signal coverage range and the target attenuation area based on the switching coordination parameter, judges whether the existing compensation behavior meets the demand, adjusts the coverage parameter and issues the switching instruction, and obtains the node switching compensation instruction set.

Images