Multi-base station cooperative data transmission system

Abstract

The invention discloses a multi-base-station cooperative data transmission system, and relates to the technical field of communication, a multi-base-station digital twinborn model is constructed, a signal transmission quality allowable confidence interval is obtained by using the multi-base-station digital twinborn model, and then a candidate base station set is screened out through the signal transmission quality allowable confidence interval; predicting future position coordinates of the user by using the user track parameters, and outputting a user track prediction map; outputting a time distribution characteristic and a space distribution characteristic of the candidate base station set by using a multi-base-station digital twin model, and outputting a space-time channel quality map of the candidate base station set; and screening the candidate base station set based on the user track prediction map and the space-time channel quality map of the candidate base station set, obtaining a comprehensive score of the screened base station, and outputting a cooperative base station list and a switching strategy. Intelligent switching and efficient cooperation based on digital twin modeling and dynamic prediction are achieved, and the reliability and user experience of a communication network are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and particularly to a multi-base station cooperative data transmission system. Background Art

[0002] With the continuous development of mobile communication technologies, users have increasingly higher requirements for the stability, speed, and quality of data transmission. Multi-base station cooperative data transmission has become an important means to improve system performance, and more advanced technologies and methods are needed to achieve effective cooperation between base stations and optimize resource allocation.

[0003] The existing technologies lack accurate evaluation and comprehensive consideration of signal transmission quality when selecting base stations. At the same time, due to the inability to accurately predict user behavior, it is difficult for the system to make resource allocation and handover preparations in advance. Moreover, the existing technologies do not comprehensively analyze the spatio-temporal characteristics of channels, and the cooperation and handover between base stations are not flexible and intelligent enough.

[0004] Therefore, in view of the above problems, there is an urgent need for a multi-base station cooperative data transmission system. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technologies, the present invention provides a multi-base station cooperative data transmission system, which solves the problems of handover delay, unstable signal quality, and uneven resource allocation caused by dynamic environmental changes in traditional multi-base station cooperation.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A multi-base station cooperative data transmission system includes a digital twin modeling module, a dynamic confidence interval design module, a user behavior prediction module, a channel prediction and optimization module, and an intelligent handover and cooperation module. Among them: The digital twin modeling module is used to obtain base station geographical coordinates, transmission power, antenna parameters, and environmental topology data, and construct a multi-base station digital twin model; The dynamic confidence interval design module is used to obtain the allowable confidence interval of signal transmission quality by using the data output by the multi-base station digital twin model, and then screen out a candidate base station set through the allowable confidence interval of signal transmission quality; The user behavior prediction module is used to obtain user trajectory parameters, predict the future position coordinates of the user by using the user trajectory parameters, and then output a user trajectory prediction map; The channel prediction and optimization module is used to utilize the time distribution characteristics and space distribution characteristics of the candidate base station set output by the multi-base station digital twin model, and output a spatio-temporal channel quality map of the candidate base station set; The intelligent handover and cooperation module is used to screen the candidate base station set based on the user trajectory prediction map and the spatio-temporal channel quality map of the candidate base station set, obtain the comprehensive score of the screened base stations, and then output a cooperative base station list and a handover strategy.

[0007] Further, the confidence interval of the signal transmission quality is specifically the comprehensive evaluation confidence interval of transmission delay, path loss, and channel quality.

[0008] Further, the dynamic confidence interval design module is specifically analyzed as follows: extracting the transmission delay, path loss, and channel quality related data of each base station by using the multi-base station digital twin model, preprocessing the related data, and then determining the quality evaluation values of each base station through the transmission delay, path loss, and channel quality; identifying the confidence interval of the allowable signal transmission quality based on the sample data of the quality evaluation values, and then screening out the base stations whose quality evaluation values are within the confidence interval of the allowable signal transmission quality to form a candidate base station set.

[0009] Further, the user behavior prediction module is specifically analyzed as follows: the user trajectory parameters include the base station information and movement parameters connected by the user before. The base station information connected by the user before includes the geographical coordinates and connection time of the base station, and the movement parameters include the movement speed and movement direction; inputting the historical user trajectory parameters into the recurrent neural network, training the recurrent neural network to identify the change law of the trajectory, and then inputting the current user trajectory parameters into the trained recurrent neural network to output the position coordinates of the user at each future time point; integrating the predicted position coordinates of the user at each future time point with the corresponding time points to form complete trajectory prediction data, and then displaying and updating the trajectory prediction data in the form of a graph to output the user trajectory prediction graph.

[0010] Further, the channel prediction and optimization module is specifically analyzed as follows: obtaining the time distribution characteristic parameters and space distribution characteristic parameters of the candidate base station set by using the multi-base station digital twin model, and then identifying the time distribution characteristic values and space distribution characteristic values of each candidate base station by using the time distribution characteristic parameters and space distribution characteristic parameters respectively; combining the time distribution characteristic values and space distribution characteristic values of each candidate base station to determine the spatio-temporal channel quality evaluation values of each candidate base station, and then generating a spatio-temporal channel quality graph according to the positions of the candidate base stations and the spatio-temporal channel quality evaluation values. The spatio-temporal channel quality graph is marked with a high-quality base station area and a low-quality base station area. The high-quality base station area is specifically the base station area where the spatio-temporal channel quality evaluation value is greater than or equal to the spatio-temporal channel quality threshold, and the low-quality base station area is specifically the base station area where the spatio-temporal channel quality evaluation value is lower than the spatio-temporal channel quality threshold.

[0011] Further, the time distribution characteristic value is specifically calculated by using the STARMA model, and the space distribution characteristic value is specifically calculated by using the Kriging interpolation method.

[0012] Further, the intelligent switching and collaboration module is specifically analyzed as follows: superimpose the future position coordinate information in the user trajectory prediction map on the spatio-temporal channel quality map, and mark the base stations in the superimposed area; according to the positional relationship between the user's future position and each marked base station, as well as the distribution of high-quality base station areas and low-quality base station areas in the spatio-temporal channel quality map, select the base stations whose distance from the user's future position coordinates is less than the distance threshold and whose spatio-temporal channel quality evaluation result is a high-quality base station area as the preferred base stations; for each preferred base station, identify the load situation of the base station, and then combine the distance between the base station and the user's future position and the spatio-temporal channel quality evaluation value of the base station to determine the comprehensive score of each preferred base station, and select and form a collaborative base station list based on the high and low comprehensive scores; monitor the user's mobile state and the signal quality of the currently connected base station in real time, and then determine whether to trigger the switching of the base station connection according to the user's mobile state and the signal quality of the currently connected base station.

[0013] The present invention has the following beneficial effects: This multi-base station collaborative data transmission system reflects the actual situation of the multi-base station system by acquiring various base station and environmental data to build a model, providing an accurate basis for subsequent analysis; uses the data output by the digital twin model to determine the confidence interval allowed for signal transmission quality, and screens the candidate base station set with this, which can improve the accuracy and effectiveness of base station selection and ensure signal transmission quality; predicts the future position coordinates by analyzing the user trajectory parameters and outputs a map, which helps to plan data transmission in advance and improve the adaptability of the system to user movement; outputs the spatio-temporal channel quality map according to the spatio-temporal and spatial distribution characteristics of the candidate base station set, intuitively showing the channel quality distribution and facilitating the system to make reasonable decisions; based on the user trajectory prediction and the channel quality map for screening and scoring, outputs the collaborative base station list and the handover strategy, which can realize the intelligent collaboration and handover between base stations and improve the stability and efficiency of data transmission.

[0014] Of course, it is not necessary for any product implementing the present invention to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a structural diagram of a multi-base station collaborative data transmission system of the present invention.

[0016] Figure 2 It is a method flow chart of a multi-base station collaborative data transmission system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] In the embodiments of the present application, a multi-base station collaborative data transmission system realizes intelligent switching and efficient collaboration based on digital twin modeling and dynamic prediction, significantly improving the reliability of the communication network and the user experience.

[0018] The overall idea of the embodiments of this application is as follows: comprehensively obtain data related to base stations and the environment, construct a multi-base station digital twin model to provide accurate virtual model support for the system; based on the data output by the digital twin model, determine the allowable confidence interval of signal transmission quality, screen out the candidate base station set, and initially narrow down the range of base station selection; predict the future location of users by analyzing user trajectory parameters to provide a basis for the system to know in advance the user movement trend; use the digital twin model to analyze the spatio-temporal and spatial distribution characteristics of the candidate base station set, output the spatio-temporal channel quality map, and visually present the channel quality situation; comprehensively consider the user trajectory prediction map and the spatio-temporal channel quality map, further screen and score the candidate base station set, determine the collaborative base station list and handover strategy, and achieve intelligent collaborative data transmission of multi-base stations.

[0019] Please refer to Figure 1 、 Figure 2 An embodiment of the present invention provides a technical solution: a multi-base station collaborative data transmission system, including a digital twin modeling module, a dynamic confidence interval design module, a user behavior prediction module, a channel prediction and optimization module, and an intelligent handover and collaboration module, where: The digital twin modeling module is used to obtain base station geographical coordinates, transmission power, antenna parameters, and environmental topology data, and construct a multi-base station digital twin model; the dynamic confidence interval design module is used to obtain the allowable confidence interval of signal transmission quality by using the data output by the multi-base station digital twin model, and then screen out the candidate base station set through the allowable confidence interval of signal transmission quality; the user behavior prediction module is used to obtain user trajectory parameters, predict the future location coordinates of users by using the user trajectory parameters, and then output a user trajectory prediction map; the channel prediction and optimization module is used to utilize the time distribution characteristics and spatial distribution characteristics of the candidate base station set output by the multi-base station digital twin model to output the spatio-temporal channel quality map of the candidate base station set; the intelligent handover and collaboration module is used to screen the candidate base station set based on the user trajectory prediction map and the spatio-temporal channel quality map of the candidate base station set, and obtain the comprehensive score of the screened base stations, and then output the collaborative base station list and handover strategy.

[0020] Specifically, the allowable confidence interval of signal transmission quality is specifically the comprehensive evaluation confidence interval of transmission delay, path loss, and channel quality.

[0021] The specific analysis of the dynamic confidence interval design module is as follows: use the multi-base station digital twin model to extract data related to transmission delay, path loss, and channel quality of each base station, and preprocess the relevant data, and then determine the quality evaluation value of each base station through transmission delay, path loss, and channel quality; identify the allowable confidence interval of signal transmission quality based on the sample data of the quality evaluation value, and then screen out the base stations whose quality evaluation values are within the allowable confidence interval of signal transmission quality to form a candidate base station set.

[0022] In this implementation plan, the specific steps for constructing the multi - base - station digital twin model are as follows: Obtain relevant information such as the geographical coordinates, transmission power, antenna parameters, and environmental topology data of the base stations; Determine the architecture of the digital twin model according to the characteristics of the collected data and system requirements, for example, select appropriate mathematical models or algorithms to describe the behavior and interrelationships of the base stations; Use the collected data such as the geographical coordinates, transmission power, and antenna parameters of the base stations as the initial parameters of the model and substitute them into the model for initialization settings; Train the model using the actual operation data, and by adjusting the parameters and structure of the model, enable the model to simulate the actual operation of the multi - base - station system, continuously optimizing the performance and accuracy of the model; Use the validation dataset to verify and evaluate the trained model to ensure that the model can accurately reflect the true characteristics of the multi - base - station system and meet the design requirements.

[0023] The geographical coordinates of the base stations are obtained through the construction planning documents of the base stations, Geographic Information System (GIS) data, or Global Positioning System (GPS). When building base stations, operators will record the accurate location information of the base stations, and this information is obtained from relevant databases or documents; The transmission power is obtained from the device configuration file or management system of the base station. The transmission power of the base station is clearly specified in the device settings, and by interacting with the base - station device or accessing the relevant management interface, the transmission - power value can be obtained; The antenna parameters include the gain, radiation pattern, polarization mode, etc. of the antenna, which are provided by the antenna manufacturer and recorded in the technical specification documents of the antenna. During the construction and maintenance of the base stations, relevant technical personnel will master this antenna - parameter information, and it can also be obtained from the operator's device management system; The environmental topology data describes the environmental characteristics around the base stations, such as buildings, terrain, etc., and is obtained through Geographic Information System (GIS) data, urban planning data, or specialized environmental surveys. For example, by using remote - sensing technology, map data, and on - site surveys and other means, the environmental information around the base stations is obtained to construct an environmental topology model.

[0024] Assume that the data output by the multi - base - station digital twin model is: Transmission - delay set: , where represents the transmission delay of the th base station, is the total number of base stations; Path - loss set: , where represents the path loss of the th base station; Channel - quality set: , where represents the channel quality of the th base station.

[0025] The pre - processing of the relevant data is specifically as follows: Remove the outliers in the data. For example, for the transmission delay , if or (where is the mean value of the transmission delay, is the standard deviation of the transmission delay), then this value is regarded as an outlier and removed; the min-max normalization method is adopted to normalize the transmission delay, path loss, and channel quality data to the interval [0, 1]. Taking the transmission delay as an example: , represents the transmission delay after normalization, represents the minimum transmission delay, represents the maximum transmission delay. Similarly, for the path loss and the channel quality are normalized to obtain and , represents the path loss after normalization, represents the channel quality after normalization.

[0026] Weights , and are assigned to the transmission delay, path loss, and channel quality respectively, and satisfy . Calculate the comprehensive value of the signal transmission quality of the th base station, , where is because the smaller the transmission delay, the better, so it is negated to ensure monotonicity consistency with other indicators.

[0027] Assume that the comprehensive value follows a normal distribution, and calculate the sample mean and the sample standard deviation ; for a given confidence level (e.g., 95%), according to the properties of the normal distribution, calculate the confidence interval: ; where, is the quantile of the standard normal distribution. For a 95% confidence level, .

[0028] The transmission delay refers to the time delay experienced by the signal from the sending end to the receiving end, including the propagation time of the signal in the transmission medium, the time for the base station to process the signal, and the delay caused by network congestion and other factors. By recording the signal sending time and receiving time at the sending end and receiving end respectively, the difference between the two is calculated to obtain the transmission delay. In an actual system, a network test tool or a timestamp is added to the communication protocol to measure the transmission delay.

[0029] Path loss refers to the attenuation of signal strength during signal transmission due to factors such as propagation distance, signal reflection, scattering, and diffraction. It reflects the energy loss of the signal on the transmission path. Path loss is calculated by measuring the transmit power at the transmitter and the received signal power at the receiver, and taking the difference between the two. In actual measurements, devices such as signal strength testers are used to measure at different locations, or the path loss is calculated through theoretical models based on parameters such as the distance between the base station and the receiving point and environmental characteristics.

[0030] Channel quality is used to describe the quality of a communication channel in transmitting signals, taking into account factors such as the bit error rate, signal-to-noise ratio, and fading of the signal. Channel quality directly affects the reliability and efficiency of data transmission. The channel quality is evaluated by the receiver analyzing and processing the received signal. For example, by measuring the signal-to-noise ratio (SNR), bit error rate (BER) of the signal or using channel estimation algorithms to obtain relevant parameters of the channel, and then evaluating the channel quality.

[0031] By comprehensively considering transmission delay, path loss, and channel quality to determine the quality evaluation value of the base station, which measures the signal transmission quality of the base station and avoids the one-sidedness of single-index evaluation; screening the candidate base station set based on the confidence interval allowed for signal transmission quality, excluding base stations that do not meet the requirements, improving the efficiency and accuracy of base station selection, and providing a high-quality base station selection range for subsequent data transmission.

[0032] Specifically, the user behavior prediction module is specifically analyzed as follows: The user trajectory parameters include the base station information and movement parameters that the user has connected to before. The base station information that the user has connected to before includes the geographical coordinates and connection time of the base station, and the movement parameters include the movement speed and direction; using the historical user trajectory parameters as input to the recurrent neural network, training the recurrent neural network to identify the change rules of the trajectory, and then inputting the current user trajectory parameters into the trained recurrent neural network to output the position coordinates of the user at each future time point; integrating the predicted position coordinates of the user at each future time point with the corresponding time points to form complete trajectory prediction data, and then displaying and updating the trajectory prediction data in the form of a graph to output the user trajectory prediction graph.

[0033] In this implementation plan, the geographical coordinates and connection time of the base station are obtained from the base station management system or relevant database of the communication network. When the user connects to the base station, the base station will record the connection time of the user and its own geographical coordinate information, and these data can be read and utilized by the system.

[0034] The moving speed and direction are measured by sensors built into mobile devices such as mobile phones, such as accelerometers and gyroscopes, to measure the motion state of the device, and then the moving speed and direction are calculated. It is also possible to estimate the moving speed and direction based on the change in the positions of the base stations connected by the user at different times. For example, by calculating the distance and direction between two base stations connected at adjacent time points and combining the time interval, the average moving speed and direction can be obtained.

[0035] The specific steps to identify the trajectory change pattern are as follows: Preprocess the collected historical trajectory parameters, including operations such as data cleaning and normalization, to ensure the quality and consistency of the data and facilitate the processing of the neural network. For example, normalize data such as the moving speed and the geographical coordinates of the base stations, map them to a specific numerical range, and prevent certain features from affecting the training effect of the model due to overly large or small numerical values; Initialize the recurrent neural network, set the structure and parameters of the network, including the number of neurons, the number of layers, activation functions, etc. For example, select activation functions such as ReLU or Sigmoid functions to introduce non-linearity and improve the expressive ability of the model. At the same time, randomly initialize the weights and biases of the model to provide initial values for subsequent training; Input the preprocessed historical trajectory parameters into the recurrent neural network, and perform forward propagation calculations according to the structure of the network and the set algorithm. At each time step, the network calculates a new hidden state based on the current input and the previous hidden state and gradually passes it forward, finally obtaining the output result, that is, the predicted value of the user's future position coordinates; Compare the prediction result with the actual user position coordinates, and measure the difference between the predicted value and the true value by defining an appropriate loss function. Common loss functions include mean squared error (MSE), mean absolute error (MAE), etc.; According to the calculation result of the loss function, calculate the gradient of each parameter with respect to the loss function through the backpropagation algorithm, and use optimization algorithms such as stochastic gradient descent, Adagrad, Adadelta, etc. to update the weights and biases of the model, so that the loss function gradually decreases. In each iteration, adjust the parameters in the opposite direction of the gradient to gradually optimize the performance of the model and enable it to better fit the change pattern in the historical trajectory data; During the training process, regularly evaluate the model using the validation dataset, observe the change trend of the loss function and the accuracy of the prediction result. If the loss function no longer decreases significantly or reaches the preset stopping conditions such as the number of iterations, accuracy, etc., it is considered that the model training is completed. Otherwise, continue the iterative training, continuously adjust the weights and biases of the model until the stopping conditions are met; Through continuous iterative optimization, the model gradually learns the change pattern of the user's trajectory, so as to make predictions on new input data.

[0036] By considering the information of the base stations previously connected by the user and the movement parameters, the recurrent neural network can learn the changing pattern of the user's trajectory, thereby predicting the future position coordinates of the user, providing a reliable basis for subsequent base station handover and cooperation; integrating the predicted position coordinates with the time information and presenting the update in the form of a map can reflect the user's dynamic trajectory in real time, facilitating the system to make corresponding adjustments and decisions in a timely manner, and improving the real-time performance and adaptability of the system.

[0037] Specifically, the channel prediction and optimization module is specifically analyzed as follows: using the multi-base station digital twin model to obtain the time distribution characteristic parameters and space distribution characteristic parameters of the candidate base station set, and then using the time distribution characteristic parameters and space distribution characteristic parameters to identify the time distribution characteristic values and space distribution characteristic values of each candidate base station respectively; combining the time distribution characteristic values and space distribution characteristic values of each candidate base station to determine the spatio-temporal channel quality evaluation values of each candidate base station, and then generating a spatio-temporal channel quality map according to the positions of the candidate base stations and the spatio-temporal channel quality evaluation values. The spatio-temporal channel quality map is marked with high-quality base station areas and low-quality base station areas. The high-quality base station area is specifically the base station area where the spatio-temporal channel quality evaluation value is greater than or equal to the spatio-temporal channel quality threshold, and the low-quality base station area is specifically the base station area where the spatio-temporal channel quality evaluation value is lower than the spatio-temporal channel quality threshold.

[0038] In this implementation scheme, the time distribution characteristic value is specifically calculated using the STARMA model, and the space distribution characteristic value is specifically calculated using the Kriging interpolation method.

[0039] For the time distribution characteristic value: calculated using the STARMA model, assuming that the observed spatio-temporal sequence data is , where represents the spatial position, represents the time, and the general form of the STARMA model is expressed as: , where and are polynomials about the time lag operator and the space lag operator respectively, is a white noise sequence. By fitting and estimating the historical spatio-temporal data, the parameters of the model are obtained, and then the model is used to predict the future time distribution characteristic value. The specific calculation process involves the estimation and solution of the model parameters, using methods such as maximum likelihood estimation.

[0040] For the space distribution characteristic value: calculated using the Kriging interpolation method, collecting the space distribution characteristic parameter values (such as signal strength, propagation loss) of the known position points. Let these known points be , and the corresponding parameter values be . For the position point to be estimated , the numerical values of its spatial distribution characteristics can be calculated by the Kriging interpolation formula: , where are the weight coefficients obtained by solving the Kriging equations, represents the number of the spatial distribution characteristic parameter values at the known position points, represents the total number of the spatial distribution characteristics at the known position points. This system of equations takes into account the spatial correlation between the known points and the distance relationship between the point to be estimated and the known points.

[0041] The time distribution characteristic parameters include parameters related to the time dimension such as the intensity change of the base station signal at different time points, the change of the signal propagation delay over time, and the time correlation of the channel fading. These parameters reflect the dynamic change characteristics of the channel quality in time and space.

[0042] The spatial distribution characteristic parameters include parameters only related to the spatial position such as the distance between the base station and the user, the propagation loss of the signal in space, the signal intensity difference at different positions, and the spatial correlation of the signal, which describe the distribution characteristics of the channel quality in space.

[0043] The time distribution characteristic parameters are obtained by combining the digital twin model of the base station with the actual signal monitoring data. The digital twin model simulates the propagation of the base station signal under different time and space conditions, and at the same time combines the real-time signal data collected by the sensors or monitoring devices actually deployed in the network to calibrate and optimize the model, so as to obtain accurate time distribution characteristic parameters.

[0044] The spatial distribution characteristic parameters are directly obtained by measuring the distance between the base station and the user equipment and the signal intensity. The digital twin model can also be used to simulate the propagation of the signal in space, considering the influence of environmental factors such as terrain and buildings on the signal propagation, so as to obtain the spatial distribution characteristic parameters.

[0045] The specific steps to generate the spatio-temporal channel quality map according to the positions of the candidate base stations and the spatio-temporal channel quality evaluation values are as follows: According to the position information of each candidate base station, mark it on the geographical space, compare the calculated spatio-temporal channel quality evaluation value with the preset spatio-temporal channel quality threshold, determine the area to which each base station belongs (high-quality base station area or low-quality base station area), and distinguish the high-quality base station area and the low-quality base station area with different colors or marks to form the spatio-temporal channel quality map.

[0046] The method for obtaining the spatio-temporal channel quality threshold is as follows: Based on past communication system operation experience, experimental data in similar scenarios, or industry standards, set an initial spatio-temporal channel quality threshold. For example, in certain specific communication environments, through long-term testing and evaluation, it is found that when the channel quality evaluation value reaches a certain specific range, the system can ensure good communication quality and stability, so the boundary values of this range are used as the initial threshold; It is also possible to obtain a theoretical threshold through mathematical calculations based on communication theory and channel models. For example, according to relevant theories such as Shannon's formula, combined with parameters such as the system's bandwidth and noise level, calculate the channel quality threshold that can meet certain data transmission rate and bit error rate requirements; An adaptive method can also be adopted to dynamically adjust the spatio-temporal channel quality threshold according to the current operating state and performance indicators of the system. For example, the system can monitor indicators such as the success rate of data transmission, delay, and packet loss rate in real time. When these indicators show abnormal changes, the threshold is adjusted accordingly to optimize the system performance. The specific adjustment algorithm is based on feedback control theory and adjusts according to the deviation between the performance indicator and the target value; Machine learning algorithms such as clustering algorithms and decision tree algorithms can also be used to analyze and process a large amount of channel quality data, automatically learn and determine appropriate thresholds. For example, use the clustering algorithm to divide the channel quality data into different categories, and then determine the threshold according to the boundary between the categories, or use the decision tree algorithm to divide the channel quality levels according to different features and conditions, so as to determine the corresponding threshold.

[0047] By comprehensively considering the time distribution characteristics and spatial distribution characteristics of the candidate base station set, the channel quality of each candidate base station can be evaluated, providing more accurate channel state information for users; The generation of the spatio-temporal channel quality map helps the system to perform reasonable resource allocation according to the channel quality of the base stations, allocate data transmission tasks to high-quality base station areas, and improve the overall performance and data transmission efficiency of the system; Considering the time distribution characteristics enables the system to adapt to the dynamic changes of the network environment, such as the impact of factors such as user movement and signal interference on channel quality, thereby improving the robustness and adaptability of the system.

[0048] Specifically, the intelligent switching and collaboration module is specifically analyzed as follows: superimpose the future position coordinate information in the user trajectory prediction map on the spatio-temporal channel quality map, and mark the base stations in the superimposed area; according to the positional relationship between the user's future position and each marked base station, as well as the distribution of high-quality base station areas and low-quality base station areas in the spatio-temporal channel quality map, select the base stations whose distance from the user's future position coordinate is less than the distance threshold and whose spatio-temporal channel quality evaluation result is in the high-quality base station area as the preferred base stations; for each preferred base station, identify the load situation of the base station, and then combine the distance between the base station and the user's future position and the spatio-temporal channel quality evaluation value of the base station to determine the comprehensive score of each preferred base station, and select and form a collaborative base station list based on the high and low comprehensive scores; monitor the user's mobile state and the signal quality of the currently connected base station in real time, and then determine whether to trigger a handover of the base station connection according to the user's mobile state and the signal quality of the currently connected base station.

[0049] In this implementation plan, the specific steps for superimposing the user trajectory prediction map and the spatio-temporal channel quality map and marking the base stations are as follows: ensure that the user trajectory prediction map contains future position coordinate information, the spatio-temporal channel quality map contains base station positions and channel quality evaluation information, and the two have a unified geographic coordinate system; traverse each future position coordinate in the user trajectory prediction map, and check whether the coordinate falls within the coverage area of a certain base station in the spatio-temporal channel quality map, and the coverage area can be determined according to information such as the signal propagation range and geographic coordinates of the base station; if the future position coordinate falls within the coverage area of a certain base station, mark the base station as the base station in the superimposed area, and highlight these base stations in the map with a specific color, symbol or identifier.

[0050] The distance threshold is set as follows: according to past communication network operation experience and test data, set a fixed distance threshold. For example, in densely populated urban areas, the distance threshold can be set to several hundred meters, and in suburban or rural areas, the distance threshold can be appropriately increased; the distance threshold can also be dynamically adjusted according to factors such as the user's moving speed and communication service type. For example, for users moving at high speed, the distance threshold is appropriately increased to prepare for base station handover in advance, and for communication services with high real-time requirements, the distance threshold is reduced to ensure signal quality; the distance threshold can also be adjusted by continuously monitoring network performance indicators such as communication success rate and packet loss rate to optimize network performance, and machine learning algorithms can be used to automatically optimize the distance threshold according to historical data and real-time monitoring results.

[0051] The method for identifying the load situation of a base station is as follows: The base station's own management system records and statistics the load information of the base station in real time, such as the number of currently connected users, data transmission rate, bandwidth occupancy, etc. By communicating with the base station management system, these load data can be directly obtained; traffic monitoring devices can also be deployed in the network to monitor the data traffic between the base station and users in real time, and estimate the load situation of the base station according to the traffic volume; or collect historical base station load data and related influencing factors (such as time, location, user behavior, etc.), use machine learning algorithms to establish a load prediction model, and predict the load situation of the base station according to the current environmental information and real-time data.

[0052] The specific steps for selecting and forming a cooperative base station list based on the comprehensive score are as follows: Sort the preferred base stations in descending order according to the comprehensive score; according to the system requirements and set rules, select a number of base stations with higher rankings to form a cooperative base station list.

[0053] The method for monitoring the user's movement state and the signal quality of the currently connected base station is as follows: The user equipment (such as a mobile phone) monitors its own movement state (such as acceleration, speed, direction, etc.) in real time and feeds this information back to the base station and the communication system. At the same time, the user equipment can also measure the signal strength and signal quality (such as signal-to-noise ratio, bit error rate, etc.) of the currently connected base station and upload them to the system; the base station evaluates the user equipment's movement state and signal quality by monitoring the communication parameters between the base station and the user equipment, such as signal strength, transmission rate, packet loss rate, etc.

[0054] The specific steps for determining whether to trigger a handover of the base station connection according to the user's movement state and the signal quality of the currently connected base station are as follows: Preset signal quality thresholds (such as signal strength threshold, signal-to-noise ratio threshold, etc.) and movement state thresholds (such as movement speed threshold); obtain the user's movement state and the signal quality of the currently connected base station in real time; if the signal quality of the currently connected base station is lower than the signal quality threshold and the user's movement speed exceeds the movement state threshold, trigger a base station handover; if the signal quality of the currently connected base station continues to be lower than the signal quality threshold within a preset time, even if the user's movement speed does not exceed the threshold, trigger a base station handover; if the user moves to a position far from the currently connected base station and there is a base station with a higher comprehensive score in the cooperative base station list, trigger a base station handover.

[0055] By combining user trajectory prediction and spatio-temporal channel quality maps, high-quality base stations suitable for the user's future location are selected to ensure that the user maintains a communication connection at all times during movement, reduce signal interruptions and interference, and improve communication stability and data transmission rate; the selected cooperative base stations comprehensively consider distance, channel quality, and load conditions, which helps to allocate base station resources, avoid overloading of some base stations, and improve the resource utilization rate of the entire communication network; real-time monitoring of the user's movement status and the signal quality of the current base station, and determining whether to switch base stations accordingly, can achieve intelligent and timely base station switching, avoid communication failures caused by signal quality degradation, and improve the user's communication experience.

[0056] In summary, the present application has at least the following effects: Based on the geographical coordinates, transmission power, antenna parameters, and environmental topology data of the base station, a virtual model is constructed to simulate the signal propagation characteristics in real time, create a "digital twin" for the base station, and predict signal coverage blind spots or interference hotspots in advance; by dynamically screening the candidate base station set through the model output data, it avoids misjudgment or missed judgment caused by fixed thresholds, adapts to the dynamic impact of the complex environment on the signal, and improves the robustness of the handover decision; uses user trajectory parameters to predict the future location and generate a trajectory prediction map to provide forward-looking guidance for base station handover; combines the time distribution characteristics of candidate base stations to generate a channel quality map, quantifies signal strength, delay and other indicators between different base stations and users, provides data support for base station selection, and preferentially selects the base station with the best channel quality to improve the user experience; based on user trajectory prediction and channel quality map, multi-dimensional scoring of candidate base stations is performed, and a list of cooperative base stations and handover strategies are output to reduce the handover interruption time and avoid video stuttering or call interruption; through the digital twin model and dynamic confidence interval, the system can real-time perceive environmental changes and quickly adjust the base station cooperation strategy to avoid over-occupying high-load base stations.

[0057] Those skilled in the art should understand that the embodiments of the present invention can be provided as a system. Therefore, the present invention can be in the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0058] The present invention is described with reference to the structural diagrams of the systems according to the embodiments of the present invention. It should be understood that the combination of each structure in the structural diagram can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in each structure of the structural diagram.

[0059] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to work in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in each structure of the structural diagram.

[0060] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable apparatus to generate a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in each structure of the structural diagram.

[0061] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0062] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A multi-base station cooperative data transmission system, characterized in that: It includes digital twin modeling module, dynamic confidence interval design module, user behavior prediction module, channel prediction and optimization module, and intelligent switching and collaboration module, among which: The digital twin modeling module is used to obtain the geographic coordinates, transmission power, antenna parameters and environmental topology data of the base station to build a digital twin model of multiple base stations; The dynamic confidence interval design module is used to obtain the allowable confidence interval of signal transmission quality by using the output data of the multi-base station digital twin model, and then screen out the candidate base station set by the allowable confidence interval of signal transmission quality; The user behavior prediction module is used to obtain user trajectory parameters, use the user trajectory parameters to predict the user's future location coordinates, and then output a user trajectory prediction map; The channel prediction and optimization module is used to output the time distribution characteristics and spatial distribution characteristics of the candidate base station set by using the multi-base station digital twin model, and output the spatiotemporal channel quality spectrum of the candidate base station set; The intelligent switching and collaboration module is used to screen the candidate base station set based on the user trajectory prediction map and the spatiotemporal channel quality map of the candidate base station set, obtain the comprehensive scores of the screened base stations, and then output the collaborative base station list and the switching strategy.

2. A multi-base station cooperative data transmission system according to claim 1, characterized in that: The signal transmission quality allowable confidence interval is specifically a comprehensive evaluation confidence interval of transmission delay, path loss and channel quality.

3. A multi-base station cooperative data transmission system according to claim 2, characterized in that: The dynamic confidence interval design module is specifically analyzed as follows: The multi-base station digital twin model is used to extract the transmission delay, path loss and channel quality related data of each base station, and the relevant data is preprocessed to determine the quality assessment value of each base station based on the transmission delay, path loss and channel quality; Based on the sample data of the quality evaluation value, the allowed confidence interval of the signal transmission quality is identified, and then the base stations whose quality evaluation values ​​are within the allowed confidence interval of the signal transmission quality are screened out to form a set of candidate base stations.

4. The multi-base station cooperative data transmission system according to claim 1, characterized in that: The user behavior prediction module is specifically analyzed as follows: the user trajectory parameters include the base station information and movement parameters previously connected by the user, the base station information previously connected by the user includes the geographical coordinates of the base station and the connection time, and the movement parameters include the movement speed and the movement direction; The historical user trajectory parameters are input into the recurrent neural network, the recurrent neural network is trained, and the changing rules of the trajectory are identified. Then the current user trajectory parameters are input into the trained recurrent neural network, and the position coordinates of the user at each future time point are output; The predicted position coordinates of the user at each future time point are integrated with the corresponding time point to form complete trajectory prediction data, which is then displayed and updated in the form of a graph to output the user trajectory prediction graph.

5. The multi-base station cooperative data transmission system according to claim 1, characterized in that: The channel prediction and optimization module is specifically analyzed as follows: using the multi-base station digital twin model to obtain the time distribution characteristic parameters and space distribution characteristic parameters of the candidate base station set, and then using the time distribution characteristic parameters and space distribution characteristic parameters to respectively identify the time distribution characteristic values ​​and space distribution characteristic values ​​of each candidate base station; Combine the time distribution characteristic value and the spatial distribution characteristic value of each candidate base station to determine the space-time channel quality evaluation value of each candidate base station, and then generate a space-time channel quality map based on the position and space-time channel quality evaluation value of the candidate base station. The space-time channel quality map includes high-quality base station areas and low-quality base station areas. The high-quality base station area is specifically a base station area whose space-time channel quality evaluation value is greater than or equal to the space-time channel quality threshold, and the low-quality base station area is specifically a base station area whose space-time channel quality evaluation value is lower than the space-time channel quality threshold.

6. A multi-base station cooperative data transmission system according to claim 5, characterized in that: The temporal distribution characteristic value is specifically obtained by calculation using the STARMA model, and the spatial distribution characteristic value is specifically obtained by calculation using the Kriging interpolation method.

7. A multi-base station cooperative data transmission system according to claim 5, characterized in that: The intelligent switching and collaboration module is specifically analyzed as follows: superimposing the future position coordinate information in the user trajectory prediction map with the spatiotemporal channel quality map, and marking the base stations in the superimposed area; According to the positional relationship between the user's future location and each marked base station, and the distribution of high-quality base station areas and low-quality base station areas in the spatiotemporal channel quality map, base stations whose distance to the user's future location coordinates is less than a distance threshold and whose spatiotemporal channel quality assessment results are high-quality base station areas are selected as preferred base stations; For each preferred base station, identify the load of the base station, and then determine the comprehensive score of each preferred base station in combination with the distance between the base station and the future location of the user and the spatiotemporal channel quality assessment value of the base station, and select and form a list of cooperative base stations based on the comprehensive score; Monitor the user's mobility status and the signal quality of the currently connected base station in real time, and then determine whether to trigger the switching of base station connection based on the user's mobility status and the signal quality of the currently connected base station.

Images