6g communication signal quality optimization method, device and medium based on intelligent transportation

Abstract

The application discloses a 6G communication signal quality optimization method and device based on intelligent traffic, and a medium, relates to the field of 6G terahertz communication technology, and can identify environment-related interference that cannot be detected by traditional methods through joint analysis of meteorological data and communication propagation path data, can track the change of interference relationship in real time, can extract the space-time characteristics of interference through the time delay extension and Doppler frequency shift of channel attenuation data, optimize the space-time joint prediction performance, model real-time space topology and time evolution based on space-time graph neural network, enhance the prediction accuracy of the model, generate an optimization strategy through an optimization strategy, can support multi-objective optimization, and can monitor the optimization result in real time during the optimization process and feed back to the model, quickly correct the model deviation, generate an interference optimization strategy, improve the comprehensiveness and accuracy of interference optimization, solve the problems of poor communication signal quality caused by static modeling, neglecting meteorology, prediction lag and strategy rigidity, and improve the communication signal quality.

Description

Technical Field

[0001] This invention relates to the field of 6G terahertz communication technology, specifically to a method, device, and medium for optimizing 6G communication signal quality based on intelligent transportation. Background Technology

[0002] For intelligent transportation infrastructure (such as traffic signal control systems, roadside units, high-definition cameras and video surveillance systems, sensor networks, etc.), using 6G terahertz communication links to communicate with vehicles and monitoring centers can achieve real-time transmission of traffic flow data, control commands, and safety warnings. However, the application of 6G communication technology in intelligent transportation infrastructure faces challenges from various weather interferences: rain, fog, and snow can increase signal attenuation or communication link interruption, leading to decreased reliability and communication quality; weather changes can cause changes in atmospheric refractive index, resulting in changes in signal propagation paths and increased multipath effects.

[0003] Therefore, in order to improve the communication quality of 6G terahertz communication in intelligent transportation infrastructure under extreme weather conditions, interference management is required. Existing technologies mainly rely on static interference modeling and rule-driven optimization strategies, which have significant limitations.

[0004] In terms of interference detection and modeling, traditional methods (such as spectrum sensing and static interference matrix) can only reflect the current interference state and cannot predict dynamic changes, and completely ignore the impact of meteorological factors (such as rainfall attenuation and atmospheric refraction) on wireless propagation.

[0005] Interference prediction techniques often employ single-dimensional time series analysis (such as ARIMA, LSTM) or static graph neural networks (GCN / GAT), lacking the ability to model spatiotemporal joint features, resulting in insufficient prediction accuracy in mobile scenarios (such as vehicle-to-everything (V2X) and drones).

[0006] Optimization strategy generation is usually based on fixed rules (such as threshold-triggered power adjustment) or offline optimization algorithms, which makes it difficult to adapt to sudden interference (such as signal fading caused by thunderstorms) and differentiated service requirements (QoS differences between eMBB / URLLC).

[0007] Meteorological data applications are limited to independent compensation (such as the Rain Attenuation model for satellite links) and are not linked with interference management in a closed loop, resulting in delayed response.

[0008] Therefore, the shortcomings of existing solutions can be summarized as follows: lack of environmental awareness, insufficient dynamic topology adaptability, rigid strategies, and lack of feedback calibration in open-loop control. Specifically, in 6G millimeter-wave communication, traditional methods cannot predict the attenuation of signals due to weather changes, requiring manual intervention to adjust power, resulting in a high call drop rate; while static graph models are difficult to track real-time topology changes in UAV swarms, leading to interference coordination failure. Summary of the Invention

[0009] The technical problem this invention aims to solve is that static modeling, neglecting meteorological conditions, predictive lag, and rigid strategies lead to poor communication signal quality. The goal is to provide a method, device, and medium for optimizing 6G communication signal quality based on intelligent transportation. By jointly analyzing meteorological data and communication propagation path data, it identifies environmentally relevant interference that traditional methods cannot detect, and can track changes in interference relationships in real time. Through delay spread and Doppler shift of channel attenuation data, it extracts the spatiotemporal characteristics of interference, optimizes the spatiotemporal joint prediction performance, and enhances model prediction accuracy by jointly modeling real-time spatial topology and temporal evolution based on spatiotemporal graph neural networks. Through optimization strategy generation, it supports multi-objective optimization and monitors optimization results in real time during the optimization process, feeding them back to the model to quickly correct model biases and generate interference optimization strategies. This improves the comprehensiveness and accuracy of interference optimization, thereby enhancing communication signal quality.

[0010] This invention is achieved through the following technical solution:

[0011] The first aspect of this invention provides a method for optimizing the quality of 6G communication signals based on intelligent transportation, comprising the following specific steps:

[0012] Real-time collection of meteorological data and communication propagation path data;

[0013] Meteorological attenuation is calculated based on meteorological data, and channel attenuation data is obtained by combining it with communication propagation path data.

[0014] Calculate the delay spread and Doppler shift of channel attenuation data to extract the spatiotemporal features of interference;

[0015] Based on the spatiotemporal characteristics of interference, the interference period is identified, and a dynamic interference relationship topology is generated.

[0016] An interference prediction model is constructed based on dynamic interference relationship topology.

[0017] Based on the output of the interference prediction model, the communication propagation path data is optimized and adjusted to generate an interference optimization strategy.

[0018] The system monitors the data quality of the optimized communication propagation path in real time, and re-inputs the monitoring results into the interference prediction model through closed-loop feedback. The model is then compared with the dynamic threshold that adapts to business requirements, triggering the strategy engine to select the optimal strategy from the interference optimization strategy library.

[0019] Furthermore, the real-time meteorological data includes rainfall intensity and temperature and humidity; the real-time communication propagation path data includes channel state information, path loss, multipath delay spread, and Doppler shift.

[0020] Furthermore, the step of calculating meteorological-related attenuation based on meteorological data and combining it with communication propagation path data to obtain channel attenuation data specifically includes:

[0021] The base attenuation is determined based on the communication distance and signal frequency;

[0022] Based on the base attenuation, calculate free space loss, atmospheric absorption attenuation, precipitation attenuation, and cloud and fog attenuation;

[0023] By integrating free space loss, atmospheric absorption attenuation, precipitation attenuation, and cloud and fog attenuation, and combining them with segmented correction of the propagation path, meteorological-related attenuation is obtained.

[0024] Non-meteorological attenuation is obtained from communication propagation path data. Meteorological attenuation and non-meteorological attenuation are then fused to obtain channel attenuation data.

[0025] Furthermore, the calculation of the delay spread and Doppler shift of the channel attenuation data to extract the spatiotemporal features of interference specifically includes:

[0026] Obtain the channel impulse response containing channel attenuation data;

[0027] Delay spread and Doppler shift are calculated based on channel impulse response;

[0028] Extracting temporal features based on time delay extension;

[0029] Frequency domain features are extracted based on Doppler frequency shift;

[0030] Spatial features are extracted based on the spatial distribution of interference sources estimated by direction of arrival.

[0031] Furthermore, the step of identifying the interference period and generating a dynamic interference relationship topology based on the spatiotemporal characteristics of the interference specifically includes:

[0032] By analyzing the spectrum of channel attenuation data using Fourier transform, periodic components in the spectrum are identified, and the spectral interference period is obtained.

[0033] Based on the analysis of the time-domain characteristics of channel attenuation data using short-time Fourier transform, periodic patterns in the time domain are identified, and the time-domain interference period is obtained.

[0034] Based on the spectral interference period, the temporal interference period, and the spatiotemporal characteristics of interference, an initial topology graph of the communication network is constructed, where nodes represent communication nodes and edges represent communication interference between communication nodes.

[0035] The extracted spatiotemporal features of interference are embedded as node features into the initial topology graph, and time encoding is used to embed time information into the node features to generate a dynamic interference relationship topology.

[0036] Furthermore, the construction of the interference prediction model based on the dynamic interference relationship topology specifically includes:

[0037] The dynamic interference relationship topology is input into a spatiotemporal graph neural network model, which includes a graph convolution module, a temporal prediction module, and a joint prediction layer.

[0038] The graph convolution module is used to aggregate interference information from neighboring nodes;

[0039] The timing prediction module captures the timing dependency of node interference;

[0040] The joint prediction layer is used to stitch together spatiotemporal features and output prediction results through a fully connected layer. The prediction results include interference intensity, interference hotspots, or key interference paths.

[0041] Furthermore, the interference optimization strategy specifically includes:

[0042] Based on the interference intensity, the transmission power of the communication node is adjusted through a dynamic power control algorithm to reduce the signal coverage strength in high interference areas;

[0043] Based on the interference hotspot areas, channel allocation algorithms or frequency band switching strategies are used to switch the interfered nodes to low-interference channels or frequency bands.

[0044] Based on the key interference paths, optimize the data routing strategy and select alternative paths to avoid high-interference links.

[0045] Furthermore, the process of selecting the optimal strategy specifically includes:

[0046] The data quality of the optimized communication propagation path is monitored in real time. The monitoring results are input into the interference prediction model through a closed-loop feedback mechanism to calculate the prediction error value and trigger model retraining when the error exceeds the threshold.

[0047] The predicted error value is compared with a dynamically adjusted business demand threshold, which is automatically adapted according to the real-time business type.

[0048] When the prediction error exceeds the dynamic threshold, the policy engine is triggered to select the optimal policy from the interference optimization policy library, which includes: a power control policy sub-library, a channel switching policy sub-library, a routing optimization policy sub-library, and a load balancing policy sub-library.

[0049] The selected strategy is executed and continuously monitored to form an adaptive optimization closed loop.

[0050] A second aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a 6G communication signal quality optimization method based on intelligent transportation.

[0051] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for optimizing the quality of 6G communication signals based on intelligent transportation.

[0052] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0053] Traditional interference analysis ignores the impact of meteorological factors on wireless propagation, leading to prediction bias. By fusing meteorological and communication data to establish a cross-domain interference correlation model, the transmit power and beam direction can be dynamically adjusted based on real-time meteorological data. This effectively compensates for signal attenuation, reduces the impact of multipath effects, and significantly improves the reliability of communication links. Mobility and time-varying weather cause rapidly changing interference relationships, rendering static topology models ineffective. A dynamic interference relationship topology is constructed based on a spatiotemporal graph neural network, updating the weights of nodes and edges in real time to achieve real-time representation of the dynamic interference topology. Open-loop optimization strategies are difficult to adapt to sudden interference (such as signal fading caused by thunderstorms). Closed-loop feedback re-inputs the optimized network state into the model to achieve dynamic policy calibration, forming a closed-loop optimization of interference prediction and policy generation. Due to the large differences in interference tolerance among different services, fixed threshold strategies are not applicable. The interference threshold is dynamically adjusted according to service requirements, and the optimal solution is selected from the policy library through reinforcement learning. This invention enables adaptive selection of multi-objective strategies; furthermore, through an innovative architecture of "data fusion-dynamic modeling-closed-loop optimization," it significantly improves interference management capabilities in complex environments, laying the foundation for 6G intelligent anti-interference communication; the 6G communication signal quality optimization method based on intelligent transportation can significantly improve the reliability, signal coverage, and data transmission efficiency of the communication system, while optimizing traffic management and services, reducing energy consumption and costs, and enhancing emergency response capabilities, providing strong support for the efficient operation of intelligent transportation systems. Attached Figure Description

[0054] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0055] Figure 1 This is the signal quality optimization process in an embodiment of the present invention. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0057] As one possible implementation method, such as Figure 1 As shown, this embodiment provides a 6G communication signal quality optimization method based on intelligent transportation, including the following specific steps: real-time collection of meteorological data and communication propagation path data; calculation of meteorological-related attenuation based on meteorological data, and obtaining channel attenuation data by combining communication propagation path data; the calculation of channel attenuation data is used to provide basic data for subsequent interference identification and optimization; calculation of delay spread and Doppler shift of channel attenuation data, extraction of spatiotemporal interference features, analysis of interference characteristics based on spatiotemporal features, and provision of basis for identifying interference cycles and generating dynamic interference relationship topology; identification of interference cycles and generation of dynamic interference relationship topology based on interference spatiotemporal features; construction of an interference prediction model based on the dynamic interference relationship topology, the interference model is used to predict future interference situations, and provides directional guidance for the optimization and adjustment of communication propagation path data; optimization and adjustment of communication propagation path data based on the output of the interference prediction model, generating interference optimization strategies. When optimizing and adjusting communication propagation path data, the interference prediction model is used to predict potential interference and adjust propagation path parameters in advance, reducing the communication interference caused by traditional passive response. In case of signal interruption or quality degradation, the system monitors the data quality of the optimized communication propagation path in real time. The monitoring results are then re-inputted into the interference prediction model through closed-loop feedback. This model is compared with a dynamic threshold that adapts to business needs, triggering the strategy engine to select the optimal strategy from the interference optimization strategy library. This closed-loop feedback forms a "prediction-optimization-verification" loop, significantly shortening the network anomaly response time. Simultaneously, the closed-loop feedback mechanism can identify prediction model deviations, triggering strategy reselection or manual intervention to reduce systemic risks. It automatically adjusts performance thresholds according to business needs, ensuring that the optimized strategy matches the business scenario and avoiding over-optimization or under-optimization issues of static strategies. This enables on-demand allocation of resources such as spectrum and power. By training the model with historical optimization results and feedback data, the strategy library continuously expands the set of optimal solutions. The strategy engine can comprehensively consider indicators such as latency, bandwidth, and energy consumption to select the Pareto optimal solution. In summary, when applied to 5G / 6G mobile networks, this embodiment can dynamically avoid neighboring cell interference and improve edge user rates. When applied to the Industrial Internet of Things, it can ensure low-latency and high-reliability transmission of critical control commands.

[0058] In some possible implementations, weather changes significantly impact the channel during 6G communication data propagation. Therefore, it's necessary to consider the quantitative impact of weather on the channel. Rainfall intensity leads to channel path loss, and temperature and humidity cause atmospheric absorption attenuation. The loss is typically updated dynamically using the ITU-R P.838 rain attenuation model, while the ITU-R P.676 model calculates oxygen molecule and water vapor absorption losses, particularly affecting millimeter-wave high-frequency bands. Adding weather correction terms to real-time path loss data yields an environmentally perceptible equivalent path loss. Furthermore, heavy rainfall can increase the reflectivity of scatterers; sudden increases in delay spread can identify scattering interference caused by rain and fog. Therefore, acquiring both meteorological and communication propagation path data is crucial to address these impacts. Thus, real-time meteorological data includes rainfall intensity and temperature / humidity; real-time communication propagation path data includes channel state information, path loss, multipath delay spread, and Doppler shift.

[0059] In some possible implementations, meteorological-related attenuation is calculated based on meteorological data, and channel attenuation data is obtained by combining it with communication propagation path data. Specifically, this includes: determining the base attenuation based on communication distance and signal frequency; calculating free space loss, atmospheric absorption attenuation, rainfall attenuation, and cloud / fog attenuation based on the base attenuation; fusing the free space loss, atmospheric absorption attenuation, rainfall attenuation, and cloud / fog attenuation, and combining them with propagation path segmentation correction to obtain meteorological-related attenuation; obtaining non-meteorological-related attenuation based on communication propagation path data, and fusing the meteorological-related and non-meteorological-related attenuations to obtain channel attenuation data. Specifically, free space loss A... FS =20log10( d )+20log10( f )+20log10(4 π / c ), d Indicates communication distance. f Indicates signal frequency. c Represents the speed of light; atmospheric absorption attenuation A atm = c o • d + c w • d , c o Indicates the oxygen attenuation coefficient. c w Indicates the water vapor attenuation coefficient; rainfall attenuation A rain = k • R α • L eff , R Indicates rainfall intensity, k and αIndicates frequency / polarization related parameters, L eff Indicates the effective path length; cloud attenuation A cloud = K l • M • d , K l Indicates the water content coefficient of a specific liquid. M The density of liquid water in clouds and fog is expressed as a function of humidity, and the meteorological attenuation A is obtained. p =∑(A FS +A atm +A rain +A cloud +A other );

[0060] For non-meteorologically related attenuation: Channel State Information (CSI) provides amplitude and phase information of the channel, which can be used to assess the impact of weather conditions on signal propagation. Path loss is the signal attenuation caused by factors such as increased distance and obstruction during propagation. Weather conditions increase path loss. Multipath delay spread is caused by phenomena such as reflection, refraction, and scattering encountered by the signal during propagation, resulting in different arrival times at the receiver. Weather conditions affect multipath delay spread. Doppler shift is caused by the relative motion between the transmitter and receiver, resulting in a change in the frequency of the received signal.

[0061] By adding meteorological-related attenuation and non-meteorological-related attenuation (such as path loss, multipath delay spread, and Doppler shift), the total attenuation, i.e., channel attenuation data, can be obtained.

[0062] In some possible implementations, path loss beyond weather-related attenuation also includes the impact of building obstruction on the channel in urban road environments, which significantly increases path loss. Signals encounter more reflection, refraction, and absorption as they pass through buildings, leading to a substantial decrease in signal strength. In urban environments, due to environmental complexity, factors such as building height, density, and material properties must be considered, and the path loss is typically described using empirical models (such as the Okumura-Hata model) or two-ray ground reflection models.

[0063] In some possible implementations, the delay spread and Doppler shift of the channel attenuation data are calculated to extract the spatiotemporal features of the interference, specifically including:

[0064] Obtain the channel impulse response containing channel attenuation data;

[0065] Delay spread and Doppler shift are calculated based on the channel impulse response. Delay spread refers to the time delay between the arrival of the first received signal component and the arrival of the last received signal component associated with a single transmitted pulse. The channel frequency response is obtained by performing a Fourier transform on the channel impulse response, and then the bandwidth of the frequency response is analyzed to estimate the delay spread. A larger delay spread indicates more severe frequency-selective fading. Doppler shift is the change in signal frequency caused by the relative motion between the signal source and receiver. In the channel impulse response, Doppler shift affects the frequency components of the signal. The Doppler shift is estimated by analyzing the rate of change of the channel impulse response. A larger Doppler shift indicates more significant signal fading.

[0066] Extracting time-domain features based on time delay spread: Time delay spread causes the signal to broaden in the time domain. The time-domain features of time delay spread can be extracted by calculating the autocorrelation function or cross-correlation function of the signal.

[0067] Frequency domain feature extraction based on Doppler frequency shift: Doppler frequency shift causes a frequency shift in the signal in the frequency domain. The frequency domain features of the Doppler frequency shift can be extracted by performing a Fourier transform on the signal.

[0068] Spatial features can be extracted based on the direction of arrival (DOA) estimation of the spatial distribution of interference sources: DOA estimation uses the signal received by the antenna array to estimate the incident direction of the signal. After estimating the DOA of the interference source, spatial features can be extracted.

[0069] In some possible implementations, based on the spatiotemporal characteristics of the interference, the interference period is identified, and a dynamic interference relationship topology is generated, specifically including:

[0070] The Fast Fourier Transform (FFT) is used to transform channel attenuation data from the time domain to the frequency domain, allowing for spectrum analysis. FFT efficiently converts time-domain signals to frequency-domain signals, facilitating the identification of periodic components in the spectrum. Periodic components in the spectrum typically manifest as distinct peaks; these peaks are identified by setting amplitude thresholds, thereby determining the spectral interference period.

[0071] Short-time Fourier transform (STFT) provides local information about the signal in both time and frequency by windowing the signal in the time domain and performing Fourier transform within each window. This results in a time-frequency graph, which can be used to observe the frequency changes of the signal at different time points. Since periodic patterns appear as repeating frequency distribution patterns in the time-frequency graph, the time-domain interference period can be extracted by analyzing these patterns.

[0072] By combining the spectral interference period, the time-domain interference period, and the spatiotemporal characteristics of interference, as well as other possible interference characteristics (such as interference intensity and duration), a spatiotemporal characteristic description of interference is formed. Each node in the communication network is regarded as a communication entity, and the edges between nodes represent the communication interference between communication nodes. The initial topology graph of the communication network is constructed.

[0073] The extracted spatiotemporal features of interference are embedded as node attributes into the topology graph. Each node's features can include interference period, interference intensity, and interference frequency, thus initializing the location and communication characteristics of each node. Based on the spatiotemporal features of the interference, the interference relationships between nodes are determined. Weights and labels are assigned to each edge to represent the intensity and type of interference. Simultaneously, to account for the impact of time on interference, time encoding is added to the features of each node. Then, based on the node's time state, the nodes are periodically updated, dynamically adjusting the connections between nodes to obtain a dynamic interference relationship topology. The node feature time encoding can be used to verify the temporal accuracy of the topology, ensuring that the updated topological relationships of each node are correctly represented.

[0074] In some possible implementations, an interference prediction model is constructed based on the dynamic interference relationship topology, specifically including:

[0075] The dynamic disturbance relationship topology is input into the spatiotemporal graph neural network model, which includes a graph convolution module, a temporal prediction module, and a joint prediction layer.

[0076] The graph convolution module is mainly used to aggregate interference information from neighboring nodes, which can effectively handle spatial dependencies between nodes.

[0077] The time series prediction module captures the time series dependencies of node interference, which is usually implemented using a Long Short-Term Memory (LSTM) network or a gated recurrent unit (GRU) and can handle long-term dependencies in time series data.

[0078] The joint prediction layer is used to stitch together spatiotemporal features and output the prediction results through a fully connected layer. This layer integrates spatial and temporal information to predict interference intensity, interference hotspots, or key interference paths.

[0079] In some possible implementations, the interference optimization strategy specifically includes:

[0080] Based on the interference intensity, the transmission power of communication nodes is dynamically adjusted. In high-interference areas, the transmission power of nodes is reduced to decrease signal coverage strength, thereby reducing interference to other nodes. This step can reduce signal coverage strength in high-interference areas, effectively reducing interference to surrounding nodes. Reducing transmission power can save energy consumption of communication nodes and extend the service life of equipment. By reasonably controlling the power, more nodes can communicate in the same frequency band, improving the utilization rate of spectrum resources. This is especially suitable for scenarios such as wireless sensor networks and cellular networks that require dynamic power adjustment to cope with interference.

[0081] Based on the interference hotspots, channel allocation algorithms or frequency band switching strategies are used to switch the interfered nodes to low-interference channels or frequency bands. By switching to low-interference channels or frequency bands, interference can be effectively avoided and communication quality can be improved. Dynamic channel allocation can flexibly adjust channel resources according to current traffic demand and channel conditions, improve spectrum utilization, and respond in real time to changes in communication traffic, fluctuations in channel quality, and the diversity of user needs.

[0082] Based on the critical interference path, optimize the data routing strategy and select alternative paths to avoid high-interference links. By selecting low-interference alternative paths, the impact of critical interference paths on data transmission can be effectively reduced, improving communication reliability. Optimizing data routing can reduce data transmission latency and packet loss rate, improving the overall network performance. When problems occur on the critical interference path, selecting alternative paths can ensure normal data transmission and enhance the network's fault tolerance.

[0083] In some possible implementations, the process of selecting the optimal strategy specifically includes:

[0084] The system monitors the data quality of the optimized communication propagation path in real time, inputs the monitoring results into the interference prediction model through a closed-loop feedback mechanism, calculates the prediction error value, and triggers model retraining when the error exceeds the threshold. By inputting the monitoring results into the interference prediction model, a complete feedback loop is formed. This mechanism can adjust the system parameters in real time to adapt to the constantly changing communication environment, thereby improving the stability and performance of the system.

[0085] The predicted error value is compared with the dynamically adjusted service demand threshold. The dynamic threshold is automatically adapted according to the real-time service type. This dynamic adjustment mechanism can better adapt to different service needs. For example, in high-priority services, a stricter threshold can be set to ensure communication quality, while in low-priority services, the threshold can be appropriately relaxed to improve resource utilization.

[0086] When the prediction error exceeds the dynamic threshold, the policy engine is triggered to select the optimal policy from the interference optimization policy library. This policy library is designed to provide multiple optimization schemes to cope with different interference situations. For example, the power control policy sub-library can reduce interference by adjusting the transmit power; the channel switching policy sub-library can switch to a clearer channel; the routing optimization policy sub-library can replan the communication path; and the load balancing policy sub-library can reasonably allocate the communication load. The prediction error value is compared with the dynamically adjusted service demand threshold, and the dynamic threshold is automatically adapted according to the real-time service type.

[0087] The selected strategy is executed and continuously monitored to form an adaptive optimization closed loop. This closed-loop mechanism can continuously adjust the optimization strategy to adapt to the real-time changing communication environment. For example, if a certain strategy is not effective in the current environment, the system will automatically switch to other strategies and continue to monitor and optimize, thereby achieving continuous performance improvement.

[0088] As one possible implementation, this embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements a 6G communication signal quality optimization method based on intelligent transportation.

[0089] As one possible implementation, this embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements a 6G communication signal quality optimization method based on intelligent transportation.

[0090] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for optimizing 6G communication signal quality based on intelligent transportation, characterized in that, The specific steps include: real-time acquisition of meteorological data and communication propagation path data; Meteorological attenuation is calculated based on meteorological data, and channel attenuation data is obtained by combining it with communication propagation path data. Specifically, this calculation includes: determining the base attenuation based on communication distance and signal frequency; calculating free space loss, atmospheric absorption attenuation, rainfall attenuation, and cloud / fog attenuation based on the base attenuation; fusing the free space loss, atmospheric absorption attenuation, rainfall attenuation, and cloud / fog attenuation, and combining them with propagation path segmentation correction to obtain meteorological attenuation; and obtaining non-meteorological attenuation based on communication propagation path data, fusing the meteorological attenuation and non-meteorological attenuation to obtain channel attenuation data. The method involves calculating the delay spread and Doppler frequency shift of channel attenuation data to extract spatiotemporal features of interference. Specifically, this calculation includes: obtaining the channel impulse response containing channel attenuation data; calculating the delay spread and Doppler frequency shift based on the channel impulse response; extracting time-domain features based on the delay spread; extracting frequency-domain features based on the Doppler frequency shift; and estimating the spatial distribution of interference sources based on the direction of arrival (DOA) to extract spatial features. Based on the spatiotemporal characteristics of interference, interference periods are identified, and a dynamic interference relationship topology is generated. Specifically, this process includes: analyzing the spectrum of channel attenuation data using Fourier transform to identify periodic components and obtain the spectral interference period; analyzing the temporal characteristics of channel attenuation data using short-time Fourier transform to identify periodic patterns and obtain the temporal interference period; constructing an initial topology graph of the communication network based on the spectral interference period, temporal interference period, and spatiotemporal interference characteristics, where nodes represent communication nodes and edges represent communication interference between nodes; embedding the extracted spatiotemporal interference characteristics as node features into the initial topology graph, and using time encoding to embed time information into the node features to generate the dynamic interference relationship topology. An interference prediction model is constructed based on dynamic interference relationship topology. Specifically, this construction includes: inputting the dynamic interference relationship topology into a spatiotemporal graph neural network model, which includes a graph convolution module, a temporal prediction module, and a joint prediction layer. The graph convolution module aggregates interference information from neighboring nodes; the temporal prediction module captures the temporal dependencies of node interference; and the joint prediction layer concatenates the spatiotemporal features and outputs the prediction result through a fully connected layer. The prediction result includes interference intensity, interference hotspots, or key interference paths. Based on the output of the interference prediction model, the communication propagation path data is optimized and adjusted to generate an interference optimization strategy. The system monitors the data quality of the optimized communication propagation path in real time, and re-inputs the monitoring results into the interference prediction model through closed-loop feedback. The model is then compared with the dynamic threshold that adapts to business requirements, triggering the strategy engine to select the optimal strategy from the interference optimization strategy library.

2. The 6G communication signal quality optimization method based on intelligent transportation according to claim 1, characterized in that, The real-time meteorological data collected includes rainfall intensity and temperature and humidity; the real-time communication propagation path data collected includes channel state information, path loss, multipath delay spread, and Doppler shift.

3. The 6G communication signal quality optimization method based on intelligent transportation according to claim 1, characterized in that, The interference optimization strategy specifically includes: adjusting the transmission power of the communication node according to the interference intensity using a dynamic power control algorithm to reduce the signal coverage intensity in high interference areas; Based on the interference hotspot areas, channel allocation algorithms or frequency band switching strategies are used to switch the interfered nodes to low-interference channels or frequency bands. Based on the key interference paths, optimize the data routing strategy and select alternative paths to avoid high-interference links.

4. The 6G communication signal quality optimization method based on intelligent transportation according to claim 1, characterized in that, The process of selecting the optimal strategy specifically includes: real-time monitoring of the data quality of the optimized communication propagation path, inputting the monitoring results into the interference prediction model through a closed-loop feedback mechanism, calculating the prediction error value, and triggering model retraining when the error exceeds the threshold. The predicted error value is compared with a dynamically adjusted business demand threshold, which is automatically adapted according to the real-time business type. When the prediction error exceeds the dynamic threshold, the policy engine is triggered to select the optimal policy from the interference optimization policy library, which includes: a power control policy sub-library, a channel switching policy sub-library, a routing optimization policy sub-library, and a load balancing policy sub-library. The selected strategy is executed and continuously monitored to form an adaptive optimization closed loop.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the 6G communication signal quality optimization method based on intelligent transportation as described in any one of claims 1 to 4.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the 6G communication signal quality optimization method based on intelligent transportation as described in any one of claims 1 to 4.

Images