Research On Time-varying Directional Channel Measurements And Modeling For Intelligent Transportation

Transportation is one of the pillar industries of national economic development,and road transportation is the most important part,which bears most of the transportation demand.Road transport serves people's travel and economic development and brings increasingly serious negative impacts such as traffic safety,traffic congestion,and environmental pollution.Under this background,the vehicular communications represented by C-V2X(Cellular Vehicle-to-Everything)has gradually become a hot research field.C-V2 X can realize comprehensive and efficient information transmission among traffic participants,which is helpful to improve the operational efficiency of road traffic and alleviate traffic congestion.A deep understanding of the radio propagation mechanism of wireless channels and the establishment of accurate channel models are the cornerstones of the design,performance evaluation,and standardization of wireless communication systems.Vehicular communication has the characteristics of a unique propagation environment,high moving speed,and low antenna height,which makes the wireless channel of C-V2 X significantly different from traditional cellular communication channels.Therefore,for the research of radio wave propagation and channel modeling in C-V2 X,we should not only continue to pay attention to propagation mechanism,system design,and model building but also fully consider the specificity of C-V2 X channels itself.And focus on the technical difficulties brought by unique communication environment and demand,and investigate the key problems that need to be solved urgently in channel modeling of vehicular communications.Based on the above requirements,this thesis studies the radio wave propagation characteristics and modeling methods of C-V2 X wireless channels from various aspects.The main contributions and novelties of this thesis are as follows:1)A multi-antenna dynamic channel measurement system is designed for the channel measurement of C-V2 X communications.The measurement system can realize channel measurement of the vehicular environment below 6GHz with 30 MHz bandwidth.Then,the channel measurement system is used to carry out the channel measurement in typical scenarios such as urban and highways,and complex scenarios such as underground garages and tunnels.A large number of measured data collected based on channel measurement constitute the basis of research on C-V2 X channel propagation characteristics and channel modeling.2)Aiming at the intelligent sensing of C-V2 X communication scenarios,the real-time scenario identification and the fast direction-of-arrival estimation algorithms based on machine learning are proposed.The real-time C-V2 X scenario identification algorithm uses machine learning as a training tool to get the scenario identification model and verifies the identification accuracy in various scenarios.Besides,the proposed fast direction-of-arrival estimation algorithm uses the output signals of array antennas as training data to construct the direction-of-arrival estimation model,which can realize the real-time estimation of direction-of-arrival and angular dispersion of signals in C-V2 X scenarios.3)Aiming at the research of channel modeling method in dynamic C-V2 X environment,a three-dimensional directional dynamic channel model based on multipath cluster structure is proposed,and the specific model parameters and channel simulation methods in different scenarios are given.The proposed channel model can accurately describe the influence of environmental differences on channel characteristics.In addition,the proposed channel modeling method has good universality and can be extended to various communication scenarios with the support of measured data.On the other hand,the proposed model supports dynamic channel simulation,which is suitable for non-stationary scenarios.4)Aiming at the complex and diverse channel propagation environment in C-V2 X communications,measurements are carried out on several important complex C-V2 X scenarios,such as an underground garage,viaduct,tunnel,cutting,and vehicle obstruction,and propagation characteristics in the corresponding propagation scenarios are studied.Besides,a dynamic cluster vehicular channel model is proposed for obstruction scenarios.The related research is helpful to deepen the understanding of different C-V2 X scenarios,and the model can also directly serve the channel simulation and communication system design.To sum up,this thesis focuses on the wireless channel in C-V2 X communications,provides a new solution for intelligent C-V2 X scenario sensing,and provides a new modeling theory for directional non-stationary channels.Also,this thesis deeply depicts and models the channel characteristics in diversified C-V2 X scenarios,makes up for some shortcomings in the current research field of vehicular channels,and enriches the related propagation theory and channel modeling research.