Research On Key Technologies Of Channel Measurement Assistance And Target Tracking In Vehicular Heterogeneous Networks

With the high-speed construction of vehicular communications network and the gradual landing of many applications in Internet of Vehicles,the demand for wide-area coverage,and the stability and reliability of vehicular signals has also increased.In the practical scenario of vehicular communications,the air-ground coordinated vehicle-mounted heterogeneous network architecture effectively complements the existing communication technologies.This thesis relies on the technical architecture and based on the channel data actually measured in the scenarios of V2V and V2I to analyze the channel characteristics and channel correlations in different scenarios in detail,and uses the analysis results to assist the channel measurement process.At the same time,in order to ensure the stability of vehicular heterogeneous network in complex environment,this thesis also studies the problem of real-time and efficient target tracking when UAV is used as an auxiliary communication node.In order to effectively carry out channel measurement in vehicular heterogeneous network,compared with establishment of a channel model with weak generalization,this thesis analyzes the characteristics of the vehicular channels from the perspective of real data,and analyzes the characteristics based on the spatial correlation and time-domain correlation of vehicular antennas.Based on neural network in deep learning,this thesis innovatively implements real-time channel state prediction based on channel data,which plays a positive auxiliary role for the unmeasured channel at a certain moment.The simulation results demostrate that although the vehicular channel varys quickly and the environmental factors are complex,certain channel correlation still exists.At the same time,the data-driven channel prediction algorithm proposed in this thesis has good accuracy,robustness and execution efficiency,and can effectively adapt to the rapid change of vehicular channels.Additionally,in order to ensure the proper functioning of the auxiliary communication node UAV,this paper designs a target tracking algorithm based on the return signal to monitor its flight status in real time.Aiming at the problem of balancing weight degradation and particle diversity in the resampling process,this thesis proposes an improved particle filter algorithm based on system resampling and additional random perturbance.This method ensures that in environments with different types of noise,particle filter algorithm maintains particle diversity and reduces weight degradation.Simulation results demonstrate that the proposed algorithm can predict flight trajectory more accurately than traditional particle filter.