Research On Iov Based On Communication Technology Of High Frequency And Low Frequency Cooperation

With the rapid development of the Internet of Things,5G and mobile multimedia services,the demands of smart cities and intelligent transportation have been arisen.The fulfillment of the above requirements need new wireless networks technology.The massive connections need higher communication rate in the future.However,traditional mobile communication networks and low-frequency wireless networks such as WIFI will be unable to meet the requirements.With the in-depth study of millimeter-wave bands,the academia and industriy have confidence in the future.Millimeter-wave bands have rich frequency band resources,and it can support many applications with high communication rate requirements,such as AR and VR.However,if the current beamforming technologies in802.11ad/ay standard are applied to some specific scenarios directly,it will result in a lot of unnecessary overhead.Therefore,based on the high frequency and low frequency cooperation network architecture,this thesis tries to solve two major issues of how to reduce the beam training overhead and how to optimize the handover process in the internet of vehicle.Firstly,the thesis analyzes the existing related standard technologies,including the SNR performance simulation.Based on the simulation results,we point out the shortcomings of the existing standards.To be more specific,in the Io V scenario,these methods take too long time to train the beams,which will cause the vehicle nodes long delay receiving information,and the potential dangerous.Therefore,we need to study other methods or algorithms to improve existing methods.Based on this background,this thesis proposes two optimization schemes based on high frequency and low frequency cooperation network architecture: the one is a method to quickly find a beam to establish a communication link,and the other is to use machine learning algorithms to reduce beam training overhead.Through simulation,we find that both methods can reduce access delay and improve network performance effectively.Then the thesis studies the handover problem when a vehicle moves between two or more network overlaping areas.In the Io V scenario,when a vehicle moves during the overlaping areas,if using the existing methods,the vehicle is frequently triggered to do handover,which may increase the risk of network interruption.Thus,this thesis proposes a reinforcement learning algorithm based on the high frequency and low frequency cooperation network architecture to let the Io V system make the best decision.According to simulation analysis,this algorithm allows the system to optimize network resources based on the utility function to reduce the times of unnecessary handover,and to improve the system's performance such as Qo S and throughput.