Research On Cellular-vehicle To Everything Communication And Edge Service Optimization

With the large-scale deployment of 5G network,the rapid development of Cellular Vehicle-to-Everything(C-V2X)industry has been promoted.C-V2 X has become the strategic development direction of China,which can not only meet the demand of high reliability and low latency of C-V2 X,but also has the potential to meet the diversified demand of users for quality of service.The thesis focused on Vehicle to Everything(V2X)communication model for intelligent networked vehicles,V2 X resource allocation,edge computing and cache collaboration algorithm,which can provide theoretical foundation and technical support for the next edge network deployment and vehicle-road-cloud collaboration application,and the main innovative research work was as follows:Firstly,the thesis was focused on the C-V2 X communication model,and proposed a V2 X simulation model for wireless channels to model the core communication nodes based on the highway scenario and urban scenario,which can provide reference for the simulation and validation of the cellular Telematics mobile communication model by taking full advantage of the C-V2 X converged communication.Then,a channel model based on direct link communication and Uu communication was constructed to address the problems of high power consumption,spectrum constraint and energy inefficiency in 5G scenario.Based on the model,a resource allocation algorithm was proposed for maximizing the downlink throughput of the system and ensuring the connectivity of V2 V links.In view of the dynamic,time-varying and diverse characteristics of C-V2 X,the Kuhn Munkras(KM)algorithm was used to dynamically schedule channel resources.The simulation results show that the proposed algorithm improves the downlink throughput by about 3% compared with the comparison algorithm,and maximizes the reliability of each V2 V link.Finally,aiming at the contradiction between the demand of network-connected vehicle users and limited computing resources in the V2 X environment of intelligent network-connected transportation,an edge computing and caching cooperative network architecture with multi-user,multi-service,and multi-edge server coexistence was constructed to optimize the comprehensive cooperation of computing and caching resources.Based on this architecture,we proposed a collaborative offloading and caching mechanism for V2 X services of intelligent networked transportation based on multi-access edge computing;introduce the Least Recently Used(LRU)algorithm to identify and make decisions on edge offloading tasks,utilize edge caching results and optimize computational workload;based on the Deep Deterministic Policy Gradient(DDPG)algorithm for distributed offloading of edge computing tasks to achieve collaborative decision making between local and edge offloading.The simulation results show that the proposed joint offloading model can obtain better computational offloading performance than the traditional vehicle local offloading model,server offloading model and Local or Offloading(LOO)offloading model.