Friendly Coexistence And Resource Management Of V2V Communications And Wi-Fi In Unlicensed Bands

With the advent of the Internet of things,vehicle networking technology raises more concern by all walks of life.On the one hand,in September 2016,vehicle-to-vehicle communication(V2V)based on the LTE(Long Term Evolution)communications system took the lead in standardization,indicating that LTE-V will become the key technology of the next generation communication system.On the other hand,the number of global vehicular communicating devices is very large,which aggravates the burden of licensed spectrum whose resources are strain by itself.Therefore,to develop a wealth of unlicensed band resources and unload a part of V2V communication services to unlicensed band becomes the main research direction to solve the problem of strain resources in authorized spectrum.Considering that Wi-Fi users are the primary users of unlicensed bands,the research,on the co-existence of V2V with Wi-Fi in unlicensed bands,has become more significant.The main innovations and contributions are briefly summarized as follows:Firstly,this paper presents a communication scenario in which V2V communication and Wi-Fi systems coexist in the unlicensed bands:the base station with the LTE-U function occupies the unlicensed band resource for data transmission under the premise of guaranteeing the quality of service(QoS)of the Wi-Fi users to relieve the burden on the licensed spectrum resource.The time domain sharing strategies adopted by the research project are LBT(Listen-Before-Talk),Duty-Cycle and ABS(Almost Black Subframes),and implement radio resource management based on the above three sharing mechanisms to improve system performance.Secondly,considering LBT and Duty-Cycle mechanism,in order to ensure the high reliability and low latency of V2V devices,targeting at maximizing the throughput of V2V system,a reasonable mathematical model is established,and the heuristic Lagrangian dual algorithm is used to solve this model to obtain the optimal resource allocation Program,making the system performance optimal.The simulation analysis of the convergence of the algorithm,the superiority of the performance and dual-system performance improvements comparing to single-system.Finally,considering ABS collaborative sharing strategy,a dynamic blank subframes allocation algorithm based on reinforcement learning is proposed.The ABS cooperation strategy solves the problem of coexistence by configuring a certain number of blank subframes.During the blank subframes,LTE is in slienced state and Wi-Fi will use these subframes for data transmission.In this paper,The the number of LTE-U subframes and blank subframes are regarded as action strategy.The delay of the V2V device,the rate satisfaction of the V2V device and the satisfaction of the Wi-Fi device are regarded as learning state of agent.Q-learning algorithm is used to dynamically configure the number of sub-frames and blank sub-frames in order to achieve the goal of making the accumulative reward maximized.The convergence characteristics of the algorithm and the system performance of LTE and Wi-Fi are analyzed by simulation.