Research And Implementation Of Mobile Edge Computing Offloading Algorithm For The Internet Of Vehicles

With the development of Internet of Vehicles(Io V)technology and intelligent transportation,a large amount of data with low time delay needs to be processed in real time.The Mobile Edge Computing(MEC)offloading algorithm,which divides task data to make full use of local and cloud computing resources,achieves the goal of processing task data rapidly.Efficient MEC offloading algorithm can reduce task data processing time,system energy consumption and improve task data processing efficiency.This thesis aimes at the research of computing offloading algorithm in the Io V environment,which mainly does the following three aspects:First of all,to solve the problem of the high time delay and high energy consumption of single MEC server computing offloading algorithm,a multi-objective computing offloading algorithm based on interior point optimization method is proposed.The interior point method optimization method is used to solve the problem of the joint optimization of energy consumption and cost function,which is then used to solve the data offloading probability problem and to obtain the optimal data offloading probability.Experimental results show that,given the same data processing capacity of the MEC server and the total data volume of the system,this algorithm reduces the time delay and energy consumption and improves the efficiency of data processing comparing with the traditional algorithms as time increases.Secondly,in view of the increase of the data backlog in computing offloading algorithm in the multi-MEC server Io V system,which reduces the stability of the system and leads to low algorithm efficiency,a time-varying computing offloading algorithm based on Lyapunov optimization is proposed.Lyapunov optimization method is used to minimize the expected upper bound of each queue in the system,reduce the system data backlog,and further combine with the loss function to solve the amount of real-time optimal offloading data.Experimental results show that,given the same data processing capacity of the MEC server,when the amount of task data increases,this algorithm reduces the data backlog and reduces the task data processing time and energy consumption compared with traditional algorithms.Finally,combined with the multi-objective computing offloading algorithm based on the interior point optimization method and time-varying computing offloading algorithm based on Lyapunov optimization,a MEC offloading prototype system for Io V is designed and implemented.The experimental results show that the MEC offloading system for Io V proposed in this thesis can effectively realize the computation offloading of task data and enhance the efficiency of the data processing,and the utility of the multi-objective computing offloading algorithm based on the interior point optimization method and time-varying computing offloading algorithm based on Lyapunov optimization are verified.