Research On Task Offloading Strategy For Ultra Dense Network

With the rapid development of the mobile Internet,the computation intensive tasks of many new applications have put higher requirements on the computing capability of mobile terminals.The emergence of Mobile Edge Computing(MEC)technology can effectively solve the contradiction between the high-performance requirements of new applications and the limited resources of mobile terminals.MEC supports mobile terminals to offload complex computing tasks to the MEC server at the edge of the network for execution,which can significantly improve user service quality.In addition,considering that in the future 5G era wireless networks will inevitably present a trend of intensive deployment,this paper combines MEC with ultra-dense networks(UDN),comprehensively considering task demand characteristics and network resource status,and formulating the optimal joint task offloading and resource allocation strategies to minimize the user's task computing time or energy consumption.The paper mainly studies task offloading and resource allocation in UDN scenarios.The specific contents are as follows:For a multi-user multi-MEC server system in UDN,the computing time is defined as the maximum time required by users to perform tasks in the system,and the situation in which the computing tasks can be split is considered,and the minimum task computing time problem is established.In order to solve the optimization problem,the original problem is transformed into the offloading proportion subproblem and the task offloading decision subproblem,and the conditions for the existence of the optimal offloading proportion are proposed and proved.Based on the game theory,the task offloading decision subproblem is transformed into a non-cooperative game,and proves that there is a Nash equilibrium in the game.Finally,an algorithm for minimizing computation time is proposed to determine the optimal offloading strategy of all users.For the MEC network that supports Unmanned Aerial Vehicle(UAV)communication,the system energy consumption is defined as the energy consumed by all UAV in the network,and the task offloading proportion and transmission power are considered to establish the energy consumption minimization problem.In order to solve the optimization problem,the original problem is converted into task offloading proportion subproblem,transmission power subproblem and task decision subproblem.The different methods were used to solve the different subproblem,and iteratively solved in order to obtain an optimaltask offloading strategy that can minimize UAV energy consumption.Simulation results show that the proposed algorithm can minimize the energy consumption of UAVs and improve the system resource utilization.In the UDN-MEC network,a task offloading algorithm that jointly optimizes task computing time and energy consumption is proposed.The system overhead is the weighted sum of task computing time and energy consumption.By jointly optimizing task offloading proportion and transmission power,a task overhead minimization system overhead problem model is established.In order to solve this optimization problem,a minimum overhead algorithm based on particle swarm optimization is proposed to obtain the optimal offloading strategy for each user.Simulation results show that the proposed algorithm can converge to the global optimal solution,and significantly saves the overall system overhead.