The Research On Energy Consumption Cost And Safety Performance Optimization Of UAV Cooperative MEC Network

With the rapid development of wireless communication technology and intelligent devices,data traffic is increasing exponentially.The explosive data traffic and mobile terminals with different service requirements have caused fatal defects such as data congestion and complicated architecture to centralized cloud computing.Mobile edge computing extends cloud computing resources to the edge of the network near the user side,which is a new decentralized paradigm,providing task offloading,storage and application services for mobile internet of things devices,greatly shortening the transmission distance and effectively reducing energy consumption.However,it is difficult to ensure the efficient services required in smart cities with fixed infrastructure.In recent years,unmanned aerial vehicle(UAV)has been widely used in the field of wireless communication because of its advantages of on-demand deployment,high maneuverability and low cost.UAV with server can not only help Io T devices to complete data calculation and reduce energy consumption,but also serve as a relay node to cooperate with Io T devices and let base stations provide services.This paper focuses on the offloading resource allocation model in UAV cooperative mobile communication network,From the user’s and system’s perspectives,the goal of reducing user’s energy consumption and improving system’s revenue and safety performance is achieved by optimizing wireless resources,offloading strategy and UAV’s location.The specific research contents and work of this paper are summarized as follows:1.In the scenario of single UAV offloading with communication module and server,a UAV cooperative task offloading scheme is proposed,and the optimization of users’ offloading task volume,UAV power and UAV trajectory under the condition of ensuring downlink user experience is studied,and the total energy consumption of offloading users is optimized.Aiming at the non-convex problem of objective optimization,a multi-objective joint optimization algorithm for optimizing user energy consumption is proposed.The simulation results show that the proposed algorithm can converge quickly,and the computing energy consumption of offloading users is significantly reduced.2.In the scenario of multi-UAV offloading with communication module and server,a revenue scheme of multi-UAV cooperation is proposed,and the optimization problems of power control,resource allocation and user association are designed under the premise of ensuring system performance.Because the goal is a mixed integer nonlinear problem.In order to solve this problem effectively,a multi-agent reinforcement deep learning algorithm based on deep learning is proposed.Through simulation analysis,the scheme of this paper can effectively improve the system revenue.3.In the scenario of cooperative mobile edge computing with a single UAV as a relay,aiming at the security problem of user information in the mobile offloading scenario,a two-level offloading security strategy is proposed,and the two-level offloading security model in the case of eavesdropping on multiple UAVs is studied.The multi-objective joint optimization scheme solves the optimization objective problem of maximizing the security capacity of the system and effectively improves the security performance of the system.