Research On Efficient And Energy-saving Task Offloading Mechanism Based On Edge Computing

Due to the gradually mature Internet of Things(Io T)and 5G technology,various applications in an endless stream.Cloud computing has strong cloud servers,which can provide computing services for massive data.However,the service form of emerging applications is increasingly close to users,and the remote service mode of cloud computing can not satisfy the basic needs of such applications.As a multi-functional service platform on the edge of the network,edge computing can effectively offset the shortcomings of cloud computing,and provide real-time and efficient service experience for users.At present,the research on edge computing is growing,but for,there is a lack of relevant consideration for the sustainability of terminal devices battery,the rational utilization of network resources and the processing of time-delay sensitive tasks.This thesis studies efficient and energysaving task offloading mechanism based on edge computing,which mainly covers the following three aspects:1)Edge Computing Offloading Mechanism Based on D2 D Collaboration: In order to relieve the communication pressure and reduce the load on the service nodes,this paper proposes an edge computing offloading mechanism based on D2 D collaboration.Specifically,based on the comprehensive consideration of offloading decisions,and transmission power allocation,an optimization problem that minimizes the total energy consumption of task completion is formulated.Furthermore,the motivation measurement constraint of D2 D devices is defined to promote collaboration between cooperation users and common users.At the same time,an efficient computation offloading algorithm based on dynamic sensing of bat population is proposed.The algorithm combines the ideas of classical bat algorithm,and introduces an adaptive dynamic inertia weight.Through sensing environment in real time to adjust the movement direction and speed of bat population,and the chaotic mapping theory is used to initialize the population.Finally,the simulation results show that the proposed scheme can converge at a faster speed,and achieve the optimal offloading and power allocation strategy.Compared with other benchmark schemes,the proposed scheme has significant advantages in reducing system energy consumption.2)Efficient and Energy-saving Computation Offloading Mechanism with Energy Harvesting: In order to effectively extend the lifetime of Io T devices,improve the energy efficiency of tasks processing,and build a self-sustaining and green edge computing system,this paper proposes an efficient and energy-saving computation offloading mechanism with energy harvesting.Specifically,based on the comprehensive consideration of local computing resources,energy harvesting time allocation ratio,and offloading decision,an optimization problem that minimizes the total energy consumption of all user devices is formulated.Meanwhile,a deep learning based efficient and energysaving offloading decision and resource allocation algorithm is proposed.The design of deep neural network architecture incorporating with regularization method and the employment of the stochastic gradient descent method can accelerate the convergence speed of developed algorithm and improve its generalization performance.Furthermore,it can minimize the total energy consumption of task processing by integrating the momentum gradient descent to solve the resource optimization allocation problem.Finally,the simulation results show that the proposed mechanism in this paper has significant advantage in convergence speed,and can achieve an optimal offloading and resource allocation strategy which is approximate to the solution of greedy algorithm.3)Intelligent Resource Allocation and Computation Offloading Mechanism for Edge Computing:In order to satisfy the differentiated demands of different users in Io T scenarios,improve resource utilization,and build an efficient edge computing service system,this paper proposes a priority-based edge computing offloading mechanism for Io T.Through sensing the priority,edge nodes can provide corresponding computing services with reasonable computing resource allocation,thereby avoiding execution failure due to long-time task waiting,and improve user service quality.Specifically,under the comprehensive consideration of computation offloading decision,bandwidth resource allocation ratio,and computing resource of edge node ratio,an optimization problem that minimizes the total energy consumption of all tasks is formulated.At the same time,a priority-based intelligent resource allocation and offloading algorithm is proposed to solve the above optimization problems.The algorithm combines the idea of deep deterministic policy gradient algorithm,and designs a dual“actor-critic” network architecture,which accelerates the convergence speed of the training process,and,to make the algorithm suitable for the mixed integer optimization proposed in this paper problem,discretize the continuous actions to generate binary offloading decisions.Finally,the simulation results show the effectiveness of the proposed scheme in this paper,and can achieve the optimal offloading and resource allocation strategy which is approximate to the solution of greedy algorithm.Compared with the local computing and full offloading schemes,the proposed scheme reduces the total system energy consumption by about 52% and 13%,respectively.