| The rapid development of wireless networks and the popularization of portable mobile devices have changed people’s work and lifestyle.Due to the explosive growth of the number of smart devices,mobile users have put forward higher requirements for the computing and storage capacity of the device.When performing computation-intensive and delay-sensitive tasks,the limited computing and storage resources of mobile devices will lead to problems such as high energy consumption,long response time and high network bandwidth occupation.The solution of traditional cloud computing is to transmit local computing tasks to the cloud for processing,and then return the results to the local computer,which has the disadvantages of high latency and network congestion in the process of processing terminal computing tasks.Mobile edge computing is a new type of data processing paradigm,which places servers at the edge of the network to reduce the computing pressure of cloud service centers,provide users with resources such as high fluency and high bandwidth,effectively speed up the operation of devices,alleviate the shortage of local resources,and improve user experience.Mobile edge computing can reduce the load on the network,but there are still shortcomings in task offloading and resource allocation: mobile devices are limited by physical size,and battery energy is insufficient to support the high energy consumption and capacity requirements of smart services;In the multi-user and multi-server scenario,there are problems of resource competition and server selection.Mobile devices can be in multiple network signal ranges at the same time,creating serious interference problems.In view of the above problems,this paper conducts in-depth research and discussion,and the main work is as follows:(1)Aiming at the task offloading problem under multi-user and multi-server mobile edge computing system,this paper takes the average execution cost of mobile devices as the performance index,and improves the online task unloading algorithm based on Lyapunov optimization theory in order to reduce the drop rate of computing tasks and reduce the execution cost of mobile devices during task unloading.Faced with multiple mobile devices issuing task unloading requests at the same time,the algorithm establishes a task queue model according to the task unloading situation,and makes dynamic unloading decisions based on the generation of mobile device tasks in each time slot under the premise of ensuring the stability of the system queue.The simulation results show that the algorithm effectively reduces the average execution cost of mobile devices while ensuring the stability of the system.(2)In the interference network of resource multiplexing,in order to improve the throughput of the network,under the premise of ensuring the delay and reliability requirements of D2 D users,a scheme suitable for joint channel allocation and power control in D2 D communication is proposed,a low-complexity heuristic algorithm is used to allocate channel resources,and the DQN algorithm is used to determine the optimal transmission power of D2 D pairs in each mode,and adaptive decisions are made based on local observations.Experimental simulation results show that the scheme can reduce communication interference,improve the performance of cellular network and D2 D link,and improve system throughput. |
