Research On Computational Offloading Technology Based On Improved Particle Swarm Optimization Algorith

With the continuous development of Internet technology,the number of smart terminals and the amount of application data has grown rapidly,and users’ demand for computing resources has also become larger and larger.At the same time,various new applications have emerged one after another,such as virtual reality,augmented reality,autonomous driving,and online games.These new applications are characterized by computationally intensive and time-delay sensitivity,requiring not only strong computing power as operational support but also feedback of operational results to requesting users with extremely low latency.In mobile edge computing(MEC)users offload their computation tasks to servers with high computing capacity so that their tasks can be executed successfully by invoking powerful computing power.At the same time,it deploys servers at the edge of the network,which minimizes the communication distance between users and servers,reduces the total delay of tasks,and solves two major problems of new applications.In addition,since the smart terminal offloads task to servers for computation instead of computing by itself,the energy consumption of the smart terminal can be greatly reduced.MEC involves many technologies,and computing offloading technology is its key technology.Computing offloading technology mainly solves important issues such as whether tasks are offloaded,where tasks are offloaded to,and the resource allocation in servers.Aiming at the problem of computing offloading,a multi-user multi-task multi-server mobile edge computing offloading model is built in this paper,and this model comprehensively considers the server task balance and the total cost of task delay and energy consumption,while establishing a random mobility model based on the mobility characteristics of intelligent terminals,taking into account the impact of mobility on computational offloading.The main works of this paper are as follows:(1).To solve the problem of computing offload in the multi-user and multi-server scenario,a mobile edge computing offload model of multi-user,multi-task and multi-server is established.Considering the task delay and energy consumption,the total cost expression is formulated as the adaptive weighted sum of task delay and energy consumption.At the same time,the task allocation balance is considered to avoid the occurrence that the server cannot unload subsequent tasks to the server due to high load.To minimize the total cost of the system,the proposed improved particle swarm optimization algorithm is used to optimize and solve the problem.The simulation results show that compared to other comparative algorithms,the improved particle swarm optimization algorithm can effectively reduce the total cost of the system and obtain a better computing offloading strategy.(2).For the mobility characteristics of intelligent terminals,establish a multi-user and multi-server system model for joint mobility management,consider the impact of mobile device mobility on offloading decisions and resource allocation,and establish a multi mobile device and multi server random mobility model.When establishing the model,the time is discretized into multiple time slots,and the mobile device task calculation offloading problem for one of the time slots is considered.At the same time,the total cost expression is set as the weighted sum of the total task delay and energy consumption,and the time delay and energy consumption weights are adaptively adjusted based on the remaining energy information and charging information of the mobile device.To minimize the total cost of the system,an improved particle swarm optimization algorithm is used to solve this problem.The simulation results show that,considering the random movement of terminal devices,the improved particle swarm optimization algorithm can obtain a computational offloading method that minimizes the total cost of the system.