| The rapid emergence of emerging services such as Io T and Telematics has placed increasing demands on the network’s transmission bandwidth and data distribution and processing capabilities.Traditional cloud computing architectures can no longer meet the high reliability,low latency and low power requirements of users.To meet this challenge,mobile edge computing(MEC),a new network architecture,has emerged.The basic idea of Mobile Edge Computing is to reduce network latency,improve data security and reduce bandwidth usage by placing computing and data storage close to the source of data.The uneven distribution of population density between regions has led to a serious imbalance in the load on base stations and difficulties in meeting the real-time requirements of users.To address the above issues,this paper is dedicated to the study of edge server placement and computation offloading strategies,with the following main research elements.Firstly,for the server placement problem in the mobile cellular network scenario,a system model of the server placement problem is first established,while the energy consumption and load balancing of the edge servers are considered and modelled separately,and the problem is defined as a multi-objective optimization problem.An adaptive server placement algorithm based on the enhanced whale optimization algorithm(EWOA)is proposed for this problem.The algorithm improves the step size of the whale shrinkage envelope to an adaptive step size with increasing number of iterations,based on the traditional whale swarm algorithm,in order to solve the shortcomings of low convergence accuracy and prematureness,in addition to introducing the idea of cross-variance,from which the optimal solution is selected.Simulation results show that the improved whale swarm algorithm achieves significant progress in reducing the total energy consumption of the system and achieving load balancing compared to the other three optimization algorithms.Secondly,for the study of computational offloading strategies,as most current studies only consider single-edge or edge-of-things cloud architectures without utilizing offsite edge server resources,this paper proposes a multi-edge collaborative network architecture.Tasks in this architecture can be selected to be executed locally,by a local server,by an offsite server or in the cloud,and a mathematical model is developed for the weighted sum of delay and energy consumption for each of the four offloading modes.The immune particle swarm algorithm(IPSO)is used to solve the optimization objective to address the shortcomings of traditional particle swarm algorithms,which tend to be premature and fall into local optimality.Simulation results show that the offloading strategy proposed in this paper has the lowest total cost and can improve the execution efficiency of the task compared with other three offloading strategies.Thirdly,an edge server placement and computation offload system is designed and implemented.The system can help users to perform server placement and compute offload operations with a visual interface and obtain simulation results while ensuring user data security.The functional modules of the system include registration and login module,user management module,device management module,server placement module and computing offloading module. |
