| Smart devices themselves have computing performance bottlenecks,which may not meet the business needs of some specific scenarios,and need to rely on highperformance computing devices to assist in completing some or all computing tasks.Ultra-Dense Heterogeneous Networks(UDHNs)support many intelligent terminals and diversified services by deploying a large number of micro base stations within the coverage of macro base stations and integrating wireless technologies such as 4G LTE,5G NR,and Wi-Fi.Mobile edge computing(MEC)can significantly reduce the service delay of applications by deploying servers at network edge nodes close to users.For users,computing offload technology breaks through the limitations of mobile devices in computing power,battery resources,and storage availability.In this context,the cost of task offloading has become the primary concern of users.It is very important to propose a solution that can effectively reduce the cost of user offloading according to the geographical location of edge devices and user devices,task demand,network status,and other conditions.In order to optimize the user’s task calculation offload cost,the main research content of this paper includes the following two aspects:(1)Aiming at the multi-user and single-task computational offload scenario in ultra-dense heterogeneous edge computing networks,the problem of minimizing task processing cost under the constraints of computing resources and delay is studied.Firstly,the principle of separation of control bearer and user bearer under 5G network architecture is followed to complete the construction of ultra-dense heterogeneous network model.Secondly,in the binary offloading mode,the allocation of mobile terminal association,frequency band resources,and computing resources is jointly optimized,and a mathematical model is constructed to minimize the total cost of task processing.Finally,the complex constraint problem is transformed into an unconstrained problem by the penalty function method,and then based on the respective advantages of genetic algorithm and particle swarm optimization,the Hybrid Particle Swarm Optimization(HPSO)is improved and designed to solve it.The simulation results show that compared with other offloading algorithms,the algorithm designed in this paper has significant advantages in reducing task processing costs.(2)Aiming at the multi-user and multi-task computational offload scenario in ultra-dense heterogeneous edge computing networks,the diversity of tasks in the device and the flexibility of partial offloading increase the complexity of the problem.This paper adopts a Hierarchical Algorithm Used For Computation Offloading(HACO)to solve it.For the multi-task offloading problem,the HACO uses the improved artificial fish swarm algorithm and the improved particle swarm algorithm for alternating optimization.In the initialization stage of the algorithm,a scheme based on the proportion of offloaded tasks is proposed for computing resource allocation,which ensures the fairness and effectiveness of computing resource allocation.In addition,in order to accelerate the convergence speed of the traditional artificial fish swarm algorithm,we dynamically adjust the field of view and step size of the artificial fish,and add a twice judgment mechanism to the behavior of the artificial fish.Through the experimental results,it is not difficult to find that the computational offloading strategy proposed in this paper can reduce the task processing cost and improve the utilization rate of computing resources. |
