基于新型组网的无人机巡检任务处理优化方案

Task Processing Optimization Scheme of Unmanned Aerial Vehicle Inspection Based on New Networking

考虑用户移动性,特别是在无人机巡检场景下,由于多接入边缘计算(Multi-access Edge Computing, MEC)节点服务范围小,大概率发生任务迁移。任务迁移将增加任务处理的时延和成本。为了降低任务迁移概率,提出了一种新的MEC组网架构——MEC POOL。在此基础上,构建了MEC POOL组网架构下的任务处理效益模型,将任务处理效益最优问题转换成受限条件下最优解问题。为了解决上述最优解问题,设计了一种基于任务处理效益最优的粒子群算法。实验结果表明,在发生任务迁移的场景下,MEC POOL组网方案相比MEC传统组网方案,任务处理效益和任务处理时延效益均可提升10%以上。对于多用户场景,平均任务处理效益提升8%以上,平均任务处理时延减少10%以上。

Considering user mobility,especially in the scenario of unmanned aerial vehicle inspection,it is likely that task migration will occur due to the small service range of Multi-access Edge Computing(MEC)node.Task migration will increase task processing time delay and cost.To reduce the probability of task migration,a new MEC structure-MEC POOL is proposed.On this basis,a task processing benefit model of MEC POOL structure is constructed.The problem of optimal task processing benefit is converted into the problem of optimal solution under restricted conditions.In order to solve the above optimal solution problem,a particle swarm optimization algorithm based on task processing benefit optimization is designed.The experimental results show that the task processing benefit and task processing delay benefit of the MEC POOL scheme are improved by more than 10%compared to the traditional MEC scheme in the scenario of task migration.For multi-user scenario,the average task processing benefit is improved by more than 8%and the average task processing time delay is reduced by more than 10%.