FANET中时延感知的跨层优化方法  

Delay-aware cross-layer optimization method for FANET

作　　者：文少杰 黄传河[1] WEN Shaojie;HUANG Chuanhe(Computer School,Wuhan University,Wuhan 430072,China)

机构地区：[1]武汉大学计算机学院,湖北武汉430072

出　　处：《通信学报》2018年第4期1-12,共12页Journal on Communications

基　　金：国家自然科学基金资助项目(No.61373040;No.61572370)~~

摘　　要：针对FANET中的实时路由、速率分配和功率控制问题,提出异步的分布式跨层优化(ADCO)方法。为了解决所提问题,首先设计时延约束的跨层优化框架,然后利用拉格朗日松弛和对偶分解方法把联合优化问题分解为几个复杂度较低的子问题。ADCO允许每一个中继节点仅通过局部信息来完成对不同子问题的优化,同时,中继节点利用异步更新机制实现对偶变量的更新。仿真结果表明,所提方法有效地增加了吞吐量,减少了数据分组的超时率和功率消耗。An asynchronous distributed cross-layer optimization(ADCO)method was proposed to solve the problem of jointly considering real-time routing,rate allocation and power control in FANET(flying ad hoc network).And a delay-constrained cross-layer optimization framework was designed to formally represent proposed problem.Then Lagrangian relaxation and dual decomposition methods was used to divide joint optimization problem into several sub-problems.ADCO allowed each relay node to perform the optimization operation for different sub-problems with local information,and the relay nodes could update the dual variables based on asynchronous update mechanism.The simulation results show that the proposed algorithm can improve the network performance effectively in terms of energy efficiency,packet timeout ratio and network throughput.

关 键 词：FANET 时延感知 跨层优化 对偶分解 异步更新 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]