面向能效的无人机群移动中继服务系统分层优化研究

In the rescue and surveillance scenarios, Unmanned Aerial Vehicles (UAVs) have been a preferable choice to assist the packets transmissions from the distant user groups to their access point (AP). To prolong the UAVs flight duration as well as embrace green communication resolutions, it is of strategic significance to realize energy efficient transmissions while guarantee the delay and reliability performance. Current researches on UAV-relaying networks only focus on immediate information transmission scheme, however, it usually happens that packets arrive randomly in the practical communication scenarios while the service rates at the servers are limited. Hence, the packets are in a queue and await their service turn before being forwarded. By integrating the queue theory, information theory and layered-optimization theory, this project will perform investigations on the energy efficiency theory and energy efficient control and optimization for the multi-hop multi-queue (MHMQ) service chain networks. Specifically, we will study 1) queue length state transition of multi-dimensional Markov chain, 2) causes and effects between the energy efficiency and running mechanisms of the MHMQ queueing system, 3) energy efficient service chain control and resource allocation policies. The above researches will reveal the running mechanisms that can flexibly adjust to various channel conditions and packets arrival rates, including UAV number and flying speed, UAV service policy, the initial relaying position of UAVs, power control policy. The outcomings of the project will provide theoretical guides for the engineering designs.

在救灾、侦察等通信场景中，远处的用户群需要多级无人机中继协助将数据包前传至目标接入点。在保证传输时效性和可靠性的前提下，实现能效最优对延长无人机续航时间、响应“绿色”通信革命具有重大战略意义。目前无人机中继研究仅从信息论方向研究信息即到即传的通信场景，而实际网络中数据包往往随机到达而服务台服务速率有限，必须考虑数据包排队、等待前传的问题。本项目针对无人机充当“移动服务台”的多级—多队列服务链网络，综合排队论、信息论和分层凸优化理论，围绕能效理论及高能效优化设计开展研究，包括：1）不同中继接入方式（TDMA、NOMA）下，多维Markov链队列长度状态转换模式；2）能效与排队系统运营机制间的成因关系；3）高能效服务链控制和资源匹配机理。通过以上理论研究，揭示适变信道条件和数据包到达速率的运行机制，包括无人机数量和飞行速度控制、中继服务启动条件、中继起点位置、功率控制，为工程设计提供理论指导。

在保证传输时效性和可靠性的前提下，实现能效最优对延长无人机续航时间、响应“绿色”通信革命具有重大战略意义。本项目针对能量受限的无人机中继网络，深入展开高能效的传输策略设计和资源分层优化研究。本项目围绕能效理论及优化设计开展研究，深入挖掘和分析了能效与排队系统运营机制间的成因关系，研究了高能效服务链控制和资源匹配机理，实现了排队论、分层凸优化和博弈优化理论在高能效的无人机传输中的应用。. 申请书中设定的工作目标全部完成，包括4方面的研究工作：1）无人机中继服务系统的能效理论构建；2）无人机中继服务传输平均时延理论构建；3）无人机系统能效分层优化与无人机分层博弈；4）无人机中继群服务高能效传输编码策略设计。. 受项目支持，相关的研究成果发表在IEEE Transactions on Communications、IEEE Transactions on Vehicular Technology、IEEE Transactions on Mobile Computing等国际权威期刊和IEEE VTC等领域旗舰会议；授权发明专利4项；软件著作权2项；培养博士、硕士研究生8名。超额完成项目计划书任务。