| 田文* **,刘敏* **,孙胜*,陈亚丽*,王煜炜*,曹宁***.基于协作通信的无人机集群网络连通性恢复机制[J].高技术通讯(中文),2026,36(6):564~575 |
| 基于协作通信的无人机集群网络连通性恢复机制 |
| Cooperative communication based connectivity recovery for UAV swarm networks |
| |
| DOI：10. 3772 / j. issn. 1002 - 0470. 2026. 06. 002 |
| 中文关键词: 无人机； 无人机集群网络； 协作通信； 连通性恢复 |
| 英文关键词: unmanned aerial vehicle, unmanned aerial vehicle swarm networks, cooperative communication, connectivity recovery |
| 基金项目: |
| 作者 | 单位 | | 田文* ** | (*中国科学院计算技术研究所北京 100190)
(**中国科学院大学北京 101408)
(***中国电信天翼视联科技有限公司杭州 310023） | | 刘敏* ** | | | 孙胜* | | | 陈亚丽* | | | 王煜炜* | | | 曹宁*** | |
|
| 摘要点击次数: 85 |
| 全文下载次数: 71 |
| 中文摘要: |
| 无人机（unmanned aerial vehicle，UAV）集群网络由于节点高动态性和链路的不稳定性，常常被割裂为多个无法互相通信的分区。因此，网络连通性恢复是无人机领域的一个重要问题。现有的解决方案需要无人机节点长距离移动以恢复通信连接，导致时间和能量开销大。本文研究了利用协作通信技术提高无人机网络连通性恢复效率的问题。为此，设计了一种基于协作通信的无人机网络连通性恢复算法（cooperative communication based connectivity recovery algorithm for UAV swarm networks, C3RUN）。该算法利用协作通信扩大无人机节点的通信范围，并将无人机节点移动至更合适的位置以进一步提升协作通信的性能，从而实现网络连通性的快速恢复。最后，本文进行仿真实验来评估C3RUN的性能，结果表明，与基于图卷积神经网络和虚拟势场的2种连通性恢复算法相比，C3RUN在恢复时间上分别加快了24.1%和60.5%，在节点平均移动距离上分别减少了19.2%和54.4%。 |
| 英文摘要: |
| Unmanned aerial vehicle （UAV） swarm networks are often divided into multiple separated clusters due to the high dynamics of nodes and the instability of links. As a result, network connectivity recovery is an important issue in this area. Existing solutions often require UAV nodes to move long distances to restore communication connections, resulting in significant time and energy costs. In this paper, we study the issue of utilizing cooperative communication technology to improve the connectivity recovery efficiency in UAV networks. Then, we design a cooperative communication based connectivity recovery algorithm for UAV swarm networks, named C3RUN. The algorithm utilizes cooperative communication to expand the communication range of UAV nodes and moves them to more suitable positions to further improve the performance of cooperative communication, thereby achieving rapid recovery of network connectivity. Finally, we conduct extensive simulations to evaluate the performance of C3RUN. Simulation results reveal that compared with two connectivity recovery algorithms based on graph convolutional neural networks and virtual potential fields, the C3RUN speeds up the recovery time by 24.1% and 60.5% respectively, and reduces the average moving distance of nodes by 19.2% and 54.4% respectively. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
