摘要
在对抗环境下,捕获无人机通信网络的通联拓扑有助于我们高效发现并破坏其集群功能。然而,在非合作条件下,传统的拓扑先验信息难以获取,通联拓扑推理面临着巨大的挑战。现有相关研究总体上仍处于起步阶段,系统模型和推理机理不清晰,各类方法在同一数据维度下的对比较少。因此,针对非合作的物理场景,该文首先构建了系统模型,揭示了推理机理。然后,分别对相关性、格兰杰因果、转移熵和多维霍克斯过程4种方法进行了仿真对比分析。最后,对该研究方向的发展前景进行了展望。
In confrontational environment,capturing the communication topology of the Unmanned Aerial Vehicles(UAV)communication network helps us to discover efficiently and destroy its cluster function.However,under non-cooperative conditions,the traditional priori information of topology is difficult to obtain,and communication topology inference faces huge challenges.Existing related research is still in its infancy as a whole,the system model and inference mechanism are not clear,and the comparison of various methods in the same data dimension is also rare.Therefore,for the non-cooperative physical scene,firstly the system model is constructed and the inference mechanism is revealed.Then,the four methods of correlation,Granger causality,transfer entropy and multidimensional Hawkes process are simulated and compared.Finally,the prospects for the development of this research direction are prospected.
作者
宋叶辉
丁国如
徐承龙
孙佳琛
汤鹏
SONG Yehui;DING Guoru;XU Chenglong;SUN Jiachen;TANG Peng(College of Communications,Army Engineering University of PLA,Nanjing 210007,China)
出处
《电子与信息学报》
EI
CSCD
北大核心
2022年第3期924-939,共16页
Journal of Electronics & Information Technology
基金
国家自然科学基金(61871398,61901520,61931011,U20B2038,62171462)
江苏省自然科学基金杰出青年项目(BK20190030)。
关键词
无人机通信网络
通联拓扑推理
格兰杰因果
转移熵
霍克斯过程
UAV communication network
Communication topology inference
Granger causality
Transfer entropy
Hawkes process
