A Neighbor Discovery Algorithm for UAV Networking Based on Directional Antennas

一种基于定向天线的蜂群组网邻居发现算法

¹ School of Cyberspace Security, Northwestern Polytechnical University, Xi'an 710072, China
² Science and Technology on Communication Networks Laboratory, Shijiazhuang 050081, China

Received: 12 March 2019

Abstract

Using directional antennas in Unmanned Aerial Vehicle (UAV) swarm has many advantages, such as longer transmission range, spatial reuse, anti-jamming and low probability of intercept. Neighbor discovery is a crucial step in the initialization of UAV networking. We introduce a neighbor discovery algorithm based on iterative common neighbors (ICN-ND), which can reduce the time of neighbor discovery process and the networking delay. To avoid the huge exchange-data quantity, we use the orientation and distance to represent the neighbor's location instead of latitude and longitude. Compared with the scan-based algorithm, the ICN-ND has better performance on the time to complete the neighbor discovery process and the convergence speed, which is validated the practicality by QualNet simulations.

摘要

采用定向天线进行蜂群组网，具有空间复用度高、信号传输距离远、抗干扰以及低截获的先天优势。无人机蜂群在编队前期，需要快速发现相邻节点进行组网，因此邻居发现是组网的必要前提，对MAC层和网络层的相关设计有着重要的影响。针对采用定向天线的组网模式，在基于扫描方式的邻居发现规划型算法（scan-base algorithm-deterministic，SBA-D）的基础上，提出了基于邻居交集迭代发现的方案（neighbor discovery algorithm based on iterative common neighbors，ICN-ND），充分利用已知的邻居信息，在相邻节点之间寻找邻居集合中的交集，利用公共邻居来提高邻居发现的效率，加快邻居发现的过程，进而降低无人机前期组网的时延，此外为了降低邻居信息交互的数据量，对表征邻居位置的数据结构进行了优化。最后的仿真实验表明，在不同节点密度和不同天线波束宽度下，ICN-ND算法收敛速度以及发现全部邻居所需时隙数远远优于SBA-D。