Vehicular ad-hoc networks (VANETs) use clustering to manage data dissemination among vehicles. However, the high dynamic mobility of vehicles can lead to frequent re-clustering and decreased cluster stability. In previous work, cluster heads are selected generally based on mobility characteristics of vehicles, failing to take into account spatial and temporal dependency. Several studies have proposed cluster management mechanisms that increase cluster stability. However, in doing so these mechanisms generate excessive communication packets. In this paper, we propose CATRB, a stable clustering algorithm that uses the traffic regularity of buses while considering the mobility of vehicles such as velocity, position, and direction. In particular, we use the fixed routes of buses in urban areas as a reference index. In a regular triangle, its centroid, incenter, and circumcenter are all at the same point. Our scheme applies this idea to choose the most appropriate cluster header in VANETs. We develop an analytic model and a simulation model to evaluate the performance of CATRB. Simulation results show that the proposed scheme significantly improves cluster stability by decreasing the number of cluster head changes.

车载自组织网络（VANET）使用群集来管理车辆之间的数据分发。但是，车辆的高动态机动性可能导致频繁的重新聚类并降低群集稳定性。在先前的工作中，通常基于车辆的机动性特征来选择簇头，而没有考虑空间和时间的依赖性。几项研究提出了提高集群稳定性的集群管理机制。但是，这样做时，这些机制会生成过多的通信数据包。在本文中，我们提出了一种CATRB，一种稳定的聚类算法，该算法利用公交的交通规则性，同时考虑车辆的移动性（例如速度，位置和方向）。特别是，我们以市区公交固定路线为参考指标。在正三角形中，其质心 内心和外心都在同一点。我们的方案将这种想法应用于选择VANET中最合适的群集头。我们开发了一个分析模型和一个仿真模型来评估CATRB的性能。仿真结果表明，该方案通过减少簇头变化次数，显着提高了簇的稳定性。