Outdoor mmWave Channel Propagation Models using Clustering Algorithms
Outdoor mmWave Channel Propagation Models using Clustering Algorithms
复制标题
DOI：
10.1109/icnc47757.2020.9049734
复制
发表时间：
2020-02
期刊：
影响因子：
--
通讯作者：
B. Antonescu;Miead Tehrani Moayyed;S. Basagni
中科院分区：
文献类型：
--
作者：
B. Antonescu;Miead Tehrani Moayyed;S. Basagni
本文关注的任务，生成更简单，但准确的毫米波信道模型的基础上聚类到达接收机的所有多径分量。我们的工作重点是28 GHz的通信在城市室外场景模拟的光线跟踪工具。我们调查的有效性k-means和k-power-means聚类算法预测的最佳数量的集群，通过使用集群有效性指数（CVI）和得分融合技术。我们的研究结果显示了这些技术的联合使用如何生成毫米波大尺度和小尺度信道模型的精确近似，大大简化了在任何接收器位置分析大量射线的复杂性。
This paper concerns the task of generating simpler yet accurate mmWave channel models based on clustering all multipath components arriving at the receiver. Our work focuses on 28 GHz communications in urban outdoor scenarios simulated with a ray-tracer tool. We investigate the effectiveness of k-means and k-power-means clustering algorithms in predicting the optimal number of clusters by using cluster validity indices (CVIs) and score fusion techniques. Our results show how the joint use of these techniques generate accurate approximation of the mmWave large-scale and small-scale channel models, greatly simplifying the complexity of analyzing large amount of rays at any receiver location.