Statistical Characterization of 28GHz V2X Channels via Autonomous Beam-Steered Measurements

The capabilities of the millimeter wave (mmWave) spectrum to fulfill the ultra high data rate demands of V2X (Vehicle-to-Everything) communications necessitates the need for accurate channel modeling to facilitate the efficient development of next-generation network and device design strategies. Ergo, this work describes the design of a novel fully autonomous robotic beam-steering platform, equipped with a custom broadband sliding correlator channel sounder, for 28GHz V2X propagation modeling activities on the NSF POWDER experimental testbed. The compiled datasets constitute geo-positioning logs, alignment specifics, and signal propagation measurements, along unplanned vehicular routes in urban, suburban, and foliage environments. Leveraging a closed-form design exhibiting uninhibited rotational mobility, this beam-alignment platform facilitates the collection of a continuous series of measurements, a distinct yet critical necessity for mmWave channel modeling in vehicular networks. Consequently, the calibrated and post-processed datasets enable crucial propagation analyses necessary for the efficient design and deployment of next-generation V2X networks. Specifically, this paper first studies the pathloss behavior of 28GHz signals along various routes onsite and empirically evaluates the validity of popular outdoor large-scale micro- and macro-cellular pathloss standards---namely, 3GPP TR38.901, ITU-R M.2135, METIS, and mmMAGIC. Next, analyzing the spatial autocorrelation coefficient under distance and antenna alignment accuracy effects delivers unique insights on the decoherence characteristics of 28GHz signals. In addition to shadow fading studies, this paper investigates the fading properties of the obstructed mmWave signal, in terms of its average fade depth and duration, under both static and dynamic blockages. Lastly, using the SAGE algorithm, multipath clustering analyses, centered around the Kolmogorov-Smirnov statistic, facilitate empirical validations of the favored Saleh-Valenzuela, Quasi-Deterministic, and stochastic mmWave channel models vis-à-vis cluster inter-arrival times, cluster decay attributes, and RMS delay and direction spreads.

毫米波（mmWave）频谱在满足车联网（V2X）通信的超高数据速率需求方面展现出巨大潜力，因此，精确的信道建模对于促进下一代网络和设备设计策略的高效开发至关重要。鉴于此，本研究阐述了设计一种新型完全自主的机器人波束转向平台，该平台配备有定制的宽带滑动相关器信道声纳，用于在NSF POWDER实验测试平台上进行28GHz V2X传播建模活动。所编制的数据集包括地理定位日志、对齐细节以及在城市、郊区和植被环境中的信号传播测量，覆盖了非计划的车辆路线。借助一种展现出无障碍旋转灵活性的封闭式设计，该波束对准平台便于收集一系列连续的测量数据，这对于毫米波信道在车载网络中的建模具有显著且关键的意义。因此，经过校准和后处理的数据集能够支持对下一代V2X网络进行高效设计和部署所必需的关键传播分析。具体而言，本文首先研究了现场不同路线上的28GHz信号的路径损耗行为，并经验性地评估了流行户外大规模微蜂窝和宏蜂窝路径损耗标准（即3GPP TR38.901、ITU-R M.2135、METIS和mmMAGIC）的有效性。接着，分析距离和天线对齐精度对空间自相关系数的影响，揭示了28GHz信号的去相干特性。此外，本文还研究了遮挡毫米波信号的衰落特性，包括其在静态和动态遮挡下的平均衰落深度和持续时间。最后，利用SAGE算法，围绕柯尔莫哥洛夫-斯米尔诺夫统计量进行的多径聚类分析，为Saleh-Valenzuela、准确定性和随机毫米波信道模型相对于簇到达时间、簇衰减属性以及均方根延迟和方向散布的实证验证提供了便利。