Blockage is a substantial problem at millimeter wave (mmWave) frequencies, much more so than sub-6 GHz. Blockage events caused by objects such as cars and humans occur quickly and have substantial attenuation; typical attenuations in this work range from 11 to 22 dB, with a maximum of 32 dB. This problem is overcome with spatial diversity. During blockage events, mmWave transceivers can beamform towards an unblocked path. In this paper we explore this idea in depth: if the primary path is blocked, which alternate paths will be available? To answer this question, we have built a measurement system that uses 60 GHz phased arrays to rapidly obtain spatially resolved measurements of the channel. Scans are repeated every 3.2 ms, allowing us to observe blockage events that occur on the different paths in the channel. The design of this system is presented, and a description is given of three human-body blockage measurement campaigns. Paths are assigned states of blocked/unblocked, and each measurement is expressed as a time series of states which specify the status of each path. The evolution between states is described with a Markov model, and the measurement results are used to derive transition probabilities between states.

中文翻译：

---

基于测量的室内毫米波阻塞模型

在毫米波 (mmWave) 频率下，阻塞是一个严重问题，比低于 6 GHz 的频率要严重得多。汽车、人等物体造成的堵塞事件发生迅速，衰减较大；该工作的典型衰减范围为 11 到 22 dB，最大为 32 dB。空间多样性克服了这个问题。在阻塞事件期间，毫米波收发器可以朝向畅通路径进行波束成形。在本文中，我们深入探讨了这个想法：如果主路径被阻塞，哪些备用路径可用？为了回答这个问题，我们构建了一个测量系统，该系统使用 60 GHz 相控阵来快速获得通道的空间分辨测量。每 3.2 毫秒重复一次扫描，使我们能够观察通道中不同路径上发生的阻塞事件。介绍了该系统的设计，并描述了三个人体阻塞测量活动。路径被分配了阻塞/未阻塞状态，并且每个测量都表示为状态的时间序列，这些状态指定了每个路径的状态。状态间的演化用马尔可夫模型描述，测量结果用于推导状态间的转移概率。