Wireless mesh networking using millimeter-Wave (mmWave) bands provides a promising option for connecting densely deployed small base stations (BSs) to the core network. To fully exploit the potential of both wireless mesh networking and mmWave communications, hybrid beamforming is indispensable to help maintain connections with multiple neighbors, while providing directional antenna gain for BSs in mmWave mesh backhaul networks (MMBNs). Hybrid beamforming can form multiple directional links and each consists of a number of antenna subarrays. In this paper, we study the antenna subarray management problem for hybrid beamforming in MMBNs. To this end, a joint static and dynamic subarray scheduling (SDSS) scheme is developed to simultaneously achieve high-throughput performance and fluctuation-adaptive capability for each backhaul link. Specifically, static subarrays are jointly allocated with transmit power on a large time scale. Based on the static subarray allocation, dynamic subarrays are scheduled on a small time scale. The effectiveness of the proposed scheme is verified via extensive simulations. Compared to other nonadaptive schemes, our scheme reduces the total power consumption and packet delivery delay significantly while satisfying the capacity demand of each backhaul link.

使用毫米波（mmWave）频段的无线网状网络为将密集部署的小型基站（BS）连接到核心网络提供了一个有希望的选择。为了充分利用无线网状网络和mmWave通信的潜力，混合波束成形必不可少，以帮助维持与多个邻居的连接，同时为mmWave网状回传网络（MMBN）中的BS提供定向天线增益。混合波束成形可以形成多个方向链接，每个方向链接都由许多天线子阵列组成。在本文中，我们研究了MMBN中混合波束成形的天线子阵列管理问题。为此，静态和动态子数组联合调度（SDSS）方案旨在为每个回程链路同时实现高吞吐量性能和自适应波动能力。具体而言，静态子阵列在较大的时间尺度上与发射功率联合分配。基于静态子阵列分配，可以在较小的时间范围内调度动态子阵列。通过广泛的仿真验证了所提方案的有效性。与其他非自适应方案相比，我们的方案在满足每个回程链路的容量需求的同时，显着降低了总功耗和数据包传递延迟。