To deal with the explosive growth of mobile traffic, millimeter-wave (mmWave) communications with abundant bandwidth resources have been applied to vehicular networks. As mmWave signal is sensitive to blockage, we introduce the unmanned aerial vehicle (UAV)-aided two-way relaying system for vehicular connection enhancement and coverage expansion. How to improve transmission efficiency and to reduce latency time in such a dynamic scenario is a challenging problem. In this paper, we formulate the joint optimization problem of relay selection and transmission scheduling, aiming to reduce transmission time while meeting the throughput requirements. To solve this problem, two schemes are proposed. The first one is the random relay selection with concurrent scheduling (RCS), a low-complexity algorithm implemented in two steps. The second one is the joint relay selection with dynamic scheduling (JRDS), which fully avoids relay contentions and exploits potential concurrent ability, to obtain further performance enhancement over RCS. Through extensive simulations under different environments with various flow numbers and vehicle speeds, we demonstrate that both RCS and JRDS schemes outperform the existing schemes significantly in terms of transmission time and network throughput. We also analyze the impact of threshold selection on achievable performance.

翻译：为应对移动流量的爆炸性增长，具备丰富带宽资源的毫米波通信已应用于车载网络。由于毫米波信号易受遮挡影响，本文引入无人机辅助的双向中继系统以增强车载连接并扩展覆盖范围。如何在动态场景中提升传输效率并降低延迟是极具挑战性的问题。本文构建了中继选择与传输调度的联合优化问题，旨在满足吞吐量需求的同时最小化传输时间。针对该问题，提出两种方案：第一种是随机中继选择与并发调度（RCS）算法，这是一种两步式低复杂度算法；第二种是联合中继选择与动态调度（JRDS）算法，通过完全规避中继冲突并充分利用潜在并发能力，在RCS基础上进一步提升性能。通过在不同流量数量及车辆速度的多种环境下进行大量仿真，结果表明RCS与JRDS方案在传输时间与网络吞吐量方面均显著优于现有方案。此外，本文还分析了阈值选择对可实现性能的影响。