UAV ground-to-air channel prediction during blockages using six dimensional channel data

Ground-to-air (GA) communication using unmanned aerial vehicles (UAVs) has gained popularity in recent years and is expected to be part of 5G networks and beyond. However, the GA links are susceptible to frequent blockages at millimeter wave (mmWave) frequencies. During a link blockage, the channel information cannot be obtained reliably. In this work, we provide a novel method of channel prediction during the GA link blockage at 28 GHz. In our approach, the multipath components (MPCs) along a UAV flight trajectory are arranged into independent path bins based on the minimum Euclidean distance among the channel parameters of the MPCs. After the arrangement, the channel parameters of the MPCs in individual path bins are forecasted during the blockage. An autoregressive model is used for forecasting. The results obtained from ray tracing simulations indicate a close match between the actual and the predicted mmWave channel.

近年来，地面至空中（GA）通信采用无人机（UAVs）的方式日益受到青睐，并被预期将成为第五代（5G）及以后网络技术的重要组成部分。然而，在毫米波（mmWave）频段，GA链路容易遭受频繁的阻断。在链路阻断期间，无法可靠地获取信道信息。在本研究中，我们提出了一种在28 GHz频段GA链路阻断期间进行信道预测的新方法。在我们的方法中，根据无人机飞行轨迹上的多径分量（MPCs）的信道参数之间的最小欧几里得距离，将这些MPCs排列成独立的路径分箱。排列完成后，在阻断期间对各个路径分箱中的MPCs信道参数进行预测。我们使用自回归模型进行预测。来自光线追踪模拟的结果表明，实际与预测的毫米波信道之间存在紧密匹配。