This paper proposes a mobile unmanned aerial vehicle (UAV)-enabled jamming scheme under the probabilistic line-of-sight (LoS) channel model to improve the secrecy of ground wiretap channels in urban areas, where a friendly UAV is deployed to cooperatively transmit jamming signals to confuse the suspicious ground eavesdroppers. Our goal is to maximize the average (expected) secrecy rate by jointly optimizing the user scheduling, the source's transmit power, the UAV's jamming power and its three-dimensional (3D) trajectory for a given flight time. Since the expected secrecy rate is highly complicated with respect to the 3D UAV trajectory, we derive a more tractable lower bound for it. Nevertheless, the resulting optimization problem is still non-convex and difficult to solve optimally due to the imperfect location information of the eavesdroppers. To tackle such an intractable problem, we first derive a worst-case (expected) secrecy rate, and then we propose an efficient iterative algorithm to obtain a suboptimal solution to it by applying the block coordinate descent (BCD) and successive convex approximation (SCA) techniques. Simulation results show that the proposed algorithm under the probabilistic LoS channel model significantly outperforms various benchmark algorithms.

本文提出了一种基于概率视距（LoS）信道模型的移动无人机（UAV）启用干扰方案，以提高城市地区地面窃听信道的保密性，其中部署友好的无人机以协同传输干扰信号以迷惑可疑的地面窃听者。我们的目标是通过联合优化用户调度、源的发射功率、无人机的干扰功率及其给定飞行时间的三维 (3D) 轨迹来最大化平均（预期）保密率。由于相对于 3D 无人机轨迹而言，预期保密率非常复杂，因此我们为其推导出了一个更易于处理的下限。尽管如此，由于窃听者的位置信息不完善，由此产生的优化问题仍然是非凸的并且难以最优解决。为了解决这样一个棘手的问题，我们首先推导出最坏情况（预期）保密率，然后我们提出了一种有效的迭代算法，通过应用块坐标下降（BCD）和逐次凸逼近（SCA）来获得次优解) 技术。仿真结果表明，该算法在概率视距信道模型下明显优于各种基准算法。然后我们提出了一种有效的迭代算法，通过应用块坐标下降 (BCD) 和连续凸逼近 (SCA) 技术来获得次优解。仿真结果表明，该算法在概率视距信道模型下明显优于各种基准算法。然后我们提出了一种有效的迭代算法，通过应用块坐标下降 (BCD) 和连续凸逼近 (SCA) 技术来获得次优解。仿真结果表明，该算法在概率视距信道模型下明显优于各种基准算法。