Joint trajectory design and power allocation for unmanned aerial vehicles aided secure transmission in the presence of no-fly zone
Joint trajectory design and power allocation for unmanned aerial vehicles aided secure transmission in the presence of no-fly zone
复制标题
DOI：
10.1049/cmu2.12326
复制
发表时间：
2022-01
期刊：
影响因子：
--
通讯作者：
Guo Yin;Xiaotian Zhou;Piming Ma;Fang Fang-Fang;Peng Zhao
中科院分区：
文献类型：
--
作者：
Guo Yin;Xiaotian Zhou;Piming Ma;Fang Fang-Fang;Peng Zhao
无人机（UAV）由于其高移动性和灵活部署等优点而被广泛认为是未来无线网络的关键推动者。本文研究了无人机辅助保密通信系统，其中无人机作为移动的干扰机，以防止窃听者偷听机密信息。以保密率最大化为目标，建立了无人机轨迹、发射功率和源节点发射功率的联合优化问题。此外，还考虑了禁飞区（NFZ）和不完美估计对窃听者位置的影响。由于原始问题是难以跟踪的，最坏情况下的保密率（WCSR）的假设，首先绕过估计误差带来的不确定性。然后提出了一种基于块坐标下降（BCD）的算法，将问题分解为三个子问题，通过迭代的方式得到无人机的轨迹、无人机的发射功率和源的发射功率。仿真结果表明，该算法可以显着提高保密性能。此外，该算法在估计误差下的鲁棒性也得到了验证。
Unmanned aerial vehicles (UAVs) are widely considered as key enablers for future wireless networks due to their advantages, such as high mobility and flexible deployment. In this paper, the UAV assisted secure communication system is investigated, where the UAV is deployed as the mobile jammer to prevent the eavesdropper from overhearing the confidential message. With the objective of maximizing the secrecy rate, a joint optimization problem involving the trajectory and transmit power of UAV, as well as the transmit power of source node is formulated. Moreover, the effects of Non‐Fly Zone (NFZ) and imperfect estimation on the location of eavesdropper are also taken into consideration. As the original problem is hardly trackable, the worst case secrecy rate (WCSR) assumption is first employed to bypass the uncertainty brought by the estimation error. Then a block coordinate descent (BCD) based algorithm is proposed to decompose the problem into three sub‐ones, where the trajectory of UAV, the transmit power of UAV and the transmit power of source can be obtained in an iterative manner. Simulation results reveal that the proposed algorithm can improve the secrecy performance significantly. In addition, the robustness of the proposed algorithm under the estimation error can also be verified.