无人机辅助物理层安全下的保密性能优化

信息安全是影响物联网(IoT)应用的关键因素之一，物理层安全是解决物联网信息通信安全问题的有效技术。该文针对物联网中带有主动攻击的全双工窃听者，利用无人机(UAV)辅助发射人工噪声的方法，提升系统物理层安全性能。为了跟踪窃听者位置移动，首先采用贝叶斯测距和最小二乘法迭代估计窃听者位置，然后提出基于Q-learning的无人机轨迹优化算法，以达到在窃听者移动情况下系统保密性能最优。仿真结果表明，该算法能快速收敛，并且无人机能够跟踪窃听者移动来确定自身最佳位置，对窃听信道实施有效干扰，从而保证系统可达安全速率最大。

Secrecy Performance Optimization of Unmanned Aerial Vehicle -aided Physical Layer Security

Physical layer security is effective to solve the problem of communication security of Internet of Things (IoT). As a full duplex eavesdropper with active attack and passive eavesdropping exists in the IoT, Unmanned Aerial Vehicle (UAV) transmits artificial noise to the eavesdropper for improving secrecy performance in this paper. Based on estimating eavesdropper position, a trajectory optimization algorithm based on Q-learning is proposed to track the eavesdropper mobility and obtain the optimal secrecy performance. The results show that the proposed algorithm converges quickly. UAV can track the eavesdropper mobility to determine the best position of jamming the eavesdropping channel, which guarantees maximum achievable secrecy rate.