arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2502.01003eess.SP
收藏

用于安全无人机网络的近场集成感知与通信

Near-Field Integrated Sensing and Communications for Secure UAV Networks

Jingjing Zhao, Songtao Xue, Kaiquan Cai, Xidong Mu, Yuanwei Liu, Yanbo Zhu

AI总结:

提出一种近场集成感知与通信框架,通过近场定位与轨迹跟踪实时估计窃听无人机状态,并联合优化波束成形与轨迹设计最大化保密速率,实现高时间效率的安全无人机网络。

摘要:

提出了一种新型近场集成感知与通信框架,用于具有高时间效率的安全无人机(UAV)网络。一个大孔径地面基站(GBS)在存在一个窃听无人机(E-UAV)的情况下与一个通信无人机(C-UAV)通信,其中人工噪声(AN)被用于干扰和感知目的。鉴于E-UAV的运动模型在GBS处未知,我们首先提出了一种近场定位和轨迹跟踪方案。具体而言,利用近场空间域上变化的Doppler频移观测,从回波信号中估计E-UAV的三维(3D)速度。为了及时纠正位置预测误差,采用扩展Kalman滤波器(EKF)融合预测状态和测量状态。随后,基于E-UAV的实时预测位置,我们进一步提出了一种联合GBS波束成形和C-UAV轨迹设计方案,以最大化瞬时保密速率,同时保证感知精度约束。为了解决由此产生的非凸问题,开发了一种交替优化方法,其中通过利用逐次凸逼近方法迭代求解近场GBS波束成形和C-UAV轨迹设计子问题。最后,我们的数值结果表明:1)通过利用近场球面波传播,我们提出的框架可以实时准确估计E-UAV的3D速度和位置;2)与基准方案相比,所提框架实现了优越的保密速率,并接近E-UAV轨迹完全已知时的性能。

英文摘要:

A novel near-field integrated sensing and communications framework for secure unmanned aerial vehicle (UAV) networks with high time efficiency is proposed. A ground base station (GBS) with large aperture size communicates with one communication UAV (C-UAV) under the existence of one eavesdropping UAV (E-UAV), where the artificial noise (AN) is employed for both jamming and sensing purpose. Given that the E-UAV's motion model is unknown at the GBS, we first propose a near-field localization and trajectory tracking scheme. Specifically, exploiting the variant Doppler shift observations over the spatial domain in the near field, the E-UAV's three-dimensional (3D) velocities are estimated from echo signals. To provide the timely correction of location prediction errors, the extended Kalman filter (EKF) is adopted to fuse the predicted states and the measured ones. Subsequently, based on the real-time predicated location of the E-UAV, we further propose a joint GBS beamforming and C-UAV trajectory design scheme for maximizing the instantaneous secrecy rate, while guaranteeing the sensing accuracy constraint. To solve the resultant non-convex problem, an alternating optimization approach is developed, where the near-field GBS beamforming and the C-UAV trajectory design subproblems are iteratively solved by exploiting the successive convex approximation method. Finally, our numerical results unveil that: 1) the E-UAV's 3D velocities and location can be accurately estimated in real time with our proposed framework by exploiting the near-field spherical wave propagation; and 2) the proposed framework achieves superior secrecy rate compared to benchmark schemes and closely approaches the performance when the E-UAV trajectory is perfectly known.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑