Fluid-Antenna-aided AAV Secure Communications in Eavesdropper Uncertain Location

For autonomous aerial vehicle (AAV) secure communications, traditional designs based on fixed position antenna (FPA) lack sufficient spatial degrees of freedom (DoF), which leaves the line-of-sight-dominated AAV links vulnerable to eavesdropping. To overcome this problem, this paper proposes a framework that effectively incorporates the fluid antenna (FA) and the artificial noise (AN) techniques. Specifically, the minimum secrecy rate (MSR) among multiple eavesdroppers is maximized by jointly optimizing AAV deployment, signal and AN precoders, and FA positions. In particular, the worst-case MSR is considered by taking the channel uncertainties due to the uncertainty about eavesdropping locations into account. To tackle the highly coupled optimization variables and the channel uncertainties in the formulated problem, an efficient and robust algorithm is proposed. Particularly, the uncertain regions of eavesdroppers, whose shapes can be arbitrary, are disposed by constructing convex hull. In addition, two movement modes of FAs are considered, namely, free movement mode and zonal movement mode, for which different optimization techniques are applied, respectively. Numerical results show that, the proposed FA schemes boost security by exploiting additional spatial DoF rather than transmit power, while AN provides remarkable gains under high transmit power. Furthermore, the synergy between FA and AN results in a secure advantage that exceeds the sum of their individual contributions, achieving a balance between security and reliability under limited resources.

A Learning Aided Gradient Descent for MISO Beamforming

This paper proposes a learning aided gradient descent (LAGD) algorithm to solve the weighted sum rate (WSR) maximization problem for multiple-input single-output (MISO) beamforming. The proposed LAGD algorithm directly optimizes the transmit precoder through implicit gradient descent based iterations, at each of which the optimization strategy is determined by a neural network, and thus, is dynamic and adaptive. At each instance of the problem, this network is initialized randomly, and updated throughout the iterative solution process. Therefore, the LAGD algorithm can be implemented at any signal-to-noise ratio (SNR) and for arbitrary antenna/user numbers, does not require labelled data or training prior to deployment. Numerical results show that the LAGD algorithm can outperform of the well-known WMMSE algorithm as well as other learning-based solutions with a modest computational complexity. Our code is available at https://github.com/XiaGroup/LAGD.