In this article, we study the secure communication in unmanned aerial vehicle (UAV) aided and reconfigurable intelligent surface (RIS) empowered multiuser networks. In the proposed configuration, we consider UAV-mounted RIS to reflect the signal between the base station and the legitimate users in the presence of eavesdroppers. The problem is formulated as secrecy rate maximization with channel uncertainty constraints, which is nonconvex. Therefore, we decouple it into beamforming, phase shift, and UAV trajectory subproblems which are jointly optimized. For the beamforming and phase shift subproblems, we apply S-procedure and general sign-definiteness to transform them into tractable forms. And for the UAV trajectory, we unroll the constraints and apply first-order Taylor transform and the difference of two convex functions (DC) to obtain a relaxed convex problem that can be numerically solved. The proposed solution show effectiveness and improved the secrecy rate of the network.

Energy Efficiency (EE) has become an essential metric in Device-to-Device (D2D) communication underlaying Unmanned Aerial Vehicles (UAVs) Among the several technologies that provide significant energy, simultaneous wireless information and power transfer (SWIPT) has been proposed as a promising solution to improve EE. However, it is a challenging task to study the EE under nonlinear energy harvesting (EH) due to the limited sensitivity and the composition of the non-linear circuit. Moreover, when D2D users transmit information using the EH from UAVs, interferences to cellular users occur and deteriorate the throughput. To tackle these problems, we leverage concepts from artificial intelligence (AI) to optimize EE of UAV-assisted D2D communication. Specifically, multi-agent deep reinforcement learning was proposed to jointly maximize throughput and EE, where the reward function is defined in terms of the introduced goal. Simulation results verify the supremacy of proposed approach over traditional algorithms.

Integrating unmanned aerial vehicles (UAVs) as the base station is a recent progress and a promising solution to assist cellular networks to improve coverage. In this paper, an analytical framework is proposed to analyze coverage probability and capacity, using stochastic geometry. We incorporate the air-to-ground (A2G), and two-piece path loss model that undergoes millimeter wave (mmWave) channel fading in both line-of-site (LOS) and non-line-of-site (NLOS). Furthermore, we assume that all users’ equipment’s (UEs) are clustered around Terrestrial Base Stations (TBS) and UAVs, according to Poisson Cluster Process (PCP), i.e., Matern cluster process. We also propose Multiple-Input-Multiple-Output (MIMO) transmission where multiple single-antenna UAVs simultaneously communicate with TBS. Initially, we obtain expressions for association probability with each tier. After that, we derive downlink coverage expression based on signal-to-interference-plus-noise-ratio (SINR), by assuming 3D directional beamforming with UAV height being not fixed. Numerical results demonstrate the efficiency and accuracy of the proposed approach.

Unmanned aerial vehicles (UAV) have been applied in many civilian and commercial applications, including agriculture mapping, military surveillance, reconnaissance activities, and wireless communication. Due to the exponential increase in traffic demand and coverage hole problem, UAVs can be immediately deployed in the network to provide maximum wireless coverage and improve the quality of service (QoS) for users. However, to maintain communication while moving, the UAVs harvest energy from the RF signal of base stations in the ground. In this context, we propose a directional wireless power transfer (WPT) for charging UAVs by applying multi-beam energy in this letter. Two sub-problems are considered in our optimization problem which are energy efficiency (EE) and throughput maximization while taking into account interference from adjacent UAVs. Thus, to solve the formulated problem a deep deterministic policy gradient (DDPG) algorithm is implemented and compared to other deep reinforcement learning (DRL) methods.