In low-altitude Internet of Things (IoT) scenarios, achieving integrated sensing and communication (ISAC) with multi-base station (BS) cooperation is essential for building intelligent air–ground networks. In such system, user fairness becomes increasingly important, especially in heterogeneous environments with diverse channel conditions and service demands. However, incorporating rate-splitting multiple access (RSMA) into multi-user MIMO system presents new challenges for fairness-aware resource allocation, due to the coupling between communication and sensing signals, the high dimensionality of optimization variables, and the non-convexity of the joint design. This paper proposes a fairness-oriented alternating optimization framework for coordinated precoding in multi-BS RSMA-enabled ISAC system. The objective is to maximize the minimum common stream rate across users to enhance fairness, while meeting private stream quality of service (QoS) and sensing performance constraints. The precoding subproblem is convexified using semidefinite relaxation (SDR) and the Schur complement, while the reconfigurable intelligent surface (RIS) phase optimization is performed on the complex unit-modulus manifold via the Riemannian conjugate gradient (RCG) method. Numerical results demonstrate that the proposed method effectively ensures fairness, achieves a well-balanced trade-off between sensing and communication, and improves the overall user rate compared to conventional multiple access schemes.

Unmanned aerial vehicles (UAVs) have been widely employed for emergency communications in Internet of Things (IoT) networks. The temporal freshness of data in disaster rescue, represented by the Age of Information (AoI), holds critical significance. Due to UAVs' limited energy capacity, it is crucial to optimize their energy consumption while ensuring effective assistance in the IoT system's fresh data collection tasks. Therefore, our work sets out to minimize the AoI and energy consumption to meet time-sensitive and priority-aware demands in resource-constrained environments. We formulate a multi-objective optimization problem by jointly optimizing UAV flight trajectory, IoT device transmitting power, and UAV radio frequency (RF) power, with constraints such as IoT devices' priority levels and UAVs' locations. This problem is modeled as a Markov Decision Process (MDP), and we propose a novel reinforcement learning (RL)-based method, termed the MultiAgent Data Collection Energy Transmitting Scheme (MADCET), to solve it. Extensive simulations demonstrate that the proposed scheme significantly outperforms benchmarks by reducing AoI, satisfying priority demands, minimizing energy consumption, and achieving good convergence.

With the explosive growth of user data in 6G networks, existing infrastructures face significant scalability and latency challenges. Mobile Edge Computing (MEC) partially alleviates these issues by deploying computational resources closer to users, but still struggles to fully meet the growing demands. Re-configurable Intelligent Surfaces (RIS) enhance communication performance by improving channel quality. However, optimizing resource allocation in RIS-aided MEC systems remains a critical challenge due to the complexity of real-time optimization of multiple parameters. Although Deep Reinforcement Learning (DRL) methods like Deep Deterministic Policy Gradient (DDPG) have been applied, they often suffer from Q-value overestimation and instability, resulting in suboptimal performance in dynamic environments. This paper focuses on a single-cell RIS-aided MEC network where multiple user devices offload computational tasks to an edge server. We propose an optimized resource allocation scheme using an improved Distributed Soft Actor-Critic (DSAC-T) algorithm. This approach jointly optimizes power control, computation offloading, edge computing resource allocation, and RIS phase shifts, aiming to minimize total offloading delay to ensure real-time service requirements. Simulation results demonstrate that DSAC-T outperforms multiple baseline methods (e.g., DDPG and SAC) in reducing offloading delay by 46.39% and 16.70%, respectively, while significantly enhancing system stability and convergence speed.

Integrated sensing and communication (ISAC)-enabled low-altitude Internet of Things (IoT) networks hold significant potential for applications in smart cities and emergency communication systems. However, achieving secure and energy-efficient communication under complex environments, particularly in the presence of the mobile full-duplex eavesdropper (MFDE), presents significant challenges. This study investigates the optimization of secure rate energy efficiency (SREE) in ISAC-enabled low-altitude IoT networks, where the problem is further complicated by the strong coupling between unmanned aerial vehicles (UAV) trajectory design, power allocation, and artificial noise (AN) generation, leading to an optimization issue marked by significant dimensionality and a lack of convexity. To tackle this challenge, a power cost factor-based Twin Delayed Deep Deterministic Policy Gradient (CTD3) algorithm is developed, which incorporates a game-theoretic power allocation strategy into the TD3 framework to efficiently handle the high-dimensional coupled optimization problem. The algorithm reformulates part of the high-dimensional continuous optimization process into a strategy interaction problem and introduces a power cost factor into the utility function, effectively reducing the dimensionality of optimization variables and the overall computational burden. Furthermore, an adaptive dynamic sensing mechanism is introduced to enhance resource utilization while effectively countering the dynamic behavior of eavesdroppers. The effectiveness of the proposed strategy in enhancing SREE performance amidst environmental uncertainties is validated through extensive simulations, where it consistently outperforms baseline methods.

With the advancement of unmanned aerial vehicles (UAVs) technology, UAV-assisted cellular networks (UACNs) have emerged as a new communication paradigm aimed at enhancing the coverage and capacity of ground networks. Unfortunately, the limited energy capacity of UAVs significantly restricts their operational duration, so optimizing energy efficiency is of importance. However, existing optimization schemes often overlook the impact of ground user mobility on user association, lacking ability to achieve optimal energy efficiency. In this paper, the K-Means method is applied to optimize user association by periodically clustering users. Additionally, given the dynamic nature of the wireless channels, we utilize the Multi-Agent Twin Delayed Deep Deterministic Policy Gradient (MATD3) approach to jointly optimize 3D trajectory and power allocation. The objective is to maximize the sum energy efficiency while meeting the constraints included maximum power, minimum achievable data rate and spatial limitation. Simulation results demonstrate the effectiveness of the proposed algorithm compared with other benchmark algorithms.

Unmanned aerial vehicle (UAV)-assisted video streaming is gaining growing interests in satellite-terrestrial networks due to the mobility and caching capability. However, it is challenging to perform trajectory planning and cache management towards maximizing quality of experience (QoE) for video streaming due to a dynamic network topology and a class of hybrid control actions. In this paper, we consider a QoE-oriented video streaming transport system in satellite-UAV-terrestrial network. Our goal is to design a transmission scheduling policy that can maximize the QoE received by the ground users (GUs) under the cache capacity constraints. In this regard, we formulate a scheduling problem as a cache-constrained Markov decision process (CMDP). To tackle the CMDP, we propose a novel hybrid reinforcement learning algorithm with risk sensibility. Extensive simulations show that our proposed scheme improves QoE by more than 50% over the conventionally configured schemes.

When ground base stations are not available in the aftermath of a disaster, unmanned aerial vehicle (UAV) acting as flying relay is a promising option. The UAVs with limited energy as flying relays allow for wider data coverage and more stable data transmission. However, with the changes of ground devices topology and channel, it is challenging to consider quality of service (QoS) and the age of information (AoI) in UAV communication under the energy constraint. In this paper, we propose a dynamic priority packet scheduling for UAV assisted AoI-aware network whose utility is maximized subject to QoS to get the best tradeoff of the energy consumption and the weighted AoI. Specifically, the dynamics of devices are characterized by Gauss-Markov mobility model. Dynamic priority is affected by devices' movement, channel changes and others. We optimize the trajectory of the UAV and the scheduling scheme of the packets by the Dueling Double Deep Q Network (D3QN) algorithm. Simulations show that the scheme significantly improves the utility of the system compared to the benchmarks.

Unmanned aerial vehicles (UAVs) have shown great potential in live video streaming applications, especially in surveillance and reconnaissance. However, ensuring high quality of service (QoS) remains a challenge due to the dynamic nature of wireless channels. In this paper, we tackle the crucial challenge of energy-efficient and secure UAV-enabled live video streaming. To maximize long-term energy efficiency, we propose a cross-layer optimization framework that coordinates the adjustment of video coding parameters, wireless resource allocation, and UAV trajectory planning. We formulate the joint optimization as a constrained Markov decision process (CMDP) to capture the complex interdependencies between video quality, energy usage, and security risks. We introduce a new performance metric that captures the trade-off between video quality and energy consumption. The core of our method is a customized first-order constrained policy optimization, which efficiently handle complex real-world constraints like UAV battery capacities and end-to-end transmission delays. Our approach achieves scalability and sample efficiency with minimal gradient information. Through extensive system modeling and simulations under various network conditions, we validate the effectiveness of the proposed method compared with existing reinforcement learning algorithms.

Unmanned aerial vehicles (UAVs) are integrated into existing networks to enhance coverage, increase network capacity and provide ubiquitous access service. However, the channel in the UAV network is prone to noise and interference due to the complex environments. Reconfigurable intelligent surface (RIS), as an emerging technology in recent years, can be applied to the UAV network to establish the transmission environment by intelligibly adjusting signal characteristics, which can achieve significant gains in coverage and spectral efficiency. Thus, we consider RIS aided UAV networks for virtual reality (VR) content transmission under the presence of eavesdroppers, and maximize the time average sum secrecy energy efficiency (SEE) via adjusting UAV trajectory, beamforming matrix of UAV and RIS jointly by the deep reinforcement learning (DRL) approach. To eliminate the time correlation and the coupling of variables, we propose a Lyapunov guided decay twin-delayed deep deterministic policy gradient (TD3) scheme to tackle the decoupled problem. Simulations demonstrate the effectiveness of the proposed scheme and its outperformance in SEE compared with other benchmarks.

Unmanned aerial vehicles (UAVs) are anticipated to be integrated into the next generation wireless networks as new aerial mobile users, which can provide various live streaming applications such as surveillance, reconnaissance, etc. For such applications, due to the dynamic characteristics of traffic and wireless channels, how to guarantee the quality of service (QoS) is a challenging task. In this paper, with recent advances in scalable video coding (SVC), we study secure video streaming in wireless networks with UAVs. By jointly optimizing video levels selection and power allocation, the research tries to maximize the energy efficiency, which is the ratio of video quality to power consumption, while satisfying the secrecy timeout probability (STP) requirement. The aforementioned problem is modeled as a constrained Markov decision process (CMDP). And then, the study employs a state-of-the-art reinforcement learning algorithm, namely safe deep Q-learning network (safe-DQN), to solve the CMDP problem, in which a safety policies set is induced by constructing a Lyapunov function. Extensive simulation results with different system parameters show the effectiveness of the proposed algorithm compared with other existing reinforcement learning algorithms.

Due to the limited energy of the Internet of Things (IoT) device, unmanned aerial vehicle (UAV) as a mobile fusion center can effectively prolong the lifetime of IoT device via supporting communication with the device directly. Moreover, since UAV's energy constrained, it will be a good measure to take multiple UAVs to collect data from devices in large areas. In this paper, we investigate multi-UAV collaborative data collection system, where multiple UAVs collect data from two-dimensional distributed devices on flying mode or hovering mode. The objective is to minimize UAVs' total flight time while allowing each device to complete data upload successfully with limited energy. To this end, firstly, a cell partition based on Voronoi diagram is used to allocate the collection areas of each UAV. Then, in each associated area, UAV determines the whole trajectory to serve devices. Lastly, given load requirement of ground devices and energy limitation, the optimal data collection mode of each device is decided to minimize flight time of each UAV. Simulation results show that the proposed multi-UAV data collection scheme can shorten collection task completion time significantly.