Uncrewed aerial vehicles (UAVs)-assisted low-altitude communication networks have emerged as a promising solution for extending air-to-ground communication coverage and services. However, UAV-assisted communications are highly susceptible to jamming attacks due to its high probability of line-of-sight links. In this paper, we design an embodied intelligence-enhanced low-altitude communication network under malicious jammers, where multiple UAVs act as embodied intelligent agents to collaborate and jointly optimize power allocation and spectrum allocation to minimize transmission delay, while guaranteeing quality of service requirements against jamming attacks. Considering the non-convex problem and highly dynamic wireless environments, we propose an embodied multiagent deep reinforcement learning (E-MA-DRL)-based intelligent resource allocation approach to jointly optimize the communication resource, where embodied intelligent agents (UAVs) sense communication states, learn to make decisions and perform resource allocation actions. To enhance learning efficiency and performance, we then design prioritized experience replay (PER) and transfer learning (TL) in a double deep Q-network (DDQN) algorithm, to smartly schedule the communication resource and reduce the effect of jamming attacks and inter-channel interference. Simulation results show that the proposed approach significantly reduces communication delay and improves the probability of successful transmission in low-altitude communication networks against jamming attacks.

The integration of mobile edge computing (MEC) with semantic communication (SemCom) has emerged as a promising solution to address the growing demand for computing services. However, challenges remain in balancing task delay and energy consumption for unmanned aerial vehicles (UAVs) providing dynamic edge services, as the diversity of tasks complicates offloading and resource allocation. This paper investigates a multi-modal task offloading problem for Task-Oriented SemCom (TOSC)-based low-altitude MEC networks. We aim to minimize the weighted sum of ground devices’ average task delay and UAV’s energy consumption by jointly optimizing task offloading indicators, as well as the three-dimensional (3D) trajectory and computing resources of the UAV. To solve the formulated nonconvex optimization problem, which involves discrete and continuous variables, we propose an exact Penalty-based Alternative and soft Penalty-based Proximal Policy Optimization (PA-P3O) algorithm. Specifically, we develop an Exact Penalty (EP)-based algorithm to address the equilibrium-constrained task offloading problem with the discrete offloading indicators. Then, the coupling between the 3D trajectory and computing resources of the UAV is modeled as a Markov decision process (MDP), and a Soft Penalty (SP)-based deep reinforcement learning (DRL) algorithm is proposed to address the high-dimensional action space. Simulation results demonstrate that our proposed algorithm outperforms benchmark schemes and reveal an elevation-angle-distance tradeoff.

Unmanned aerial vehicles (UAVs) have been widely employed as airborne base stations to enhance terrestrial communications. However, the growing demand for communications, spectrum and energy resources are increasingly in short supply, while malicious jamming from jammers threatens the communications reliability. To address these challenges, we propose a joint reliable channel selection and power control approach for multi-UAV-enabled communications networks under malicious jamming attacks, with the goal of maximizing user communication capacity under limited energy constraint and avoiding malicious jamming from jammers. Due to the dynamic and time-varying nature of communication environments, we propose an intelligent resource optimization algorithm based on game theory guided reinforcement learning. To be specific, we employ a hierarchical learning algorithm based on the Stackelberg game to help users in the follower layer cooperatively select channels to against co-channel interference and jamming, and develop a deep reinforcement learning-based algorithm for dynamic power control to maintain communication efficiency. Simulation results demonstrate that our proposed approach can significantly improve the user communication rate and achieves faster convergence compared with existing algorithms.

The space-air-ground integrated network (SAGIN) is a crucial technology for sixth-generation (6G) wireless communication networks to achieve seamless coverage and high throughput. In this paper, we propose an unmanned aerial vehicle (UAV)-assisted SAGIN structure, where the UAV is responsible for collecting data from ground users (GUs) and transmitting it to low-earth orbit (LEO) satellites. This paper also formulates a joint energy-efficient and fair resource scheduling optimization problem under jamming attacks and limited energy constraints, where the line-of-sight (LoS) links between the UAV and GUs are susceptible to being jammed. Due to the non-convex problem and dynamic environments, a deep reinforcement learning (DRL)-based twin delayed deep deterministic policy gradient (TD3) is developed to search optimal UAV trajectory to maximize energy efficiency (EE) and fairness against jamming. Simulation results verify that the proposed intelligent resource scheduling algorithm outperforms the baseline algorithms in terms of EE and fairness index in different settings.

Due to the characteristic of wide coverage, flexible deployment, and low cost, unmanned aerial vehicles (UAVs) have been employed to provide mobile crowdsensing and edge computing. However, the limited computation and onboard battery capacities of UAVs impose a changeling for timely computation and endurance. In this article, we consider an aerial edge computing system where multiple cellular-connected UAVs are employed to perform sensing and computation tasks over target subregions, and the UAVs can offload their computation tasks to the ground base station (BS) with the binary offloading scheme. We aim to minimize the maximum energy consumption of all UAVs by optimizing 3-D UAV trajectories jointly with the binary offloading indicator as well as computation resource allocation, subject to the target sensing constraints and the computation completion time constraints. The optimization problem we formulated is nonconvex and involves binary design variables, making it difficult to find the optimal solution. To address this challenge, we propose an efficient alternating optimization algorithm that can obtain a high-quality suboptimal solution, where the exact penalty method with equilibrium constraints is adopted to tackle the binary constraints. To tackle the nonconvexity of the optimization subproblems, we utilize the successive convex approximation approach to obtain a suboptimal solution. Extensive simulations are conducted and the results demonstrate that the proposed design significantly reduces the energy consumption of the UAVs over several baseline methods.

Due to its high mobility, flexible deployment, and low cost, unmanned aerial vehicles (UAVs) have attracted wide attention in wireless communication in recent years. However, the delay requirements (e.g., video streaming, online game, etc.) may limit the UAV's mobility. In this paper, we consider a three-dimensional (3D) UAV communication network, where a UAV is employed to fly flexibly in 3D space to serve ground users with delay requirements. To characterize the fundamental tradeoff between throughput and delay, we introduce the minimum required rate for users and aim to maximize the minimum weighted sum of throughput and required rate for each user, via joint optimization of the 3D UAV trajectory as well as communication time and rate allocation. The formulated problem is a non-convex optimization problem, which is generally intractable. By decomposing the formulated problem into two subproblems, we propose an iterative algorithm by block coordinate descent and difference of two convex (D.C.) optimization as well as successive convex approximation (SCA) techniques. Finally, extensive simulation results show that our proposed solution outperforms baseline schemes and unveils the interesting insights and tradeoff between throughput and delay over 3D UAV communication networks.

Mobile edge computing (MEC) has been considered to provide computation services near the edge of mobile networks, while the unmanned aerial vehicle (UAV) is becoming an important integrated component to extend service coverage. In this paper, we consider a UAV-enabled MEC with binary computation offloading, where a UAV serves as an aerial edge server and each task of devices is either executing locally or offloading to the aerial edge server as a whole. To provide fairness among different ground devices, we aim to maximize the minimum computation throughput for all devices via the joint design of computing mode selection and UAV trajectory as well as resource allocation. The optimization problem is formulated as a mixed-integer nonlinear problem consisting of binary variables, which is difficult to tackle. The influence of non-binary solutions is penalized with a penalty function, based on which we develop an efficient iteration algorithm to obtain a suboptimal solution via leveraging the penalty successive convex approximation (P-SCA) method and difference of two convex (D.C.) optimization framework, where the algorithm is guaranteed to converge. Extensive simulations are conducted and the results with different system parameters show the effectiveness of the proposed joint design algorithm compared with other benchmark schemes.