Unmanned aerial vehicles (UAVs) have been extensively employed in the low-altitude intelligent network (LAIN) on data collection and transmission, enabling predictive maintenance, enhanced safety, and improved operational efficiency. However, the openness of wireless communication networks makes UAVs vulnerable to numerous security threats. To secure the critical transmitted data, many authenticated key agreement (AKA) schemes have been developed. Nevertheless, most existing AKA schemes fail to efficiently and securely authenticate communications between a single user and multiple UAVs in IIoT environments. To this end, we propose an extensible multi-party AKA scheme for LAINs. Specifically, we employ the physical unclonable functions and the Chinese remainder theorem to facilitate efficient authentication and data aggregation. Furthermore, by leveraging the additive homomorphic cryptography and blockchain, our scheme ensures privacy even in the presence of semi-trusted mobile operators. Formal security analyses and performance evaluations indicate that the proposed scheme meets the security requirements for LAINs while maintaining lightweight and extensible energy consumption.

Flying ad hoc networks (FANETs) offer flexible, real-time wireless communication solutions for multi-drone systems by utilizing drones as network routers. However, FANETs’ unique characteristics, including high mobility, unstable network topology, and intermittent connectivity, pose significant challenges in designing efficient and reliable routing protocols. Traditional routing protocols for mobile ad hoc networks often fall short in highly dynamic airborne environments due to excessive control overhead, increased latency, and inefficient route maintenance. To address these issues, this paper proposes an enhanced on-demand predictive (EDP) routing protocol that integrates a neighbor coverage-based predictive flooding mechanism and an adaptive link quality-based route maintenance strategy. The flooding mechanism mitigates directional deafness by using a Kalman filter-based probabilistic forwarding model, while the route maintenance method optimizes path selection based on distance, traffic load, and link lifetime. Simulation results show that EDP significantly improves packet delivery rate, reduces network delay, and lowers overhead compared to benchmarks, making it well-suited for applications in FANETs.

Current mainstream multiagent reinforcement learning (MARL) algorithms primarily focus on acquiring the global maximum reward throughout the entire training process, from the initial to the final stage. Whereas directly pursuing the global maximum return tends to be inefficient, particularly in environments with sparse rewards or the large-scale multiagent system. To address these challenges, previous algorithms have been developed to maintain individual policies to guide global training. Nevertheless, these approaches generally neglect either efficiency or the potential for local collaboration at the early stage of training. In this article, we propose the role-policy enhanced global policy (RPEGP) algorithm, which integrates the concept of distinct roles within the actor–critic-based MARL framework. RPEGP simultaneously considers both collaborative behaviors among agents and efficient global policy training. Specifically, RPEGP exploits the similarities among agents to assign distinct roles, training role-policies and the global policy concurrently. Through the initialization and enhancement of the role-policies, the global policy is trained more efficiently and effectively. Empirical experiments are conducted in well-known cooperative multiagent environments, including StarCraft II micromanagement (SMAC) and multiagent particle environment (MPE). Experimental results demonstrate that RPEGP outperforms baseline algorithms across various evaluation metrics and training efficiency, confirming its ability to address complex cooperative tasks generically and efficiently.

Collaborative underwater target hunting, facilitated by multiple autonomous underwater vehicles (AUVs), plays a significant role in various domains, especially military missions. Existing research predominantly focuses on designing efficient and high-success-rate hunting policy, particularly addressing the target's evasion capabilities. However, in real-world scenarios, the target can not only adjust its evasion policy based on its observations and predictions but also possess eavesdropping capabilities. If communication among hunter AUVs, such as hunting policy exchanges, is intercepted by the target, it can adapt its escape policy accordingly, significantly reducing the success rate of the hunting mission. To address this challenge, we propose a covert communication-guaranteed collaborative target hunting framework, which ensures efficient hunting in complex underwater environments while defending against the target's eavesdropping. To the best of our knowledge, this is the first study to incorporate the confidentiality of inter-agent communication into the design of target hunting policy. Furthermore, given the complexity of coordinating multiple AUVs in dynamic and unpredictable environments, we propose an adaptive multi-agent diffusion policy (AMADP), which incorporates the strong generative ability of diffusion models into the multi-agent reinforcement learning (MARL) algorithm. Experimental results demonstrate that AMADP achieves faster convergence and higher hunting success rates while maintaining covertness constraints.

Flying ad hoc networks (FANETs) provide high flexibility and real-time wireless communication solutions for multiunmanned aerial vehicle (UAV) systems by utilizing UAVs as routers. However, conventional routing protocols are inadequate for FANETs due to high mobility and dynamic topology of UAV networks. To address these challenges, this paper proposes an enhanced on-demand predictive (EDP) routing protocol for UAV networks. The EDP protocol incorporates a neighbor-coverage-based predictive flooding mechanism and an adaptive link-quality-based route maintenance method. The flooding mechanism utilizes Kalman filter theory to predict mobility and the route maintenance method selects optimal route by evaluating multiple factors. Simulation results demonstrate that the EDP protocol significantly improves the packet delivery rate while reducing network delay and overhead under varying environments, outperforming benchmark routing protocols for FANETs.

Semantic communication (SC), by compressing raw data at the semantic level, significantly improves the information entropy of transmitted data and is considered as one of the key enabling technologies for the next-generation communication. However, most current research underestimates the impact of channel interference on SC systems. As an innovative generative artificial intelligence technique, the diffusion model (DM) has demonstrated remarkable performance in image denoising and enhancement. In this paper, we focus on the effects of wireless channels on SC image transmission and propose an unmanned aerial vehicle (UAV)-enhanced SC framework, termed diffusion joint source-channel coding (D-JSCC). Initially, we deploy a ground-to-air SC system on UAVs, utilizing the aerial advantage to provide favorable channels. Subsequently, we employ DM for intelligent signal processing, adaptively denoising channel interferences and optimizing received images with respect to numerical errors and perceptual loss. The results show that DJSCC consistently exhibits superior performance across various metrics over different channel conditions.

Machine learning (ML) enhanced unmanned aerial vehicle (UAV) networks are envisioned to facilitate extensive applications in next-generation wireless networks. Due to the privacy concern and communication overhead in cloud-centric ML, federated reinforcement learning (FRL) enables UAVs to collaboratively train a policy model without disclosing raw observation data. However, the model aggregator in centralized FRL architecture poses various potential threats such as a single point of failure and is inappropriate to distributed networks with unreliable links and nodes. In this paper, we propose AirFRL, a topology-aware decentralized federated reinforcement learning framework for UAV-enabled networks. In AirFRL, we consider the topology dynamics influenced by nodes' mobility and communication quality and its impact on AirFRL. To accelerate training process and guarantee the model performance, we also incorporate the model compression to lighten the local model and introduce the consensus distance and data correlation to reflect the discrepancy between local models and local data. Furthermore, we propose an efficient algorithm to decide the optimal neighbour node selection and model compression ratio. A case study and numerical results demonstrate that AirFRL can achieve linear training speedup and guarantee the learning performance for UAV-enabled networks.

The integration of blockchain technology with unmanned aerial vehicles (UAVs) offers considerable potential, enhancing cybersecurity and driving innovation within the UAV industry. However, due to the dynamic nature of UAVs and limited resources, existing blockchain consensus technologies cannot be directly applied to UAVs. To this end, we propose a chainbased optimized blockchain consensus protocol designed for UAV ad hoc networks, which employs the particle swarm optimization (PSO) algorithm to optimize chain consensus. We design several sub-protocols to cope with malicious nodes in the UAV network, node changes during UAV missions, topology changes. Numerical results show that our protocol increases throughput, reduces communication overhead, and enhances operation efficiency in UAV networks.

Flying ad hoc networks (FANETs) composed of small unmanned aerial vehicles (UAVs) are flexible, inexpensive, and fast to deploy, which have been used in an increasing number of mission scenarios. However, unstable link quality and frequently changing network topology pose significant challenges for adopting existing routing protocols in mobile ad hoc networks (MANETs). In this paper, we propose a time-slotted on-demand predictive (TSDP) routing protocol designed specifically for UAV networks. The TSDP protocol introduces a novel approach to route selection by incorporating multiple criteria, including delivery ratio, adjacent degree, and mobility prediction factor, to ensure reliable and efficient data transmission. By addressing high latency in route discovery and excessive broadcast overhead, TSDP employs a time-slotted communication mechanism that reduces packet drop rates and enhances route stability. Simulation results demonstrate that TSDP consistently outperforms ad hoc on-demand distance vector (AODV) and dynamic source routing (DSR) protocols in terms of throughput, packet delivery ratio, end-to-end delay, and overhead, particularly in highly dynamic network environments.

With the advancement of large model technologies, AI -generated content is gradually emerging as a mainstream method for content creation. The metaverse, as a key application scenario for the next-generation communication technologies, heavily depends on advanced content generation technologies. Nevertheless, the diverse types of metaverse applications and their stringent real-time requirements constrain the full potential of AIGC technologies within this environment. In order to tackle with this problem, we construct a priority-aware multi-UAV aided metaverse system and formulate it as a Markov decision process (MDP). We propose a diffusion-based reinforcement learning algorithm to solve the resource allocation problem and demonstrate its superiority through enough comparison and ablation experiments.

Autonomous aerial vehicles (AAVs), as the key Internet of Things (IoT) devices, play a dominant position in low-altitude environments. Semantic communication (SC), as the next-generation communication technology, serves as a bridge for surpassing the Shannon limit toward the 6G wireless network. Establishing air-ground SC to provide intelligent IoT services is a crucial initiative for building future smart cities. In this article, we propose an AAV-based SC framework, named diffusion joint source-channel coding (D-JSCC). Abandoning traditional convolutional neural networks, we use transformers as the backbone and innovatively incorporate the diffusion model (DM) for image enhancement, achieving an optimal balance between image distortion and human perception. To accurately capture the numerical and perceptual loss induced by wireless channels and seamlessly amalgamate the DM with SC, we integrate channel states as strong prior information to refine the sampling process. Furthermore, we employ the gradient guidance strategy, which counteracts the randomness of sampling ensuring high robustness in harsh communication conditions. Additionally, we strike a balance between performance and sampling steps, ensuring both efficient computation and high-quality image enhancement. Comprehensive experiments demonstrate the advantages of D-JSCC across different communication environments.

The extensive deployment of Internet of Things (IoT) devices across diverse industries has introduced substantial challenges in information collection, which exactly hinders its further advancement. By virtue of its flexibility and mobility, mobile crowdsensing (MCS), particularly drone-assisted mobile crowdsensing, is regarded as a new paradigm for addressing information collection problems in IoT environments. Nonetheless, owing to the inherent size and energy constraints of drones, planning their trajectories to efficiently perform the crowdsensing tasks from a large number of heterogeneous and spatiotemporally distributed IoT devices presents a significant problem. In this paper, we develop a multi-drone assisted crowdsensing social IoT (SIoT) system that integrates the social attributes of IoT devices and enables performing social community-oriented crowdsensing. Given the significance of information freshness in crowdsensing tasks, we jointly optimize the age of information (AoI) and drone energy consumption within the problem of multi-drone trajectory planning. We formulate the aforementioned problem as a decentralized partially observable Markov decision process (Dec-POMDP) and propose a deep-reinforcement-learning-based algorithm to address this problem. A series of experiments is conducted and the simulation results demonstrate the superiority and robustness of the proposed algorithm in effectively balancing the AoI of the whole SIoT system and energy consumption of drones during the crowdsensing tasks.

In the Internet of Things (IoT) networks, monitoring information collection is critical for intelligent decision-making, which is a significant challenge for the sensors deployed at remote locations. Relay can effectively improve the transmission quality and transmission range of sensors by means of multi-hop transmission. It is an effective method for remote data collection in IoT networks. However, the lifetime of the relay may be dramatically reduced due to the heavy resource overhead for frequent short packet delivery. In this paper, we present an efficient relay transmission scheme for IoT networks, in which the frame aggregation technology is employed at the relay to reduce the resource overhead by sharing a common frame header and tail. Meanwhile, for the delay caused by frame aggregation, we analyze the freshness of the sensing data in terms of age of information (AoI) and take it as a constraint for the frame aggregation system. Besides, the optimal frame aggregation period is determined based on the closed-form expressions derived for the average AoI and transmission efficiency. Simulation results show that the theoretical analysis closely matches the simulations, and the proposed method significantly improves transmission efficiency compared to the traditional decode-and-forward method.

The popular multiagent reinforcement learning (MARL) methods primarily focus on exploring the capability of value functions to facilitate multiagent coordination. These MARL methods, following the centralized training with decentralized execution (CTDE) paradigm, tend to design ingenious network architectures while overlooking the impact of coordination through expanding local observation information. To tackle this deficiency, we model the multiagent systems (MASs) as a graph and use a graph neural network (GNN) to extract rich information between one agent and the others efficiently. Moreover, we propose a multigraph-neural-network information representation (MGIR) method that uses the power of GNN in local observation information extraction, enabling the acquisition of higher quality information. Specifically, multiple GNNs are used during centralized training to characterize the MAS from different perspectives and extract representations of latent variables. During distributed execution, these latent variables are leveraged to expand local observation information. Extensive comparative experiments substantiate that our proposed MGIR demonstrates superior coordination performance when compared with baseline methods. In addition, it can be flexibly integrated into various value function decomposition methods of MARL.

Under the advancement of emerging communication technologies, the utilization of Internet of Things (IoT) is progressively expanding across diverse domains. The age of information (AoI) stands as a crucial measurement in evaluating the efficiency of IoT networks. For efficient and reliable data collection, Unmanned aerial vehicle (UAV) have been extensively applied in IoT networks. However, the escalating number of sensor nodes and random data sampling mode within IoT networks have made it challenging for UAV trajectory planning with the constraint of energy consumption. In response to this challenge, our solution entails an attention-based actor-critic algorithm for multi-UAV path planning in a pre-clustered IoT network, which takes into account both the average AoI of clusters and the energy consumption of each UAV. The simulation outcomes validate that our algorithm achieves a trade-off between the information freshness and energy consumption in the multi-UAV data gathering scenario.

While symbol-level (SL) precoders have been shown to improve transmission performance by treating multi-user interference (MUI) as a useful resource, the SL precoders only employ uniform modulation for all downlink users, and the incurred complexity increases linearly with the block length. In this letter, we investigate the possibility of mixed-modulations interference exploitation (IE) for satellite communications, at a sub-block level. By exploiting the specific detection regions of constellation points of different modulations, a novel sub-block level mixed-modulation (BL-MIE) design is proposed, guaranteeing that MUI is always constructive in each sub-block duration, regardless of the users’ heterogeneous modulation schemes. Compared to the classic SL design, it is proved that the BL-MIE provides complexity reduction on the order of square root of the sub-block length, i.e., ${\mathcal{O}}(\sqrt{n})$, with n denoting the number of symbols per sub-block. Hence, it well strikes the balance between the performance and complexity. Simulation demonstrates that the proposed designs significantly outperform the benchmarks in terms of power consumption and throughput performance.

With the rapid development of wireless communication networks, UAVs serving as base stations are increasingly being applied in various scenarios which not only include edge computation and task offloading, but also involve emergency communication, vehicular network enhancement, etc. In order to enhance the utility of UAV base stations’ allocation and deployment, a series of algorithms have been proposed, utilizing heuristic methods, learning-based algorithms or optimization approaches. However, it is intractable for current algorithms to handle the exponential computation increment with UAV base stations increasing, and complicated application scenarios with high dynamic demands. To solve the above issues, we formulate a decision problem with a long sequence to optimize the deployment of multi-UAV base stations for maximizing vehicular networks’ communication coverage ratio, which needs to be subject to co-constraints consisting of moving velocity, energy consumption and communication coverage radius. To solve this optimization problem, we creatively propose an algorithm named dense multi-agent reinforcement learning (DMARL), which is under the dual-layer nested decision-making framework, centralized training with decentralized deployment, and accelerates training by only collecting critical states into the dense sampling buffer. To prove our proposed algorithm’s effectiveness and generalization ability, we conduct experimental simulations in scenarios with different scales. Corresponding results have been provided to verify our algorithm’s superiority in training efficiency and performance metrics, including coverage ratio and energy consumption, compared with other algorithms.

Efficient data processing is crucial for industrial Internet of Things (IIoT) applications, but the limited energy and computing resources in IIoT devices (IIoT-Ds) pose constraints. This article utilizes a unmanned aerial vehicle (UAV) as a computing server for enhanced IIoT mission execution. Specifically, the energy consumption of IIoT-Ds and the UAV, as well as the weighted cost of the communication and computing scheduling strategy in the UAV-aided IIoT, are jointly taken into account. An optimization problem based on the system energy consumption is built under the constraints of UAV motion, computing offloading, and transmitting power allocation. A problem decoupling-based alternating optimization method is proposed to solve the minimization problem by decomposing it into three subproblems: 1) UAV motion optimization; 2) computing offloading configuration; and 3) transmitting power allocation. Through comparing the proposed communication and computing scheduling strategy with existing methods, simulation results illustrate its attainment of quasi-optimal performance, thereby validating the effectiveness of the alternating optimization method.

Federated learning (FL), utilizing data from the edge devices (EDs) while protecting user privacy has gained much attention. Its efficacy is substantially influenced by both the quantity of connected devices and the quality of wireless communications. Network congestion, resulting from multiple access and signal attenuation caused by physical obstacles may severely impact the convergence of the FL model. To address these issues, this article employs nonorthogonal multiple access (NOMA) for uplink transmission and designs a two-tier FL framework consisting of ground devices and unmanned aerial vehicles (UAVs) to ensure the construction of Line of Sight (LoS) channels from EDs to the base station. Moreover, we construct a multiobjective joint optimization problem to minimize the FL convergence time considering constraints, such as the NOMA uplink latency, ED selection strategy, local training latency, and energy consumption. We also deduce the theoretical upper bound of the convergence time and transform the proposed multiobjective problem into a solvable form by eliminating the discrete variables determined by the ED selection. In turn, we utilize the proximal policy optimization (PPO) algorithm to solve this optimization problem. Finally, the extensive experimental results demonstrate the advantages of our proposed algorithm in terms of latency and energy consumption, while yielding a high robustness and scalability.

Unmanned aerial vehicles (UAVs) empowered Internet of things (IoT) networks have emerged as a burgeoning paradigm in the era of 6G. However, due to substantial data volume and privacy concerns, the conventional UAV backhaul to cloud center framework is not applicable to various latency and privacy-sensitive applications. Therefore, we propose a blockchain-based hierarchical federated learning (FL) framework for UAV-enabled IoT networks. Specifically, we utilize the total data distance-aware device association to mitigate model impairment arising from imbalanced data distribution. Besides, we introduce a lightweight blockchain into FL to tackle the trust deficit caused in decentralized global model aggregation. Furthermore, we design an optimization framework that jointly orchestrating device association, wireless resource allocation, and UAV deployment, aiming at a balance between the learning latency and model accuracy. To address the formulated optimization problem, we proposed a two-stage algorithm that integrates both greedy strategy and soft actor-critic algorithm. Extensive experiments show that our proposed scheme outperforms contemporary relative to state-of-the-art alternatives.

Metaverse is envisioned to merge the actual world with a virtual world to bring users unprecedented immersive feelings. To ensure user experience, federated learning (FL) has been expected as a critical enabler to provide metaverse users with high-quality sensing, communicating, and rendering. However, considering the limitation of wireless communication resources and the stringent requirements of users, collaborating with massive metaverse users to realize FL still has tremendous challenges. Most pioneer works on improving the performance of FL assume that the system states are static, which is unsuitable in the metaverse. Because the FL in the metaverse is always a complicated long-term iteration process, where the fluctuations of channel status and available computing resources of users are inevitable, a changeless strategy may lead to poor results. Therefore, this paper proposes an efficient FL scheme relying on dynamic user selection, gradient quantization, and resource allocation. Specifically, we derive the convergence error bound to reveal the impact of user selection, wireless transmission error, and gradient quantization error of each iteration on FL’s convergence. Based on the theoretical analysis, we jointly and dynamically optimize the user selection, gradient quantization, and resource allocation to minimize the error bound with time and energy consumption budgets. Furthermore, to make the formulated sequential decision-making problem tractable, we transform it into a Markov decision process and design a soft actor-critic-based solution. Extensive experiments validate that our proposed scheme has superior performance compared to conventional schemes in dynamic-changing network environments.

Underwater covert communication (UCC) technology can prevent legitimate transmission from being intercepted upon by potential eavesdroppers while ensuring a certain rate at the receiver under the condition of underwater acoustic channels. Previous studies have focused on UCC designs that rely on fixed transmitters and receivers, with limited attention given to dynamic moving senders, such as the widely-used autonomous underwater vehicle (AUV). Therefore, the establishment of a secure link between the mobile AUV and the receiver remains unexplored. In this paper, we construct an AUV-aided UCC architecture. Specifically, leveraging the unique characteristics of the underwater environment i.e., time-variant channel, severe attenuation, and ambient noise, the AUV plans its trajectory from the settled start point to the destination, adjusting its transmission power for covert communications. Accounting for both green energy consumption and communication security, we develop a novel multi-objective deep deterministic policy gradient (MODDPG) framework for jointly optimizing AUV’s diving energy consumption as well as effective throughput under the covertness constraint. Moreover, we propose an active-trust mechanism at the receiving side to pose an extra safe guard. To handle this, an evolutionary game model between the receiver and eavesdropper is built. Simulations and numerical results demonstrate that our proposed method can achieve a Pareto-optimal solution for covert communications with rapid convergence speed. The evolutionary stable strategy (ESS) enables the receiver to attain superior benefits and security compared to other strategies.

In this paper, we consider a wireless network enabled by multiple unmanned aerial vehicles (UAVs), which observe physical processes and transmit status updates to a monitor node over an error-prone communication channel. The communication scenarios are classified into two modes based on monitor types: UAV-to-UAV (U2U) and UAV-to-network (U2N) scenarios. Specifically, the U2N scenario is capable of covering a larger area compared to U2U scenario with a high signal-to-noise ratio (SNR). However, the U2U scenario constructs communication links more quickly, resulting in lower latency and higher rates. To evaluate the timeliness of the UAV-aided network, we utilize the age of information (AoI) as a fundamental indicator. AoI measures the time delay between the most recent data generation and the current moment. To compensate for the error-prone channel, this study employs a combination of hybrid automatic repeat request (HARQ) protocols, which include fixed-redundancy HARQ (FR-HARQ) and infinite incremental redundancy HARQ (IIR-HARQ) protocols. Furthermore, the average and peak Age of Information (AAoI and PAoI) of UAV-aided networks are derived for both U2U and U2N scenarios, and the theoretical expressions agree with simulation results. Additionally, FR-HARQ performs a better time performance than IIR-HARQ, and such observation van be verified through simulations.

Given the increasing attention paid to the security of underwater communications, covert communication system has been envisaged as a key enabler for empowering the marine information networks to address the challenge of ever-increasing demand of anti-eavesdropping. However, dynamic underwater hydrology environment and ambient noise make it substantially difficult to reduce the decoding error probability as much as possible on the premise of meeting the concealment requirements. In this paper, we first build up underwater covert communication (UCC) system and analyze its secrecy performance at the physical (PHY) layer. Moreover, we propose a dynamic power-threshold based bargaining game model to preserve the receiver’s concealment, while ensuring high receiver-side signal-to-interface-noise ratio (SINR). The simulation results show the detection probability at the equilibrium of our model is optimal, and thus it is capable of both overcoming the dynamic changes and of increasing the system’s stability significantly by analyzing the time discount factor.