Unmanned aerial vehicles (UAVs) can use the collected data to perform machine-learning tasks and enhance their intelligence level. The distributed framework of federated learning (FL) is suitable for resource-constrained UAV swarms. However, the dynamic channel conditions caused by the mobility of the UAVs impact the wireless FL performance. Additionally, data insufficiency and heterogeneity also affect the performance of the trained models. Thus, in this paper, we propose a joint training, trajectory, and beamforming optimization for an energy-efficient FL scheme that leverages the mobility of UAVs to enhance the model training efficiency, namely TTBO-FL. We adopt a two-level FL for the UAV network, where a high-level UAV (H-UAV) is for model aggregation and a set of low-level UAVs (L-UAVs) is for model training. Considering the dynamic channel conditions, we design a three-dimensional uniform linear array (3D ULA) and implement 3D analog beamforming to increase communication efficiency among UAVs. Moreover, we introduce transfer learning techniques and regularization constraints to mitigate the problem of data insufficiency and heterogeneity. Then, we formulate an optimization problem for UAV trajectory planning, local epoch adjustment, and beamforming, and adopt a soft actor-critic (SAC)-based algorithm to solve it. The simulation results show that, compared to the baseline schemes, the proposed schemes achieve higher model accuracy and lower energy consumption for the UAV swarm.

Semantic communication (SemCom), as a novel paradigm for future communication systems, has recently attracted much attention due to its superiority in communication efficiency. However, similar to traditional communication, it also suffers from eavesdropping threats. Intelligent eavesdroppers could launch advanced semantic analysis techniques to infer secret semantic information. Therefore, some researchers have designed Semantic Steganography Communication (SemSteCom) schemes to confuse semantic eavesdroppers. However, the state-of-the-art SemSteCom schemes for image transmission rely on the pre-selected cover image, which limits the generalization. To address this issue, we propose a Generative Diffusion Model-based Coverless Semantic Steganography Communication (SemSteDiff) scheme to hide secret images into generated stego images. The semantic related private and public keys enable legitimate receiver to decode secret images correctly while the eavesdropper without the completely correct key-pairs fail to obtain them. Simulation results demonstrate the effectiveness of the plug-and-play design in different Joint Source-Channel Coding (JSCC) frameworks. Results under different eavesdropping settings show that, when Signal-to-Noise Ratio (SNR) = 0 dB, the peak signal-to-noise ratio (PSNR) of the legitimate receiver is 4.14 dB higher than that of the eavesdropper.

Federated learning (FL) is suitable for unmanned aerial vehicles (UAVs) and ground devices to exchange model parameters periodically and learn a shared model without transmitting raw data. However, the existing UAV-assisted FL systems consider all learning phases to be equally important, which is inconsistent with the critical learning period (CLP) that exists in the FL training process, leading to significant overheads and inefficient resource utilization. Besides, data heterogeneity and insufficiency among devices also affect the performance of the trained models. To address the above issues, in this paper, we propose a CLP-aware FL framework that identifies unequally important learning stages and implements corresponding FL training strategies. Given the significant differences between global and local model parameter distributions in the early training epochs, we introduce a Federated Kullback-Leibler divergence (KLD) Norm (FKN) metric that measures the KLD between these distributions for efficient CLP detection. To capture the data drift caused by environmental shift in UAV swarms, we also develop a computationally efficient Federated Drift Norm (FDN) metric to enable online detection of CLP. We formulate an optimization problem for CLP-based participating device selection, UAV visit frequencies to different devices, and model aggregation period, then adopt a deep reinforcement learning (DRL)-based algorithm to solve it. Simulation results show that our strategy reduces energy consumption while maintaining model accuracy compared to baselines, i.e., CriticalFL, pFedBayes, and FedExp.

While cooperative perception in the Internet of Vehicles (IoV) is crucial for enhancing autonomous driving safety, the required massive data exchange imposes a significant burden on communication systems. Although Semantic Communication (SemCom) is a promising solution, its efficacy is severely compromised under adverse weather conditions, due to the concurrent corruption of source visual data and deterioration of the wireless channel. To address these issues, this paper proposes a Dual-Modal Semantic Communication (DMSC) framework based on the Joint Source-Channel Coding (JSCC) paradigm. The proposed framework uses the superior penetration of infrared (IR) to recover fine-grained details that are obscured in the RGB image by adverse weather. This fusion preserves essential semantic information at the source, providing a robust input for the semantic encoder to process. To enable adaptive fusion and transmission, we introduce a SNR-Aware Dual-Attention Module (SNR-DAM). This module utilizes attention mechanisms to dynamically adjust the weights of feature channels and the attention of spatial regions in response to changing channel conditions. Simulation results show that the proposed framework significantly outperforms both conventional separate source-channel coding and cascaded JSCC-dehazing baselines, achieving higher image reconstruction fidelity, particularly under the adverse weather and low Signal-To-Noise Ratio (SNR) channel conditions.

To support the large-scale connectivity of massive intelligent devices in the Internet of Things (IoT) scenario, in this paper, an uplink multi-user semantic communication system based on model division multiple access (MDMA) is investigated. A model resource pool consisting of multiple mutually exclusive semantic models is built as one new kind of access resource, and a global optimization problem is formulated to make the mapping of different semantic models have stronger mutual exclusion. To solve this problem, a joint deep adversarial semantic decomposition (JDASD) algorithm is proposed to enhance the ability of the semantic models to eliminate interference from other devices. Simulation results demonstrate that the proposed JDASD algorithm achieves higher anti-interference capability in the MDMA system than the traditional independent training scheme applied in most works, showing the advantages of the proposed scheme for the IoT scenario with massive devices.

In this paper, we investigate the joint communication-motion planning problem for unmanned aerial vehicle (UAV) swarm in the presence of jammers. Specifically, we consider a cluster-based UAV swarm architecture, where multiple cluster member (CM) UAVs transmit messages to a cluster head (CH) UAV through air-to-air links affected by malicious jammers. Our objective is to maximize the sum uplink rate of the UAV swarm by optimizing the trajectories and the transmit power of all UAVs. To achieve this goal, we formulate a joint multi-UAV trajectories and transmit power optimization problem under speed, transmit power, trajectories and received signal-to-interference-plus-noise ratio (SINR) constraints. In order to solve the problem, we establish a Markov decision process (MDP). For the multi-agent environment and the high-dimensional continuous action space, we adopt a multi-agent twin delayed deep deterministic (MATD3) policy gradient-based algorithm. Simulation results show that the proposed scheme can effectively improve the sum uplink rate of the UAV swarm compared to the baseline schemes.

Unmanned aerial vehicle (UAV)-assisted mobile edge computing (MEC) has been touted as a promising solution for providing computing services in disaster relief and other settings due to its flexibility and ease of deployment. Nevertheless, providing computing services for a large number of mobile devices is challenged by UAVs’ limited computation and energy resources. To this end, we propose a scheme for joint trajectory optimization and task offloading that aims to minimize the total delay of all computing tasks. Our proposed scheme involves formulating the scheduling of mobile devices and computing tasks, adjustment of UAV flight angle and speed, and transmission power control as a non-convex mixed integer programming problem. In order to address the issue, we establish a Markov decision process (MDP) for UAV-assisted MEC systems. Given the high-dimensional continuous action space, we adopt a reinforcement learning algorithm based on Deep Deterministic Policy Gradient (DDPG). The results of the simulation indicate that our suggested scheme outperforms the baseline schemes in processing delay, and the DDPG-based algorithm exhibits rapid convergence.

Federated learning (FL) provides an efficient distributed learning framework for computing-constrained Unmanned aerial vehicles (UAVs) swarms. However, due to the dynamic channel condition and limited resources in UAV swarm, the efficiency of FL requires to be further improved. In this paper, by leveraging the motion characteristics of UAVs, we propose an energy-efficient FL framework based on trajectory planning to train machine learning (ML) models. In the proposed framework, a two-level UAV swarm is established, consisting of a high-level UAV (H-UAV) and a group of low-level UAVs (L-UAVs). Considering the limited energy resources of the UAVs, our primary objective is to minimize the flight energy consumption of the H-UAV as it is much higher than the communication and computation energy consumption. To achieve this goal, we formulate the optimization problem jointly considering the adjustment of FL training parameters and the trajectory planning of H-UAV. Then, we reformulate the problem as a Markov decision process (MDP) and employ soft actor-critic (SAC) and deep deterministic policy gradient (DDPG) algorithms to tackle it. Simulation results show that the proposed approach and the proposed algorithms have good performance.

Abstract Unmanned aerial vehicles (UAVs) are employed as mobile relay nodes to enable timely remote monitoring by collecting information from monitoring devices and transferring the collected information to base station. The freshness of delivered information is critical to system performance, which can be quantified by the concept of age of information (AoI). Nevertheless, the fresher information comes at the cost of higher energy consumption at UAVs. Considering the limited onboard energy, it is essential to strike a balance between the age of delivered information and the required energy budget. here, both straight trajectory and circular trajectory of UAV are considered, and study a problem with the goal of supporting timely data collection while minimizing the energy consumption at UAV. The problem is formulated as a multi‐criteria optimization problem by jointly considering the aging of collected information, the trajectory planning of UAV and energy consumption at UAV. To solve the formulated problem, a solution procedure to find a sequence of Pareto‐optimal points is proposed. Simulation results demonstrate the Pareto‐optimal curve, which yields the energy‐efficient UAV trajectory for timely data collection.

The fast development of mobile communication and artificial intelligence (AI) technologies greatly promotes the prosperity of the Internet of Things (IoT), where various types of IoT devices can perform more intelligent tasks. Considering the privacy leakage and limited communication resources, federated learning (FL) has emerged to enable devices to collaboratively train AI models based on their local data without raw data exchanges. Nevertheless, it is still challenging for guaranteeing any FL models to be effective due to the sluggish willingness of IoT devices and the model poisoning attacks in the FL. To address these issues, in this article, we introduce blockchain technology and propose a blockchain-based FL framework for supporting a trustworthy and reliable FL paradigm in IoT. In the proposed framework, we design a committee-based participant selection mechanism that selects the aggregate node and local model updates dynamically to construct the global model. Moreover, considering the tradeoff between the energy consumption and the convergence rate of the FL model, we perform the channel allocation, block size adjustment, and block producer selection jointly. Since the remaining resources, handling transactions, and channel conditions are dynamically varying (i.e., stochastic environment), we formulate the problem as a Markov decision process (MDP) and adopt a deep reinforcement learning (DRL)-based algorithm to solve it. The simulation results demonstrate the effectiveness of the proposed framework and show the superior performance of the DRL-based resource allocation algorithm compared with other baseline methods in terms of energy consumption.

The real-time intelligent perception and prediction of traffic situation can assist connected automated vehicles (CAVs) in path planning and reduce traffic congestion in Cognitive Internet of Vehicles (CIoVs). The centralized traffic congestion prediction solutions generally fail to adapt to the dynamic traffic environment and lead to significant communication overheads. Blockchain technology has attracted great attention in the information sharing of vehicular networks for its advantages in decentralization, transparency, traceability, and tamper-proof capability. However, due to the bottlenecks, such as high computational cost, current blockchains are incapable actuate on efficient online traffic situational cognition and prediction for CIoVs. Motivated by this, we propose a blockchain-enabled cognitive segments sharing framework for online multistep congestion duration prediction. We design a cognitive model of traffic situation based on anomaly detection and filtering mechanism to guarantee the accuracy of the cognitive segments before being packaged into the block. Furthermore, to improve the consensus efficiency, we design a credit evaluation mechanism and propose a credit-based delegated Byzantine fault tolerance (CDBFT) algorithm. Finally, we propose an online multistep prediction algorithm based on long short-term memory (LSTM) to predict future traffic congestion duration. Experimental results demonstrate that the proposed algorithms achieve shorter consensus latency and higher predictive accuracy than the existing algorithms.

The digital twin (DT) bridges the physical world with the digital world in real-time for the Industrial Internet of Things (IIoT) and federated learning (FL) enables edge intelligence services for IIoT under the premise of avoiding privacy leakage. The fusion of two technologies can extremely accelerate the development of Industry 4.0 by enabling instant intelligence services. However, in the resource-constrained IIoT, the energy consumption of performing FL and maintaining the virtual object in the digital space by DT technology become the bottlenecks and can not be ignored. To address these issues, in this paper, we proposed an energy-efficient FL framework for DT-enabled IIoT. In the proposed framework, IIoT devices choose different training methods considering dynamic time-varying environment status to achieve energy-efficient FL, i.e., either train locally or connect to the virtual object by DT in the corresponding server of a small base station (SBS) to train mapped data using computing resources of SBS. Then, we investigate the joint training method selection and resource allocation problem to minimize the energy consumption while satisfying the convergence rate of the training model. Considering the problem is intractable using traditional approaches, we use a deep reinforcement learning (DRL)-based algorithm to solve it. Simulation results show that the proposed framework decreases greatly energy consumption compared with the static framework while satisfying the convergence rate of FL.

In wireless networks, network load can be highly unbalanced due to the mobility of user equipments (UEs). Unmanned Aerial Vehicles (UAVs) supported base station with the advantage of flexible deployment, ubiquitous wireless coverage and high speed data rate, is a promising approach to handle with the foregoing problem. However, how to achieve cost-effective UAV deployment in an autonomous and dynamic manner is a significant challenge. Facing this problem, we propose a novel UAV base station intelligent deployment scheme based on machine learning and evaluate its performance on a realworld dataset. First, we conduct data preprocessing to process, clean, and transform raw data into formatted data. Missing values are filled by Conditional Mean Imputation (CMI) method and outliers are corrected by pauta criterion. Then, we use hybrid approach which contains ARIMA model and XGBoost model. Linear predictions are carried out by ARIMA model and later nonlinear model XGBoost are applied on residue of ARIMA. Resultant prediction is obtained by adding linear and nonlinear prediction, hybrid model is estimated by Root Mean Square Error (RMSE) and R2 score. Finally, according to predicted results, UAV base stations can be deployed to cater for dynamically changing demands in the hotspot areas and achieve cost-effective deployment. Simulation results show that the propose scheme is superior to other benchmark schemes in load balancing.

By bringing computing capabilities to the network edge, mobile edge computing (MEC) has emerged as a promising technique to enable low-latency video streaming services. However, due to the rapid growth of the number of devices and the heterogeneous formats of the video streams, the traditionally centralized content delivery schemes are insufficient to provide secure, adaptive video services with low complexity. To achieve a decentralized content market among untruthful parties (e.g., users and operators), in this paper, we propose an effective video transcoding and delivery approach based on MEC and blockchain. In the proposed approach, we envision a set of blockchain-based smart contracts to build an autonomous content delivery market, where all the participants are financially enforced by smart contract terms. Then, users, small base stations (SBSs), and content provider (CP) are able to autonomously adjust their strategies according to the content market statistics. Moreover, we formulate the optimization problem, including resource allocation, determining content price and quality levels of contents, as a three-stage Stackelberg game. We analyze the subgame equilibrium for each stage and the interplays of the three-stage game. Lastly, an iterative algorithm is proposed to obtain the solution. Simulation results are presented to show the effectiveness of the proposed approach.

Driven by the large-scale video traffic, mobile edge computing (MEC) has emerged as a promising technique that extends cloud-computing capabilities to the proximate small base stations (SBSs) in wireless networks, especially in ultra-dense networks (UDNs). With MEC, video transcoding, which processes the adaptive bitrates of a video and provides the adaptive video streaming to users, can significantly release the backhaul burden of networks. However, video transcoding is a time-consuming task, and how to guarantee quality-of- service (QoS) for large video data with MEC is still challenging. To address this issue, in this paper, we propose a joint SBSs selection, tasks scheduling, and resource allocation approach for achieving a delay- optimal transcoding under the constraints of network cost. Specifically, to reduce the delay, a set of SBSs are formed into a Virtual SBSs Group (VSG) to perform the video transcoding and delivering in parallel for a given user. Then, the joint tasks scheduling and feasible resource allocation are performed to minimizing total delay while maintaining a low network cost. The optimization problem is formulated as a mixed integer non- convex programming problem and a three-stage search solution is proposed to solve it. Simulation results show that our proposed approach can significantly improve the transcoding performance while satisfying the resource consumption constraint.

Ultra-dense network (UDN), as well as nonorthogonal multiple access (NOMA), has been emerging as promising techniques to meet the growing demand of data traffic in next-generation wireless networks. However, due to the spectrum sharing among SBSs and users, interference management (IM) is becoming a more important issue in NOMA-based UDN. Moreover, the massive small base stations (SBSs) with various types and overlapped coverage require more intelligent and efficient mechanisms for the IM problem. Thus, in this paper, to reduce interference and improve operation efficiency, we propose a self-optimizing resource allocation (SORA) scheme for IM with joint consideration of the dynamic interference conditions and fierce resource competition among SBSs. Concretely, each SBS constructs the interfering SBSs group adaptively to represent the potential interference from other SBSs. Then, to reduce interference and meet users' requirements, each SBS performs the resource allocation including sub-band and power allocation independently. Moreover, we formulate the problem as a non-cooperation satisfaction game, where a satisfaction function is established for evaluating each SBS's utility. When every SBS's utility is above a preset threshold, the game is considered to reach the satisfaction equilibrium. A distributed algorithm is designed to enable each SBS to learn the satisfaction equilibrium and allocate the resource autonomously. Simulation results show the effectiveness of the proposed scheme compared with the traditional schemes.

A user-centric ultra-dense network (UUDN) is proposed as one of the promising solutions to provide very high area throughput density and flexible access service for users in the fifth-generation systems. On the one hand, network densification provides opportunities to cooperate among a large number of access points (APs) for serving a given user. On the other hand, the limited radio resources cause the serious competition among numerous APs and may degrade the network performance. Therefore, to support large number of connections and break through the restriction of limited frequency resource, non-orthogonal multiple access (NOMA), which supports multiple signals to transmit on the same frequency resource, is introduced into the UUDN. However, NOMA with network densification arises a series of challenges. And the method to group APs efficiently on the same frequency to support for a given user is a critical problem. Thus, in this paper, we propose a user-centric access framework for providing efficient access service and the flexible resource management in NOMA-based UUDN. Under the proposed framework, we then investigate the access scheme that organizes multiple APs into respective AP group (APG) cooperatively to provide access service for each user, aiming at maximizing the system energy efficiency. First, considering the users' requirement and network environment, a grouping evaluation model is set up to organize APG efficiently. Then, we formulate the resource allocation problem of APG as a mix-integer non-linear programming problem, which is hard to tackle. For tractability purpose, we transform this problem and propose low-complexity algorithms based on matching and differ of convex programming theories to obtain a feasible solution. Extensive simulation results are presented to demonstrate the significant performance improvement compared with the existing schemes.