A Secure Hierarchical Federated Learning Architecture for UAV-Enabled Edge Computing Systems

Cooperative Covert Communication and Task Offloading for AI-Empowered Multi-UAV Networks

Multi-UAV edge networks, as an effective supplement to ground sensor systems, can significantly improve perception coverage and data processing efficiency. However, as UAV networks scale up, effective multi-UAV cooperation becomes increasingly critical and challenging, especially for coupled deployment, task allocation, and resource management. Meanwhile, due to the openness and broadcast nature of wireless channels, UAV transmissions are vulnerable to eavesdropping, making cooperative security protection essential for reliable UAV edge computing. To address these issues, this paper investigates a multi-UAV secure edge computing scenario in which UAVs cooperate both for self-jamming to thwart the aerial eavesdropper and for distributed edge computing to assist task processing. We establish the digital models of the UAV secure edge computing workflow and formulate the latency minimization problem under covert communication constraints. Then, a particle swarm optimization (PSO) + block coordinate descent (BCD) method for discrete state spaces and a multi-agent deep deterministic policy gradient (MADDPG)-based scheme suitable for continuous real-world environments are proposed, respectively. Extensive analysis and simulations demonstrate the effectiveness of our methods, achieving covert task offloading while significantly reducing task processing latency.

Communication-Aware Federated Learning for Energy Management in Edge-Cloud Autonomous Systems

Partially-Observable Transmission Control for UAV-Enabled Federated Learning in IoT Networks

Uncrewed aerial vehicle (UAV)-enabled federated learning (FL) can provide flexible, on-demand edge intelligence for large-scale IoT deployments, but operating in shared unlicensed bands makes uplink update delivery interference-coupled and unreliable. In this paper, we develop a packet-level transmission framework that captures buffer overflow, delay violations, and transmission errors, and uses the resulting packet delivery ratio (PDR) to represent partial-update reception through a packetized, Bernoulli-masked FL aggregation process. We then formulate a fairness-consensus bilevel (FCB) optimization that jointly controls (i) transmission thresholds to maximize the average PDR while reaching consensus under partial observability and (ii) transmission powers to improve the worst PDR and enforce fairness across IoT learners. To solve this problem, we propose an alternating FCB optimizer composed of a consensus-based threshold controller (CTC), which drives the IoT learners toward a PDR-efficient consensus on transmission thresholds, and a fairness-based power controller (FPC), which updates transmission powers to improve the worst PDR and ensure fairness under the resulting consensus thresholds. Numerical results on CNN-based FL tasks show that the FCB optimizer improves FL aggregation and training performance by enhancing packet-level update delivery, consistently outperforming baseline transmission policies.

A Simulation-Driven Trust-Aware Federated Learning Framework for Robust Intelligent IoT Networks

The experimental results demonstrate that the proposed Trust-FedAvg framework substantially improves robustness over conventional FedAvg and remains competitive with established robust aggregation strategies, particularly under directional model-manipulation attacks and intermittent-connectivity conditions.

An Option-Based Hierarchical Approach for Dynamic Mobile Crowdsensing Over Edge-Assisted UAV Networks

With the rapid advancement of edge computing and uncrewed aerial vehicle (UAV) technologies, edge-assisted UAV networks have emerged as a promising solution for efficient data collection in mobile crowdsensing (MCS). This paper addresses the problem of decentralized dynamic data collection in UAV networks supported by edge systems. In this setting, UAVs autonomously collect data from ground sensor nodes (SNs), using shared task state information (TSI) exchanged through the edge system to enhance cooperation and improve overall efficiency. However, two key challenges arise in such dynamic environments: 1) Continuous data generation at SNs requires timely and non-redundant collection under limited UAV communication and energy constraints; and 2) UAVs face a fundamental trade-off between collecting fresh data at SNs and updating TSI via the edge system. To address the unique challenges posed by such dynamic environments, we propose the weighted age of data queue (WAoDQ), a novel data freshness indicator that quantitatively captures the time-varying nature of data freshness. Building upon WAoDQ, we further develop Fresh-DCEA, an option-based hierarchical multi-agent deep reinforcement learning (HMADRL)-based algorithm that integrates high-level strategic planning (e.g., data collection or TSI update decisions) with low-level action execution (e.g., UAV trajectory adjustments), enabling efficient and adaptive dynamic task assignment. Simulation results demonstrate that Fresh-DCEA outperforms benchmark methods in terms of data freshness, collection effectiveness, and energy efficiency, thereby validating its scalability and adaptability in dynamic MCS environments.

AoI-Aware UAV-Assisted Secure Status Updating: An Agentic AI-Enabled DRL Approach

The rapid expansion of real-time Internet of Things (IoT) applications has positioned uncrewed aerial vehicles (UAVs) as a promising solution for flexible and timely data collection in areas lacking robust infrastructure. This paper investigates a UAV-assisted secure status updating system, where a UAV serves as a mobile relay to forward status updating packets from ground devices (GDs) under the threat of a potential eavesdropper. To ensure information freshness and operational sustainability, we formulate a long-term stochastic optimization problem to minimize the cumulative average age-of-information (AoI) of all GDs and energy consumption of the UAV. The formulated optimization problem is an online mixed-integer non-linear programming problem, which involves the joint optimization of the flight speed, direction, and transmission power of the UAV as well as the binary scheduling indicator of GDs. To tackle the inherent non-convexity and complex spatial-temporal coupling, we propose an agentic artificial intelligence (AI)-enabled deep reinforcement learning (DRL) approach, named adaptive truncated quantile critics with large language models (LLM)-enabled state representation and reward function design (ATQC-L). Specifically, an adaptive truncated quantile mechanism is incorporated to mitigate distributional overestimation in dynamic environments. Furthermore, we leverage the reasoning capability of LLMs as an offline design-time agent to generate task-aware state representation and intrinsic reward functions. Simulation results demonstrate that the proposed ATQC-L algorithm outperforms representative DRL baselines in balancing information freshness and energy consumption of the UAV, while maintaining stable performance under different network scales, LLM backbones, truncation-parameter settings, imperfect eavesdropping channel state information, and mobile eavesdropping scenarios.