ABSTRACT High‐volume data transmission in Internet of Things (IoT) networks demands the establishment of high‐quality multi‐hop communication paths at the network layer. To address this challenge, this paper proposes a joint routing and resource allocation framework that simultaneously optimizes relay node selection and transmit power allocation under decentralized manner. The proposed approach leverages deep reinforcement learning (DRL) to enable hop‐by‐hop decision‐making with only local observations. To mitigate the partial observability stemmed from limited perception range of node, a transformer‐based architecture is integrated into the DRL agent to enable the fusion of observations and actions collected along the established path. Specifically, the encoder module aggregates historical and frontier observations along the forwarding path, while the decoder module models temporal dependencies among historical actions. The fused representation is utilized to generate optimal decisions for next‐hop node selection and power allocation. After each action execution, a one‐hot encoded placeholder of the chosen action is fed back into the decoder module for subsequent decisions. Finally, numerical simulations demonstrate that the proposed transformer‐enhanced DRL framework significantly outperforms state‐of‐the‐art baselines in terms of end‐to‐end path quality, and robustness under decentralized IoT network configuration.

Unmanned aerial vehicle (UAV)-assisted federated edge learning (FEL) has emerged as a promising paradigm for privacy-preserving data processing in resource-constrained environments. However, the reliance on open wireless communication inherently exposes the system to eavesdroppers, who can eavesdrop and exploit shared model updates to reconstruct sensitive data, posing serious threats to the privacy and security. To address this challenge, we propose a privacy-aware UAV-assisted FEL framework that integrates adaptive local differential privacy (DP) into the model upload process, where user-specific noise is injected into local updates to prevent eavesdroppers from reconstructing sensitive data. To further enhance security and privacy performance, an indicator named value of privacy and security (VoPS) is designed to characterize the combined connection between training cost and privacy leakage. Furthermore, limited system resources including bandwidth allocation, user CPU frequency, DP noise scale, and UAV CPU frequency are collaboratively optimized under considering leakage threshold and heterogeneous computing constraints. Then, a deep deterministic policy gradient (DDPG)-based resource collaboration and secure aggregation scheme is proposed to solve the problem, in which the continuous optimization strategy is intelligently generated through the interaction between the agent and the dynamic privacy-aware UAV-assisted FEL system. Simulation results validate the effectiveness of the proposed scheme in enhancing the security and privacy performance of the system.

This paper proposes a joint secure sensing and communication framework for full-link covert transmissions, which integrates an intelligent reflecting surface (IRS)-assisted non-orthogonal multiple access (NOMA) system and compliant distributed cooperative jammers to enhance the communication quality of legitimate users while promoting the efficient utilization of limited resources. In the proposed scheme, upon sensing potential eavesdropper embodied by unmanned aerial vehicle (UAV), the dual-functional base station (BS) covertly transmits the acquired UAV state information to friendly jammers within relevant coverage area and issues activation commands promptly. Simultaneously, with IRS assistance, reconfigurable parameters such as signal phase in NOMA transmissions are adjusted to satisfy public user’s service requirements while facilitating covert communications for legitimate user. To ensure dynamic adaptability and link sustainability, the BS leverages historical sensing data to predict the UAV’s flight trajectory in real time and infer its movement intent. If the UAV exhibits a tendency to deviate from the currently effective jamming zone, the BS proactively activates friendly jammers in adjacent regions to maintain covert transmission rates and ensure robust system operation. To address the non-convex optimization challenge arising from jointly optimizing sensing beamforming, communication beamforming, and the IRS reflection matrix with highly coupled variables, we disassemble the problem into three subproblems. Correspondingly, an alternating optimization framework is designed by employing the semidefinite relaxation (SDR), Gaussian randomization, penalty-based methods, and Dinkelbach transformation to jointly maximize covert transmission rates while guaranteeing both sensing accuracy and communication quality of service (QoS). Simulation results demonstrate that the proposed scheme achieves superior covert transmission rates compared with benchmark schemes. Moreover, the dual-covertness mechanisms for sensing and communication further enhance the system security, validating the framework’s robustness in dynamic resource-constrained environments.

Unmanned Aerial Vehicles (UAVs) face significant challenges in autonomous navigation due to their limited energy and computational resources. This paper introduces a knowledge-driven meta-learning framework specifically designed for ultra-fast path planning in lightweight UAVs. The proposed approach integrates domain-specific knowledge across three core domains-environment, network, and behavior-with visual data to enable adaptive learning from unlabeled data and rapid model retraining in various scenarios. To evaluate this framework, we created the Meta-UAV Optimal Path Dataset, a unique dataset tailored for complex, multi-domain path planning tasks. Additionally, a knowledge-driven loss function incorporating physics-based constraints ensures that the model's predictions align with real-world conditions. Experimental results demonstrate that our model achieves superior path efficiency, cross-domain adaptability, and lower resource consumption compared to traditional models, making it a suitable choice for real-world UAV applications.

This paper investigates optimal distributed Medium Access Control (MAC) strategies for wireless Internet of Things (IoT) networks, incorporating both terrestrial and non-terrestrial relays, including UAVs. While prior research has primarily focused on single-relay forwarding, we address the complexities of energy-efficient multi-relay data forwarding. This involves managing the challenges of relay probing, optimized relay utilization, and balancing energy trade-offs to maximize system energy efficiency (EE). To address these challenges, we propose a novel strategy called Distributed Data Collection with Opportunistic Relaying (DDC/OR), designed to optimize MAC performance in multi-relay IoT networks. Our approach leverages a decision-theoretic framework based on optimal sequential planning, extending traditional cooperative MAC models to support multiple relays. The proposed DDC/OR strategy offers a statistically optimized solution that maximizes average EE, with a rigorous proof of optimality. Additionally, we present a low-complexity implementation of the DDC/OR algorithm, suitable for practical deployments. For scenarios involving dynamic non-terrestrial movements and varying relay inter-distances, we introduce a two-timescale, self-organized algorithm to adaptively reconfigure relay strategies. To ensure scalability for large relay networks, several optimizations are introduced to enhance the feasibility of the DDC/OR approach. We validate the effectiveness of the DDC/OR strategy through extensive simulations, demonstrating significant EE gains in both terrestrial and non-terrestrial relay configurations. Notably, it also achieves comparatively better performance in latency, throughput, and overall energy consumption.

Due to the flexibility in 3-D space and high probability of line-of-sight (LoS) in air-to-ground communications, unmanned aerial vehicles (UAVs) have been considered as means to support energy-efficient data collection. However, in emergency applications, the mission completion time should be main concerns. In this article, we propose a UAV-aided data collection design to gather data from a number of ground users (GUs). The objective is to optimize the UAV’s trajectory, altitude, velocity, and data links with GUs to minimize the total mission time. However, the difficulty lies in that the formulated time minimization problem has mutual effect with trajectory variables. To tackle this issue, we first transform the original problem equivalently to the trajectory length problem and then decompose the problem into three subproblems: 1) altitude optimization; 2) trajectory optimization; and 3) velocity and link scheduling optimization. In the altitude optimization, the aim is to maximize the transmission region of GUs which can benefit trajectory designing; then, in the trajectory optimization, we propose a segment-based trajectory optimization algorithm (STOA) to avoid repeat travel; besides, we also propose a group-based trajectory optimization algorithm (GTOA) in large-scale high-density GU deployment to relieve massive computation introduced by STOA. Then, the velocity and link scheduling optimization is modeled as a mixed-integer nonlinear programming (MINLP) and block coordinate descent (BCD) is employed to solve it. Simulations show that both STOA and GTOA achieve shorter trajectory compared with the existing algorithm and GTOA has less computational complexity; besides, the proposed time minimization design is valid by comparing to the benchmark scheme.

With the emergence of Internet of Things (IoT), where any device is able to connect to the Internet and monitor/control physical elements, several applications were made possible, such as smart cities, smart health care, and smart transportation. The wide range of the requirements of these applications drives traditional IoT to cognitive IoT (CIoT) that supports smart resource allocation, automatic network operation and intelligent service provisioning. To enable CIoT, there is a need for flexible and reliable wireless communication. In this paper, we propose to combine cognitive radio (CR) with a biological mechanism called reaction–diffusion to provide efficient spectrum allocation for CIoT. We first formulate the quantization of qualitative connectivity-flexibility tradeoff problem to determine the optimal cluster size (i.e., number of cluster members) that maximizes clustered throughput but minimizes communication delay. Then, we propose a bio-inspired algorithm which is used by CIoT devices to form cluster distributedly. We compute the optimal values of the algorithm’s parameters (e.g., contention window) of the proposed algorithm to increase the network’s adaption to different scenarios (e.g., spectrum homogeneity and heterogeneity) and to decrease convergence time, communication overhead, and computation complexity. We conduct a theoretical analysis to validate the correctness and effectiveness of proposed bio-inspired algorithm. Simulation results show that the proposed algorithm can achieve excellent clustering performance in different scenarios.

In this study, security schemes at the physical layer are proposed for an amplify‐and‐forward relay system in the presence of a passive eavesdropper. We focus on maximising the transmit power of the artificial noise (AN) to degrade the minimum mean square error (MMSE) level at the eavesdropper, while achieving MMSE constraint for the desired receiver. Assuming that the desired receiver and the passive eavesdropper possess perfect channel state information (CSI), AN is designed to lie in the null space of legitimate channels. However, the available CSI is impossible to be perfect in practice. Therefore the authors present a robust beamforming scheme which maximises the worst‐case transmit power of the AN under the condition of imperfect CSI and recovers a large fraction of the performance in the perfect CSI case. A classical alternating optimisation algorithm is proposed to obtain the relay beamforming weights, and simulation results show the benefit of proposed schemes.