Satellite-enabled multi-access edge computing (MEC) networks have emerged as a promising solution for low-latency data processing in areas lacking infrastructure. However, these satellite MEC networks face significant security vulnerabilities and high communication latency due to the open-air interface and large-scale data transmission. To address these challenges, we propose a secrecy-aware adaptive federated learning (AFL) approach for a satellite MEC network. In this network, terrestrial devices perform local model training using their own data and periodically transmit updated model parameters to a satellite server in the presence of an eavesdropper. To secure both model uploading and downloading, idle devices act as friendly jammers, transmitting jamming signals to disrupt eavesdropping attempts. Our goal is to minimize the overall federated learning latency by jointly optimizing the number of quantization bits, the transmit power of MEC devices, the satellite’s transmit power, and the jammer selection strategy. To solve this problem, we first propose an AFL framework that minimizes the model uploading size while ensuring the required model accuracy. Building on this, the problem is divided into two subproblems of model uploading and model downloading, which are solved using a successive convex approximation (SCA)-based algorithm. Additionally, to improve secrecy performance, we introduce a low-complexity jammer selection strategy that significantly enhances the secrecy rate for both model uploading and downloading. Simulation results demonstrate that the proposed scheme significantly outperforms baseline methods in terms of AFL convergence, secrecy performance, and overall latency.

Multiagent systems (MASs) can be exploited for malicious activities, which pose significant threats to public safety and national security. While current literature has explored countermeasures for individual or several agents, these approaches are inadequate for large-scale malicious MASs due to the lack of a systematic, global perspective. Additionally, the heterogeneity of MASs, wherein agents exhibit varying system weights, necessitates a strategy that prioritizes agents with high system weight to maximize disruption. To address these challenges, a consensus breakdown algorithm based on critical node identification and network topology decomposition is proposed. The proposed algorithm decomposes the network topology of both large-scale homogeneous and heterogeneous MASs by disabling critical nodes, thereby splitting MASs into multiple smaller agent clusters and preventing MASs from achieving global consensus. In scenarios involving both homogeneous and heterogeneous MASs, this approach transforms the critical node identification problem into a node regression problem. The algorithm leverages GraphSAGE, a highly efficient graph neural network (GNN) with a sampling mechanism, making it well-suited for feature extraction in large-scale networks while addressing potential computational constraints common in real-world applications. Relying on the sampling mechanism, GraphSAGE enhances computational efficiency when processing large-scale network topologies. To better fit the need for consensus breakdown, the information dissemination capabilities of nodes are considered when defining node importance. Furthermore, to extend the algorithm to the scenarios of heterogeneous MASs where agents have different system weights, the node importance is combined with the system weight of each agent to determine the final node criticality. Extensive simulation results validate the superior performance of the proposed algorithm across various aspects. Comparative experiments further demonstrate the accuracy and efficiency of the algorithm.

The integration of Digital Twin (DT) technology into the 6G-enabled Internet of Everything (IoE) has revolutionized real-time monitoring and maintenance in the Industrial Internet of Things (IIoT). However, DT models, particularly time-series regression models, are increasingly vulnerable to adversarial attacks that compromise their stability and reliability. This study proposes a reinforcement learning-based adversarial attack framework for time-series regression models, enabling the generation of highly targeted and effective adversarial examples. The method optimizes a perturbation generation strategy through reinforcement learning, leveraging the temporal dynamics of time-series data to maximize its cumulative impact on the target model’s outputs under predefined perturbation constraints. Experiments on NASA’s N-CMAPSS dataset validate the method on DNN and KAN twin models using PPO and SAC algorithms, demonstrating superior attack effectiveness and stealth over FGSM, PGD, and CW, with Attack Intensity (AtI) scaling with perturbation magnitude. The method achieves higher computational efficiency by requiring only forward computation. Unlike gradient-based methods (e.g., APGD), the proposed approach remains effective against TRADES-trained models, showing notable adaptability. However, this advantage diminishes under hybrid adversarial training. This study exposes security risks in DT models under adversarial attacks and underscores the urgent need for advanced defense mechanisms to safeguard IoE systems.

The satellite-enabled Internet of Things (S-IoT) plays a crucial role by providing stable and global connectivity. However, its rapid growth brings many challenges in managing massive devices and addressing security threats. In this paper, we propose a new distributed network architecture combined with non-orthogonal multiple access (NOMA) for S-IoT. We consider a secure NOMA transmission scenario, where an appropriate legitimate device in an NOMA cluster is chosen as a jammer. To maximize the system’s total secrecy rate, we formulate an optimal secure NOMA clustering and power allocation, which is non-convex and difficult to solve directly. To solve the joint optimization problem, the original problem is transformed into two subproblems, and we propose staged algorithms to solve them efficiently. Firstly, a distributed iterative NOMA clustering algorithm is proposed to iteratively group S-IoT devices into multiple NOMA clusters. Then, a particle swarm optimization (PSO)-based optimal secure power allocation (OSPA) algorithm and a soft actor critic (SAC)-based OSPA are proposed to allocate powers for intra-cluster devices. The PSO/SAC-based OSPA algorithm can be performed among different NOMA clusters in a parallel way, which greatly improves the efficiency of power allocation. Finally, an optimal dynamic jammer strategy is proposed to dynamically select an idle device to act as a jammer, greatly improving the security performance of the systems. The simulation results demonstrate the advantages of the proposed distributed algorithms and also show that the proposed scheme greatly outperforms the state-of-the-art schemes in terms of the secrecy rate.

In this paper, we study a joint user pairing and power allocation problem in the 5th generation (5G) satellite random access (RA) networks, where some user equipments (UEs) are assisted by relay satellite UEs to establish satellite access. We aim to maximize the total sum rate of the RA system by jointly optimizing user pairing and power allocation. The above joint optimization problem is a non-convex mixed-integer problem, which is challenging to solve. To solve this problem, we decompose it into two subproblems. Firstly, a problem for optimal user pairing is formulated to find the optimal user pairing relationship. To solve this subproblem efficiently, a Q-learning based distributed user pairing algorithm (QL-DUPA) is proposed, which converts the user pairing problem to a Q-learning process. The Q-learning process can achieve a near-optimal solution and is practically feasible. Then, a problem for optimal power allocation is formulated to find the optimal power allocation coefficients in each user pair. The subproblem is convex and the optimal solution is obtained using convex optimization. Next, a satellite RA scheme with collision resolution is proposed based on the joint optimization of user pairing and power allocation, and we analyze its total sum rate. Simulation results show that the proposed satellite RA scheme with collision resolution greatly outperforms the existing schemes in terms of total sum rate.

In this paper, an enhanced energy-harvesting irregular repetition slotted ALOHA (EEH-IRSA) is proposed for wireless sensors networks (WSNs). We consider that each sensor device has a finite-sized battery that is recharged by the harvested energy from the environment in a probabilistic manner. The performance of the proposed EEH-IRSA is analyzed by using the density evolution (DE) method and the optimal degree distribution for different battery capacity is obtained by using the differential evolution algorithm. Simulation results show that the throughput of the proposed EEH-IRSA scheme can achieve 1.14 packets/slot, which is about 44.5% larger than that of the traditional EH-IRSA.

In this paper, we propose a new transmit power diversity scheme for contention resolution diversity slotted ALOHA (CRDSA) under total transmit power limitation in satellite networks. In the proposed scheme, the total transmit power of each device within a frame is considered to be the same, and allocated to multiple packet replicas, and the transmit power of each packet replica can be different, which enables the transmit power diversity. The threshold based capture model is employed at the gateway demodulator, and the recovery-error probability based on this model is derived. The threshold of capture and transmit powers of packet replicas are also jointly optimized to maximize the achievable sum rate (ASR) of CRDSA under total transmit power limitation. Simulation results show that the proposed scheme is more than 50% higher than the conventional one in terms of normalized ASR.

In this article, we propose an optimal irregular repetition slotted ALOHA (IRSA) under total transmit power constraint for random access (RA) in the Internet-of-Things (IoT)-oriented satellite networks, which is named as an optimal power-limited IRSA (PL-IRSA). In this proposal, the total transmit power for each satellite machine-type device (SMD) within a frame is considered to be the same and equally allocated into its multiple packet replicas. Due to a variant number of packet replicas, the transmit power of each packet replica for different SMDs may be different. The difference of transmit powers of packet replicas and path loss enhance the received power diversity, which is able to improve the throughput by using the capture effect. At the receiver side, we introduce a semi-analytical (SEA) model to approximate the physical-layer decoding and derive an asymptotic packet loss rate (PLR) based on this SEA model. From the derived results, an optimization problem is formulated and solved to find out the optimal degree distributions for different total transmit powers and code rates, which maximize the normalized system throughput. The simulation results demonstrate that the proposed optimal PL-IRSA can lead to significant performance improvement with respect to conventional ones.

In this paper, cooperative contention resolution diversity slotted ALOHA (CRDSA) with transmit power diversity is proposed for multi-satellite networks with partly overlapped coverage areas. The proposed scheme combines packet and power diversity transmission with efficient interference cancellation (IC) technique, which greatly improves the throughput. The packet loss ratio (PLR) of cooperative CRDSA with transmit power diversity is derived, and by using it as a fitness function, we formulate an optimization problem to find the optimized transmit power distribution, which further improves the throughput of cooperative CRDSA. Simulation results show that the cooperative CRDSA with optimized transmit power distribution remarkably outperforms the one with uniform transmit power distribution in terms of throughput.