Junyi Deng

With the advancement of artificial intelligence (AI) technology, edge networks are progressively evolving toward intelligence, and mobile intelligent agents such as uncrewed aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) are playing an increasingly vital role in this process. However, existing studies have addressed cooperation among homogeneous agents, such as multiple UAVs, without exploring collaboration between heterogeneous mobile intelligent agents. For this purpose, we propose a vehicle-mounted multi-UAV cooperative service system to collaboratively collect and process multisource heterogeneous Internet of Things (IoTs) tasks under strict latency and resource constraints. To address the complexities of heterogeneous task structures and dynamic collaboration, the article first introduces a multisource heterogeneous task scheduling mechanism, which optimizes task prioritization and resource allocation for efficient processing. In addition, a decentralized reinforcement learning approach based on partial reward decoupling (PRD) heterogeneous agent proximal policy optimization (PRD-HAPPO) is employed to enhance collaboration and trajectory planning between UAVs and vehicles. Simulation results demonstrate that the proposed framework significantly improves task completion efficiency, reduces system latency, and outperforms existing deep reinforcement learning (DRL) algorithms in terms of convergence and scalability.

Unmanned Aerial Vehicles (UAVs), which are a crucial element of the future air-space-ground integrated network, can serve as potential mobile edge computing (MEC) nodes due to their onboard capabilities for storage, communication, and computation. However, current UAV-assisted MEC collaborative offloading methods primarily focus on addressing network requirements for computing tasks, while neglecting the heterogeneity in task demands due to the heterogeneity of task types and UAV service types. To address this, we propose a task demand-oriented collaborative offloading and deployment strategy in software-defined UAV-assisted edge networks. Specifically, to enhance the cache utilization of UAVs, we introduce an SD-UEN (Software-Defined UAV-Edge Network) architecture, which facilitates cooperation among UAVs under the guidance of an SDN (Software-Defined Networking) controller. In light of the heterogeneity of task demands, we employ the Tabu Search-based Matching (TSM) algorithm to accurately match computing tasks with the appropriate UAV modes. Furthermore, to enable intelligent UAV mode switching and dynamic UAV location deployment, we leverage the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm. By centrally training the MADDPG model offline, MEC servers and UAVs, acting as learning agents, can efficiently adjust UAV modes and deploy UAVs during online execution. This algorithm dynamically optimizes UAV actions to minimize task completion time and energy consumption. The simulation results highlight that our algorithm substantially reduces task response time and energy consumption compared to other algorithms, demonstrating its effectiveness.

Hadoop is one of the most popular platforms for distributed big data processing, and YARN is at the core of Hadoop 2.0. In the process of data processing, the possibility exists that malicious nodes might compromise the results, sharing incorrect outputs on the public cloud. To address this issue, this paper introduces an efficient YARN security framework, termed CMT-YARN, which incorporates an enhanced convolutional Merkle Tree to ensure the reliability of task execution results. Within a hybrid cloud environment, this framework utilizes the improved CMHT verification technique to conduct dual verification of intermediate and final results, thereby ensuring data integrity and reliability throughout the execution process. Compared to conventional MHT, CMHT reduces additional computational overhead by more than 37% during verification processes. Qualitative and quantitative analyses indicate that with a data sampling rate of 6.8%, CMT-YARN can ensure that the number of malicious acts undetected by the system does not exceed 5; when the sampling rate exceeds 26.9%, the framework can guarantee the detection of all malicious activities. Experimental results on a real Hadoop cluster demonstrate that CMT-YARN significantly enhances computational and storage performance compared to traditional solutions.

At presently, Diffusion Model (DM) has achieved state-of-the-art performanceby modeling the image synthesis process through a series of denoising networkapplications. Image Restoration (IR) is to improve the subjective image qual-ity corrupted by various kinds of degradation unlike image synthesis. However,IR for worksite images is greatly challenging in the low-level vision field due tocomplicated environmental factors. To solve this problem, we propose a Sam-pling Improved Diffusion Model for Worksite Image Restoration (SIDM-WIR).It not only improves the authenticity and representation ability for the gener-ation process, but also handles complex backgrounds and diverse object typeseffectively. SIDM-WIR has three main contributions: 1) Its framework adopts aMish-based residual module, which enhances the ability to learn complex patternsof images, and allows for the presence of negative gradients to reduce overfittingrisks during model training. 2) It employs a Mixed-head Self-Attention (MSA)mechanism, which augments the correlation among input elements at each timestep, and maintains a better balance between capturing the global structuralinformation and local detailed textures of the image. 3) It ran in a new customizeddataset named Workplace for WIR specifically. Furthermore, experiments on bothtwo datasets demonstrate that SIDM-WIR achieves better results to previousmethods both quantitatively and qualitatively.

Deep learning has proven to be effective for enhancing the accuracy and efficiency of attack detection through training with large sample sizes. However, when applied to cyber–physical systems (CPSs), it still encounters challenges such as scarcity of attack samples, the difficulty of selecting features for high-dimensional data, and weak model-generalization ability. In response, this paper proposes ResADM, a transfer-learning-based attack detection method for CPSs. Firstly, an intentional sampling method was employed to construct different sets of samples for each class, effectively balancing the distribution of CPS-attack samples. Secondly, a feature-selection method based on importance was designed to extract the meaningful features from attack behaviors. Finally, a transfer-learning network structure based on ResNet was constructed, and the training parameters of the source model were optimized to form the network-attack detection method. The experimental results demonstrated that ResADM effectively balanced the data classes and extracted 32-dimensional attack-behavior features. After pre-training on the UNSW-NB15 dataset, ResADM achieved a detection accuracy of up to 99.95% for attack behavior on the CICIDS2017 dataset, showcasing its strong practicality and feasibility.

The novel Vehicular Cloud Network (VCN) is a close combination of Vehicular Ad hoc Network, Cloud Computing, and Edge Computing. However, VCN is facing the enormous challenge of security and privacy before it is generalized. To address these issues, VCN communication characteristics and security requirements are summarized first. Secondly, the paper conducts an in-depth analysis of the existing security problems and solutions in VCN based on several attack models, including routing, data, identity, DoS, and hostile attacks. And we summarize the security solutions into two categories: security defense measures based on identity authentication and access control, and security protection measures based on traffic detection and anomaly detection. Thirdly, this paper proposes a reliable VCN security architecture consisting of three types of clouds: vehicular, edge, and central clouds. Moreover, the VCN architecture combines a Vehicle Intrusion Detection and Prevention System with a Vehicular Cloud Security Management Platform to ensure the security of VCN. Last but not least, some open issues worth investigating are discussed.