In the Internet of Vehicles (IoV), data privacy concerns have prompted the adoption of Federated Learning (FL). Efficiency improvements in FL remain a focal area of research, with recent studies exploring model pruning to lessen both computation and communication overhead. However, in the IoV, model pruning presents unique challenges and remains underexplored. Pruning strategy design is critical as it directly impacts each vehicle's learning latency and capacity to participate in FL. Furthermore, FL performance and model pruning are intricately connected. Additionally, the fluctuating number and mobility states of vehicles per round complicate determining the optimal pruning ratio, closely intertwining pruning with vehicle selection. This study introduces Vehicular Federated Learning with Adaptive Model Pruning (VFed-AMP) to tackle these challenges by integrating adaptive pruning with dynamic vehicle selection and resource allocation. We analyze the impact of pruning ratios on learning latency and convergence rate. Then, guided by these findings, a joint optimization problem is formulated to maximize the convergence rate concerning optimal vehicle selection, bandwidth allocation, and pruning ratios. Finally, a low-complexity algorithm for joint adaptive pruning and vehicle scheduling is proposed to address this problem. Through theoretical analysis and system design, VFed-AMP enhances FL efficiency and scalability in the IoV, offering insights into optimizing FL performance through strategic model adjustments. Numerical results on various datasets show VFed-AMP achieves superior training accuracy (e.g., at least 13.4% improvement for BelgiumTS) and significantly reduces training time (e.g., at least up to$1.8\times$for CIFAR-10) compared to traditional FL methods.

SUMMARY: TAD boundaries are essential for organizing the chromatin spatial structure and regulating gene expression in eukaryotes. However, for large-scale pan-3D genome research, identifying conserved and specific TAD boundaries across different species or individuals is computationally challenging. Here, we present Tcbf, a rapid and powerful Python/R tool that integrates gene synteny blocks and homologous sequences to automatically detect conserved and specific TAD boundaries among multiple species, which can efficiently analyze huge genome datasets, greatly reduce the computational burden and enable pan-3D genome research. AVAILABILITY AND IMPLEMENTATION: Tcbf is implemented by Python/R and is available at https://github.com/TcbfGroup/Tcbf under the MIT license.

Abstract Developing an accurate onboard‐camera pose estimation is one major challenge of satellite systems, and the attempt of improving remote sensing camera pose accuracy never ceases. The camera pose can be recovered by aligning a captured 2D image and a 3D digital surface model of the corresponding scene. In this paper, a novel camera pose estimation method from captured images with the over known real scene 3D products is proposed to enhance remote sensing camera attitude accuracy. The purpose of this estimation is to determine the pose of a camera purely from an image based on a known 3D model, where 3D products of very high spatial resolution are projected onto image space by virtual camera system with error‐contained initial exterior orientation parameters, and whether the pose of the camera can be determined precisely depends on the 2D–3D registration result. The process consists of two steps: (1) feature extraction and (2) similarity measure and registration. Furthermore, the proposed method revises the rotation matrix and translation vector by utilizing formulation based on quaternion representation of rotation, respectively. We evaluate our method on challenging simulation data and results show that acceptable accuracy of camera pose can be achieved.

Cloud 5G and Cloud 6G technologies are strong backbone infrastructures to provide high data rate and data storage with low latency for preserving QoS (Quality of Service) and QoE (Quality of Experience) in applications such as driverless vehicles, drone-based deliveries, smart cities and factories, remote medical diagnosis and surgery, and artificial-intelligence-based personalized assistants. There are many techniques to support the aforementioned applications, but for privacy preservation of Cloud 5G, the existing methods are still not sufficient. Public key encryption (PKE) scheme is an important means to protect user data privacy in Cloud 5G. Currently, the most common PKE used in Cloud 5G is CCA or CPA secure ones. However, its security level maybe not enough. SOA security is a stronger security standard than CPA and CCA. Roughly speaking, PKE with SOA security means that the adversary is allowed to open a subset of challenger ciphertexts and obtains the corresponding encrypted messages and randomness, but the unopended messages and randomness remain secure in the rest of the challenger ciphertexts. Security against SOA in PKEs has been a research hotspot, especially with the wide discussion in Cloud 5G. We revisited the SOA-CLE and proposed a new security proof, which is more concise and user friendly to understand privacy preservation in Cloud 5G applications.

With the development of staring array cameras onboard, LEO satellites can work as surveillance cameras by monitoring a particular area for dozens of seconds. In this paper, a new method named as DBM (Detection Box Model) for moving targets detection is proposed for satellite based surveillance video. Compared with previous work done by other researchers by detecting moving target on roads only, DBM is a generalized method to detect moving targets in different size for the whole scene. DBM consist of two main steps: data preparation and detection of DBM. Our experiments with SkySat-1 data show that, the detection performance can reach a satisfactory level in terms of both precision rate and recall rate.