Search

Digital twin technology provides a reliable paradigm to address the high trial-and-error costs and limited perception capabilities in satellite networking. However, the dynamic constellation topology and real-time twin applications remain significant challenges in satellite network design. This paper proposes a network topology simulation approach that dynamically analyzes the inter-satellite topology based on pre-calculated ephemeris and orbital information. Furthermore, the paper introduces a digital twin algorithm based on network virtualization, cloud platform management, and software-defined networking to validate and analyze the twin requirements at different stages. Finally, a low Earth orbit (LEO) constellation twin validation environment is constructed to verify the networking protocols at various stages. The experimental results demonstrate the performance of the proposed twin systems at different stages. [ABSTRACT FROM AUTHOR]

Researchers from different fields have studied the effects of the COVID-19 pandemic and published their results in peer-reviewed journals indexed in international databases such as Web of Science (WoS), Scopus, PubMed. Focusing on efficient methods for navigating the extensive literature on COVID-19 pandemic research, our study conducts a content analysis of the top 1000 cited papers in WoS that delve into the subject by using elements of natural language processing (NLP). Knowing that in WoS, a scientific paper is described by the group Paper = {Abstract, Keyword, Title}; we obtained via NLP methods the word dictionaries with their frequencies of use and the word cloud for the 100 most used words, and we investigated if there is a degree of similarity between the titles of the papers and their abstracts, respectively. Using the Python packages NLTK, TextBlob, VADER, we computed sentiment scores for paper titles and abstracts, analyzed the results, and then, using Azure Machine Learning-Sentiment analysis, extended the range of comparison of sentiment scores. Our proposed analysis method can be applied to any research topic or theme from papers, articles, or projects in various fields of specialization to create a minimal dictionary of terms based on frequency of use, with visual representation by word cloud. Complementing the content analysis in our research with sentiment and similarity analysis highlights the different or similar treatment of the topics addressed in the research, as well as the opinions and feelings conveyed by the authors in relation to the researched issue. [ABSTRACT FROM AUTHOR]

Paper recommendation systems are important for alleviating academic information overload. Such systems provide personalized recommendations based on implicit feedback from users, supplemented by their subject information, citation networks, etc. However, such recommender systems face problems like data sparsity for positive samples and uncertainty for negative samples. In this paper, we address these two issues and improve upon them from the perspective of metric learning. The algorithm is modeled as a push–pull loss function. For the positive sample pull-out operation, we introduce a context factor, which accelerates the convergence of the objective function through the multiplication rule to alleviate the data sparsity problem. For the negative sample push operation, we adopt an unbiased global negative sample method and use an intermediate matrix caching method to greatly reduce the computational complexity. Experimental results on two real datasets show that our method outperforms other baseline methods in terms of recommendation accuracy and computational efficiency. Moreover, our metric learning method that introduces context improves by more than 5% over the element-wise alternating least squares method. We demonstrate the potential of metric learning in addressing the problem of implicit feedback recommender systems with positive and negative sample imbalances. [ABSTRACT FROM AUTHOR]

This paper aims to structure and semantically describe the information within the industrial chain by constructing an Industry Chain Knowledge Graph (ICKG), enabling more efficient and intelligent information management and analysis. In more detail, this paper constructs a multi-domain industrial chain dataset and proposes a method that combines the top-down establishment of a semantic expression framework with the bottom-up establishment of a data layer to build an ICKG. In the data layer, a deep learning algorithm based on BERT-BiLSTM-CRF is used to extract industry chain entities from relevant literature and reports. The results indicate that the model can effectively identify industry chain entities. These entities and relationships populate a Neo4j graph database, creating a large-scale ICKG for visual display and aiding cross-domain applications. [ABSTRACT FROM AUTHOR]

Recently, the phenomenon of switching back from microservice to monolith has increased in frequency, leading to intense debate in the industry. In this paper, we conduct a multivocal literature review to investigate reasoning and key aspects to pay attention to when switching back and analyze other practitioners' opinions. The results show four cases of switching back from microservice to monolith: Istio control plane, Amazon Prime Video monitoring service, Segment, and InVision. The five main reasons that led to switching back are cost, complexity, scalability, performance, and organization. During the switching back process, six key aspects need to be addressed: (1) stopping the development of more services, (2) consolidating and testing paths, (3) unifying data storage, (4) implementing the message bus principle, (5) giving up diverse techniques, and (6) learning to use modular design principles. As to the practitioners' opinions, they had mixed views about the switching back phenomenon. However, most thought that switching back required consideration of the actual system situation and principles. These results pave the way for further research and guide researchers and companies through the process of switching back from microservice to monolith. [ABSTRACT FROM AUTHOR]

In the dynamic visual cabinet, the occurrence of motion blur when consumers take out commodities will reduce the accuracy of commodity detection. Recently, although Transformer-based video deblurring networks have achieved results compared to Convolutional Neural Networks in some blurring scenarios, they are still challenging for the non-uniform blurring problem that occurs when consumers pick up the commodities, such as the problem of difficult alignment of blurred video frames of small commodities and the problem of underutilizing the effective information between the video frames of commodities. Therefore, an improved Transformer video deblurring network is proposed. Firstly, a multi-scale Transformer feature extraction method is utilized for non-uniform blurring. Secondly, for the problem of difficult alignment of small-item-blurred video frames, a temporal interactive attention mechanism is designed for video frame alignment. Finally, a feature recurrent fusion mechanism is introduced to supplement the effective information of commodity features. The experimental results show that the proposed method has practical significance in improving the accuracy of commodity detection. Moreover, compared with the recent Transformer deblurring algorithm Video Restoration Transformer, the Peak Signal-to-Noise Ratio of this paper's algorithm is higher than that of the Deep Video Deblurring dataset and the Fuzzy Commodity Dataset by 0.23 dB and 0.81 dB, respectively. [ABSTRACT FROM AUTHOR]

Mobile phone backplanes are an important part of mobile phones, and are often affected by a wide range of factors during the manufacturing process, resulting in defects of various scales and similar backgrounds. Therefore, accurately identifying these defects is crucial for improving mobile phone quality. To address this challenge, this paper proposes a multi-scale and dynamic attention fusion UNet (MDAF-UNet) model. The model innovatively combines normal convolution with dilated convolution. This allows the model to capture subtle features of defects and to perceive a larger range of feature variations. Moreover, an improved attention mechanism is introduced in this paper. It fuses channel attention and spatial attention, and dynamically adjusts the feature fusion strategy with learnable weights. This allows the model to increase the attention of important features and improve the effectiveness of feature representation. Experimental results on a publicly available dataset show that the MDAF-UNet model achieves 66.9% Mean Intersection over Union (MIoU), outperforming other state-of-the-art models. This result provides an effective solution to the mobile phone backplane defect segmentation problem. [ABSTRACT FROM AUTHOR]

The video game Factorio is a popular real-time strategy game that hosts a variety of gameplay scenarios analogous to real-world problems such as factory building, pipeline optimization, and conveyor belt logistics. It also considers environmental costs such as pollution and energy consumption. This game focuses on automation and resource management within the context of creating and maintaining factories. Players are challenged to design efficient factories, automate production, control energy consumption, and balance resource utilization, all while defending against environmental challenges such as pollution and hostile enemies. Factorio is a complex software; however, studies have demonstrated its ability to enhance student satisfaction when integrated as an educational module at the university level. This paper aims to utilize the power of Petri Net theory and the practical tool GPenSIM to model and simulate several Factorio pipelines and score them against one another, exploring which pipeline configuration is optimal. The primary objective is to make a mathematical model (Petri Net) to better understand the game's complex mechanisms, identify bottlenecks within the system, and explore opportunities for optimization within the network of modules. The insights gained from this paper may find practical applications in real-world scenarios. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) is a rapidly growing network that shares information over the Internet via interconnected devices. In addition, this network has led to new security challenges in recent years. One of the biggest challenges is the impact of denial-of-service (DoS) attacks on the IoT. The Information-Centric Network (ICN) infrastructure is a critical component of the IoT. The ICN has gained recognition as a promising networking solution for the IoT by supporting IoT devices to be able to communicate and exchange data with each other over the Internet. Moreover, the ICN provides easy access and straightforward security to IoT content. However, the integration of IoT devices into the ICN introduces new security challenges, particularly in the form of DoS attacks. These attacks aim to disrupt or disable the normal operation of the ICN, potentially leading to severe consequences for IoT applications. Machine learning (ML) is a powerful technology. This paper proposes a new approach for developing a robust and efficient solution for detecting DoS attacks in ICN-IoT networks using ML technology. ML is a subset of artificial intelligence (AI) that focuses on the development of algorithms. While several ML algorithms have been explored in the literature, including neural networks, decision trees (DTs), clustering algorithms, XGBoost, J48, multilayer perceptron (MLP) with backpropagation (BP), deep neural networks (DNNs), MLP-BP, RBF-PSO, RBF-JAYA, and RBF-TLBO, researchers compare these detection approaches using classification metrics such as accuracy. This classification metric indicates that SVM, RF, and KNN demonstrate superior performance compared to other alternatives. The proposed approach was carried out on the NDN architecture because, based on our findings, it is the most used one and has a high percentage of various types of cyberattacks. The proposed approach can be evaluated using an ndnSIM simulation and a synthetic dataset for detecting DoS attacks in ICN-IoT networks using ML algorithms. [ABSTRACT FROM AUTHOR]

With the development of communication technology, the interference from passive intermodulation to the communication system has become increasingly severe. Locating the source of passive intermodulation and then repairing or replacing the device experiencing passive intermodulation is a reliable method. Therefore, this paper proposes a dual-frequency signal localization method based on the principle of phase ranging to accurately locate the passive intermodulation sources in RF cables. This method switches the test signal frequency to obtain the phase of passive intermodulation signals at different frequencies to localize the passive intermodulation source. This paper analyzes the principle of the passive intermodulation localization method based on dual-frequency signals, establishes a localization system for experiments, and finally, analyzes the experimental error to propose an optimization strategy. Through experimental verification, this method can stabilize the positioning error of a passive intermodulation source in a transmission line less than 0.06 m. Meanwhile, this method has high positioning accuracy and the advantages of slight computational complexity, fast positioning speed, and only a small number of requirements on hardware resources. It can work in communication scenarios like satellite and base stations, providing a new technical solution for localizing passive intermodulation interference sources in 5G and 6G communication systems. [ABSTRACT FROM AUTHOR]