Article
Computer Science, Artificial Intelligence
Pengfei Jiao, Tianpeng Li, Huaming Wu, Chang-Dong Wang, Dongxiao He, Wenjun Wang
Summary: This article proposes a new approach for modeling dynamic networks that can simulate both node-level and community-level dynamic behavior. Experimental results demonstrate that the approach achieves state-of-the-art performance in community detection and evolution, and can effectively identify abnormal behavior and events.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Mathematics, Interdisciplinary Applications
Quang Nguyen, Ngoc-Kim-Khanh Nguyen, Davide Cassi, Michele Bellingeri
Summary: The study compared a new node attack strategy with other known attack strategies on complex weighted networks, finding that this new strategy can significantly reduce the weighted efficiency of the network in some cases.
Article
Computer Science, Information Systems
Yingying Huangfu, Liang Zhou, Fan Zhou
Summary: In this letter, a new method based on the dynamic Hoeffding test is proposed to distinguish abnormal relay nodes from those with normal packet loss in wireless communications. An asymptotic analysis of the detection performance on the Hoeffding test is conducted to determine a new dynamic detection threshold for reducing the probability of errors. Theoretical guarantees are provided, showing that both false alarm and missed detection probabilities of the proposed scheme exponentially converge as the number of detection periods increases. Simulation results validate the analysis and illustrate the superior detection performance of the proposed method.
IEEE WIRELESS COMMUNICATIONS LETTERS
(2022)
Article
Computer Science, Theory & Methods
Michele Coscia, Andres Gomez-Lievano, James Mcnerney, Frank Neffke
Summary: This paper discusses the Node Vector Distance (NVD) problem in complex networks and surveys algorithms capable of addressing it. NVD solutions have broader applications beyond computer vision, such as in economics, epidemiology, viral marketing, and sociology. The study showcases differences and similarities of various solution classes on synthetic and real world network data, providing a roadmap for computationally tractable solutions.
ACM COMPUTING SURVEYS
(2021)
Article
Computer Science, Artificial Intelligence
Oguz Findik, Emrah Ozkaynak
Summary: Link prediction is crucial for forecasting future links in complex networks, with traditional methods often falling short due to limited consideration of node weighting. This study proposes a novel model based on node weighting, showing superior success rates compared to current technology methods.
Article
Computer Science, Artificial Intelligence
Renny Marquez, Richard Weber
Summary: Community detection is a crucial task in social network analysis, but static networks may not capture the dynamics of real-world problems. We propose CoDeDANet, an algorithm that detects communities in dynamic attributed networks by considering both link and node information. By optimizing the importance of attributes based on spectral clustering and incorporating tensors to capture past information, CoDeDANet outperforms other state-of-the-art community detection algorithms in tests on synthetic and real-world networks.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Emanuele Pio Barracchia, Gianvito Pio, Albert Bifet, Heitor Murilo Gomes, Bernhard Pfahringer, Michelangelo Ceci
Summary: In many real-world domains, data can naturally be represented as networks. These networks often have a dynamic nature, and considering this dynamism is crucial for accurate analysis. This work proposes LP-ROBIN, a novel method that uses incremental embedding to capture the dynamics of network structure and predicts new links. Experimental results demonstrate that LP-ROBIN achieves impressive performance and competitive running times.
INFORMATION SCIENCES
(2022)
Article
Automation & Control Systems
Ronghua Shang, Weitong Zhang, Licheng Jiao, Xiangrong Zhang, Rustam Stolkin
Summary: In this article, a dynamic node immune model based on community structure and threshold is proposed to better restrain the spread of harmful information. Experimental results show that this model performs well in real networks.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Computer Science, Information Systems
Fabio Barbosa, Amaro de Sousa, Agostinho Agra, Krzysztof Walkowiak, Roza Goscien
Summary: In this study, we address the issue of multiple node failure events in dynamic Elastic Optical Networks (EONs) by proposing RMSA algorithms that combine the path disaster availability metric with spectrum usage metrics. This combination allows for a dynamic adjustment of resource utilization goals based on network load levels, aiming to achieve a good balance between spectrum usage efficiency and resilience to multiple node failures. Simulation results show that these algorithms offer the best trade-off in terms of spectrum efficiency and network resilience.
OPTICAL SWITCHING AND NETWORKING
(2021)
Article
Engineering, Industrial
Sebastian Wandelt, Yifan Xu, Xiaoqian Sun
Summary: This study uses complex network analysis techniques to evaluate the importance of airports in airline operations and proposes a recovery baseline model for airline recovery under node disruptions. The research found that existing node importance methods often underestimate the impact of node failures.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Green & Sustainable Science & Technology
Xiangyu Ma, Huijie Zhou, Zhiyi Li
Summary: This paper provides a comprehensive literature review on the application of complex network theories in resilience evaluation and enhancement of modern power systems. It decomposes resilience into structural and operational aspects, discussing structural resilience through graph modeling and analyzing static and dynamic characteristics, and investigating operational resilience through the progression of preventive, corrective, and restorative strategies in extreme events. It also extends the discussion to multilayer networks as modern power systems are increasingly interconnected with communication networks and other energy carriers. Overall, complex network theories are found to be effective in understanding and improving the structural and operational resilience of modern power systems.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2021)
Article
Environmental Sciences
Deniz Berfin Karakoc, Megan Konar
Summary: Global food trade plays a crucial role in food security and availability. Research shows that there is a competitive relationship between efficiency and resilience in food trade networks when only network topology is considered, but a cooperative relationship exists when trade connection intensity is taken into account. This complex network framework can assist policymakers in evaluating the relationship between efficiency and resilience in critical supply chains.
ENVIRONMENTAL RESEARCH LETTERS
(2021)
Article
Physics, Multidisciplinary
Fan Wang, Gaogao Dong, Lixin Tian, H. Eugene Stanley
Summary: Percolation behavior and critical phenomena are important characteristics of complex networks. Studying the percolation behavior of finite components provides insights into the overall network from a microscopic perspective and offers a new approach to determine the critical threshold.
NEW JOURNAL OF PHYSICS
(2022)
Article
Mathematics, Interdisciplinary Applications
Xing-Zhang Wen, Yue Zheng, Wen-Li Du, Zhuo-Ming Ren
Summary: This article investigates the impact of network topology on classical node centralities, including degree, closeness, betweenness, and eigenvector. Through constructing two types of growing scale-free networks with adjustable clustering coefficient and assortativity, and simulating three null models on ten real networks to adjust cluster and assortativity, the study finds that cluster and assortativity have a significant impact on node centrality in complex networks. Therefore, it is important to consider network topology when selecting node centralities to identify the significance or influence of nodes in complex networks.
CHAOS SOLITONS & FRACTALS
(2023)
Article
Multidisciplinary Sciences
Mark M. Dekker, Debabrata Panja
Summary: Many socio-technical systems, such as supply chains, international trade and human mobility, are vulnerable to large-scale disruptions due to the dynamical intertwining of heterogeneous operational elements, agents and locations over complex networks. The spreading of train delays on the railway network reveals that large-scale disruptions rely on dynamic interdependencies among multiple layers of operational elements. The cascading delay mechanism amplifies delays locally and spreads them over the network, leading to new constraints elsewhere.
Article
Computer Science, Information Systems
Yan Zong, Xuewu Dai, Zhuangkun Wei, Mengbang Zou, Weisi Guo, Zhiwei Gao
Summary: A robust packet-coupled oscillators (R-PkCOs) protocol is proposed to reduce the effects of disturbances in clock drift, timestamping uncertainty, and delays. A static output controller is used to adjust the drifting clock, and the H-8 robust control design solution ensures a limited impact of drifting frequency and delays on synchronization accuracy. Experimental results demonstrate a time synchronization precision of 6 μs in a 21-node IEEE 802.15.4 network.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Engineering, Chemical
Yanxin Liu, Weisi Guo, Philip Longhurst, Ying Jiang
Summary: The pseudo-parallel first-order model and the first-order autoregressive (AR (1)) model showed high accuracy in predicting the residual biogas potential (RBP) result using experimental biogas production data of 15 days. Multivariate regression with decision trees (DTs) successfully predicted the model parameters for the AR (1) model from substrate physicochemical parameters. The prediction accuracy of DTs can be further improved with a larger training dataset.
Article
Computer Science, Information Systems
Mengbang Zou, Weisi Guo
Summary: Synchronization is critical for system-level behavior in various systems, and the relationship between network topology and synchronizability remains an open challenge. We propose an analytical method using perturbation theory to accurately determine synchronizability by estimating extreme eigenvalues. Our method reveals the combined influence of global and local topology on synchronizability and shows a negative relationship between the smallest nonzero eigenvalue lambda((2)) and the local assortativity of nodes with the smallest degree value. We validate our framework in a scale-free network driven by ordinary differential equations.
IEEE SYSTEMS JOURNAL
(2023)
Article
Engineering, Electrical & Electronic
Zhuoxiao Lin, Bin Li, Zhuangkun Wei, Yu Huang, Weisi Guo, Chenglin Zhao
Summary: In this paper, a non-coherent signal detector is proposed to take advantage of the biochemical diversity property of multiple reacting molecules for reliable nano-scale communications. The detector utilizes the dynamical transient characteristics of messenger, reactant, and product molecules to achieve diversity detection in a single-input multiple-output (SIMO) system. Theoretical analysis and numerical simulations demonstrate the advantages of the detector, which outperform conventional detectors and even surpass the coherent maximum a posteriori (MAP) detector.
IEEE TRANSACTIONS ON COMMUNICATIONS
(2023)
Article
Computer Science, Information Systems
Adolfo Perrusquia, Weisi Guo
Summary: In this paper, a reward inference algorithm for discrete-time expert controllers is proposed, which utilizes the complementary mechanisms of the striatum, neocortex, and hippocampus to infer the reward function. The proposed approach combines data-driven and online learning methods and shows stability and convergence using Lyapunov stability theory.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Chen Li, Schyler C. C. Sun, Zhuangkun Wei, Antonios Tsourdos, Weisi Guo
Summary: Increased drone proliferation poses new threats to airports and national infrastructures, with estimated economic damages of millions of dollars per day. Training accurate drone detection algorithms under scarce data is challenging. We propose a method using GAN and TDA to understand the general data distribution and acquire under-represented data, achieving a significant improvement in accuracy compared to existing methods.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Automation & Control Systems
Adolfo Perrusquia, Weisi Guo
Summary: This article presents a closed-loop output error approach for trajectory inference of a class of linear systems. The approach combines the advantages of state estimation and parameter identification algorithms, using online data and an estimated model to infer a noise-free trajectory. A composite update rule based on a least-squares rule is proposed to improve robustness and convergence.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Automation & Control Systems
Adolfo Perrusquia, Weisi Guo
Summary: This article proposes a physics informed trajectory inference method for nonlinear systems. The approach combines state and parameter estimation algorithms to infer the trajectory of the nonlinear system using noisy state measurements. The algorithm utilizes a parallel estimated model with a low-pass filter parameterization, which allows for noise attenuation and avoids biased estimation by using estimated states instead of noisy measurements.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Weisi Guo, Zhuangkun Wei, Oscar Gonzalez, Adolfo Perrusquia, Antonios Tsourdos
Summary: Autonomous systems (ASs) cooperate for safe navigation by using centralized or distributed coordination mechanisms that consist of observations, unobservable states, and control variables. The security of data transfer between ASs is crucial for safety, and both cryptography and physical layer security (PLS) methods are employed to secure communication surfaces, each with their own limitations and dependencies.
IEEE VEHICULAR TECHNOLOGY MAGAZINE
(2023)
Article
Information Science & Library Science
Cian Naik, Cassidy R. R. Sugimoto, Vincent Lariviere, Chenlei Leng, Weisi Guo
Summary: Diversity in human capital is crucial for high-quality research, especially in areas involving diverse cultures, environments, and challenges. However, there is a lack of research on the quantification of diverse academic collaborations and their impact on research quality, particularly at an international scale and across different domains. This study measures the impact of geographic diversity in coauthorships on paper citation across different academic fields. The results demonstrate that geographic coauthor diversity improves paper citation, but collaborations involving very long distances have varying impacts and there are well-established collaboration circles that yield less impact than expected based on travel distances. These findings apply to different subject areas with varying strengths and can help researchers identify new opportunities and inform funders on areas that require additional support.
QUANTITATIVE SCIENCE STUDIES
(2023)
Article
Computer Science, Theory & Methods
Zhuangkun Wei, Bin Li, Weisi Guo
Summary: The development of reconfigurable intelligent surfaces (RIS) has advanced the research of physical layer security (PLS). However, there is a lack of research on how adversarial RIS can be used to attack and obtain legitimate secret keys generated by PL-SKG.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2023)
Article
Engineering, Civil
Kai-Fung Chu, Weisi Guo
Summary: This research proposes a federated deep deterministic policy gradient (FDDPG) algorithm with privacy preservation to enhance the profitability and passenger satisfaction of Mobility-as-a-service (MaaS). Experimental results demonstrate that this method can increase MaaS profit and passenger satisfaction by approximately 90% and 15% respectively, while maintaining stable training against agent dropout. The approach and findings of this study could enhance MaaS utility and promote passenger trust and participation in MaaS and other data-driven transportation systems.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Benjamin Fraser, Brendan Copp, Gurpreet Singh, Orhan Keyvan, Tongfei Bian, Valentin Sonntag, Yang Xing, Weisi Guo, Antonios Tsourdos
Summary: This paper explores the use of multi-person pose estimation techniques to reduce the risk of airborne pathogen transmission. The developed techniques analyze CCTV inputs for crowd analysis and utilize pose feature positions and mask detection to assess interpersonal distance and behavior. By combining multiple models and assessing transmission risk based on scientific literature, a real-time risk density heat-map is displayed. This system has the potential to improve public space management and reduce transmission in future pandemics.
2022 IEEE INTERNATIONAL CONFERENCE ON MULTISENSOR FUSION AND INTEGRATION FOR INTELLIGENT SYSTEMS (MFI)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Adeyemi Osigbesan, Solene Barrat, Harkeerat Singh, Dongzi Xia, Siddharth Singh, Yang Xing, Weisi Guo, Antonios Tsourdos
Summary: Fall-related injuries in the workplace account for a significant percentage of global work accident claims. This paper proposes a modern method using computer vision technology to detect and classify fall events in real-time by analyzing body poses.
2022 IEEE INTERNATIONAL CONFERENCE ON MULTISENSOR FUSION AND INTEGRATION FOR INTELLIGENT SYSTEMS (MFI)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Adolfo Perrusquia, Weisi Guo
2022 8TH INTERNATIONAL CONFERENCE ON CONTROL, DECISION AND INFORMATION TECHNOLOGIES (CODIT'22)
(2022)
