Pub Date : 2025-01-06DOI: 10.1016/j.ins.2024.121851
Libang Yin , Yining Qian , An-Yang Lu
This paper is concerned with the tracking control problem for nonlinear multi-agent systems susceptible to DoS attacks based on I/O data. First, tailored to DoS attacks characterized by limited attack frequency and duration, an adaptive compensation scheme adjusting input signals based on DoS attack intervals is designed within the framework of the dynamic threshold event-triggered model free adaptive control strategy, which mitigates the impact of communication disruptions. Besides, by reformulating the tracking control problem as a feasibility problem, an algorithm is designed to obtain two variable parameters of controller by employing the linear matrix inequality (LMI) technique, thereby enhancing system performance and reducing the times of event-triggering. And the boundedness of tracking errors is proven using the contraction mapping principle. Finally, the validity of the proposed method is validated through simulation comparisons.
{"title":"Data-driven control with event-triggered dynamic compensation for multi-agent systems under DoS attacks","authors":"Libang Yin , Yining Qian , An-Yang Lu","doi":"10.1016/j.ins.2024.121851","DOIUrl":"10.1016/j.ins.2024.121851","url":null,"abstract":"<div><div>This paper is concerned with the tracking control problem for nonlinear multi-agent systems susceptible to DoS attacks based on I/O data. First, tailored to DoS attacks characterized by limited attack frequency and duration, an adaptive compensation scheme adjusting input signals based on DoS attack intervals is designed within the framework of the dynamic threshold event-triggered model free adaptive control strategy, which mitigates the impact of communication disruptions. Besides, by reformulating the tracking control problem as a feasibility problem, an algorithm is designed to obtain two variable parameters of controller by employing the linear matrix inequality (LMI) technique, thereby enhancing system performance and reducing the times of event-triggering. And the boundedness of tracking errors is proven using the contraction mapping principle. Finally, the validity of the proposed method is validated through simulation comparisons.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"701 ","pages":"Article 121851"},"PeriodicalIF":8.1,"publicationDate":"2025-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143105413","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-06DOI: 10.1016/j.ins.2024.121852
Dewei Bai , Dawen Xia , Xiaoping Wu , Dan Huang , Yang Hu , Youliang Tian , Weihua Ou , Yantao Li , Huaqing Li
Traffic Flow Forecasting (TFF) is crucial for various Intelligent Transportation System (ITS) applications, including route planning and emergency management. TFF is challenging due to the dynamic spatiotemporal patterns exhibited by traffic flow. However, existing TFF methods rely on the “average” spatiotemporal patterns for forecasting. To this end, this study investigates a heuristic-aware model named “Future-heuristic Differential Graph Transformer” (FDGT) for TFF with dynamic spatiotemporal patterns. Specifically, we define a kind of heuristic knowledge, called “future statistic” which provides reference information to describe the status of an object in the future. Then, we embed these statistics as coding features in the temporal domain of inputs. Next, we utilize Higher-order Differential Neural Networks (HDNNs) to enhance the perception of variation trends in the series. Moreover, we employ a Dual Spatiotemporal Convolutional Module (DSCM) to simultaneously learn global and local spatiotemporal dependencies. Finally, the Future-heuristic Fusion (FF) adaptively optimizes the weight distribution of each component, dynamically fuses the decoder's initial prediction and future statistics, and improves the model's generalization ability to capture spatiotemporal heterogeneities at different periods. Experimental results on four public datasets demonstrate that FDGT outperforms existing state-of-the-art TFF methods while maintaining superior execution efficiency.
{"title":"Future-heuristic differential graph transformer for traffic flow forecasting","authors":"Dewei Bai , Dawen Xia , Xiaoping Wu , Dan Huang , Yang Hu , Youliang Tian , Weihua Ou , Yantao Li , Huaqing Li","doi":"10.1016/j.ins.2024.121852","DOIUrl":"10.1016/j.ins.2024.121852","url":null,"abstract":"<div><div>Traffic Flow Forecasting (TFF) is crucial for various Intelligent Transportation System (ITS) applications, including route planning and emergency management. TFF is challenging due to the dynamic spatiotemporal patterns exhibited by traffic flow. However, existing TFF methods rely on the “average” spatiotemporal patterns for forecasting. To this end, this study investigates a heuristic-aware model named “Future-heuristic Differential Graph Transformer” (FDGT) for TFF with dynamic spatiotemporal patterns. Specifically, we define a kind of heuristic knowledge, called “future statistic” which provides reference information to describe the status of an object in the future. Then, we embed these statistics as coding features in the temporal domain of inputs. Next, we utilize Higher-order Differential Neural Networks (HDNNs) to enhance the perception of variation trends in the series. Moreover, we employ a Dual Spatiotemporal Convolutional Module (DSCM) to simultaneously learn global and local spatiotemporal dependencies. Finally, the Future-heuristic Fusion (FF) adaptively optimizes the weight distribution of each component, dynamically fuses the decoder's initial prediction and future statistics, and improves the model's generalization ability to capture spatiotemporal heterogeneities at different periods. Experimental results on four public datasets demonstrate that FDGT outperforms existing state-of-the-art TFF methods while maintaining superior execution efficiency.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"701 ","pages":"Article 121852"},"PeriodicalIF":8.1,"publicationDate":"2025-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143181311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
An effective and robust health management strategy for battery power systems is essential for ensuring the reliable operation of Unmanned Aerial Vehicles (UAVs). This paper presents an adaptive Decay Regularized Stochastic Configuration Network (DRSCN) with multi-level data processing for predicting the Remaining Useful Life (RUL) of UAV batteries. We first introduce a Multisource Signal Enhancement Analysis Framework (MSEAF) designed to efficiently extract critical battery health indicators from complex signals. A key contribution is the enhancement of the SCN model's output layer using decay regularization, which sparsifies the weights and significantly reduces the risk of overfitting in later prediction stages. To further optimize DRSCN, the Convex Lens and Dual-Mechanism Enhanced Sand Cat Swarm Optimization (CLDM-SCSO) algorithm is employed for precise hyperparameter tuning, resulting in improved prediction accuracy. Extensive experiments using the NASA HIRF battery dataset demonstrate the framework's superior accuracy and reliability compared to existing methods, offering an efficient and dependable solution for UAV battery health monitoring.
{"title":"Decay regularized stochastic configuration networks with multi-level data processing for UAV battery RUL prediction","authors":"Zihao Liao , Shaobo Li , Peng Zhou , Chenglong Zhang","doi":"10.1016/j.ins.2024.121840","DOIUrl":"10.1016/j.ins.2024.121840","url":null,"abstract":"<div><div>An effective and robust health management strategy for battery power systems is essential for ensuring the reliable operation of Unmanned Aerial Vehicles (UAVs). This paper presents an adaptive Decay Regularized Stochastic Configuration Network (DRSCN) with multi-level data processing for predicting the Remaining Useful Life (RUL) of UAV batteries. We first introduce a Multisource Signal Enhancement Analysis Framework (MSEAF) designed to efficiently extract critical battery health indicators from complex signals. A key contribution is the enhancement of the SCN model's output layer using decay regularization, which sparsifies the weights and significantly reduces the risk of overfitting in later prediction stages. To further optimize DRSCN, the Convex Lens and Dual-Mechanism Enhanced Sand Cat Swarm Optimization (CLDM-SCSO) algorithm is employed for precise hyperparameter tuning, resulting in improved prediction accuracy. Extensive experiments using the NASA HIRF battery dataset demonstrate the framework's superior accuracy and reliability compared to existing methods, offering an efficient and dependable solution for UAV battery health monitoring.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"701 ","pages":"Article 121840"},"PeriodicalIF":8.1,"publicationDate":"2025-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143105412","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-06DOI: 10.1016/j.ins.2024.121843
Mingde Huang , Yiming Wu , Qiuxia Huang
Privacy concerns and cyber-attacks are two typical threats in networked multi-agent systems (MASs), while little research has properly addressed both. To fill this gap, we investigate a privacy-preserving consensus strategy against cyber-attacks for MASs. First, a novel network eavesdropper model and a cyber-attack model that is more strategic than existing literature are proposed. Then, a homomorphic encryption-based mean subsequence reduced (HE-MSR) consensus algorithm equipped with a privacy protection strategy is designed for each normal agent. The results reveal that the privacy of states of all normal agents and the accurate consensus are guaranteed under mild network topology conditions. Furthermore, these results are extended to the case of a time-varying MAS network environment. Finally, numerical simulations and hardware experiments on Raspberry Pi are conducted to verify the theoretical results.
{"title":"Resilient and privacy-preserving consensus for multi-agent systems","authors":"Mingde Huang , Yiming Wu , Qiuxia Huang","doi":"10.1016/j.ins.2024.121843","DOIUrl":"10.1016/j.ins.2024.121843","url":null,"abstract":"<div><div>Privacy concerns and cyber-attacks are two typical threats in networked multi-agent systems (MASs), while little research has properly addressed both. To fill this gap, we investigate a privacy-preserving consensus strategy against cyber-attacks for MASs. First, a novel network eavesdropper model and a cyber-attack model that is more strategic than existing literature are proposed. Then, a homomorphic encryption-based mean subsequence reduced (HE-MSR) consensus algorithm equipped with a privacy protection strategy is designed for each normal agent. The results reveal that the privacy of states of all normal agents and the accurate consensus are guaranteed under mild network topology conditions. Furthermore, these results are extended to the case of a time-varying MAS network environment. Finally, numerical simulations and hardware experiments on Raspberry Pi are conducted to verify the theoretical results.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"700 ","pages":"Article 121843"},"PeriodicalIF":8.1,"publicationDate":"2025-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143151381","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-06DOI: 10.1016/j.ins.2024.121848
Yongfu Wang , Yang Liu , Ran Yi , Yanchen Jiang
In computer vision and intelligent transportation, object detection algorithms are a major research hotspot for improving the perception of autonomous vehicles. Although deep learning-based object identification algorithms perform well in terms of traffic object detection, they have low accuracy in complex road settings and poor real-time performance. In order to address these issues, this paper proposes a real-time traffic object recognition technique, namely the Toward Our Dream (TOD)-You Only Look Once version 7 (YOLOv7) method, that makes use of a lightweight network model with an improved Deep Stochastic Configuration Networks (DeepSCN). Firstly, the Dilatation Mingle (DM)-Spatial Pyramid Pooling Cross-Stage Partial Convolution (SPPCSPC) module boosts object identification performance for multi-scale objects, unifies semantic information from several size feature maps, and improves the network architecture of the YOLOv7 algorithm. Secondly, an improved DeepSCN is proposed, which improves the classification performance of the detecting head. Lastly, we carried out experiments on ablation, comparison, and visual validation. The experimental findings show that our lightweight technique is better than state-of-the-art methods in terms of accuracy and real-time performance for object detection.
{"title":"Real-time traffic object detection algorithm with deep stochastic configuration networks","authors":"Yongfu Wang , Yang Liu , Ran Yi , Yanchen Jiang","doi":"10.1016/j.ins.2024.121848","DOIUrl":"10.1016/j.ins.2024.121848","url":null,"abstract":"<div><div>In computer vision and intelligent transportation, object detection algorithms are a major research hotspot for improving the perception of autonomous vehicles. Although deep learning-based object identification algorithms perform well in terms of traffic object detection, they have low accuracy in complex road settings and poor real-time performance. In order to address these issues, this paper proposes a real-time traffic object recognition technique, namely the Toward Our Dream (TOD)-You Only Look Once version 7 (YOLOv7) method, that makes use of a lightweight network model with an improved Deep Stochastic Configuration Networks (DeepSCN). Firstly, the Dilatation Mingle (DM)-Spatial Pyramid Pooling Cross-Stage Partial Convolution (SPPCSPC) module boosts object identification performance for multi-scale objects, unifies semantic information from several size feature maps, and improves the network architecture of the YOLOv7 algorithm. Secondly, an improved DeepSCN is proposed, which improves the classification performance of the detecting head. Lastly, we carried out experiments on ablation, comparison, and visual validation. The experimental findings show that our lightweight technique is better than state-of-the-art methods in terms of accuracy and real-time performance for object detection.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"700 ","pages":"Article 121848"},"PeriodicalIF":8.1,"publicationDate":"2025-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143150714","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-06DOI: 10.1016/j.ins.2024.121849
Xiaoli Chen , Youliang Tian , Shuai Wang , Kedi Yang , Wei Zhao , Jinbo Xiong
Privacy Preserving Federated Learning (PPFL) protects the clients' local data privacy by uploading encrypted gradients to the server. However, in real-world scenarios, the heterogeneous distribution of client data makes it challenging to identify poisoning gradients. During local iterations, the models continuously move in different directions, which causes the boundary between benign and malicious gradients to persistently shift. To address these challenges, we design a Dynamic Byzantine-robust Federated Learning (DBFL) defense strategy based on Two-trapdoor Homomorphic Encryption (THE), which enables the detection of encrypted poisoning attacks in heterogeneous data scenarios. Specifically, we introduce a secure Manhattan distance method that accurately measures the differences between elements in two encrypted gradients, allowing for precise detection of poisoning attacks in heterogeneous data scenarios while maintaining privacy. Furthermore, we design a Byzantine-tolerant aggregation mechanism based on dynamic threshold, where the threshold is capable of adapting to the continuously changing boundary between poisoning gradients and benign gradients in heterogeneous data scenarios. This ensures DBFL to effectively exclude poisoning gradients even when 70% of the clients are malicious and controlled by Byzantine attackers. Security analysis demonstrates that DBFL achieves IND-CPA security. Extensive evaluations on two benchmark datasets (i.e., MNIST and CIFAR-10) show that DBFL outperforms existing defense strategies. In particular, DBFL achieves a 7%-40% accuracy improvement in the non-IID setting compared to existing solutions for defending against untargeted and targeted attacks.
{"title":"DBFL: Dynamic Byzantine-Robust Privacy Preserving Federated Learning in Heterogeneous Data Scenario","authors":"Xiaoli Chen , Youliang Tian , Shuai Wang , Kedi Yang , Wei Zhao , Jinbo Xiong","doi":"10.1016/j.ins.2024.121849","DOIUrl":"10.1016/j.ins.2024.121849","url":null,"abstract":"<div><div>Privacy Preserving Federated Learning (PPFL) protects the clients' local data privacy by uploading encrypted gradients to the server. However, in real-world scenarios, the heterogeneous distribution of client data makes it challenging to identify poisoning gradients. During local iterations, the models continuously move in different directions, which causes the boundary between benign and malicious gradients to persistently shift. To address these challenges, we design a Dynamic Byzantine-robust Federated Learning (DBFL) defense strategy based on Two-trapdoor Homomorphic Encryption (THE), which enables the detection of encrypted poisoning attacks in heterogeneous data scenarios. Specifically, we introduce a secure Manhattan distance method that accurately measures the differences between elements in two encrypted gradients, allowing for precise detection of poisoning attacks in heterogeneous data scenarios while maintaining privacy. Furthermore, we design a Byzantine-tolerant aggregation mechanism based on dynamic threshold, where the threshold is capable of adapting to the continuously changing boundary between poisoning gradients and benign gradients in heterogeneous data scenarios. This ensures DBFL to effectively exclude poisoning gradients even when 70% of the clients are malicious and controlled by Byzantine attackers. Security analysis demonstrates that DBFL achieves IND-CPA security. Extensive evaluations on two benchmark datasets (i.e., MNIST and CIFAR-10) show that DBFL outperforms existing defense strategies. In particular, DBFL achieves a 7%-40% accuracy improvement in the non-IID setting compared to existing solutions for defending against untargeted and targeted attacks.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"700 ","pages":"Article 121849"},"PeriodicalIF":8.1,"publicationDate":"2025-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143151259","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-03DOI: 10.1016/j.ins.2024.121833
Betania Silva Carneiro Campello , Sarah BenAmor , Leonardo Tomazeli Duarte , João Marcos Travassos Romano
Preference disaggregation analysis (PDA) is an approach in multicriteria decision analysis that aims to extract preferential information from holistic judgments provided by decision-makers. This paper presents an original methodology for PDA that addresses two challenges in this field. First, we consider the structure of the data as a tensor within the context of PDA to capture decision-makers' preferences based on descriptive measures of the criteria time series, such as trend and average. This approach enables an understanding of decision-makers' preferences in scenarios involving time series analysis, which is common in medium- to long-term impact decisions. Second, the paper addresses the robustness concern in PDA methods, which involves dealing with multiple compatible models reflecting the decision-maker's preferences, using a multi-objective model. This approach allows for identifying multiple preference models and provides a mechanism to converge towards the most likely preference model. The proposed method is evaluated using real data. Results show that the decision-maker's preference for a criterion can vary based on descriptive measures. This highlights the importance of considering both the criterion and the descriptive measures in the decision problem. The multi-objective analysis produces multiple solutions and, under specific conditions, can lead to a single solution reflecting the decision-maker's preferences.
{"title":"Improving preference disaggregation in multicriteria decision making: Incorporating time series analysis and a multi-objective approach","authors":"Betania Silva Carneiro Campello , Sarah BenAmor , Leonardo Tomazeli Duarte , João Marcos Travassos Romano","doi":"10.1016/j.ins.2024.121833","DOIUrl":"10.1016/j.ins.2024.121833","url":null,"abstract":"<div><div>Preference disaggregation analysis (PDA) is an approach in multicriteria decision analysis that aims to extract preferential information from holistic judgments provided by decision-makers. This paper presents an original methodology for PDA that addresses two challenges in this field. First, we consider the structure of the data as a tensor within the context of PDA to capture decision-makers' preferences based on descriptive measures of the criteria time series, such as trend and average. This approach enables an understanding of decision-makers' preferences in scenarios involving time series analysis, which is common in medium- to long-term impact decisions. Second, the paper addresses the robustness concern in PDA methods, which involves dealing with multiple compatible models reflecting the decision-maker's preferences, using a multi-objective model. This approach allows for identifying multiple preference models and provides a mechanism to converge towards the most likely preference model. The proposed method is evaluated using real data. Results show that the decision-maker's preference for a criterion can vary based on descriptive measures. This highlights the importance of considering both the criterion and the descriptive measures in the decision problem. The multi-objective analysis produces multiple solutions and, under specific conditions, can lead to a single solution reflecting the decision-maker's preferences.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"700 ","pages":"Article 121833"},"PeriodicalIF":8.1,"publicationDate":"2025-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143150712","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-03DOI: 10.1016/j.ins.2024.121853
Kamel Mebarkia , Aicha Reffad
In this research, a new supervised, non-parametric and adaptive classifier is proposed: the twigs classifier. The twigs classifier uses twigs that are nothing but the boundary vectors (BVs) and their corresponding twin support vectors (SVs) found by a novel, simple and intuitive algorithm: the boundary vector bisection-based algorithm (BVB). The BVB algorithm pushes iteratively a population of scattered seeds to converge toward the boundaries between classes independently to the classes number and to the data dimensionality. Some limitations of the BVB algorithm were presented and treated. A modified version of the BVB algorithm, the BVB circle-based algorithm (BVBC), is proposed to solve the like spiral problems. The twigs classifier uses a simple dot product between the nearest twig to the object to be classified and the twig-object vector. The adaptation of the twig’s orientation and the pruning of twigs have significantly improved the classification accuracy (CA). The BVB/BVBC algorithm and the twigs classifier are evaluated and validated using synthetic and 20 UCI datasets. The efficiency of the twigs classifier is shown for multi-classification problems, for imbalanced and high-dimension data. The twigs classifier outperforms the majority of the compared classifiers among them the deep learning classifier in some cases, and achieves a misclassification rate of less than 1% in most cases.
{"title":"Twigs classifiers based on the boundary vectors Machine (BVM): A novel approach for supervised learning","authors":"Kamel Mebarkia , Aicha Reffad","doi":"10.1016/j.ins.2024.121853","DOIUrl":"10.1016/j.ins.2024.121853","url":null,"abstract":"<div><div>In this research, a new supervised, non-parametric and adaptive classifier is proposed: the <em>twigs classifier</em>. The twigs classifier uses twigs that are nothing but the boundary vectors (BVs) and their corresponding twin support vectors (SVs) found by a novel, simple and intuitive algorithm: the boundary vector bisection-based algorithm (BVB). The BVB algorithm pushes iteratively a population of scattered seeds to converge toward the boundaries between classes independently to the classes number and to the data dimensionality. Some limitations of the BVB algorithm were presented and treated. A modified version of the BVB algorithm, the BVB circle-based algorithm (BVBC), is proposed to solve the like spiral problems. The twigs classifier uses a simple dot product between the nearest twig to the object to be classified and the twig-object vector. The adaptation of the twig’s orientation and the pruning of twigs have significantly improved the classification accuracy (CA). The BVB/BVBC algorithm and the twigs classifier are evaluated and validated using synthetic and 20 UCI datasets. The efficiency of the twigs classifier is shown for multi-classification problems, for imbalanced and high-dimension data. The twigs classifier outperforms the majority of the compared classifiers among them the deep learning classifier in some cases, and achieves a misclassification rate of less than 1% in most cases.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"701 ","pages":"Article 121853"},"PeriodicalIF":8.1,"publicationDate":"2025-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143105414","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-03DOI: 10.1016/j.ins.2024.121835
Zhe Yang , Libao Deng , Chunlei Li , Yifan Qin , LiLi Zhang
With the rapid advancement of low-Earth-orbit (LEO) satellite technology, satellite phone calls have become increasingly widespread. However, this progress introduces new challenges: the high velocity of LEO satellites necessitates frequent reconnections between terminals and satellites, which can adversely affect communication quality. Moreover, the frequent signal measurements required to maintain connectivity significantly increase the energy consumption of the terminals. To address these challenges, this paper proposes a terminal measurement and satellite-switching strategy. The strategy aims to minimize energy consumption during switching while preserving a high level of user experience. The strategy is based on predictions of satellite cell visibility time and beam visibility time. To further optimise the hyperparameters of this method, we established a simulation-based hyperparameter optimisation model and designed a domain-transformed surrogate-assisted differential evolutionary algorithm (DT-SADE). The algorithm utilises a radial basis function (RBF) model as a surrogate for global search and a Kriging model for local search, and uses domain transformation to address discrepancies between the simulation and surrogate models. Experimental results demonstrate that the proposed method outperforms comparative algorithms across multiple performance metrics, and that hyperparameter optimisation further enhances its performance, highlighting the significance and effectiveness of hyperparameter optimisation.
{"title":"A domain-transformed surrogate-assisted differential evolutionary algorithm for hyperparameter optimisation of satellite handover strategy","authors":"Zhe Yang , Libao Deng , Chunlei Li , Yifan Qin , LiLi Zhang","doi":"10.1016/j.ins.2024.121835","DOIUrl":"10.1016/j.ins.2024.121835","url":null,"abstract":"<div><div>With the rapid advancement of low-Earth-orbit (LEO) satellite technology, satellite phone calls have become increasingly widespread. However, this progress introduces new challenges: the high velocity of LEO satellites necessitates frequent reconnections between terminals and satellites, which can adversely affect communication quality. Moreover, the frequent signal measurements required to maintain connectivity significantly increase the energy consumption of the terminals. To address these challenges, this paper proposes a terminal measurement and satellite-switching strategy. The strategy aims to minimize energy consumption during switching while preserving a high level of user experience. The strategy is based on predictions of satellite cell visibility time and beam visibility time. To further optimise the hyperparameters of this method, we established a simulation-based hyperparameter optimisation model and designed a domain-transformed surrogate-assisted differential evolutionary algorithm (DT-SADE). The algorithm utilises a radial basis function (RBF) model as a surrogate for global search and a Kriging model for local search, and uses domain transformation to address discrepancies between the simulation and surrogate models. Experimental results demonstrate that the proposed method outperforms comparative algorithms across multiple performance metrics, and that hyperparameter optimisation further enhances its performance, highlighting the significance and effectiveness of hyperparameter optimisation.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"700 ","pages":"Article 121835"},"PeriodicalIF":8.1,"publicationDate":"2025-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143150775","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-03DOI: 10.1016/j.ins.2024.121854
Han Wang , Jin-Long Lin , Zhen-Song Chen , Zengqiang Wang
Given the potential severe impact of supply chain disruptions on production capacity and delivery times, such disruptions represent a significant risk for manufacturing enterprises. In this context, it is crucial to enhance the resilience of the supply chain and ensure business continuity during crises. This paper introduces a novel simulation-driven model that identifies the best locations for disaster recovery centers (DRCs) to reduce these risks. Using the anyLogistix software, we simulate an enterprise supply chain and evaluate potential DRC locations based on six key characteristics that were derived from a comprehensive literature review. Then, to evaluate these places and find the best DRC site, decision-makers combined large-scale group decision-making (LSGDM) with improved interval-valued two-tuple linguistic evaluation. anyLogistix simulation confirms the effectiveness of the selected DRCs in mitigating distribution center disruptions, providing crucial management guidance. The innovative integration of LSGDM and simulation demonstrates that establishing DRCs significantly reduces the negative impacts of supply chain disruptions. Comparative analyses validate the model’s rationality and feasibility, offering a valuable framework for enhancing supply chain resilience in manufacturing enterprises. This research contributes to the field by presenting a comprehensive approach to DRC location selection that considers both quantitative and qualitative factors and highlights the importance of proactive risk management strategies in ensuring the continuity of manufacturing supply chains.
{"title":"Large-scale group decision making for simulation-guided disaster recovery center location selection","authors":"Han Wang , Jin-Long Lin , Zhen-Song Chen , Zengqiang Wang","doi":"10.1016/j.ins.2024.121854","DOIUrl":"10.1016/j.ins.2024.121854","url":null,"abstract":"<div><div>Given the potential severe impact of supply chain disruptions on production capacity and delivery times, such disruptions represent a significant risk for manufacturing enterprises. In this context, it is crucial to enhance the resilience of the supply chain and ensure business continuity during crises. This paper introduces a novel simulation-driven model that identifies the best locations for disaster recovery centers (DRCs) to reduce these risks. Using the anyLogistix software, we simulate an enterprise supply chain and evaluate potential DRC locations based on six key characteristics that were derived from a comprehensive literature review. Then, to evaluate these places and find the best DRC site, decision-makers combined large-scale group decision-making (LSGDM) with improved interval-valued two-tuple linguistic evaluation. anyLogistix simulation confirms the effectiveness of the selected DRCs in mitigating distribution center disruptions, providing crucial management guidance. The innovative integration of LSGDM and simulation demonstrates that establishing DRCs significantly reduces the negative impacts of supply chain disruptions. Comparative analyses validate the model’s rationality and feasibility, offering a valuable framework for enhancing supply chain resilience in manufacturing enterprises. This research contributes to the field by presenting a comprehensive approach to DRC location selection that considers both quantitative and qualitative factors and highlights the importance of proactive risk management strategies in ensuring the continuity of manufacturing supply chains.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"700 ","pages":"Article 121854"},"PeriodicalIF":8.1,"publicationDate":"2025-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143151726","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号