Wireless Sensor Networks(WSN)have gained significant attention over recent years due to their extensive applications in various domains such as environmentalmonitoring,healthcare systems,industrial automation,and smar...Wireless Sensor Networks(WSN)have gained significant attention over recent years due to their extensive applications in various domains such as environmentalmonitoring,healthcare systems,industrial automation,and smart cities.However,such networks are inherently vulnerable to different types of attacks because they operate in open environments with limited resources and constrained communication capabilities.Thepaper addresses challenges related to modeling and analysis of wireless sensor networks and their susceptibility to attacks.Its objective is to create versatile modeling tools capable of detecting attacks against network devices and identifying anomalies caused either by legitimate user errors or malicious activities.A proposed integrated approach for data collection,preprocessing,and analysis in WSN outlines a series of steps applicable throughout both the design phase and operation stage.This ensures effective detection of attacks and anomalies within WSNs.An introduced attackmodel specifies potential types of unauthorized network layer attacks targeting network nodes,transmitted data,and services offered by the WSN.Furthermore,a graph-based analytical framework was designed to detect attacks by evaluating real-time events from network nodes and determining if an attack is underway.Additionally,a simulation model based on sequences of imperative rules defining behaviors of both regular and compromised nodes is presented.Overall,this technique was experimentally verified using a segment of a WSN embedded in a smart city infrastructure,simulating a wormhole attack.Results demonstrate the viability and practical significance of the technique for enhancing future information security measures.Validation tests confirmed high levels of accuracy and efficiency when applied specifically to detecting wormhole attacks targeting routing protocols in WSNs.Precision and recall rates averaged above the benchmark value of 0.95,thus validating the broad applicability of the proposed models across varied scenarios.展开更多
Previous studies have shown that deep learning is very effective in detecting known attacks.However,when facing unknown attacks,models such as Deep Neural Networks(DNN)combined with Long Short-Term Memory(LSTM),Convol...Previous studies have shown that deep learning is very effective in detecting known attacks.However,when facing unknown attacks,models such as Deep Neural Networks(DNN)combined with Long Short-Term Memory(LSTM),Convolutional Neural Networks(CNN)combined with LSTM,and so on are built by simple stacking,which has the problems of feature loss,low efficiency,and low accuracy.Therefore,this paper proposes an autonomous detectionmodel for Distributed Denial of Service attacks,Multi-Scale Convolutional Neural Network-Bidirectional Gated Recurrent Units-Single Headed Attention(MSCNN-BiGRU-SHA),which is based on a Multistrategy Integrated Zebra Optimization Algorithm(MI-ZOA).The model undergoes training and testing with the CICDDoS2019 dataset,and its performance is evaluated on a new GINKS2023 dataset.The hyperparameters for Conv_filter and GRU_unit are optimized using the Multi-strategy Integrated Zebra Optimization Algorithm(MIZOA).The experimental results show that the test accuracy of the MSCNN-BiGRU-SHA model based on the MIZOA proposed in this paper is as high as 0.9971 in the CICDDoS 2019 dataset.The evaluation accuracy of the new dataset GINKS2023 created in this paper is 0.9386.Compared to the MSCNN-BiGRU-SHA model based on the Zebra Optimization Algorithm(ZOA),the detection accuracy on the GINKS2023 dataset has improved by 5.81%,precisionhas increasedby 1.35%,the recallhas improvedby 9%,and theF1scorehas increasedby 5.55%.Compared to the MSCNN-BiGRU-SHA models developed using Grid Search,Random Search,and Bayesian Optimization,the MSCNN-BiGRU-SHA model optimized with the MI-ZOA exhibits better performance in terms of accuracy,precision,recall,and F1 score.展开更多
Deep neural networks(DNNs)have found extensive applications in safety-critical artificial intelligence systems,such as autonomous driving and facial recognition systems.However,recent research has revealed their susce...Deep neural networks(DNNs)have found extensive applications in safety-critical artificial intelligence systems,such as autonomous driving and facial recognition systems.However,recent research has revealed their susceptibility to backdoors maliciously injected by adversaries.This vulnerability arises due to the intricate architecture and opacity of DNNs,resulting in numerous redundant neurons embedded within the models.Adversaries exploit these vulnerabilities to conceal malicious backdoor information within DNNs,thereby causing erroneous outputs and posing substantial threats to the efficacy of DNN-based applications.This article presents a comprehensive survey of backdoor attacks against DNNs and the countermeasure methods employed to mitigate them.Initially,we trace the evolution of the concept from traditional backdoor attacks to backdoor attacks against DNNs,highlighting the feasibility and practicality of generating backdoor attacks against DNNs.Subsequently,we provide an overview of notable works encompassing various attack and defense strategies,facilitating a comparative analysis of their approaches.Through these discussions,we offer constructive insights aimed at refining these techniques.Finally,we extend our research perspective to the domain of large language models(LLMs)and synthesize the characteristics and developmental trends of backdoor attacks and defense methods targeting LLMs.Through a systematic review of existing studies on backdoor vulnerabilities in LLMs,we identify critical open challenges in this field and propose actionable directions for future research.展开更多
Unsteady aerodynamic characteristics at high angles of attack are of great importance to the design and development of advanced fighter aircraft, which are characterized by post-stall maneuverability with multiple Deg...Unsteady aerodynamic characteristics at high angles of attack are of great importance to the design and development of advanced fighter aircraft, which are characterized by post-stall maneuverability with multiple Degrees-of-Freedom(multi-DOF) and complex flow field structure.In this paper, a special kind of cable-driven parallel mechanism is firstly utilized as a new suspension method to conduct unsteady dynamic wind tunnel tests at high angles of attack, thereby providing experimental aerodynamic data. These tests include a wide range of multi-DOF coupled oscillatory motions with various amplitudes and frequencies. Then, for aerodynamic modeling and analysis, a novel data-driven Feature-Level Attention Recurrent neural network(FLAR) is proposed. This model incorporates a specially designed feature-level attention module that focuses on the state variables affecting the aerodynamic coefficients, thereby enhancing the physical interpretability of the aerodynamic model. Subsequently, spin maneuver simulations, using a mathematical model as the baseline, are conducted to validate the effectiveness of the FLAR. Finally, the results on wind tunnel data reveal that the FLAR accurately predicts aerodynamic coefficients, and observations through the visualization of attention scores identify the key state variables that affect the aerodynamic coefficients. It is concluded that the proposed FLAR enhances the interpretability of the aerodynamic model while achieving good prediction accuracy and generalization capability for multi-DOF coupling motion at high angles of attack.展开更多
Ransomware attacks pose a significant threat to critical infrastructures,demanding robust detection mechanisms.This study introduces a hybrid model that combines vision transformer(ViT)and one-dimensional convolutiona...Ransomware attacks pose a significant threat to critical infrastructures,demanding robust detection mechanisms.This study introduces a hybrid model that combines vision transformer(ViT)and one-dimensional convolutional neural network(1DCNN)architectures to enhance ransomware detection capabilities.Addressing common challenges in ransomware detection,particularly dataset class imbalance,the synthetic minority oversampling technique(SMOTE)is employed to generate synthetic samples for minority class,thereby improving detection accuracy.The integration of ViT and 1DCNN through feature fusion enables the model to capture both global contextual and local sequential features,resulting in comprehensive ransomware classification.Tested on the UNSW-NB15 dataset,the proposed ViT-1DCNN model achieved 98%detection accuracy with precision,recall,and F1-score metrics surpassing conventional methods.This approach not only reduces false positives and negatives but also offers scalability and robustness for real-world cybersecurity applications.The results demonstrate the model’s potential as an effective tool for proactive ransomware detection,especially in environments where evolving threats require adaptable and high-accuracy solutions.展开更多
The methods of network attacks have become increasingly sophisticated,rendering traditional cybersecurity defense mechanisms insufficient to address novel and complex threats effectively.In recent years,artificial int...The methods of network attacks have become increasingly sophisticated,rendering traditional cybersecurity defense mechanisms insufficient to address novel and complex threats effectively.In recent years,artificial intelligence has achieved significant progress in the field of network security.However,many challenges and issues remain,particularly regarding the interpretability of deep learning and ensemble learning algorithms.To address the challenge of enhancing the interpretability of network attack prediction models,this paper proposes a method that combines Light Gradient Boosting Machine(LGBM)and SHapley Additive exPlanations(SHAP).LGBM is employed to model anomalous fluctuations in various network indicators,enabling the rapid and accurate identification and prediction of potential network attack types,thereby facilitating the implementation of timely defense measures,the model achieved an accuracy of 0.977,precision of 0.985,recall of 0.975,and an F1 score of 0.979,demonstrating better performance compared to other models in the domain of network attack prediction.SHAP is utilized to analyze the black-box decision-making process of the model,providing interpretability by quantifying the contribution of each feature to the prediction results and elucidating the relationships between features.The experimental results demonstrate that the network attack predictionmodel based on LGBM exhibits superior accuracy and outstanding predictive capabilities.Moreover,the SHAP-based interpretability analysis significantly improves the model’s transparency and interpretability.展开更多
Future 6G communications are envisioned to enable a large catalogue of pioneering applications.These will range from networked Cyber-Physical Systems to edge computing devices,establishing real-time feedback control l...Future 6G communications are envisioned to enable a large catalogue of pioneering applications.These will range from networked Cyber-Physical Systems to edge computing devices,establishing real-time feedback control loops critical for managing Industry 5.0 deployments,digital agriculture systems,and essential infrastructures.The provision of extensive machine-type communications through 6G will render many of these innovative systems autonomous and unsupervised.While full automation will enhance industrial efficiency significantly,it concurrently introduces new cyber risks and vulnerabilities.In particular,unattended systems are highly susceptible to trust issues:malicious nodes and false information can be easily introduced into control loops.Additionally,Denialof-Service attacks can be executed by inundating the network with valueless noise.Current anomaly detection schemes require the entire transformation of the control software to integrate new steps and can only mitigate anomalies that conform to predefined mathematical models.Solutions based on an exhaustive data collection to detect anomalies are precise but extremely slow.Standard models,with their limited understanding of mobile networks,can achieve precision rates no higher than 75%.Therefore,more general and transversal protection mechanisms are needed to detect malicious behaviors transparently.This paper introduces a probabilistic trust model and control algorithm designed to address this gap.The model determines the probability of any node to be trustworthy.Communication channels are pruned for those nodes whose probability is below a given threshold.The trust control algorithmcomprises three primary phases,which feed themodel with three different probabilities,which are weighted and combined.Initially,anomalous nodes are identified using Gaussian mixture models and clustering technologies.Next,traffic patterns are studied using digital Bessel functions and the functional scalar product.Finally,the information coherence and content are analyzed.The noise content and abnormal information sequences are detected using a Volterra filter and a bank of Finite Impulse Response filters.An experimental validation based on simulation tools and environments was carried out.Results show the proposed solution can successfully detect up to 92%of malicious data injection attacks.展开更多
The recent interest in the deployment of Generative AI applications that use large language models (LLMs) has brought to the forefront significant privacy concerns, notably the leakage of Personally Identifiable Infor...The recent interest in the deployment of Generative AI applications that use large language models (LLMs) has brought to the forefront significant privacy concerns, notably the leakage of Personally Identifiable Information (PII) and other confidential or protected information that may have been memorized during training, specifically during a fine-tuning or customization process. We describe different black-box attacks from potential adversaries and study their impact on the amount and type of information that may be recovered from commonly used and deployed LLMs. Our research investigates the relationship between PII leakage, memorization, and factors such as model size, architecture, and the nature of attacks employed. The study utilizes two broad categories of attacks: PII leakage-focused attacks (auto-completion and extraction attacks) and memorization-focused attacks (various membership inference attacks). The findings from these investigations are quantified using an array of evaluative metrics, providing a detailed understanding of LLM vulnerabilities and the effectiveness of different attacks.展开更多
Federated Learning(FL),a burgeoning technology,has received increasing attention due to its privacy protection capability.However,the base algorithm FedAvg is vulnerable when it suffers from so-called backdoor attacks...Federated Learning(FL),a burgeoning technology,has received increasing attention due to its privacy protection capability.However,the base algorithm FedAvg is vulnerable when it suffers from so-called backdoor attacks.Former researchers proposed several robust aggregation methods.Unfortunately,due to the hidden characteristic of backdoor attacks,many of these aggregation methods are unable to defend against backdoor attacks.What's more,the attackers recently have proposed some hiding methods that further improve backdoor attacks'stealthiness,making all the existing robust aggregation methods fail.To tackle the threat of backdoor attacks,we propose a new aggregation method,X-raying Models with A Matrix(XMAM),to reveal the malicious local model updates submitted by the backdoor attackers.Since we observe that the output of the Softmax layer exhibits distinguishable patterns between malicious and benign updates,unlike the existing aggregation algorithms,we focus on the Softmax layer's output in which the backdoor attackers are difficult to hide their malicious behavior.Specifically,like medical X-ray examinations,we investigate the collected local model updates by using a matrix as an input to get their Softmax layer's outputs.Then,we preclude updates whose outputs are abnormal by clustering.Without any training dataset in the server,the extensive evaluations show that our XMAM can effectively distinguish malicious local model updates from benign ones.For instance,when other methods fail to defend against the backdoor attacks at no more than 20%malicious clients,our method can tolerate 45%malicious clients in the black-box mode and about 30%in Projected Gradient Descent(PGD)mode.Besides,under adaptive attacks,the results demonstrate that XMAM can still complete the global model training task even when there are 40%malicious clients.Finally,we analyze our method's screening complexity and compare the real screening time with other methods.The results show that XMAM is about 10–10000 times faster than the existing methods.展开更多
With the increasing number of connected devices in the Internet of Things(IoT)era,the number of intrusions is also increasing.An intrusion detection system(IDS)is a secondary intelligent system for monitoring,detectin...With the increasing number of connected devices in the Internet of Things(IoT)era,the number of intrusions is also increasing.An intrusion detection system(IDS)is a secondary intelligent system for monitoring,detecting and alerting against malicious activity.IDS is important in developing advanced security models.This study reviews the importance of various techniques,tools,and methods used in IoT detection and/or prevention systems.Specifically,it focuses on machine learning(ML)and deep learning(DL)techniques for IDS.This paper proposes an accurate intrusion detection model to detect traditional and new attacks on the Internet of Vehicles.To speed up the detection of recent attacks,the proposed network architecture developed at the data processing layer is incorporated with a convolutional neural network(CNN),which performs better than a support vector machine(SVM).Processing data are enhanced using the synthetic minority oversampling technique to ensure learning accuracy.The nearest class mean classifier is applied during the testing phase to identify new attacks.Experimental results using the AWID dataset,which is one of the most common open intrusion detection datasets,revealed a higher detection accuracy(94%)compared to SVM and random forest methods.展开更多
Modeling of unsteady aerodynamic loads at high angles of attack using a small amount of experimental or simulation data to construct predictive models for unknown states can greatly improve the efficiency of aircraft ...Modeling of unsteady aerodynamic loads at high angles of attack using a small amount of experimental or simulation data to construct predictive models for unknown states can greatly improve the efficiency of aircraft unsteady aerodynamic design and flight dynamics analysis.In this paper,aiming at the problems of poor generalization of traditional aerodynamic models and intelligent models,an intelligent aerodynamic modeling method based on gated neural units is proposed.The time memory characteristics of the gated neural unit is fully utilized,thus the nonlinear flow field characterization ability of the learning and training process is enhanced,and the generalization ability of the whole prediction model is improved.The prediction and verification of the model are carried out under the maneuvering flight condition of NACA0015 airfoil.The results show that the model has good adaptability.In the interpolation prediction,the maximum prediction error of the lift and drag coefficients and the moment coefficient does not exceed 10%,which can basically represent the variation characteristics of the entire flow field.In the construction of extrapolation models,the training model based on the strong nonlinear data has good accuracy for weak nonlinear prediction.Furthermore,the error is larger,even exceeding 20%,which indicates that the extrapolation and generalization capabilities need to be further optimized by integrating physical models.Compared with the conventional state space equation model,the proposed method can improve the extrapolation accuracy and efficiency by 78%and 60%,respectively,which demonstrates the applied potential of this method in aerodynamic modeling.展开更多
The development of Intelligent Railway Transportation Systems necessitates incorporating privacy-preserving mechanisms into AI models to protect sensitive information and enhance system efficiency.Federated learning o...The development of Intelligent Railway Transportation Systems necessitates incorporating privacy-preserving mechanisms into AI models to protect sensitive information and enhance system efficiency.Federated learning offers a promising solution by allowing multiple clients to train models collaboratively without sharing private data.However,despite its privacy benefits,federated learning systems are vulnerable to poisoning attacks,where adversaries alter local model parameters on compromised clients and send malicious updates to the server,potentially compromising the global model’s accuracy.In this study,we introduce PMM(Perturbation coefficient Multiplied by Maximum value),a new poisoning attack method that perturbs model updates layer by layer,demonstrating the threat of poisoning attacks faced by federated learning.Extensive experiments across three distinct datasets have demonstrated PMM’s ability to significantly reduce the global model’s accuracy.Additionally,we propose an effective defense method,namely CLBL(Cluster Layer By Layer).Experiment results on three datasets have confirmed CLBL’s effectiveness.展开更多
With the prevalence of machine learning in malware defense,hackers have tried to attack machine learning models to evade detection.It is generally difficult to explore the details of malware detection models,hackers c...With the prevalence of machine learning in malware defense,hackers have tried to attack machine learning models to evade detection.It is generally difficult to explore the details of malware detection models,hackers can adopt fuzzing attack to manipulate the features of the malware closer to benign programs on the premise of retaining their functions.In this paper,attack and defense methods on malware detection models based on machine learning algorithms were studied.Firstly,we designed a fuzzing attack method by randomly modifying features to evade detection.The fuzzing attack can effectively descend the accuracy of machine learning model with single feature.Then an adversarial malware detection model MaliFuzz is proposed to defend fuzzing attack.Different from the ordinary single feature detection model,the combined features by static and dynamic analysis to improve the defense ability are used.The experiment results show that the adversarial malware detection model with combined features can deal with the attack.The methods designed in this paper have great significance in improving the security of malware detection models and have good application prospects.展开更多
Watermarking is embedding visible or invisible data within media to verify its authenticity or protect copyright.The watermark is embedded in significant spatial or frequency features of the media to make it more resi...Watermarking is embedding visible or invisible data within media to verify its authenticity or protect copyright.The watermark is embedded in significant spatial or frequency features of the media to make it more resistant to intentional or unintentional modification.Some of these features are important perceptual features according to the human visual system(HVS),which means that the embedded watermark should be imperceptible in these features.Therefore,both the designers of watermarking algorithms and potential attackers must consider these perceptual features when carrying out their actions.The two roles will be considered in this paper when designing a robust watermarking algorithm against the most harmful attacks,like volumetric scaling,histogram equalization,and non-conventional watermarking attacks like the Denoising Convolution Neural Network(DnCNN),which must be considered in watermarking algorithm design due to its rising role in the state-of-the-art attacks.The DnCNN is initialized and trained using watermarked image samples created by our proposed Covert and Severe Attacks Resistant Watermarking Algorithm(CSRWA)to prove its robustness.For this algorithm to satisfy the robustness and imperceptibility tradeoff,implementing the Dither Modulation(DM)algorithm is boosted by utilizing the Just Noticeable Distortion(JND)principle to get an improved performance in this sense.Sensitivity,luminance,inter and intra-block contrast are used to adjust the JND values.展开更多
BACKGROUND Rebleeding after recovery from esophagogastric variceal bleeding(EGVB)is a severe complication that is associated with high rates of both incidence and mortality.Despite its clinical importance,recognized p...BACKGROUND Rebleeding after recovery from esophagogastric variceal bleeding(EGVB)is a severe complication that is associated with high rates of both incidence and mortality.Despite its clinical importance,recognized prognostic models that can effectively predict esophagogastric variceal rebleeding in patients with liver cirrhosis are lacking.AIM To construct and externally validate a reliable prognostic model for predicting the occurrence of esophagogastric variceal rebleeding.METHODS This study included 477 EGVB patients across 2 cohorts:The derivation cohort(n=322)and the validation cohort(n=155).The primary outcome was rebleeding events within 1 year.The least absolute shrinkage and selection operator was applied for predictor selection,and multivariate Cox regression analysis was used to construct the prognostic model.Internal validation was performed with bootstrap resampling.We assessed the discrimination,calibration and accuracy of the model,and performed patient risk stratification.RESULTS Six predictors,including albumin and aspartate aminotransferase concentrations,white blood cell count,and the presence of ascites,portal vein thrombosis,and bleeding signs,were selected for the rebleeding event prediction following endoscopic treatment(REPET)model.In predicting rebleeding within 1 year,the REPET model ex-hibited a concordance index of 0.775 and a Brier score of 0.143 in the derivation cohort,alongside 0.862 and 0.127 in the validation cohort.Furthermore,the REPET model revealed a significant difference in rebleeding rates(P<0.01)between low-risk patients and intermediate-to high-risk patients in both cohorts.CONCLUSION We constructed and validated a new prognostic model for variceal rebleeding with excellent predictive per-formance,which will improve the clinical management of rebleeding in EGVB patients.展开更多
Cyberspace is extremely dynamic,with new attacks arising daily.Protecting cybersecurity controls is vital for network security.Deep Learning(DL)models find widespread use across various fields,with cybersecurity being...Cyberspace is extremely dynamic,with new attacks arising daily.Protecting cybersecurity controls is vital for network security.Deep Learning(DL)models find widespread use across various fields,with cybersecurity being one of the most crucial due to their rapid cyberattack detection capabilities on networks and hosts.The capabilities of DL in feature learning and analyzing extensive data volumes lead to the recognition of network traffic patterns.This study presents novel lightweight DL models,known as Cybernet models,for the detection and recognition of various cyber Distributed Denial of Service(DDoS)attacks.These models were constructed to have a reasonable number of learnable parameters,i.e.,less than 225,000,hence the name“lightweight.”This not only helps reduce the number of computations required but also results in faster training and inference times.Additionally,these models were designed to extract features in parallel from 1D Convolutional Neural Networks(CNN)and Long Short-Term Memory(LSTM),which makes them unique compared to earlier existing architectures and results in better performance measures.To validate their robustness and effectiveness,they were tested on the CIC-DDoS2019 dataset,which is an imbalanced and large dataset that contains different types of DDoS attacks.Experimental results revealed that bothmodels yielded promising results,with 99.99% for the detectionmodel and 99.76% for the recognition model in terms of accuracy,precision,recall,and F1 score.Furthermore,they outperformed the existing state-of-the-art models proposed for the same task.Thus,the proposed models can be used in cyber security research domains to successfully identify different types of attacks with a high detection and recognition rate.展开更多
This study was aimed to prepare landslide susceptibility maps for the Pithoragarh district in Uttarakhand,India,using advanced ensemble models that combined Radial Basis Function Networks(RBFN)with three ensemble lear...This study was aimed to prepare landslide susceptibility maps for the Pithoragarh district in Uttarakhand,India,using advanced ensemble models that combined Radial Basis Function Networks(RBFN)with three ensemble learning techniques:DAGGING(DG),MULTIBOOST(MB),and ADABOOST(AB).This combination resulted in three distinct ensemble models:DG-RBFN,MB-RBFN,and AB-RBFN.Additionally,a traditional weighted method,Information Value(IV),and a benchmark machine learning(ML)model,Multilayer Perceptron Neural Network(MLP),were employed for comparison and validation.The models were developed using ten landslide conditioning factors,which included slope,aspect,elevation,curvature,land cover,geomorphology,overburden depth,lithology,distance to rivers and distance to roads.These factors were instrumental in predicting the output variable,which was the probability of landslide occurrence.Statistical analysis of the models’performance indicated that the DG-RBFN model,with an Area Under ROC Curve(AUC)of 0.931,outperformed the other models.The AB-RBFN model achieved an AUC of 0.929,the MB-RBFN model had an AUC of 0.913,and the MLP model recorded an AUC of 0.926.These results suggest that the advanced ensemble ML model DG-RBFN was more accurate than traditional statistical model,single MLP model,and other ensemble models in preparing trustworthy landslide susceptibility maps,thereby enhancing land use planning and decision-making.展开更多
基金the International Scientific Complex“Astana”was funded by the Committee of Science of the Ministry of Science and Higher Education of the Republic of Kazakhstan(Grant No.AP19680345).
文摘Wireless Sensor Networks(WSN)have gained significant attention over recent years due to their extensive applications in various domains such as environmentalmonitoring,healthcare systems,industrial automation,and smart cities.However,such networks are inherently vulnerable to different types of attacks because they operate in open environments with limited resources and constrained communication capabilities.Thepaper addresses challenges related to modeling and analysis of wireless sensor networks and their susceptibility to attacks.Its objective is to create versatile modeling tools capable of detecting attacks against network devices and identifying anomalies caused either by legitimate user errors or malicious activities.A proposed integrated approach for data collection,preprocessing,and analysis in WSN outlines a series of steps applicable throughout both the design phase and operation stage.This ensures effective detection of attacks and anomalies within WSNs.An introduced attackmodel specifies potential types of unauthorized network layer attacks targeting network nodes,transmitted data,and services offered by the WSN.Furthermore,a graph-based analytical framework was designed to detect attacks by evaluating real-time events from network nodes and determining if an attack is underway.Additionally,a simulation model based on sequences of imperative rules defining behaviors of both regular and compromised nodes is presented.Overall,this technique was experimentally verified using a segment of a WSN embedded in a smart city infrastructure,simulating a wormhole attack.Results demonstrate the viability and practical significance of the technique for enhancing future information security measures.Validation tests confirmed high levels of accuracy and efficiency when applied specifically to detecting wormhole attacks targeting routing protocols in WSNs.Precision and recall rates averaged above the benchmark value of 0.95,thus validating the broad applicability of the proposed models across varied scenarios.
基金supported by Science and Technology Innovation Programfor Postgraduate Students in IDP Subsidized by Fundamental Research Funds for the Central Universities(Project No.ZY20240335)support of the Research Project of the Key Technology of Malicious Code Detection Based on Data Mining in APT Attack(Project No.2022IT173)the Research Project of the Big Data Sensitive Information Supervision Technology Based on Convolutional Neural Network(Project No.2022011033).
文摘Previous studies have shown that deep learning is very effective in detecting known attacks.However,when facing unknown attacks,models such as Deep Neural Networks(DNN)combined with Long Short-Term Memory(LSTM),Convolutional Neural Networks(CNN)combined with LSTM,and so on are built by simple stacking,which has the problems of feature loss,low efficiency,and low accuracy.Therefore,this paper proposes an autonomous detectionmodel for Distributed Denial of Service attacks,Multi-Scale Convolutional Neural Network-Bidirectional Gated Recurrent Units-Single Headed Attention(MSCNN-BiGRU-SHA),which is based on a Multistrategy Integrated Zebra Optimization Algorithm(MI-ZOA).The model undergoes training and testing with the CICDDoS2019 dataset,and its performance is evaluated on a new GINKS2023 dataset.The hyperparameters for Conv_filter and GRU_unit are optimized using the Multi-strategy Integrated Zebra Optimization Algorithm(MIZOA).The experimental results show that the test accuracy of the MSCNN-BiGRU-SHA model based on the MIZOA proposed in this paper is as high as 0.9971 in the CICDDoS 2019 dataset.The evaluation accuracy of the new dataset GINKS2023 created in this paper is 0.9386.Compared to the MSCNN-BiGRU-SHA model based on the Zebra Optimization Algorithm(ZOA),the detection accuracy on the GINKS2023 dataset has improved by 5.81%,precisionhas increasedby 1.35%,the recallhas improvedby 9%,and theF1scorehas increasedby 5.55%.Compared to the MSCNN-BiGRU-SHA models developed using Grid Search,Random Search,and Bayesian Optimization,the MSCNN-BiGRU-SHA model optimized with the MI-ZOA exhibits better performance in terms of accuracy,precision,recall,and F1 score.
基金supported in part by the National Natural Science Foundation of China under Grants No.62372087 and No.62072076the Research Fund of State Key Laboratory of Processors under Grant No.CLQ202310the CSC scholarship.
文摘Deep neural networks(DNNs)have found extensive applications in safety-critical artificial intelligence systems,such as autonomous driving and facial recognition systems.However,recent research has revealed their susceptibility to backdoors maliciously injected by adversaries.This vulnerability arises due to the intricate architecture and opacity of DNNs,resulting in numerous redundant neurons embedded within the models.Adversaries exploit these vulnerabilities to conceal malicious backdoor information within DNNs,thereby causing erroneous outputs and posing substantial threats to the efficacy of DNN-based applications.This article presents a comprehensive survey of backdoor attacks against DNNs and the countermeasure methods employed to mitigate them.Initially,we trace the evolution of the concept from traditional backdoor attacks to backdoor attacks against DNNs,highlighting the feasibility and practicality of generating backdoor attacks against DNNs.Subsequently,we provide an overview of notable works encompassing various attack and defense strategies,facilitating a comparative analysis of their approaches.Through these discussions,we offer constructive insights aimed at refining these techniques.Finally,we extend our research perspective to the domain of large language models(LLMs)and synthesize the characteristics and developmental trends of backdoor attacks and defense methods targeting LLMs.Through a systematic review of existing studies on backdoor vulnerabilities in LLMs,we identify critical open challenges in this field and propose actionable directions for future research.
基金supported by the National Natural Science Foundation of China(Nos.12172315,12072304,11702232)the Fujian Provincial Natural Science Foundation,China(No.2021J01050)the Aeronautical Science Foundation of China(No.20220013068002).
文摘Unsteady aerodynamic characteristics at high angles of attack are of great importance to the design and development of advanced fighter aircraft, which are characterized by post-stall maneuverability with multiple Degrees-of-Freedom(multi-DOF) and complex flow field structure.In this paper, a special kind of cable-driven parallel mechanism is firstly utilized as a new suspension method to conduct unsteady dynamic wind tunnel tests at high angles of attack, thereby providing experimental aerodynamic data. These tests include a wide range of multi-DOF coupled oscillatory motions with various amplitudes and frequencies. Then, for aerodynamic modeling and analysis, a novel data-driven Feature-Level Attention Recurrent neural network(FLAR) is proposed. This model incorporates a specially designed feature-level attention module that focuses on the state variables affecting the aerodynamic coefficients, thereby enhancing the physical interpretability of the aerodynamic model. Subsequently, spin maneuver simulations, using a mathematical model as the baseline, are conducted to validate the effectiveness of the FLAR. Finally, the results on wind tunnel data reveal that the FLAR accurately predicts aerodynamic coefficients, and observations through the visualization of attention scores identify the key state variables that affect the aerodynamic coefficients. It is concluded that the proposed FLAR enhances the interpretability of the aerodynamic model while achieving good prediction accuracy and generalization capability for multi-DOF coupling motion at high angles of attack.
文摘Ransomware attacks pose a significant threat to critical infrastructures,demanding robust detection mechanisms.This study introduces a hybrid model that combines vision transformer(ViT)and one-dimensional convolutional neural network(1DCNN)architectures to enhance ransomware detection capabilities.Addressing common challenges in ransomware detection,particularly dataset class imbalance,the synthetic minority oversampling technique(SMOTE)is employed to generate synthetic samples for minority class,thereby improving detection accuracy.The integration of ViT and 1DCNN through feature fusion enables the model to capture both global contextual and local sequential features,resulting in comprehensive ransomware classification.Tested on the UNSW-NB15 dataset,the proposed ViT-1DCNN model achieved 98%detection accuracy with precision,recall,and F1-score metrics surpassing conventional methods.This approach not only reduces false positives and negatives but also offers scalability and robustness for real-world cybersecurity applications.The results demonstrate the model’s potential as an effective tool for proactive ransomware detection,especially in environments where evolving threats require adaptable and high-accuracy solutions.
基金supported by the National Natural Science Foundation of China Project(No.62302540)please visit their website at https://www.nsfc.gov.cn/(accessed on 18 June 2024).
文摘The methods of network attacks have become increasingly sophisticated,rendering traditional cybersecurity defense mechanisms insufficient to address novel and complex threats effectively.In recent years,artificial intelligence has achieved significant progress in the field of network security.However,many challenges and issues remain,particularly regarding the interpretability of deep learning and ensemble learning algorithms.To address the challenge of enhancing the interpretability of network attack prediction models,this paper proposes a method that combines Light Gradient Boosting Machine(LGBM)and SHapley Additive exPlanations(SHAP).LGBM is employed to model anomalous fluctuations in various network indicators,enabling the rapid and accurate identification and prediction of potential network attack types,thereby facilitating the implementation of timely defense measures,the model achieved an accuracy of 0.977,precision of 0.985,recall of 0.975,and an F1 score of 0.979,demonstrating better performance compared to other models in the domain of network attack prediction.SHAP is utilized to analyze the black-box decision-making process of the model,providing interpretability by quantifying the contribution of each feature to the prediction results and elucidating the relationships between features.The experimental results demonstrate that the network attack predictionmodel based on LGBM exhibits superior accuracy and outstanding predictive capabilities.Moreover,the SHAP-based interpretability analysis significantly improves the model’s transparency and interpretability.
基金funding by Comunidad de Madrid within the framework of the Multiannual Agreement with Universidad Politécnica de Madrid to encourage research by young doctors(PRINCE project).
文摘Future 6G communications are envisioned to enable a large catalogue of pioneering applications.These will range from networked Cyber-Physical Systems to edge computing devices,establishing real-time feedback control loops critical for managing Industry 5.0 deployments,digital agriculture systems,and essential infrastructures.The provision of extensive machine-type communications through 6G will render many of these innovative systems autonomous and unsupervised.While full automation will enhance industrial efficiency significantly,it concurrently introduces new cyber risks and vulnerabilities.In particular,unattended systems are highly susceptible to trust issues:malicious nodes and false information can be easily introduced into control loops.Additionally,Denialof-Service attacks can be executed by inundating the network with valueless noise.Current anomaly detection schemes require the entire transformation of the control software to integrate new steps and can only mitigate anomalies that conform to predefined mathematical models.Solutions based on an exhaustive data collection to detect anomalies are precise but extremely slow.Standard models,with their limited understanding of mobile networks,can achieve precision rates no higher than 75%.Therefore,more general and transversal protection mechanisms are needed to detect malicious behaviors transparently.This paper introduces a probabilistic trust model and control algorithm designed to address this gap.The model determines the probability of any node to be trustworthy.Communication channels are pruned for those nodes whose probability is below a given threshold.The trust control algorithmcomprises three primary phases,which feed themodel with three different probabilities,which are weighted and combined.Initially,anomalous nodes are identified using Gaussian mixture models and clustering technologies.Next,traffic patterns are studied using digital Bessel functions and the functional scalar product.Finally,the information coherence and content are analyzed.The noise content and abnormal information sequences are detected using a Volterra filter and a bank of Finite Impulse Response filters.An experimental validation based on simulation tools and environments was carried out.Results show the proposed solution can successfully detect up to 92%of malicious data injection attacks.
文摘The recent interest in the deployment of Generative AI applications that use large language models (LLMs) has brought to the forefront significant privacy concerns, notably the leakage of Personally Identifiable Information (PII) and other confidential or protected information that may have been memorized during training, specifically during a fine-tuning or customization process. We describe different black-box attacks from potential adversaries and study their impact on the amount and type of information that may be recovered from commonly used and deployed LLMs. Our research investigates the relationship between PII leakage, memorization, and factors such as model size, architecture, and the nature of attacks employed. The study utilizes two broad categories of attacks: PII leakage-focused attacks (auto-completion and extraction attacks) and memorization-focused attacks (various membership inference attacks). The findings from these investigations are quantified using an array of evaluative metrics, providing a detailed understanding of LLM vulnerabilities and the effectiveness of different attacks.
基金Supported by the Fundamental Research Funds for the Central Universities(328202204)。
文摘Federated Learning(FL),a burgeoning technology,has received increasing attention due to its privacy protection capability.However,the base algorithm FedAvg is vulnerable when it suffers from so-called backdoor attacks.Former researchers proposed several robust aggregation methods.Unfortunately,due to the hidden characteristic of backdoor attacks,many of these aggregation methods are unable to defend against backdoor attacks.What's more,the attackers recently have proposed some hiding methods that further improve backdoor attacks'stealthiness,making all the existing robust aggregation methods fail.To tackle the threat of backdoor attacks,we propose a new aggregation method,X-raying Models with A Matrix(XMAM),to reveal the malicious local model updates submitted by the backdoor attackers.Since we observe that the output of the Softmax layer exhibits distinguishable patterns between malicious and benign updates,unlike the existing aggregation algorithms,we focus on the Softmax layer's output in which the backdoor attackers are difficult to hide their malicious behavior.Specifically,like medical X-ray examinations,we investigate the collected local model updates by using a matrix as an input to get their Softmax layer's outputs.Then,we preclude updates whose outputs are abnormal by clustering.Without any training dataset in the server,the extensive evaluations show that our XMAM can effectively distinguish malicious local model updates from benign ones.For instance,when other methods fail to defend against the backdoor attacks at no more than 20%malicious clients,our method can tolerate 45%malicious clients in the black-box mode and about 30%in Projected Gradient Descent(PGD)mode.Besides,under adaptive attacks,the results demonstrate that XMAM can still complete the global model training task even when there are 40%malicious clients.Finally,we analyze our method's screening complexity and compare the real screening time with other methods.The results show that XMAM is about 10–10000 times faster than the existing methods.
基金The author extends the appreciation to the Deanship of Postgraduate Studies and Scientific Research atMajmaah University for funding this research work through the project number(R-2024-920).
文摘With the increasing number of connected devices in the Internet of Things(IoT)era,the number of intrusions is also increasing.An intrusion detection system(IDS)is a secondary intelligent system for monitoring,detecting and alerting against malicious activity.IDS is important in developing advanced security models.This study reviews the importance of various techniques,tools,and methods used in IoT detection and/or prevention systems.Specifically,it focuses on machine learning(ML)and deep learning(DL)techniques for IDS.This paper proposes an accurate intrusion detection model to detect traditional and new attacks on the Internet of Vehicles.To speed up the detection of recent attacks,the proposed network architecture developed at the data processing layer is incorporated with a convolutional neural network(CNN),which performs better than a support vector machine(SVM).Processing data are enhanced using the synthetic minority oversampling technique to ensure learning accuracy.The nearest class mean classifier is applied during the testing phase to identify new attacks.Experimental results using the AWID dataset,which is one of the most common open intrusion detection datasets,revealed a higher detection accuracy(94%)compared to SVM and random forest methods.
基金supported in part by the National Natural Science Foundation of China (No. 12202363)。
文摘Modeling of unsteady aerodynamic loads at high angles of attack using a small amount of experimental or simulation data to construct predictive models for unknown states can greatly improve the efficiency of aircraft unsteady aerodynamic design and flight dynamics analysis.In this paper,aiming at the problems of poor generalization of traditional aerodynamic models and intelligent models,an intelligent aerodynamic modeling method based on gated neural units is proposed.The time memory characteristics of the gated neural unit is fully utilized,thus the nonlinear flow field characterization ability of the learning and training process is enhanced,and the generalization ability of the whole prediction model is improved.The prediction and verification of the model are carried out under the maneuvering flight condition of NACA0015 airfoil.The results show that the model has good adaptability.In the interpolation prediction,the maximum prediction error of the lift and drag coefficients and the moment coefficient does not exceed 10%,which can basically represent the variation characteristics of the entire flow field.In the construction of extrapolation models,the training model based on the strong nonlinear data has good accuracy for weak nonlinear prediction.Furthermore,the error is larger,even exceeding 20%,which indicates that the extrapolation and generalization capabilities need to be further optimized by integrating physical models.Compared with the conventional state space equation model,the proposed method can improve the extrapolation accuracy and efficiency by 78%and 60%,respectively,which demonstrates the applied potential of this method in aerodynamic modeling.
基金supported by Systematic Major Project of China State Railway Group Corporation Limited(Grant Number:P2023W002).
文摘The development of Intelligent Railway Transportation Systems necessitates incorporating privacy-preserving mechanisms into AI models to protect sensitive information and enhance system efficiency.Federated learning offers a promising solution by allowing multiple clients to train models collaboratively without sharing private data.However,despite its privacy benefits,federated learning systems are vulnerable to poisoning attacks,where adversaries alter local model parameters on compromised clients and send malicious updates to the server,potentially compromising the global model’s accuracy.In this study,we introduce PMM(Perturbation coefficient Multiplied by Maximum value),a new poisoning attack method that perturbs model updates layer by layer,demonstrating the threat of poisoning attacks faced by federated learning.Extensive experiments across three distinct datasets have demonstrated PMM’s ability to significantly reduce the global model’s accuracy.Additionally,we propose an effective defense method,namely CLBL(Cluster Layer By Layer).Experiment results on three datasets have confirmed CLBL’s effectiveness.
文摘With the prevalence of machine learning in malware defense,hackers have tried to attack machine learning models to evade detection.It is generally difficult to explore the details of malware detection models,hackers can adopt fuzzing attack to manipulate the features of the malware closer to benign programs on the premise of retaining their functions.In this paper,attack and defense methods on malware detection models based on machine learning algorithms were studied.Firstly,we designed a fuzzing attack method by randomly modifying features to evade detection.The fuzzing attack can effectively descend the accuracy of machine learning model with single feature.Then an adversarial malware detection model MaliFuzz is proposed to defend fuzzing attack.Different from the ordinary single feature detection model,the combined features by static and dynamic analysis to improve the defense ability are used.The experiment results show that the adversarial malware detection model with combined features can deal with the attack.The methods designed in this paper have great significance in improving the security of malware detection models and have good application prospects.
文摘Watermarking is embedding visible or invisible data within media to verify its authenticity or protect copyright.The watermark is embedded in significant spatial or frequency features of the media to make it more resistant to intentional or unintentional modification.Some of these features are important perceptual features according to the human visual system(HVS),which means that the embedded watermark should be imperceptible in these features.Therefore,both the designers of watermarking algorithms and potential attackers must consider these perceptual features when carrying out their actions.The two roles will be considered in this paper when designing a robust watermarking algorithm against the most harmful attacks,like volumetric scaling,histogram equalization,and non-conventional watermarking attacks like the Denoising Convolution Neural Network(DnCNN),which must be considered in watermarking algorithm design due to its rising role in the state-of-the-art attacks.The DnCNN is initialized and trained using watermarked image samples created by our proposed Covert and Severe Attacks Resistant Watermarking Algorithm(CSRWA)to prove its robustness.For this algorithm to satisfy the robustness and imperceptibility tradeoff,implementing the Dither Modulation(DM)algorithm is boosted by utilizing the Just Noticeable Distortion(JND)principle to get an improved performance in this sense.Sensitivity,luminance,inter and intra-block contrast are used to adjust the JND values.
基金Supported by National Natural Science Foundation of China,No.81874390 and No.81573948Shanghai Natural Science Foundation,No.21ZR1464100+1 种基金Science and Technology Innovation Action Plan of Shanghai Science and Technology Commission,No.22S11901700the Shanghai Key Specialty of Traditional Chinese Clinical Medicine,No.shslczdzk01201.
文摘BACKGROUND Rebleeding after recovery from esophagogastric variceal bleeding(EGVB)is a severe complication that is associated with high rates of both incidence and mortality.Despite its clinical importance,recognized prognostic models that can effectively predict esophagogastric variceal rebleeding in patients with liver cirrhosis are lacking.AIM To construct and externally validate a reliable prognostic model for predicting the occurrence of esophagogastric variceal rebleeding.METHODS This study included 477 EGVB patients across 2 cohorts:The derivation cohort(n=322)and the validation cohort(n=155).The primary outcome was rebleeding events within 1 year.The least absolute shrinkage and selection operator was applied for predictor selection,and multivariate Cox regression analysis was used to construct the prognostic model.Internal validation was performed with bootstrap resampling.We assessed the discrimination,calibration and accuracy of the model,and performed patient risk stratification.RESULTS Six predictors,including albumin and aspartate aminotransferase concentrations,white blood cell count,and the presence of ascites,portal vein thrombosis,and bleeding signs,were selected for the rebleeding event prediction following endoscopic treatment(REPET)model.In predicting rebleeding within 1 year,the REPET model ex-hibited a concordance index of 0.775 and a Brier score of 0.143 in the derivation cohort,alongside 0.862 and 0.127 in the validation cohort.Furthermore,the REPET model revealed a significant difference in rebleeding rates(P<0.01)between low-risk patients and intermediate-to high-risk patients in both cohorts.CONCLUSION We constructed and validated a new prognostic model for variceal rebleeding with excellent predictive per-formance,which will improve the clinical management of rebleeding in EGVB patients.
文摘Cyberspace is extremely dynamic,with new attacks arising daily.Protecting cybersecurity controls is vital for network security.Deep Learning(DL)models find widespread use across various fields,with cybersecurity being one of the most crucial due to their rapid cyberattack detection capabilities on networks and hosts.The capabilities of DL in feature learning and analyzing extensive data volumes lead to the recognition of network traffic patterns.This study presents novel lightweight DL models,known as Cybernet models,for the detection and recognition of various cyber Distributed Denial of Service(DDoS)attacks.These models were constructed to have a reasonable number of learnable parameters,i.e.,less than 225,000,hence the name“lightweight.”This not only helps reduce the number of computations required but also results in faster training and inference times.Additionally,these models were designed to extract features in parallel from 1D Convolutional Neural Networks(CNN)and Long Short-Term Memory(LSTM),which makes them unique compared to earlier existing architectures and results in better performance measures.To validate their robustness and effectiveness,they were tested on the CIC-DDoS2019 dataset,which is an imbalanced and large dataset that contains different types of DDoS attacks.Experimental results revealed that bothmodels yielded promising results,with 99.99% for the detectionmodel and 99.76% for the recognition model in terms of accuracy,precision,recall,and F1 score.Furthermore,they outperformed the existing state-of-the-art models proposed for the same task.Thus,the proposed models can be used in cyber security research domains to successfully identify different types of attacks with a high detection and recognition rate.
基金the University of Transport Technology under the project entitled“Application of Machine Learning Algorithms in Landslide Susceptibility Mapping in Mountainous Areas”with grant number DTTD2022-16.
文摘This study was aimed to prepare landslide susceptibility maps for the Pithoragarh district in Uttarakhand,India,using advanced ensemble models that combined Radial Basis Function Networks(RBFN)with three ensemble learning techniques:DAGGING(DG),MULTIBOOST(MB),and ADABOOST(AB).This combination resulted in three distinct ensemble models:DG-RBFN,MB-RBFN,and AB-RBFN.Additionally,a traditional weighted method,Information Value(IV),and a benchmark machine learning(ML)model,Multilayer Perceptron Neural Network(MLP),were employed for comparison and validation.The models were developed using ten landslide conditioning factors,which included slope,aspect,elevation,curvature,land cover,geomorphology,overburden depth,lithology,distance to rivers and distance to roads.These factors were instrumental in predicting the output variable,which was the probability of landslide occurrence.Statistical analysis of the models’performance indicated that the DG-RBFN model,with an Area Under ROC Curve(AUC)of 0.931,outperformed the other models.The AB-RBFN model achieved an AUC of 0.929,the MB-RBFN model had an AUC of 0.913,and the MLP model recorded an AUC of 0.926.These results suggest that the advanced ensemble ML model DG-RBFN was more accurate than traditional statistical model,single MLP model,and other ensemble models in preparing trustworthy landslide susceptibility maps,thereby enhancing land use planning and decision-making.
