With an increase in internet-connected devices and a dependency on online services,the threat of Distributed Denial of Service(DDoS)attacks has become a significant concern in cybersecurity.The proposed system follows...With an increase in internet-connected devices and a dependency on online services,the threat of Distributed Denial of Service(DDoS)attacks has become a significant concern in cybersecurity.The proposed system follows a multi-step process,beginning with the collection of datasets from different edge devices and network nodes.To verify its effectiveness,experiments were conducted using the CICDoS2017,NSL-KDD,and CICIDS benchmark datasets alongside other existing models.Recursive feature elimination(RFE)with random forest is used to select features from the CICDDoS2019 dataset,on which a BiLSTM model is trained on local nodes.Local models are trained until convergence or stability criteria are met while simultaneously sharing the updates globally for collaborative learning.A centralised server evaluates real-time traffic using the global BiLSTM model,which triggers alerts for potential DDoS attacks.Furthermore,blockchain technology is employed to secure model updates and to provide an immutable audit trail,thereby ensuring trust and accountability among network nodes.This research introduces a novel decentralized method called Federated Random Forest Bidirectional Long Short-Term Memory(FRF-BiLSTM)for detecting DDoS attacks,utilizing the advanced Bidirectional Long Short-Term Memory Networks(BiLSTMs)to analyze sequences in both forward and backward directions.The outcome shows the proposed model achieves a mean accuracy of 97.1%with an average training delay of 88.7 s and testing delay of 21.4 s.The model demonstrates scalability and the best detection performance in large-scale attack scenarios.展开更多
A Distributed Denial-of-Service(DDoS)attack poses a significant challenge in the digital age,disrupting online services with operational and financial consequences.Detecting such attacks requires innovative and effect...A Distributed Denial-of-Service(DDoS)attack poses a significant challenge in the digital age,disrupting online services with operational and financial consequences.Detecting such attacks requires innovative and effective solutions.The primary challenge lies in selecting the best among several DDoS detection models.This study presents a framework that combines several DDoS detection models and Multiple-Criteria Decision-Making(MCDM)techniques to compare and select the most effective models.The framework integrates a decision matrix from training several models on the CiC-DDOS2019 dataset with Fuzzy Weighted Zero Inconsistency Criterion(FWZIC)and MultiAttribute Boundary Approximation Area Comparison(MABAC)methodologies.FWZIC assigns weights to evaluate criteria,while MABAC compares detection models based on the assessed criteria.The results indicate that the FWZIC approach assigns weights to criteria reliably,with time complexity receiving the highest weight(0.2585)and F1 score receiving the lowest weight(0.14644).Among the models evaluated using the MABAC approach,the Support Vector Machine(SVM)ranked first with a score of 0.0444,making it the most suitable for this work.In contrast,Naive Bayes(NB)ranked lowest with a score of 0.0018.Objective validation and sensitivity analysis proved the reliability of the framework.This study provides a practical approach and insights for cybersecurity practitioners and researchers to evaluate DDoS detection models.展开更多
The rapid advancement of the Internet ofThings(IoT)has heightened the importance of security,with a notable increase in Distributed Denial-of-Service(DDoS)attacks targeting IoT devices.Network security specialists fac...The rapid advancement of the Internet ofThings(IoT)has heightened the importance of security,with a notable increase in Distributed Denial-of-Service(DDoS)attacks targeting IoT devices.Network security specialists face the challenge of producing systems to identify and offset these attacks.This researchmanages IoT security through the emerging Software-Defined Networking(SDN)standard by developing a unified framework(RNN-RYU).We thoroughly assess multiple deep learning frameworks,including Convolutional Neural Network(CNN),Long Short-Term Memory(LSTM),Feed-Forward Convolutional Neural Network(FFCNN),and Recurrent Neural Network(RNN),and present the novel usage of Synthetic Minority Over-Sampling Technique(SMOTE)tailored for IoT-SDN contexts to manage class imbalance during training and enhance performance metrics.Our research has significant practical implications as we authenticate the approache using both the self-generated SD_IoT_Smart_City dataset and the publicly available CICIoT23 dataset.The system utilizes only eleven features to identify DDoS attacks efficiently.Results indicate that the RNN can reliably and precisely differentiate between DDoS traffic and benign traffic by easily identifying temporal relationships and sequences in the data.展开更多
As cyber threats become increasingly sophisticated,Distributed Denial-of-Service(DDoS)attacks continue to pose a serious threat to network infrastructure,often disrupting critical services through overwhelming traffic...As cyber threats become increasingly sophisticated,Distributed Denial-of-Service(DDoS)attacks continue to pose a serious threat to network infrastructure,often disrupting critical services through overwhelming traffic.Although unsupervised anomaly detection using convolutional autoencoders(CAEs)has gained attention for its ability to model normal network behavior without requiring labeled data,conventional CAEs struggle to effectively distinguish between normal and attack traffic due to over-generalized reconstructions and naive anomaly scoring.To address these limitations,we propose CA-CAE,a novel anomaly detection framework designed to improve DDoS detection through asymmetric joint reconstruction learning and refined anomaly scoring.Our architecture connects two CAEs sequentially with asymmetric filter allocation,which amplifies reconstruction errors for anomalous data while preserving low errors for normal traffic.Additionally,we introduce a scoring mechanism that incorporates exponential decay weighting to emphasize recent anomalies and relative traffic volume adjustment to highlight highrisk instances,enabling more accurate and timely detection.We evaluate CA-CAE on a real-world network traffic dataset collected using Cisco NetFlow,containing over 190,000 normal instances and only 78 anomalous instances—an extremely imbalanced scenario(0.0004% anomalies).We validate the proposed framework through extensive experiments,including statistical tests and comparisons with baseline models.Despite this challenge,our method achieves significant improvement,increasing the F1-score from 0.515 obtained by the baseline CAE to 0.934,and outperforming other models.These results demonstrate the effectiveness,scalability,and practicality of CA-CAE for unsupervised DDoS detection in realistic network environments.By combining lightweight model architecture with a domain-aware scoring strategy,our framework provides a robust solution for early detection of DDoS attacks without relying on labeled attack data.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
Vehicular Ad Hoc Networks(VANETs)are central to Intelligent Transportation Systems(ITS),especially for real-time communication involving emergency vehicles.Yet,Distributed Denial of Service(DDoS)attacks can disrupt sa...Vehicular Ad Hoc Networks(VANETs)are central to Intelligent Transportation Systems(ITS),especially for real-time communication involving emergency vehicles.Yet,Distributed Denial of Service(DDoS)attacks can disrupt safety-critical channels and undermine reliability.This paper presents a robust,scalable framework for detecting DDoS attacks in highway VANETs.We construct a new dataset with Network Simulator 3(NS-3)and Simulation of Urban Mobility(SUMO),enriched with real mobility traces from Germany’s A81 highway(OpenStreetMap).Three traffic classes are modeled:DDoS,Voice over IP(VoIP),and Transmission Control Protocol Based(TCP-based)video streaming(VideoTCP).The pipeline includes normalization,feature selection with SHapley Additive exPlanations(SHAP),and class balancing via Synthetic Minority Over-sampling Technique(SMOTE).Eleven classifiers are benchmarked—including eXtreme Gradient Boosting(XGBoost),Categorical Boosting(CatBoost),Adaptive Boosting(AdaBoost),Gradient Boosting(GB),and an Artificial Neural Network(ANN)—using stratified 5-fold cross-validation.XGBoost,GB,CatBoost and ANN achieve the highest performance(weighted F1-score=97%).To assess robustness under non-ideal conditions,we introduce an adversarial evaluation with packet-loss and traffic-jitter(small-sample deformation);the top models retain strong performance,supporting real-time applicability.Collectively,these results demonstrate that the proposed highway-focused framework is accurate,resilient,and well-suited for deployment in VANET security for emergency communications.展开更多
The era of big data brings new challenges for information network systems(INS),simultaneously offering unprecedented opportunities for advancing intelligent intrusion detection systems.In this work,we propose a data-d...The era of big data brings new challenges for information network systems(INS),simultaneously offering unprecedented opportunities for advancing intelligent intrusion detection systems.In this work,we propose a data-driven intrusion detection system for Distributed Denial of Service(DDoS)attack detection.The system focuses on intrusion detection from a big data perceptive.As intelligent information processing methods,big data and artificial intelligence have been widely used in information systems.The INS system is an important information system in cyberspace.In advanced INS systems,the network architectures have become more complex.And the smart devices in INS systems collect a large scale of network data.How to improve the performance of a complex intrusion detection system with big data and artificial intelligence is a big challenge.To address the problem,we design a novel intrusion detection system(IDS)from a big data perspective.The IDS system uses tensors to represent large-scale and complex multi-source network data in a unified tensor.Then,a novel tensor decomposition(TD)method is developed to complete big data mining.The TD method seamlessly collaborates with the XGBoost(eXtreme Gradient Boosting)method to complete the intrusion detection.To verify the proposed IDS system,a series of experiments is conducted on two real network datasets.The results revealed that the proposed IDS system attained an impressive accuracy rate over 98%.Additionally,by altering the scale of the datasets,the proposed IDS system still maintains excellent detection performance,which demonstrates the proposed IDS system’s robustness.展开更多
Satellite Internet(SI)provides broadband access as a critical information infrastructure in 6G.However,with the integration of the terrestrial Internet,the influx of massive terrestrial traffic will bring significant ...Satellite Internet(SI)provides broadband access as a critical information infrastructure in 6G.However,with the integration of the terrestrial Internet,the influx of massive terrestrial traffic will bring significant threats to SI,among which DDoS attack will intensify the erosion of limited bandwidth resources.Therefore,this paper proposes a DDoS attack tracking scheme using a multi-round iterative Viterbi algorithm to achieve high-accuracy attack path reconstruction and fast internal source locking,protecting SI from the source.Firstly,to reduce communication overhead,the logarithmic representation of the traffic volume is added to the digests after modeling SI,generating the lightweight deviation degree to construct the observation probability matrix for the Viterbi algorithm.Secondly,the path node matrix is expanded to multi-index matrices in the Viterbi algorithm to store index information for all probability values,deriving the path with non-repeatability and maximum probability.Finally,multiple rounds of iterative Viterbi tracking are performed locally to track DDoS attack based on trimming tracking results.Simulation and experimental results show that the scheme can achieve 96.8%tracking accuracy of external and internal DDoS attack at 2.5 seconds,with the communication overhead at 268KB/s,effectively protecting the limited bandwidth resources of SI.展开更多
Distributed Denial of Service(DDoS)attacks are one of the severe threats to network infrastructure,sometimes bypassing traditional diagnosis algorithms because of their evolving complexity.PresentMachine Learning(ML)t...Distributed Denial of Service(DDoS)attacks are one of the severe threats to network infrastructure,sometimes bypassing traditional diagnosis algorithms because of their evolving complexity.PresentMachine Learning(ML)techniques for DDoS attack diagnosis normally apply network traffic statistical features such as packet sizes and inter-arrival times.However,such techniques sometimes fail to capture complicated relations among various traffic flows.In this paper,we present a new multi-scale ensemble strategy given the Graph Neural Networks(GNNs)for improving DDoS detection.Our technique divides traffic into macro-and micro-level elements,letting various GNN models to get the two corase-scale anomalies and subtle,stealthy attack models.Through modeling network traffic as graph-structured data,GNNs efficiently learn intricate relations among network entities.The proposed ensemble learning algorithm combines the results of several GNNs to improve generalization,robustness,and scalability.Extensive experiments on three benchmark datasets—UNSW-NB15,CICIDS2017,and CICDDoS2019—show that our approach outperforms traditional machine learning and deep learning models in detecting both high-rate and low-rate(stealthy)DDoS attacks,with significant improvements in accuracy and recall.These findings demonstrate the suggested method’s applicability and robustness for real-world implementation in contexts where several DDoS patterns coexist.展开更多
In federated learning,backdoor attacks have become an important research topic with their wide application in processing sensitive datasets.Since federated learning detects or modifies local models through defense mec...In federated learning,backdoor attacks have become an important research topic with their wide application in processing sensitive datasets.Since federated learning detects or modifies local models through defense mechanisms during aggregation,it is difficult to conduct effective backdoor attacks.In addition,existing backdoor attack methods are faced with challenges,such as low backdoor accuracy,poor ability to evade anomaly detection,and unstable model training.To address these challenges,a method called adaptive simulation backdoor attack(ASBA)is proposed.Specifically,ASBA improves the stability of model training by manipulating the local training process and using an adaptive mechanism,the ability of the malicious model to evade anomaly detection by combing large simulation training and clipping,and the backdoor accuracy by introducing a stimulus model to amplify the impact of the backdoor in the global model.Extensive comparative experiments under five advanced defense scenarios show that ASBA can effectively evade anomaly detection and achieve high backdoor accuracy in the global model.Furthermore,it exhibits excellent stability and effectiveness after multiple rounds of attacks,outperforming state-of-the-art backdoor attack methods.展开更多
The exponential growth of the Internet of Things(IoT)has introduced significant security challenges,with zero-day attacks emerging as one of the most critical and challenging threats.Traditional Machine Learning(ML)an...The exponential growth of the Internet of Things(IoT)has introduced significant security challenges,with zero-day attacks emerging as one of the most critical and challenging threats.Traditional Machine Learning(ML)and Deep Learning(DL)techniques have demonstrated promising early detection capabilities.However,their effectiveness is limited when handling the vast volumes of IoT-generated data due to scalability constraints,high computational costs,and the costly time-intensive process of data labeling.To address these challenges,this study proposes a Federated Learning(FL)framework that leverages collaborative and hybrid supervised learning to enhance cyber threat detection in IoT networks.By employing Deep Neural Networks(DNNs)and decentralized model training,the approach reduces computational complexity while improving detection accuracy.The proposed model demonstrates robust performance,achieving accuracies of 94.34%,99.95%,and 87.94%on the publicly available kitsune,Bot-IoT,and UNSW-NB15 datasets,respectively.Furthermore,its ability to detect zero-day attacks is validated through evaluations on two additional benchmark datasets,TON-IoT and IoT-23,using a Deep Federated Learning(DFL)framework,underscoring the generalization and effectiveness of the model in heterogeneous and decentralized IoT environments.Experimental results demonstrate superior performance over existing methods,establishing the proposed framework as an efficient and scalable solution for IoT security.展开更多
Federated Learning(FL)protects data privacy through a distributed training mechanism,yet its decentralized nature also introduces new security vulnerabilities.Backdoor attacks inject malicious triggers into the global...Federated Learning(FL)protects data privacy through a distributed training mechanism,yet its decentralized nature also introduces new security vulnerabilities.Backdoor attacks inject malicious triggers into the global model through compromised updates,posing significant threats to model integrity and becoming a key focus in FL security.Existing backdoor attack methods typically embed triggers directly into original images and consider only data heterogeneity,resulting in limited stealth and adaptability.To address the heterogeneity of malicious client devices,this paper proposes a novel backdoor attack method named Capability-Adaptive Shadow Backdoor Attack(CASBA).By incorporating measurements of clients’computational and communication capabilities,CASBA employs a dynamic hierarchical attack strategy that adaptively aligns attack intensity with available resources.Furthermore,an improved deep convolutional generative adversarial network(DCGAN)is integrated into the attack pipeline to embed triggers without modifying original data,significantly enhancing stealthiness.Comparative experiments with Shadow Backdoor Attack(SBA)across multiple scenarios demonstrate that CASBA dynamically adjusts resource consumption based on device capabilities,reducing average memory usage per iteration by 5.8%.CASBA improves resource efficiency while keeping the drop in attack success rate within 3%.Additionally,the effectiveness of CASBA against three robust FL algorithms is also validated.展开更多
The surge in smishing attacks underscores the urgent need for robust,real-time detection systems powered by advanced deep learning models.This paper introduces PhishNet,a novel ensemble learning framework that integra...The surge in smishing attacks underscores the urgent need for robust,real-time detection systems powered by advanced deep learning models.This paper introduces PhishNet,a novel ensemble learning framework that integrates transformer-based models(RoBERTa)and large language models(LLMs)(GPT-OSS 120B,LLaMA3.370B,and Qwen332B)to enhance smishing detection performance significantly.To mitigate class imbalance,we apply synthetic data augmentation using T5 and leverage various text preprocessing techniques.Our system employs a duallayer voting mechanism:weighted majority voting among LLMs and a final ensemble vote to classify messages as ham,spam,or smishing.Experimental results show an average accuracy improvement from 96%to 98.5%compared to the best standalone transformer,and from 93%to 98.5%when compared to LLMs across datasets.Furthermore,we present a real-time,user-friendly application to operationalize our detection model for practical use.PhishNet demonstrates superior scalability,usability,and detection accuracy,filling critical gaps in current smishing detection methodologies.展开更多
Dear Editor,This letter studies the problem of stealthy attacks targeting stochastic event-based estimation,alongside proposing measures for their mitigation.A general attack framework is introduced,and the correspond...Dear Editor,This letter studies the problem of stealthy attacks targeting stochastic event-based estimation,alongside proposing measures for their mitigation.A general attack framework is introduced,and the corresponding stealthiness condition is analyzed.To enhance system security,we advocate for a single-dimensional encryption method,showing that securing a singular data element is sufficient to shield the system from the perils of stealthy attacks.展开更多
Zero-click attacks represent an advanced cybersecurity threat,capable of compromising devices without user interaction.High-profile examples such as Pegasus,Simjacker,Bluebugging,and Bluesnarfing exploit hidden vulner...Zero-click attacks represent an advanced cybersecurity threat,capable of compromising devices without user interaction.High-profile examples such as Pegasus,Simjacker,Bluebugging,and Bluesnarfing exploit hidden vulnerabilities in software and communication protocols to silently gain access,exfiltrate data,and enable long-term surveillance.Their stealth and ability to evade traditional defenses make detection and mitigation highly challenging.This paper addresses these threats by systematically mapping the tactics and techniques of zero-click attacks using the MITRE ATT&CK framework,a widely adopted standard for modeling adversarial behavior.Through this mapping,we categorize real-world attack vectors and better understand how such attacks operate across the cyber-kill chain.To support threat detection efforts,we propose an Active Learning-based method to efficiently label the Pegasus spyware dataset in alignment with the MITRE ATT&CK framework.This approach reduces the effort of manually annotating data while improving the quality of the labeled data,which is essential to train robust cybersecurity models.In addition,our analysis highlights the structured execution paths of zero-click attacks and reveals gaps in current defense strategies.The findings emphasize the importance of forward-looking strategies such as continuous surveillance,dynamic threat profiling,and security education.By bridging zero-click attack analysis with the MITRE ATT&CK framework and leveraging machine learning for dataset annotation,this work provides a foundation for more accurate threat detection and the development of more resilient and structured cybersecurity frameworks.展开更多
Graph Neural Networks(GNNs)have proven highly effective for graph classification across diverse fields such as social networks,bioinformatics,and finance,due to their capability to learn complex graph structures.Howev...Graph Neural Networks(GNNs)have proven highly effective for graph classification across diverse fields such as social networks,bioinformatics,and finance,due to their capability to learn complex graph structures.However,despite their success,GNNs remain vulnerable to adversarial attacks that can significantly degrade their classification accuracy.Existing adversarial attack strategies primarily rely on label information to guide the attacks,which limits their applicability in scenarios where such information is scarce or unavailable.This paper introduces an innovative unsupervised attack method for graph classification,which operates without relying on label information,thereby enhancing its applicability in a broad range of scenarios.Specifically,our method first leverages a graph contrastive learning loss to learn high-quality graph embeddings by comparing different stochastic augmented views of the graphs.To effectively perturb the graphs,we then introduce an implicit estimator that measures the impact of various modifications on graph structures.The proposed strategy identifies and flips edges with the top-K highest scores,determined by the estimator,to maximize the degradation of the model’s performance.In addition,to defend against such attack,we propose a lightweight regularization-based defense mechanism that is specifically tailored to mitigate the structural perturbations introduced by our attack strategy.It enhances model robustness by enforcing embedding consistency and edge-level smoothness during training.We conduct experiments on six public TU graph classification datasets:NCI1,NCI109,Mutagenicity,ENZYMES,COLLAB,and DBLP_v1,to evaluate the effectiveness of our attack and defense strategies.Under an attack budget of 3,the maximum reduction in model accuracy reaches 6.67%on the Graph Convolutional Network(GCN)and 11.67%on the Graph Attention Network(GAT)across different datasets,indicating that our unsupervised method induces degradation comparable to state-of-the-art supervised attacks.Meanwhile,our defense achieves the highest accuracy recovery of 3.89%(GCN)and 5.00%(GAT),demonstrating improved robustness against structural perturbations.展开更多
In recent years,with the rapid advancement of artificial intelligence,object detection algorithms have made significant strides in accuracy and computational efficiency.Notably,research and applications of Anchor-Free...In recent years,with the rapid advancement of artificial intelligence,object detection algorithms have made significant strides in accuracy and computational efficiency.Notably,research and applications of Anchor-Free models have opened new avenues for real-time target detection in optical remote sensing images(ORSIs).However,in the realmof adversarial attacks,developing adversarial techniques tailored to Anchor-Freemodels remains challenging.Adversarial examples generated based on Anchor-Based models often exhibit poor transferability to these new model architectures.Furthermore,the growing diversity of Anchor-Free models poses additional hurdles to achieving robust transferability of adversarial attacks.This study presents an improved cross-conv-block feature fusion You Only Look Once(YOLO)architecture,meticulously engineered to facilitate the extraction ofmore comprehensive semantic features during the backpropagation process.To address the asymmetry between densely distributed objects in ORSIs and the corresponding detector outputs,a novel dense bounding box attack strategy is proposed.This approach leverages dense target bounding boxes loss in the calculation of adversarial loss functions.Furthermore,by integrating translation-invariant(TI)and momentum-iteration(MI)adversarial methodologies,the proposed framework significantly improves the transferability of adversarial attacks.Experimental results demonstrate that our method achieves superior adversarial attack performance,with adversarial transferability rates(ATR)of 67.53%on the NWPU VHR-10 dataset and 90.71%on the HRSC2016 dataset.Compared to ensemble adversarial attack and cascaded adversarial attack approaches,our method generates adversarial examples in an average of 0.64 s,representing an approximately 14.5%improvement in efficiency under equivalent conditions.展开更多
Large language models(LLMs)have revolutionized AI applications across diverse domains.However,their widespread deployment has introduced critical security vulnerabilities,particularly prompt injection attacks that man...Large language models(LLMs)have revolutionized AI applications across diverse domains.However,their widespread deployment has introduced critical security vulnerabilities,particularly prompt injection attacks that manipulate model behavior through malicious instructions.Following Kitchenham’s guidelines,this systematic review synthesizes 128 peer-reviewed studies from 2022 to 2025 to provide a unified understanding of this rapidly evolving threat landscape.Our findings reveal a swift progression from simple direct injections to sophisticated multimodal attacks,achieving over 90%success rates against unprotected systems.In response,defense mechanisms show varying effectiveness:input preprocessing achieves 60%–80%detection rates and advanced architectural defenses demonstrate up to 95%protection against known patterns,though significant gaps persist against novel attack vectors.We identified 37 distinct defense approaches across three categories,but standardized evaluation frameworks remain limited.Our analysis attributes these vulnerabilities to fundamental LLM architectural limitations,such as the inability to distinguish instructions from data and attention mechanism vulnerabilities.This highlights critical research directions such as formal verification methods,standardized evaluation protocols,and architectural innovations for inherently secure LLM designs.展开更多
Internet of Things(IoTs)devices are bringing about a revolutionary change our society by enabling connectivity regardless of time and location.However,The extensive deployment of these devices also makes them attracti...Internet of Things(IoTs)devices are bringing about a revolutionary change our society by enabling connectivity regardless of time and location.However,The extensive deployment of these devices also makes them attractive victims for themalicious actions of adversaries.Within the spectrumof existing threats,Side-ChannelAttacks(SCAs)have established themselves as an effective way to compromise cryptographic implementations.These attacks exploit unintended,unintended physical leakage that occurs during the cryptographic execution of devices,bypassing the theoretical strength of the crypto design.In recent times,the advancement of deep learning has provided SCAs with a powerful ally.Well-trained deep-learningmodels demonstrate an exceptional capacity to identify correlations between side-channel measurements and sensitive data,thereby significantly enhancing such attacks.To further understand the security threats posed by deep-learning SCAs and to aid in formulating robust countermeasures in the future,this paper undertakes an exhaustive investigation of leading-edge SCAs targeting Advanced Encryption Standard(AES)implementations.The study specifically focuses on attacks that exploit power consumption and electromagnetic(EM)emissions as primary leakage sources,systematically evaluating the extent to which diverse deep learning techniques enhance SCAs acrossmultiple critical dimensions.These dimensions include:(i)the characteristics of publicly available datasets derived from various hardware and software platforms;(ii)the formalization of leakage models tailored to different attack scenarios;(iii)the architectural suitability and performance of state-of-the-art deep learning models.Furthermore,the survey provides a systematic synthesis of current research findings,identifies significant unresolved issues in the existing literature and suggests promising directions for future work,including cross-device attack transferability and the impact of quantum-classical hybrid computing on side-channel security.展开更多
基金supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea(NRF-2025S1A5A2A01005171)by the BK21 programat Chungbuk National University(2025).
文摘With an increase in internet-connected devices and a dependency on online services,the threat of Distributed Denial of Service(DDoS)attacks has become a significant concern in cybersecurity.The proposed system follows a multi-step process,beginning with the collection of datasets from different edge devices and network nodes.To verify its effectiveness,experiments were conducted using the CICDoS2017,NSL-KDD,and CICIDS benchmark datasets alongside other existing models.Recursive feature elimination(RFE)with random forest is used to select features from the CICDDoS2019 dataset,on which a BiLSTM model is trained on local nodes.Local models are trained until convergence or stability criteria are met while simultaneously sharing the updates globally for collaborative learning.A centralised server evaluates real-time traffic using the global BiLSTM model,which triggers alerts for potential DDoS attacks.Furthermore,blockchain technology is employed to secure model updates and to provide an immutable audit trail,thereby ensuring trust and accountability among network nodes.This research introduces a novel decentralized method called Federated Random Forest Bidirectional Long Short-Term Memory(FRF-BiLSTM)for detecting DDoS attacks,utilizing the advanced Bidirectional Long Short-Term Memory Networks(BiLSTMs)to analyze sequences in both forward and backward directions.The outcome shows the proposed model achieves a mean accuracy of 97.1%with an average training delay of 88.7 s and testing delay of 21.4 s.The model demonstrates scalability and the best detection performance in large-scale attack scenarios.
文摘A Distributed Denial-of-Service(DDoS)attack poses a significant challenge in the digital age,disrupting online services with operational and financial consequences.Detecting such attacks requires innovative and effective solutions.The primary challenge lies in selecting the best among several DDoS detection models.This study presents a framework that combines several DDoS detection models and Multiple-Criteria Decision-Making(MCDM)techniques to compare and select the most effective models.The framework integrates a decision matrix from training several models on the CiC-DDOS2019 dataset with Fuzzy Weighted Zero Inconsistency Criterion(FWZIC)and MultiAttribute Boundary Approximation Area Comparison(MABAC)methodologies.FWZIC assigns weights to evaluate criteria,while MABAC compares detection models based on the assessed criteria.The results indicate that the FWZIC approach assigns weights to criteria reliably,with time complexity receiving the highest weight(0.2585)and F1 score receiving the lowest weight(0.14644).Among the models evaluated using the MABAC approach,the Support Vector Machine(SVM)ranked first with a score of 0.0444,making it the most suitable for this work.In contrast,Naive Bayes(NB)ranked lowest with a score of 0.0018.Objective validation and sensitivity analysis proved the reliability of the framework.This study provides a practical approach and insights for cybersecurity practitioners and researchers to evaluate DDoS detection models.
基金supported by NSTC 113-2221-E-155-055NSTC 113-2222-E-155-007,Taiwan.
文摘The rapid advancement of the Internet ofThings(IoT)has heightened the importance of security,with a notable increase in Distributed Denial-of-Service(DDoS)attacks targeting IoT devices.Network security specialists face the challenge of producing systems to identify and offset these attacks.This researchmanages IoT security through the emerging Software-Defined Networking(SDN)standard by developing a unified framework(RNN-RYU).We thoroughly assess multiple deep learning frameworks,including Convolutional Neural Network(CNN),Long Short-Term Memory(LSTM),Feed-Forward Convolutional Neural Network(FFCNN),and Recurrent Neural Network(RNN),and present the novel usage of Synthetic Minority Over-Sampling Technique(SMOTE)tailored for IoT-SDN contexts to manage class imbalance during training and enhance performance metrics.Our research has significant practical implications as we authenticate the approache using both the self-generated SD_IoT_Smart_City dataset and the publicly available CICIoT23 dataset.The system utilizes only eleven features to identify DDoS attacks efficiently.Results indicate that the RNN can reliably and precisely differentiate between DDoS traffic and benign traffic by easily identifying temporal relationships and sequences in the data.
基金supported by Korea National University of Transportation Industry-Academy Cooperation Foundation in 2024.
文摘As cyber threats become increasingly sophisticated,Distributed Denial-of-Service(DDoS)attacks continue to pose a serious threat to network infrastructure,often disrupting critical services through overwhelming traffic.Although unsupervised anomaly detection using convolutional autoencoders(CAEs)has gained attention for its ability to model normal network behavior without requiring labeled data,conventional CAEs struggle to effectively distinguish between normal and attack traffic due to over-generalized reconstructions and naive anomaly scoring.To address these limitations,we propose CA-CAE,a novel anomaly detection framework designed to improve DDoS detection through asymmetric joint reconstruction learning and refined anomaly scoring.Our architecture connects two CAEs sequentially with asymmetric filter allocation,which amplifies reconstruction errors for anomalous data while preserving low errors for normal traffic.Additionally,we introduce a scoring mechanism that incorporates exponential decay weighting to emphasize recent anomalies and relative traffic volume adjustment to highlight highrisk instances,enabling more accurate and timely detection.We evaluate CA-CAE on a real-world network traffic dataset collected using Cisco NetFlow,containing over 190,000 normal instances and only 78 anomalous instances—an extremely imbalanced scenario(0.0004% anomalies).We validate the proposed framework through extensive experiments,including statistical tests and comparisons with baseline models.Despite this challenge,our method achieves significant improvement,increasing the F1-score from 0.515 obtained by the baseline CAE to 0.934,and outperforming other models.These results demonstrate the effectiveness,scalability,and practicality of CA-CAE for unsupervised DDoS detection in realistic network environments.By combining lightweight model architecture with a domain-aware scoring strategy,our framework provides a robust solution for early detection of DDoS attacks without relying on labeled attack data.
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.
文摘Vehicular Ad Hoc Networks(VANETs)are central to Intelligent Transportation Systems(ITS),especially for real-time communication involving emergency vehicles.Yet,Distributed Denial of Service(DDoS)attacks can disrupt safety-critical channels and undermine reliability.This paper presents a robust,scalable framework for detecting DDoS attacks in highway VANETs.We construct a new dataset with Network Simulator 3(NS-3)and Simulation of Urban Mobility(SUMO),enriched with real mobility traces from Germany’s A81 highway(OpenStreetMap).Three traffic classes are modeled:DDoS,Voice over IP(VoIP),and Transmission Control Protocol Based(TCP-based)video streaming(VideoTCP).The pipeline includes normalization,feature selection with SHapley Additive exPlanations(SHAP),and class balancing via Synthetic Minority Over-sampling Technique(SMOTE).Eleven classifiers are benchmarked—including eXtreme Gradient Boosting(XGBoost),Categorical Boosting(CatBoost),Adaptive Boosting(AdaBoost),Gradient Boosting(GB),and an Artificial Neural Network(ANN)—using stratified 5-fold cross-validation.XGBoost,GB,CatBoost and ANN achieve the highest performance(weighted F1-score=97%).To assess robustness under non-ideal conditions,we introduce an adversarial evaluation with packet-loss and traffic-jitter(small-sample deformation);the top models retain strong performance,supporting real-time applicability.Collectively,these results demonstrate that the proposed highway-focused framework is accurate,resilient,and well-suited for deployment in VANET security for emergency communications.
基金supported in part by the National Nature Science Foundation of China under Project 62166047in part by the Yunnan International Joint Laboratory of Natural Rubber Intelligent Monitor and Digital Applications under Grant 202403AP140001in part by the Xingdian Talent Support Program under Grant YNWR-QNBJ-2019-270.
文摘The era of big data brings new challenges for information network systems(INS),simultaneously offering unprecedented opportunities for advancing intelligent intrusion detection systems.In this work,we propose a data-driven intrusion detection system for Distributed Denial of Service(DDoS)attack detection.The system focuses on intrusion detection from a big data perceptive.As intelligent information processing methods,big data and artificial intelligence have been widely used in information systems.The INS system is an important information system in cyberspace.In advanced INS systems,the network architectures have become more complex.And the smart devices in INS systems collect a large scale of network data.How to improve the performance of a complex intrusion detection system with big data and artificial intelligence is a big challenge.To address the problem,we design a novel intrusion detection system(IDS)from a big data perspective.The IDS system uses tensors to represent large-scale and complex multi-source network data in a unified tensor.Then,a novel tensor decomposition(TD)method is developed to complete big data mining.The TD method seamlessly collaborates with the XGBoost(eXtreme Gradient Boosting)method to complete the intrusion detection.To verify the proposed IDS system,a series of experiments is conducted on two real network datasets.The results revealed that the proposed IDS system attained an impressive accuracy rate over 98%.Additionally,by altering the scale of the datasets,the proposed IDS system still maintains excellent detection performance,which demonstrates the proposed IDS system’s robustness.
基金supported by the National Key R&D Program of China(Grant No.2022YFA1005000)the National Natural Science Foundation of China(Grant No.62025110 and 62101308).
文摘Satellite Internet(SI)provides broadband access as a critical information infrastructure in 6G.However,with the integration of the terrestrial Internet,the influx of massive terrestrial traffic will bring significant threats to SI,among which DDoS attack will intensify the erosion of limited bandwidth resources.Therefore,this paper proposes a DDoS attack tracking scheme using a multi-round iterative Viterbi algorithm to achieve high-accuracy attack path reconstruction and fast internal source locking,protecting SI from the source.Firstly,to reduce communication overhead,the logarithmic representation of the traffic volume is added to the digests after modeling SI,generating the lightweight deviation degree to construct the observation probability matrix for the Viterbi algorithm.Secondly,the path node matrix is expanded to multi-index matrices in the Viterbi algorithm to store index information for all probability values,deriving the path with non-repeatability and maximum probability.Finally,multiple rounds of iterative Viterbi tracking are performed locally to track DDoS attack based on trimming tracking results.Simulation and experimental results show that the scheme can achieve 96.8%tracking accuracy of external and internal DDoS attack at 2.5 seconds,with the communication overhead at 268KB/s,effectively protecting the limited bandwidth resources of SI.
文摘Distributed Denial of Service(DDoS)attacks are one of the severe threats to network infrastructure,sometimes bypassing traditional diagnosis algorithms because of their evolving complexity.PresentMachine Learning(ML)techniques for DDoS attack diagnosis normally apply network traffic statistical features such as packet sizes and inter-arrival times.However,such techniques sometimes fail to capture complicated relations among various traffic flows.In this paper,we present a new multi-scale ensemble strategy given the Graph Neural Networks(GNNs)for improving DDoS detection.Our technique divides traffic into macro-and micro-level elements,letting various GNN models to get the two corase-scale anomalies and subtle,stealthy attack models.Through modeling network traffic as graph-structured data,GNNs efficiently learn intricate relations among network entities.The proposed ensemble learning algorithm combines the results of several GNNs to improve generalization,robustness,and scalability.Extensive experiments on three benchmark datasets—UNSW-NB15,CICIDS2017,and CICDDoS2019—show that our approach outperforms traditional machine learning and deep learning models in detecting both high-rate and low-rate(stealthy)DDoS attacks,with significant improvements in accuracy and recall.These findings demonstrate the suggested method’s applicability and robustness for real-world implementation in contexts where several DDoS patterns coexist.
文摘In federated learning,backdoor attacks have become an important research topic with their wide application in processing sensitive datasets.Since federated learning detects or modifies local models through defense mechanisms during aggregation,it is difficult to conduct effective backdoor attacks.In addition,existing backdoor attack methods are faced with challenges,such as low backdoor accuracy,poor ability to evade anomaly detection,and unstable model training.To address these challenges,a method called adaptive simulation backdoor attack(ASBA)is proposed.Specifically,ASBA improves the stability of model training by manipulating the local training process and using an adaptive mechanism,the ability of the malicious model to evade anomaly detection by combing large simulation training and clipping,and the backdoor accuracy by introducing a stimulus model to amplify the impact of the backdoor in the global model.Extensive comparative experiments under five advanced defense scenarios show that ASBA can effectively evade anomaly detection and achieve high backdoor accuracy in the global model.Furthermore,it exhibits excellent stability and effectiveness after multiple rounds of attacks,outperforming state-of-the-art backdoor attack methods.
基金supported by Princess Nourah bint Abdulrahman University Researchers Supporting Project Number(PNURSP2025R97)Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘The exponential growth of the Internet of Things(IoT)has introduced significant security challenges,with zero-day attacks emerging as one of the most critical and challenging threats.Traditional Machine Learning(ML)and Deep Learning(DL)techniques have demonstrated promising early detection capabilities.However,their effectiveness is limited when handling the vast volumes of IoT-generated data due to scalability constraints,high computational costs,and the costly time-intensive process of data labeling.To address these challenges,this study proposes a Federated Learning(FL)framework that leverages collaborative and hybrid supervised learning to enhance cyber threat detection in IoT networks.By employing Deep Neural Networks(DNNs)and decentralized model training,the approach reduces computational complexity while improving detection accuracy.The proposed model demonstrates robust performance,achieving accuracies of 94.34%,99.95%,and 87.94%on the publicly available kitsune,Bot-IoT,and UNSW-NB15 datasets,respectively.Furthermore,its ability to detect zero-day attacks is validated through evaluations on two additional benchmark datasets,TON-IoT and IoT-23,using a Deep Federated Learning(DFL)framework,underscoring the generalization and effectiveness of the model in heterogeneous and decentralized IoT environments.Experimental results demonstrate superior performance over existing methods,establishing the proposed framework as an efficient and scalable solution for IoT security.
基金supported by the National Natural Science Foundation of China(Grant No.62172123)the Key Research and Development Program of Heilongjiang Province,China(GrantNo.2022ZX01A36).
文摘Federated Learning(FL)protects data privacy through a distributed training mechanism,yet its decentralized nature also introduces new security vulnerabilities.Backdoor attacks inject malicious triggers into the global model through compromised updates,posing significant threats to model integrity and becoming a key focus in FL security.Existing backdoor attack methods typically embed triggers directly into original images and consider only data heterogeneity,resulting in limited stealth and adaptability.To address the heterogeneity of malicious client devices,this paper proposes a novel backdoor attack method named Capability-Adaptive Shadow Backdoor Attack(CASBA).By incorporating measurements of clients’computational and communication capabilities,CASBA employs a dynamic hierarchical attack strategy that adaptively aligns attack intensity with available resources.Furthermore,an improved deep convolutional generative adversarial network(DCGAN)is integrated into the attack pipeline to embed triggers without modifying original data,significantly enhancing stealthiness.Comparative experiments with Shadow Backdoor Attack(SBA)across multiple scenarios demonstrate that CASBA dynamically adjusts resource consumption based on device capabilities,reducing average memory usage per iteration by 5.8%.CASBA improves resource efficiency while keeping the drop in attack success rate within 3%.Additionally,the effectiveness of CASBA against three robust FL algorithms is also validated.
基金funded by the Deanship of Scientific Research(DSR)at King Abdulaziz University,Jeddah,under Grant No.(GPIP:1074-612-2024).
文摘The surge in smishing attacks underscores the urgent need for robust,real-time detection systems powered by advanced deep learning models.This paper introduces PhishNet,a novel ensemble learning framework that integrates transformer-based models(RoBERTa)and large language models(LLMs)(GPT-OSS 120B,LLaMA3.370B,and Qwen332B)to enhance smishing detection performance significantly.To mitigate class imbalance,we apply synthetic data augmentation using T5 and leverage various text preprocessing techniques.Our system employs a duallayer voting mechanism:weighted majority voting among LLMs and a final ensemble vote to classify messages as ham,spam,or smishing.Experimental results show an average accuracy improvement from 96%to 98.5%compared to the best standalone transformer,and from 93%to 98.5%when compared to LLMs across datasets.Furthermore,we present a real-time,user-friendly application to operationalize our detection model for practical use.PhishNet demonstrates superior scalability,usability,and detection accuracy,filling critical gaps in current smishing detection methodologies.
基金supported by the National Natural Science Foundation of China(62303353,62273030,62573320)。
文摘Dear Editor,This letter studies the problem of stealthy attacks targeting stochastic event-based estimation,alongside proposing measures for their mitigation.A general attack framework is introduced,and the corresponding stealthiness condition is analyzed.To enhance system security,we advocate for a single-dimensional encryption method,showing that securing a singular data element is sufficient to shield the system from the perils of stealthy attacks.
文摘Zero-click attacks represent an advanced cybersecurity threat,capable of compromising devices without user interaction.High-profile examples such as Pegasus,Simjacker,Bluebugging,and Bluesnarfing exploit hidden vulnerabilities in software and communication protocols to silently gain access,exfiltrate data,and enable long-term surveillance.Their stealth and ability to evade traditional defenses make detection and mitigation highly challenging.This paper addresses these threats by systematically mapping the tactics and techniques of zero-click attacks using the MITRE ATT&CK framework,a widely adopted standard for modeling adversarial behavior.Through this mapping,we categorize real-world attack vectors and better understand how such attacks operate across the cyber-kill chain.To support threat detection efforts,we propose an Active Learning-based method to efficiently label the Pegasus spyware dataset in alignment with the MITRE ATT&CK framework.This approach reduces the effort of manually annotating data while improving the quality of the labeled data,which is essential to train robust cybersecurity models.In addition,our analysis highlights the structured execution paths of zero-click attacks and reveals gaps in current defense strategies.The findings emphasize the importance of forward-looking strategies such as continuous surveillance,dynamic threat profiling,and security education.By bridging zero-click attack analysis with the MITRE ATT&CK framework and leveraging machine learning for dataset annotation,this work provides a foundation for more accurate threat detection and the development of more resilient and structured cybersecurity frameworks.
基金funded by the National Key Research and Development Program of China(Grant No.2024YFE0209000)the NSFC(Grant No.U23B2019).
文摘Graph Neural Networks(GNNs)have proven highly effective for graph classification across diverse fields such as social networks,bioinformatics,and finance,due to their capability to learn complex graph structures.However,despite their success,GNNs remain vulnerable to adversarial attacks that can significantly degrade their classification accuracy.Existing adversarial attack strategies primarily rely on label information to guide the attacks,which limits their applicability in scenarios where such information is scarce or unavailable.This paper introduces an innovative unsupervised attack method for graph classification,which operates without relying on label information,thereby enhancing its applicability in a broad range of scenarios.Specifically,our method first leverages a graph contrastive learning loss to learn high-quality graph embeddings by comparing different stochastic augmented views of the graphs.To effectively perturb the graphs,we then introduce an implicit estimator that measures the impact of various modifications on graph structures.The proposed strategy identifies and flips edges with the top-K highest scores,determined by the estimator,to maximize the degradation of the model’s performance.In addition,to defend against such attack,we propose a lightweight regularization-based defense mechanism that is specifically tailored to mitigate the structural perturbations introduced by our attack strategy.It enhances model robustness by enforcing embedding consistency and edge-level smoothness during training.We conduct experiments on six public TU graph classification datasets:NCI1,NCI109,Mutagenicity,ENZYMES,COLLAB,and DBLP_v1,to evaluate the effectiveness of our attack and defense strategies.Under an attack budget of 3,the maximum reduction in model accuracy reaches 6.67%on the Graph Convolutional Network(GCN)and 11.67%on the Graph Attention Network(GAT)across different datasets,indicating that our unsupervised method induces degradation comparable to state-of-the-art supervised attacks.Meanwhile,our defense achieves the highest accuracy recovery of 3.89%(GCN)and 5.00%(GAT),demonstrating improved robustness against structural perturbations.
文摘In recent years,with the rapid advancement of artificial intelligence,object detection algorithms have made significant strides in accuracy and computational efficiency.Notably,research and applications of Anchor-Free models have opened new avenues for real-time target detection in optical remote sensing images(ORSIs).However,in the realmof adversarial attacks,developing adversarial techniques tailored to Anchor-Freemodels remains challenging.Adversarial examples generated based on Anchor-Based models often exhibit poor transferability to these new model architectures.Furthermore,the growing diversity of Anchor-Free models poses additional hurdles to achieving robust transferability of adversarial attacks.This study presents an improved cross-conv-block feature fusion You Only Look Once(YOLO)architecture,meticulously engineered to facilitate the extraction ofmore comprehensive semantic features during the backpropagation process.To address the asymmetry between densely distributed objects in ORSIs and the corresponding detector outputs,a novel dense bounding box attack strategy is proposed.This approach leverages dense target bounding boxes loss in the calculation of adversarial loss functions.Furthermore,by integrating translation-invariant(TI)and momentum-iteration(MI)adversarial methodologies,the proposed framework significantly improves the transferability of adversarial attacks.Experimental results demonstrate that our method achieves superior adversarial attack performance,with adversarial transferability rates(ATR)of 67.53%on the NWPU VHR-10 dataset and 90.71%on the HRSC2016 dataset.Compared to ensemble adversarial attack and cascaded adversarial attack approaches,our method generates adversarial examples in an average of 0.64 s,representing an approximately 14.5%improvement in efficiency under equivalent conditions.
基金supported by 2023 Higher Education Scientific Research Planning Project of China Society of Higher Education(No.23PG0408)2023 Philosophy and Social Science Research Programs in Jiangsu Province(No.2023SJSZ0993)+2 种基金Nantong Science and Technology Project(No.JC2023070)Key Project of Jiangsu Province Education Science 14th Five-Year Plan(Grant No.B-b/2024/02/41)the Open Fund of Advanced Cryptography and System Security Key Laboratory of Sichuan Province(Grant No.SKLACSS-202407).
文摘Large language models(LLMs)have revolutionized AI applications across diverse domains.However,their widespread deployment has introduced critical security vulnerabilities,particularly prompt injection attacks that manipulate model behavior through malicious instructions.Following Kitchenham’s guidelines,this systematic review synthesizes 128 peer-reviewed studies from 2022 to 2025 to provide a unified understanding of this rapidly evolving threat landscape.Our findings reveal a swift progression from simple direct injections to sophisticated multimodal attacks,achieving over 90%success rates against unprotected systems.In response,defense mechanisms show varying effectiveness:input preprocessing achieves 60%–80%detection rates and advanced architectural defenses demonstrate up to 95%protection against known patterns,though significant gaps persist against novel attack vectors.We identified 37 distinct defense approaches across three categories,but standardized evaluation frameworks remain limited.Our analysis attributes these vulnerabilities to fundamental LLM architectural limitations,such as the inability to distinguish instructions from data and attention mechanism vulnerabilities.This highlights critical research directions such as formal verification methods,standardized evaluation protocols,and architectural innovations for inherently secure LLM designs.
基金The Key R&D Program of Hunan Province(Grant No.2025AQ2024)of the Department of Science and Technology of Hunan Province.Distinguished Young Scientists Fund(Grant No.24B0446)of Hunan Education Department.
文摘Internet of Things(IoTs)devices are bringing about a revolutionary change our society by enabling connectivity regardless of time and location.However,The extensive deployment of these devices also makes them attractive victims for themalicious actions of adversaries.Within the spectrumof existing threats,Side-ChannelAttacks(SCAs)have established themselves as an effective way to compromise cryptographic implementations.These attacks exploit unintended,unintended physical leakage that occurs during the cryptographic execution of devices,bypassing the theoretical strength of the crypto design.In recent times,the advancement of deep learning has provided SCAs with a powerful ally.Well-trained deep-learningmodels demonstrate an exceptional capacity to identify correlations between side-channel measurements and sensitive data,thereby significantly enhancing such attacks.To further understand the security threats posed by deep-learning SCAs and to aid in formulating robust countermeasures in the future,this paper undertakes an exhaustive investigation of leading-edge SCAs targeting Advanced Encryption Standard(AES)implementations.The study specifically focuses on attacks that exploit power consumption and electromagnetic(EM)emissions as primary leakage sources,systematically evaluating the extent to which diverse deep learning techniques enhance SCAs acrossmultiple critical dimensions.These dimensions include:(i)the characteristics of publicly available datasets derived from various hardware and software platforms;(ii)the formalization of leakage models tailored to different attack scenarios;(iii)the architectural suitability and performance of state-of-the-art deep learning models.Furthermore,the survey provides a systematic synthesis of current research findings,identifies significant unresolved issues in the existing literature and suggests promising directions for future work,including cross-device attack transferability and the impact of quantum-classical hybrid computing on side-channel security.
