Accurate prediction of molecular properties is crucial for selecting compounds with ideal properties and reducing the costs and risks of trials.Traditional methods based on manually crafted features and graph-based me...Accurate prediction of molecular properties is crucial for selecting compounds with ideal properties and reducing the costs and risks of trials.Traditional methods based on manually crafted features and graph-based methods have shown promising results in molecular property prediction.However,traditional methods rely on expert knowledge and often fail to capture the complex structures and interactions within molecules.Similarly,graph-based methods typically overlook the chemical structure and function hidden in molecular motifs and struggle to effectively integrate global and local molecular information.To address these limitations,we propose a novel fingerprint-enhanced hierarchical graph neural network(FH-GNN)for molecular property prediction that simultaneously learns information from hierarchical molecular graphs and fingerprints.The FH-GNN captures diverse hierarchical chemical information by applying directed message-passing neural networks(D-MPNN)on a hierarchical molecular graph that integrates atomic-level,motif-level,and graph-level information along with their relationships.Addi-tionally,we used an adaptive attention mechanism to balance the importance of hierarchical graphs and fingerprint features,creating a comprehensive molecular embedding that integrated hierarchical mo-lecular structures with domain knowledge.Experiments on eight benchmark datasets from MoleculeNet showed that FH-GNN outperformed the baseline models in both classification and regression tasks for molecular property prediction,validating its capability to comprehensively capture molecular informa-tion.By integrating molecular structure and chemical knowledge,FH-GNN provides a powerful tool for the accurate prediction of molecular properties and aids in the discovery of potential drug candidates.展开更多
This paper investigates how to achieve integrated sensing and communication(ISAC)based on a cell-free radio access network(CF-RAN)architecture with a minimum footprint of communication resources.We propose a new passi...This paper investigates how to achieve integrated sensing and communication(ISAC)based on a cell-free radio access network(CF-RAN)architecture with a minimum footprint of communication resources.We propose a new passive sensing scheme.The scheme is based on the radio frequency(RF)fingerprint learning of the RF radio unit(RRU)to build an RF fingerprint library of RRUs.The source RRU is identified by comparing the RF fingerprints carried by the signal at the receiver side.The receiver extracts the channel parameters from the signal and estimates the channel environment,thus locating the reflectors in the environment.The proposed scheme can effectively solve the problem of interference between signals in the same time-frequency domain but in different spatial domains when multiple RRUs jointly serve users in CF-RAN architecture.Simulation results show that the proposed passive ISAC scheme can effectively detect reflector location information in the environment without degrading the communication performance.展开更多
A leading position in the areas of testing,instrumentation and machine control has been established by members of the British Textile Machinery Association(BTMA)and a number of new developments in these fields will be...A leading position in the areas of testing,instrumentation and machine control has been established by members of the British Textile Machinery Association(BTMA)and a number of new developments in these fields will be showcased at this year’s ITMA Asia+CITME exhibition,which takes place in Singapore from October 28-31.“Many of our members are currently developing new technologies,either in-house or increasingly through joint projects,and there will be much to reveal by the time of ITMA Asia in Singapore,”says BTMA CEO Jason Kent.“Some of the most recent developments are really going beyond what has previously been possible.”展开更多
The integration of phytochemistry into forensic science has emerged as a groundbreaking frontier,providing unprecedented insights into nature's secrets through the precise application of phytochemical fingerprinti...The integration of phytochemistry into forensic science has emerged as a groundbreaking frontier,providing unprecedented insights into nature's secrets through the precise application of phytochemical fingerprinting of phytotoxins as a cutting-edge approach.This study explores the dynamic intersection of phytochemistry and forensic science,highlighting how the unique phytochemical profiles of toxic plants and their secondary metabolites,serve as distinctive markers for forensic investigations.By utilizing advanced techniques such as Ultra-High-Performance Liquid Chromatography(UHPLC)and High-Resolution Mass Spectrometry(HRMS),the detection and quantification of plant-derived are made more accurate in forensic contexts.Real-world case studies are presented to demonstrate the critical role of plant toxins in forensic outcomes and legal proceedings.The challenges,potential,and future prospects of integrating phytochemical fingerprinting of plant toxins into forensic science were discussed.This review aims to illuminate phytochemical fingerprinting of plant toxins as a promising tool to enhance the precision and depth of forensic analyses,offering new insights into the complex stories embedded in plant toxins.展开更多
With the rapid development of Internet of things technology,the efficiency of data transmission between devices has been significantly improved.However,the open network environment also poses serious security risks.Th...With the rapid development of Internet of things technology,the efficiency of data transmission between devices has been significantly improved.However,the open network environment also poses serious security risks.This paper proposes an innovative fingerprint template protection scheme,which generates key streams through an improved fourdimensional superchaotic system(4CSCS),uses the space-filling property of Hilbert curves to achieve pixel scrambling,and introduces dynamic DNA encoding to improve encryption.Experimental results show that this scheme has a large key space 2^(528),encrypts image information entropy of more than 7.9970,and shows excellent performance in defending against statistical attacks and differential attacks.Compared with existing methods,this scheme has significant advantages in terms of encryption performance and security,and provides a reliable protection mechanism for fingerprint authentication systems in the Internet of things environment.展开更多
Fingerprint features,as unique and stable biometric identifiers,are crucial for identity verification.However,traditional centralized methods of processing these sensitive data linked to personal identity pose signifi...Fingerprint features,as unique and stable biometric identifiers,are crucial for identity verification.However,traditional centralized methods of processing these sensitive data linked to personal identity pose significant privacy risks,potentially leading to user data leakage.Federated Learning allows multiple clients to collaboratively train and optimize models without sharing raw data,effectively addressing privacy and security concerns.However,variations in fingerprint data due to factors such as region,ethnicity,sensor quality,and environmental conditions result in significant heterogeneity across clients.This heterogeneity adversely impacts the generalization ability of the global model,limiting its performance across diverse distributions.To address these challenges,we propose an Adaptive Federated Fingerprint Recognition algorithm(AFFR)based on Federated Learning.The algorithm incorporates a generalization adjustment mechanism that evaluates the generalization gap between the local models and the global model,adaptively adjusting aggregation weights to mitigate the impact of heterogeneity caused by differences in data quality and feature characteristics.Additionally,a noise mechanism is embedded in client-side training to reduce the risk of fingerprint data leakage arising from weight disclosures during model updates.Experiments conducted on three public datasets demonstrate that AFFR significantly enhances model accuracy while ensuring robust privacy protection,showcasing its strong application potential and competitiveness in heterogeneous data environments.展开更多
Fraudulent website is an important car-rier tool for telecom fraud.At present,criminals can use artificial intelligence generative content technol-ogy to quickly generate fraudulent website templates and build fraudul...Fraudulent website is an important car-rier tool for telecom fraud.At present,criminals can use artificial intelligence generative content technol-ogy to quickly generate fraudulent website templates and build fraudulent websites in batches.Accurate identification of fraudulent website will effectively re-duce the risk of public victimization.Therefore,this study developed a fraudulent website template iden-tification method based on DOM structure extraction of website fingerprint features,which solves the prob-lems of single-dimension identification,low accuracy,and the insufficient generalization ability of current fraudulent website templates.This method uses an im-proved SimHash algorithm to traverse the DOM tree of a webpage,extract website node features,calcu-late the weight of each node,and obtain the finger-print feature vector of the website through dimension-ality reduction.Finally,the random forest algorithm is used to optimize the training features for the best combination of parameters.This method automati-cally extracts fingerprint features from websites and identifies website template ownership based on these features.An experimental analysis showed that this method achieves a classification accuracy of 89.8%and demonstrates superior recognition.展开更多
Objective:To evaluate the quality of Lygodium japonicum(Thunb.)Sw.(L.japonicum,Hai Jin Sha)by comparing its components without stewed(W)and stewed(S)using ultra-high-performance liquid chromatography(UHPLC)and chemome...Objective:To evaluate the quality of Lygodium japonicum(Thunb.)Sw.(L.japonicum,Hai Jin Sha)by comparing its components without stewed(W)and stewed(S)using ultra-high-performance liquid chromatography(UHPLC)and chemometric analysis.Additionally,network pharmacology was employed to investigate the possible mechanisms of action of L.japonicum in the urinary calculi(UC)treatment.Methods:A fingerprinting method was established to identify components through UHPLC-tandem mass spectrometry.Chemometric techniques were used to compare the L.japonicum extraction methods.Furthermore,various network pharmacological approaches were used to identify and analyze the potential targets of the identified components in relation to UC.Results:The W and S extracts were distributed into two distinct clusters.Significant differences in the levels of protocatechuic aldehyde,caffeic acid,and p-coumaric acid were observed between S and W.Network pharmacology analysis revealed that the primary targets of L.japonicum in the UC treatment were serum albumin and epidermal growth factor receptors,with potential active components including protocatechuic acid and caffeic acid.Conclusion:This study comprehensively examined the therapeutic components of L.japonicum before and after boiling,shedding light on its potential mechanisms of action in UC treatment.These findings offer valuable insights into the development and utilization of L.japonicum resources.展开更多
Website fingerprinting (WF) attacks can reveal information about the websites users browse by de-anonymizing encrypted traffic. Traditional website fingerprinting attack models, focusing solely on a single spatial fea...Website fingerprinting (WF) attacks can reveal information about the websites users browse by de-anonymizing encrypted traffic. Traditional website fingerprinting attack models, focusing solely on a single spatial feature, are inefficient regarding training time. When confronted with the concept drift problem, they suffer from a sharp drop in attack accuracy within a short period due to their reliance on extensive, outdated training data. To address the above problems, this paper proposes a parallel website fingerprinting attack (APWF) that incorporates an attention mechanism, which consists of an attack model and a fine-tuning method. Among them, the APWF model innovatively adopts a parallel structure, fusing temporal features related to both the front and back of the fingerprint sequence, along with spatial features captured through channel attention enhancement, to enhance the accuracy of the attack. Meanwhile, the APWF method introduces isomorphic migration learning and adjusts the model by freezing the optimal model weights and fine-tuning the parameters so that only a small number of the target, samples are needed to adapt to web page changes. A series of experiments show that the attack model can achieve 83% accuracy with the help of only 10 samples per category, which is a 30% improvement over the traditional attack model. Compared to comparative modeling, APWF improves accuracy while reducing time costs. After further fine-tuning the freezing model, the method in this paper can maintain the accuracy at 92.4% in the scenario of 56 days between the training data and the target data, which is only 4% less loss compared to the instant attack, significantly improving the robustness and accuracy of the model in coping with conceptual drift.展开更多
Various novel conjugated polymers(CPs)have been developed for organic photodetectors(OPDs),but their application to practical image sensors such as X-ray,R/G/B,and fingerprint sensors is rare.In this article,we report...Various novel conjugated polymers(CPs)have been developed for organic photodetectors(OPDs),but their application to practical image sensors such as X-ray,R/G/B,and fingerprint sensors is rare.In this article,we report the entire process from the synthesis and molecular engineering of novel CPs to the development of OPDs and fingerprint image sensors.We synthesized six benzo[1,2-d:4,5-d’]bis(oxazole)(BBO)-based CPs by modifying the alkyl side chains of the CPs.Several relationships between the molecular structure and the OPD performance were revealed,and increasing the number of linear octyl side chains on the conjugated backbone was the best way to improve Jph and reduce Jd in the OPDs.The optimized CP demonstrated promising OPD performance with a responsivity(R)of 0.22 A/W,specific detectivity(D^(*))of 1.05×10^(13)Jones at a bias of-1 V,rising/falling response time of 2.9/6.9μs,and cut-off frequency(f_(-3dB))of 134 kHz under collimated 530 nm LED irradiation.Finally,a fingerprint image sensor was fabricated by stacking the POTB1-based OPD layer on the organic thin-film transistors(318 ppi).The image contrast caused by the valleys and ridges in the fingerprints was obtained as a digital signal.展开更多
Aiming to ensure the consistency of quality control of Traditional Chinese Medicines(TCMs),a combination method of high-performance liquid chromatography(HPLC),ultraviolet(UV),electrochemical(EC)was developed in this ...Aiming to ensure the consistency of quality control of Traditional Chinese Medicines(TCMs),a combination method of high-performance liquid chromatography(HPLC),ultraviolet(UV),electrochemical(EC)was developed in this study to comprehensively evaluate the quality of Antiviral Mixture(AM),and Comprehensive Linear Quantification Fingerprint Method(CLQFM)was used to process the data.Quantitative analysis of three active substances in TCM was conducted.A fivewavelength fusion fingerprint(FWFF)was developed,using second-order derivatives of UV spectral data to differentiate sample levels effectively.The combination of HPLC and UV spectrophotometry,along with electrochemical fingerprinting(ECFP),successfully evaluated total active substances.Ultimately,a multidimensional profiling analytical system for TCM was developed.展开更多
The popularity of the Internet of Things(IoT)has enabled a large number of vulnerable devices to connect to the Internet,bringing huge security risks.As a network-level security authentication method,device fingerprin...The popularity of the Internet of Things(IoT)has enabled a large number of vulnerable devices to connect to the Internet,bringing huge security risks.As a network-level security authentication method,device fingerprint based on machine learning has attracted considerable attention because it can detect vulnerable devices in complex and heterogeneous access phases.However,flexible and diversified IoT devices with limited resources increase dif-ficulty of the device fingerprint authentication method executed in IoT,because it needs to retrain the model network to deal with incremental features or types.To address this problem,a device fingerprinting mechanism based on a Broad Learning System(BLS)is proposed in this paper.The mechanism firstly characterizes IoT devices by traffic analysis based on the identifiable differences of the traffic data of IoT devices,and extracts feature parameters of the traffic packets.A hierarchical hybrid sampling method is designed at the preprocessing phase to improve the imbalanced data distribution and reconstruct the fingerprint dataset.The complexity of the dataset is reduced using Principal Component Analysis(PCA)and the device type is identified by training weights using BLS.The experimental results show that the proposed method can achieve state-of-the-art accuracy and spend less training time than other existing methods.展开更多
With the rapid development of electric power systems,load estimation plays an important role in system operation and planning.Usually,load estimation techniques contain traditional,time series,regression analysis-base...With the rapid development of electric power systems,load estimation plays an important role in system operation and planning.Usually,load estimation techniques contain traditional,time series,regression analysis-based,and machine learning-based estimation.Since the machine learning-based method can lead to better performance,in this paper,a deep learning-based load estimation algorithm using image fingerprint and attention mechanism is proposed.First,an image fingerprint construction is proposed for training data.After the data preprocessing,the training data matrix is constructed by the cyclic shift and cubic spline interpolation.Then,the linear mapping and the gray-color transformation method are proposed to form the color image fingerprint.Second,a convolutional neural network(CNN)combined with an attentionmechanism is proposed for training performance improvement.At last,an experiment is carried out to evaluate the estimation performance.Compared with the support vector machine method,CNN method and long short-term memory method,the proposed algorithm has the best load estimation performance.展开更多
Objective To facilitate the quality evaluation suitable for the unique characteristics of Chinese materia medica(CMM)by developing and implementing a novel approach known as the matching frequency statistical moment(M...Objective To facilitate the quality evaluation suitable for the unique characteristics of Chinese materia medica(CMM)by developing and implementing a novel approach known as the matching frequency statistical moment(MFSM)method.Methods This study established the MFSM method.To demonstrate its effectiveness,we applied this novel approach to analyze Danxi Granules(丹膝颗粒,DXG)and its constituent herbal materials.To begin with,the ultra-performance liquid chromatography(UPLC)was applied to obtain the chromatographic fingerprints of DXG and its constituent herbal materi-als.Next,the MFSM was leveraged to compress and integrate them into a new fingerprint with fewer analytical units.Then,we characterized the properties and variability of both the original and integrated fingerprints by calculating total quantum statistical moment(TQSM)parameters,information entropy and information amount,along with their relative standard deviation(RSD).Finally,we compared the TQSM parameters,information entropy and infor-mation amount,and their RSD between the traditional and novel fingerprints to validate the new analytical method.Results The chromatographic peaks of DXG and its 12 raw herbal materials were divided and integrated into peak families by the MFSM method.Before integration,the ranges of the peak number,three TQSM parameters,information entropy and information amount for each peak or peak family of UPLC fingerprints of DXG and its 12 raw herbal materials were 95.07−209.73,9390−183064μv·s,5.928−21.33 min,22.62−106.69 min^(2),4.230−6.539,and 50530−974186μv·s,respectively.After integration,the ranges of these parameters were 10.00−88.00,9390−183064μv·s,5.951−22.02 min,22.27−104.73 min^(2),2.223−5.277,and 38159−807200μv·s,respectively.Correspondingly,the RSD of all the aforementioned pa-rameters before integration were 2.12%−9.15%,6.04%−49.78%,1.15%−23.10%,3.97%−25.79%,1.49%−19.86%,and 6.64%−51.20%,respectively.However,after integration,they changed to 0.00%,6.04%−49.87%,1.73%−23.02%,3.84%−26.85%,1.17%−16.54%,and 6.40%−48.59%,respectively.The results demonstrated that in the newly integrated fingerprint,the analytical units of constituent herbal materials,information entropy and information amount were significantly reduced(P<0.05),while the TQSM parameters remained unchanged(P>0.05).Additionally,the RSD of the TQSM parameters,information entropy,and information amount didn’t show significant difference before and after integration(P>0.05),but the RSD of the number and area of the integrated analytical units significantly decreased(P<0.05).Conclusion The MFSM method could reduce the analytical units of constituent herbal mate-rials while maintain the properties and variability from their original fingerprint.Thus,it could serve as a feasible and reliable tool to reduce difficulties in analyzing multi-compo-nents within CMMs and facilitating the evaluation of their quality.展开更多
Winter jujube(Ziziphus jujuba'Dongzao')is greatly appreciated by consumers for its excellent quality,but brand infringement frequently occurs in the market.Here,we first determined a total of 38 elements in 16...Winter jujube(Ziziphus jujuba'Dongzao')is greatly appreciated by consumers for its excellent quality,but brand infringement frequently occurs in the market.Here,we first determined a total of 38 elements in 167 winter jujube samples from the main winter jujube producing areas of China by inductively coupled plasma mass spectrometer(ICP-MS).As a result,16 elements(Mg,K,Mn,Cu,Zn,Mo,Ba,Be,As,Se,Cd,Sb,Ce,Er,Tl,and Pb)exhibited significant differences in samples from different producing areas.Supervised linear discriminant analysis(LDA)and orthogonal projection to latent structures discriminant analysis(OPLS-DA)showed better performance in identifying the origin of samples than unsupervised principal component analysis(PCA).LDA and OPLS-DA had a mean identification accuracy of 87.84 and 94.64%in the testing set,respectively.By using the multilayer perceptron(MLP)and C5.0,the prediction accuracy of the models could reach 96.36 and 91.06%,respectively.Based on the above four chemometric methods,Cd,Tl,Mo and Se were selected as the main variables and principal markers for the origin identification of winter jujube.Overall,this study demonstrates that it is practical and precise to identify the origin of winter jujube through multi-element fingerprint analysis with chemometrics,and may also provide reference for establishing the origin traceability system of other fruits.展开更多
The static and predictable characteristics of cyber systems give attackers an asymmetric advantage in gathering useful information and launching attacks.To reverse this asymmetric advantage,a new defense idea,called M...The static and predictable characteristics of cyber systems give attackers an asymmetric advantage in gathering useful information and launching attacks.To reverse this asymmetric advantage,a new defense idea,called Moving Target Defense(MTD),has been proposed to provide additional selectable measures to complement traditional defense.However,MTD is unable to defeat the sophisticated attacker with fingerprint tracking ability.To overcome this limitation,we go one step beyond and show that the combination of MTD and Deception-based Cyber Defense(DCD)can achieve higher performance than either of them.In particular,we first introduce and formalize a novel attacker model named Scan and Foothold Attack(SFA)based on cyber kill chain.Afterwards,we develop probabilistic models for SFA defenses to provide a deeper analysis of the theoretical effect under different defense strategies.These models quantify attack success probability and the probability that the attacker will be deceived under various conditions,such as the size of address space,and the number of hosts,attack analysis time.Finally,the experimental results show that the actual defense effect of each strategy almost perfectly follows its probabilistic model.Also,the defense strategy of combining address mutation and fingerprint camouflage can achieve a better defense effect than the single address mutation.展开更多
Website fingerprinting,also known asWF,is a traffic analysis attack that enables local eavesdroppers to infer a user’s browsing destination,even when using the Tor anonymity network.While advanced attacks based on de...Website fingerprinting,also known asWF,is a traffic analysis attack that enables local eavesdroppers to infer a user’s browsing destination,even when using the Tor anonymity network.While advanced attacks based on deep neural network(DNN)can performfeature engineering and attain accuracy rates of over 98%,research has demonstrated thatDNNis vulnerable to adversarial samples.As a result,many researchers have explored using adversarial samples as a defense mechanism against DNN-based WF attacks and have achieved considerable success.However,these methods suffer from high bandwidth overhead or require access to the target model,which is unrealistic.This paper proposes CMAES-WFD,a black-box WF defense based on adversarial samples.The process of generating adversarial examples is transformed into a constrained optimization problem solved by utilizing the Covariance Matrix Adaptation Evolution Strategy(CMAES)optimization algorithm.Perturbations are injected into the local parts of the original traffic to control bandwidth overhead.According to the experiment results,CMAES-WFD was able to significantly decrease the accuracy of Deep Fingerprinting(DF)and VarCnn to below 8.3%and the bandwidth overhead to a maximum of only 14.6%and 20.5%,respectively.Specially,for Automated Website Fingerprinting(AWF)with simple structure,CMAES-WFD reduced the classification accuracy to only 6.7%and the bandwidth overhead to less than 7.4%.Moreover,it was demonstrated that CMAES-WFD was robust against adversarial training to a certain extent.展开更多
With the rapid development of smart phone,the location-based services(LBS)have received great attention in the past decades.Owing to the widespread use of WiFi and Bluetooth devices,Received Signal Strength Indication...With the rapid development of smart phone,the location-based services(LBS)have received great attention in the past decades.Owing to the widespread use of WiFi and Bluetooth devices,Received Signal Strength Indication(RSSI)fingerprintbased localization method has obtained much development in both academia and industries.In this work,we introduce an efficient way to reduce the labor-intensive site survey process,which uses an UWB/IMU-assisted fingerprint construction(UAFC)and localization framework based on the principle of Automatic radio map generation scheme(ARMGS)is proposed to replace the traditional manual measurement.To be specific,UWB devices are employed to estimate the coordinates when the collector is moved in a reference point(RP).An anchor self-localization method is investigated to further reduce manual measurement work in a wide and complex environment,which is also a grueling,time-consuming process that is lead to artificial errors.Moreover,the measurements of IMU are incorporated into the UWB localization algorithm and improve the label accuracy in fingerprint.In addition,the weighted k-nearest neighbor(WKNN)algorithm is applied to online localization phase.Finally,filed experiments are carried out and the results confirm the effectiveness of the proposed approach.展开更多
Web application fingerprint recognition is an effective security technology designed to identify and classify web applications,thereby enhancing the detection of potential threats and attacks.Traditional fingerprint r...Web application fingerprint recognition is an effective security technology designed to identify and classify web applications,thereby enhancing the detection of potential threats and attacks.Traditional fingerprint recognition methods,which rely on preannotated feature matching,face inherent limitations due to the ever-evolving nature and diverse landscape of web applications.In response to these challenges,this work proposes an innovative web application fingerprint recognition method founded on clustering techniques.The method involves extensive data collection from the Tranco List,employing adjusted feature selection built upon Wappalyzer and noise reduction through truncated SVD dimensionality reduction.The core of the methodology lies in the application of the unsupervised OPTICS clustering algorithm,eliminating the need for preannotated labels.By transforming web applications into feature vectors and leveraging clustering algorithms,our approach accurately categorizes diverse web applications,providing comprehensive and precise fingerprint recognition.The experimental results,which are obtained on a dataset featuring various web application types,affirm the efficacy of the method,demonstrating its ability to achieve high accuracy and broad coverage.This novel approach not only distinguishes between different web application types effectively but also demonstrates superiority in terms of classification accuracy and coverage,offering a robust solution to the challenges of web application fingerprint recognition.展开更多
While being successful in the detection and attribution of climate change,the optimal fingerprinting method(OFM)may have some limitations from a physics-and-dynamics-based viewpoint.Here,an analysis is made on the lin...While being successful in the detection and attribution of climate change,the optimal fingerprinting method(OFM)may have some limitations from a physics-and-dynamics-based viewpoint.Here,an analysis is made on the linearity,noninteraction,and stationary-variability assumptions adopted by OFM.It is suggested that furthering OFM needs a viewpoint beyond statistical science,and the method should be combined with theoretical tools in the dynamics and physics of the Earth system,so as to be applied for the detection and attribution of nonlinear climate change including tipping elements within the Earth system.展开更多
基金supported by Macao Science and Technology Development Fund,Macao SAR,China(Grant No.:0043/2023/AFJ)the National Natural Science Foundation of China(Grant No.:22173038)Macao Polytechnic University,Macao SAR,China(Grant No.:RP/FCA-01/2022).
文摘Accurate prediction of molecular properties is crucial for selecting compounds with ideal properties and reducing the costs and risks of trials.Traditional methods based on manually crafted features and graph-based methods have shown promising results in molecular property prediction.However,traditional methods rely on expert knowledge and often fail to capture the complex structures and interactions within molecules.Similarly,graph-based methods typically overlook the chemical structure and function hidden in molecular motifs and struggle to effectively integrate global and local molecular information.To address these limitations,we propose a novel fingerprint-enhanced hierarchical graph neural network(FH-GNN)for molecular property prediction that simultaneously learns information from hierarchical molecular graphs and fingerprints.The FH-GNN captures diverse hierarchical chemical information by applying directed message-passing neural networks(D-MPNN)on a hierarchical molecular graph that integrates atomic-level,motif-level,and graph-level information along with their relationships.Addi-tionally,we used an adaptive attention mechanism to balance the importance of hierarchical graphs and fingerprint features,creating a comprehensive molecular embedding that integrated hierarchical mo-lecular structures with domain knowledge.Experiments on eight benchmark datasets from MoleculeNet showed that FH-GNN outperformed the baseline models in both classification and regression tasks for molecular property prediction,validating its capability to comprehensively capture molecular informa-tion.By integrating molecular structure and chemical knowledge,FH-GNN provides a powerful tool for the accurate prediction of molecular properties and aids in the discovery of potential drug candidates.
基金supported in part by the National Key Research and Development Program under Grant(2021YFB2900300)by the National Natural Science Foundation of China(NSFC)under Grants 61971127,61871122by the Southeast University-China Mobile Research Institute Joint Innovation Center,and by the Major Key Project of PCL(PCL2021A01-2).
文摘This paper investigates how to achieve integrated sensing and communication(ISAC)based on a cell-free radio access network(CF-RAN)architecture with a minimum footprint of communication resources.We propose a new passive sensing scheme.The scheme is based on the radio frequency(RF)fingerprint learning of the RF radio unit(RRU)to build an RF fingerprint library of RRUs.The source RRU is identified by comparing the RF fingerprints carried by the signal at the receiver side.The receiver extracts the channel parameters from the signal and estimates the channel environment,thus locating the reflectors in the environment.The proposed scheme can effectively solve the problem of interference between signals in the same time-frequency domain but in different spatial domains when multiple RRUs jointly serve users in CF-RAN architecture.Simulation results show that the proposed passive ISAC scheme can effectively detect reflector location information in the environment without degrading the communication performance.
文摘A leading position in the areas of testing,instrumentation and machine control has been established by members of the British Textile Machinery Association(BTMA)and a number of new developments in these fields will be showcased at this year’s ITMA Asia+CITME exhibition,which takes place in Singapore from October 28-31.“Many of our members are currently developing new technologies,either in-house or increasingly through joint projects,and there will be much to reveal by the time of ITMA Asia in Singapore,”says BTMA CEO Jason Kent.“Some of the most recent developments are really going beyond what has previously been possible.”
文摘The integration of phytochemistry into forensic science has emerged as a groundbreaking frontier,providing unprecedented insights into nature's secrets through the precise application of phytochemical fingerprinting of phytotoxins as a cutting-edge approach.This study explores the dynamic intersection of phytochemistry and forensic science,highlighting how the unique phytochemical profiles of toxic plants and their secondary metabolites,serve as distinctive markers for forensic investigations.By utilizing advanced techniques such as Ultra-High-Performance Liquid Chromatography(UHPLC)and High-Resolution Mass Spectrometry(HRMS),the detection and quantification of plant-derived are made more accurate in forensic contexts.Real-world case studies are presented to demonstrate the critical role of plant toxins in forensic outcomes and legal proceedings.The challenges,potential,and future prospects of integrating phytochemical fingerprinting of plant toxins into forensic science were discussed.This review aims to illuminate phytochemical fingerprinting of plant toxins as a promising tool to enhance the precision and depth of forensic analyses,offering new insights into the complex stories embedded in plant toxins.
文摘With the rapid development of Internet of things technology,the efficiency of data transmission between devices has been significantly improved.However,the open network environment also poses serious security risks.This paper proposes an innovative fingerprint template protection scheme,which generates key streams through an improved fourdimensional superchaotic system(4CSCS),uses the space-filling property of Hilbert curves to achieve pixel scrambling,and introduces dynamic DNA encoding to improve encryption.Experimental results show that this scheme has a large key space 2^(528),encrypts image information entropy of more than 7.9970,and shows excellent performance in defending against statistical attacks and differential attacks.Compared with existing methods,this scheme has significant advantages in terms of encryption performance and security,and provides a reliable protection mechanism for fingerprint authentication systems in the Internet of things environment.
基金supported by the National Natural Science Foundation of China(Nos.62002100,61902237)Key Research and Promotion Projects of Henan Province(Nos.232102240023,232102210063,222102210040).
文摘Fingerprint features,as unique and stable biometric identifiers,are crucial for identity verification.However,traditional centralized methods of processing these sensitive data linked to personal identity pose significant privacy risks,potentially leading to user data leakage.Federated Learning allows multiple clients to collaboratively train and optimize models without sharing raw data,effectively addressing privacy and security concerns.However,variations in fingerprint data due to factors such as region,ethnicity,sensor quality,and environmental conditions result in significant heterogeneity across clients.This heterogeneity adversely impacts the generalization ability of the global model,limiting its performance across diverse distributions.To address these challenges,we propose an Adaptive Federated Fingerprint Recognition algorithm(AFFR)based on Federated Learning.The algorithm incorporates a generalization adjustment mechanism that evaluates the generalization gap between the local models and the global model,adaptively adjusting aggregation weights to mitigate the impact of heterogeneity caused by differences in data quality and feature characteristics.Additionally,a noise mechanism is embedded in client-side training to reduce the risk of fingerprint data leakage arising from weight disclosures during model updates.Experiments conducted on three public datasets demonstrate that AFFR significantly enhances model accuracy while ensuring robust privacy protection,showcasing its strong application potential and competitiveness in heterogeneous data environments.
基金This research is a phased achievement of The National Social Science Fund of China(23BGL272).
文摘Fraudulent website is an important car-rier tool for telecom fraud.At present,criminals can use artificial intelligence generative content technol-ogy to quickly generate fraudulent website templates and build fraudulent websites in batches.Accurate identification of fraudulent website will effectively re-duce the risk of public victimization.Therefore,this study developed a fraudulent website template iden-tification method based on DOM structure extraction of website fingerprint features,which solves the prob-lems of single-dimension identification,low accuracy,and the insufficient generalization ability of current fraudulent website templates.This method uses an im-proved SimHash algorithm to traverse the DOM tree of a webpage,extract website node features,calcu-late the weight of each node,and obtain the finger-print feature vector of the website through dimension-ality reduction.Finally,the random forest algorithm is used to optimize the training features for the best combination of parameters.This method automati-cally extracts fingerprint features from websites and identifies website template ownership based on these features.An experimental analysis showed that this method achieves a classification accuracy of 89.8%and demonstrates superior recognition.
基金supported by Ministry of Industry and Information Technology of the People's Republic of China 2022 Industrial Technology Basic Public Service Platform Project-Traditional Chinese Medicine Whole Industry Chain Quality and Technology Service Platform(2022-230-221)Foshan Nanhai District Key Area Science and Technology Research Project[Nanke(2023)20-18].
文摘Objective:To evaluate the quality of Lygodium japonicum(Thunb.)Sw.(L.japonicum,Hai Jin Sha)by comparing its components without stewed(W)and stewed(S)using ultra-high-performance liquid chromatography(UHPLC)and chemometric analysis.Additionally,network pharmacology was employed to investigate the possible mechanisms of action of L.japonicum in the urinary calculi(UC)treatment.Methods:A fingerprinting method was established to identify components through UHPLC-tandem mass spectrometry.Chemometric techniques were used to compare the L.japonicum extraction methods.Furthermore,various network pharmacological approaches were used to identify and analyze the potential targets of the identified components in relation to UC.Results:The W and S extracts were distributed into two distinct clusters.Significant differences in the levels of protocatechuic aldehyde,caffeic acid,and p-coumaric acid were observed between S and W.Network pharmacology analysis revealed that the primary targets of L.japonicum in the UC treatment were serum albumin and epidermal growth factor receptors,with potential active components including protocatechuic acid and caffeic acid.Conclusion:This study comprehensively examined the therapeutic components of L.japonicum before and after boiling,shedding light on its potential mechanisms of action in UC treatment.These findings offer valuable insights into the development and utilization of L.japonicum resources.
基金supported by the National Defense Basic Scientific Research Program of China(No.JCKY2023602C026)the funding of Key Laboratory of Mobile Application Innovation and Governance Technology,Ministry of Industry and Information Technology(2023IFS080601-K).
文摘Website fingerprinting (WF) attacks can reveal information about the websites users browse by de-anonymizing encrypted traffic. Traditional website fingerprinting attack models, focusing solely on a single spatial feature, are inefficient regarding training time. When confronted with the concept drift problem, they suffer from a sharp drop in attack accuracy within a short period due to their reliance on extensive, outdated training data. To address the above problems, this paper proposes a parallel website fingerprinting attack (APWF) that incorporates an attention mechanism, which consists of an attack model and a fine-tuning method. Among them, the APWF model innovatively adopts a parallel structure, fusing temporal features related to both the front and back of the fingerprint sequence, along with spatial features captured through channel attention enhancement, to enhance the accuracy of the attack. Meanwhile, the APWF method introduces isomorphic migration learning and adjusts the model by freezing the optimal model weights and fine-tuning the parameters so that only a small number of the target, samples are needed to adapt to web page changes. A series of experiments show that the attack model can achieve 83% accuracy with the help of only 10 samples per category, which is a 30% improvement over the traditional attack model. Compared to comparative modeling, APWF improves accuracy while reducing time costs. After further fine-tuning the freezing model, the method in this paper can maintain the accuracy at 92.4% in the scenario of 56 days between the training data and the target data, which is only 4% less loss compared to the instant attack, significantly improving the robustness and accuracy of the model in coping with conceptual drift.
基金funded by the National Research Foundation(NRF)of Korea(2020M3H4A3081816,RS-2023-00304936,and RS-2024-00398065).
文摘Various novel conjugated polymers(CPs)have been developed for organic photodetectors(OPDs),but their application to practical image sensors such as X-ray,R/G/B,and fingerprint sensors is rare.In this article,we report the entire process from the synthesis and molecular engineering of novel CPs to the development of OPDs and fingerprint image sensors.We synthesized six benzo[1,2-d:4,5-d’]bis(oxazole)(BBO)-based CPs by modifying the alkyl side chains of the CPs.Several relationships between the molecular structure and the OPD performance were revealed,and increasing the number of linear octyl side chains on the conjugated backbone was the best way to improve Jph and reduce Jd in the OPDs.The optimized CP demonstrated promising OPD performance with a responsivity(R)of 0.22 A/W,specific detectivity(D^(*))of 1.05×10^(13)Jones at a bias of-1 V,rising/falling response time of 2.9/6.9μs,and cut-off frequency(f_(-3dB))of 134 kHz under collimated 530 nm LED irradiation.Finally,a fingerprint image sensor was fabricated by stacking the POTB1-based OPD layer on the organic thin-film transistors(318 ppi).The image contrast caused by the valleys and ridges in the fingerprints was obtained as a digital signal.
基金This study was supported by the National Natural Science Foundation of China(No.81573586).
文摘Aiming to ensure the consistency of quality control of Traditional Chinese Medicines(TCMs),a combination method of high-performance liquid chromatography(HPLC),ultraviolet(UV),electrochemical(EC)was developed in this study to comprehensively evaluate the quality of Antiviral Mixture(AM),and Comprehensive Linear Quantification Fingerprint Method(CLQFM)was used to process the data.Quantitative analysis of three active substances in TCM was conducted.A fivewavelength fusion fingerprint(FWFF)was developed,using second-order derivatives of UV spectral data to differentiate sample levels effectively.The combination of HPLC and UV spectrophotometry,along with electrochemical fingerprinting(ECFP),successfully evaluated total active substances.Ultimately,a multidimensional profiling analytical system for TCM was developed.
基金supported by National Key R&D Program of China(2019YFB2102303)National Natural Science Foundation of China(NSFC61971014,NSFC11675199)Young Backbone Teacher Training Program of Henan Colleges and Universities(2021GGJS170).
文摘The popularity of the Internet of Things(IoT)has enabled a large number of vulnerable devices to connect to the Internet,bringing huge security risks.As a network-level security authentication method,device fingerprint based on machine learning has attracted considerable attention because it can detect vulnerable devices in complex and heterogeneous access phases.However,flexible and diversified IoT devices with limited resources increase dif-ficulty of the device fingerprint authentication method executed in IoT,because it needs to retrain the model network to deal with incremental features or types.To address this problem,a device fingerprinting mechanism based on a Broad Learning System(BLS)is proposed in this paper.The mechanism firstly characterizes IoT devices by traffic analysis based on the identifiable differences of the traffic data of IoT devices,and extracts feature parameters of the traffic packets.A hierarchical hybrid sampling method is designed at the preprocessing phase to improve the imbalanced data distribution and reconstruct the fingerprint dataset.The complexity of the dataset is reduced using Principal Component Analysis(PCA)and the device type is identified by training weights using BLS.The experimental results show that the proposed method can achieve state-of-the-art accuracy and spend less training time than other existing methods.
文摘With the rapid development of electric power systems,load estimation plays an important role in system operation and planning.Usually,load estimation techniques contain traditional,time series,regression analysis-based,and machine learning-based estimation.Since the machine learning-based method can lead to better performance,in this paper,a deep learning-based load estimation algorithm using image fingerprint and attention mechanism is proposed.First,an image fingerprint construction is proposed for training data.After the data preprocessing,the training data matrix is constructed by the cyclic shift and cubic spline interpolation.Then,the linear mapping and the gray-color transformation method are proposed to form the color image fingerprint.Second,a convolutional neural network(CNN)combined with an attentionmechanism is proposed for training performance improvement.At last,an experiment is carried out to evaluate the estimation performance.Compared with the support vector machine method,CNN method and long short-term memory method,the proposed algorithm has the best load estimation performance.
基金Natural Science Foundation of Hunan province(2022JJ30453 and 2024JJ6362)the Key Research and Development Program of Hunan Province(2022SK2014).
文摘Objective To facilitate the quality evaluation suitable for the unique characteristics of Chinese materia medica(CMM)by developing and implementing a novel approach known as the matching frequency statistical moment(MFSM)method.Methods This study established the MFSM method.To demonstrate its effectiveness,we applied this novel approach to analyze Danxi Granules(丹膝颗粒,DXG)and its constituent herbal materials.To begin with,the ultra-performance liquid chromatography(UPLC)was applied to obtain the chromatographic fingerprints of DXG and its constituent herbal materi-als.Next,the MFSM was leveraged to compress and integrate them into a new fingerprint with fewer analytical units.Then,we characterized the properties and variability of both the original and integrated fingerprints by calculating total quantum statistical moment(TQSM)parameters,information entropy and information amount,along with their relative standard deviation(RSD).Finally,we compared the TQSM parameters,information entropy and infor-mation amount,and their RSD between the traditional and novel fingerprints to validate the new analytical method.Results The chromatographic peaks of DXG and its 12 raw herbal materials were divided and integrated into peak families by the MFSM method.Before integration,the ranges of the peak number,three TQSM parameters,information entropy and information amount for each peak or peak family of UPLC fingerprints of DXG and its 12 raw herbal materials were 95.07−209.73,9390−183064μv·s,5.928−21.33 min,22.62−106.69 min^(2),4.230−6.539,and 50530−974186μv·s,respectively.After integration,the ranges of these parameters were 10.00−88.00,9390−183064μv·s,5.951−22.02 min,22.27−104.73 min^(2),2.223−5.277,and 38159−807200μv·s,respectively.Correspondingly,the RSD of all the aforementioned pa-rameters before integration were 2.12%−9.15%,6.04%−49.78%,1.15%−23.10%,3.97%−25.79%,1.49%−19.86%,and 6.64%−51.20%,respectively.However,after integration,they changed to 0.00%,6.04%−49.87%,1.73%−23.02%,3.84%−26.85%,1.17%−16.54%,and 6.40%−48.59%,respectively.The results demonstrated that in the newly integrated fingerprint,the analytical units of constituent herbal materials,information entropy and information amount were significantly reduced(P<0.05),while the TQSM parameters remained unchanged(P>0.05).Additionally,the RSD of the TQSM parameters,information entropy,and information amount didn’t show significant difference before and after integration(P>0.05),but the RSD of the number and area of the integrated analytical units significantly decreased(P<0.05).Conclusion The MFSM method could reduce the analytical units of constituent herbal mate-rials while maintain the properties and variability from their original fingerprint.Thus,it could serve as a feasible and reliable tool to reduce difficulties in analyzing multi-compo-nents within CMMs and facilitating the evaluation of their quality.
基金This work was supported by the Scientific Research Foundation for High Level Talents of Qingdao Agricultural University,China(665-1120015)the National Program for Quality and Safety Risk Assessment of Agricultural Products of China(GJFP2019011)the National Natural Science Foundation of China(42207017).
文摘Winter jujube(Ziziphus jujuba'Dongzao')is greatly appreciated by consumers for its excellent quality,but brand infringement frequently occurs in the market.Here,we first determined a total of 38 elements in 167 winter jujube samples from the main winter jujube producing areas of China by inductively coupled plasma mass spectrometer(ICP-MS).As a result,16 elements(Mg,K,Mn,Cu,Zn,Mo,Ba,Be,As,Se,Cd,Sb,Ce,Er,Tl,and Pb)exhibited significant differences in samples from different producing areas.Supervised linear discriminant analysis(LDA)and orthogonal projection to latent structures discriminant analysis(OPLS-DA)showed better performance in identifying the origin of samples than unsupervised principal component analysis(PCA).LDA and OPLS-DA had a mean identification accuracy of 87.84 and 94.64%in the testing set,respectively.By using the multilayer perceptron(MLP)and C5.0,the prediction accuracy of the models could reach 96.36 and 91.06%,respectively.Based on the above four chemometric methods,Cd,Tl,Mo and Se were selected as the main variables and principal markers for the origin identification of winter jujube.Overall,this study demonstrates that it is practical and precise to identify the origin of winter jujube through multi-element fingerprint analysis with chemometrics,and may also provide reference for establishing the origin traceability system of other fruits.
基金supported by the National Key Research and Development Program of China(No.2016YFB0800601)the Key Program of NSFC-Tongyong Union Foundation(No.U1636209)+1 种基金the National Natural Science Foundation of China(61602358)the Key Research and Development Programs of Shaanxi(No.2019ZDLGY13-04,No.2019ZDLGY13-07)。
文摘The static and predictable characteristics of cyber systems give attackers an asymmetric advantage in gathering useful information and launching attacks.To reverse this asymmetric advantage,a new defense idea,called Moving Target Defense(MTD),has been proposed to provide additional selectable measures to complement traditional defense.However,MTD is unable to defeat the sophisticated attacker with fingerprint tracking ability.To overcome this limitation,we go one step beyond and show that the combination of MTD and Deception-based Cyber Defense(DCD)can achieve higher performance than either of them.In particular,we first introduce and formalize a novel attacker model named Scan and Foothold Attack(SFA)based on cyber kill chain.Afterwards,we develop probabilistic models for SFA defenses to provide a deeper analysis of the theoretical effect under different defense strategies.These models quantify attack success probability and the probability that the attacker will be deceived under various conditions,such as the size of address space,and the number of hosts,attack analysis time.Finally,the experimental results show that the actual defense effect of each strategy almost perfectly follows its probabilistic model.Also,the defense strategy of combining address mutation and fingerprint camouflage can achieve a better defense effect than the single address mutation.
基金the Key JCJQ Program of China:2020-JCJQ-ZD-021-00 and 2020-JCJQ-ZD-024-12.
文摘Website fingerprinting,also known asWF,is a traffic analysis attack that enables local eavesdroppers to infer a user’s browsing destination,even when using the Tor anonymity network.While advanced attacks based on deep neural network(DNN)can performfeature engineering and attain accuracy rates of over 98%,research has demonstrated thatDNNis vulnerable to adversarial samples.As a result,many researchers have explored using adversarial samples as a defense mechanism against DNN-based WF attacks and have achieved considerable success.However,these methods suffer from high bandwidth overhead or require access to the target model,which is unrealistic.This paper proposes CMAES-WFD,a black-box WF defense based on adversarial samples.The process of generating adversarial examples is transformed into a constrained optimization problem solved by utilizing the Covariance Matrix Adaptation Evolution Strategy(CMAES)optimization algorithm.Perturbations are injected into the local parts of the original traffic to control bandwidth overhead.According to the experiment results,CMAES-WFD was able to significantly decrease the accuracy of Deep Fingerprinting(DF)and VarCnn to below 8.3%and the bandwidth overhead to a maximum of only 14.6%and 20.5%,respectively.Specially,for Automated Website Fingerprinting(AWF)with simple structure,CMAES-WFD reduced the classification accuracy to only 6.7%and the bandwidth overhead to less than 7.4%.Moreover,it was demonstrated that CMAES-WFD was robust against adversarial training to a certain extent.
文摘With the rapid development of smart phone,the location-based services(LBS)have received great attention in the past decades.Owing to the widespread use of WiFi and Bluetooth devices,Received Signal Strength Indication(RSSI)fingerprintbased localization method has obtained much development in both academia and industries.In this work,we introduce an efficient way to reduce the labor-intensive site survey process,which uses an UWB/IMU-assisted fingerprint construction(UAFC)and localization framework based on the principle of Automatic radio map generation scheme(ARMGS)is proposed to replace the traditional manual measurement.To be specific,UWB devices are employed to estimate the coordinates when the collector is moved in a reference point(RP).An anchor self-localization method is investigated to further reduce manual measurement work in a wide and complex environment,which is also a grueling,time-consuming process that is lead to artificial errors.Moreover,the measurements of IMU are incorporated into the UWB localization algorithm and improve the label accuracy in fingerprint.In addition,the weighted k-nearest neighbor(WKNN)algorithm is applied to online localization phase.Finally,filed experiments are carried out and the results confirm the effectiveness of the proposed approach.
基金supported in part by the National Science Foundation of China under Grants U22B2027,62172297,62102262,61902276 and 62272311,Tianjin Intelligent Manufacturing Special Fund Project under Grant 20211097the China Guangxi Science and Technology Plan Project(Guangxi Science and Technology Base and Talent Special Project)under Grant AD23026096(Application Number 2022AC20001)+1 种基金Hainan Provincial Natural Science Foundation of China under Grant 622RC616CCF-Nsfocus Kunpeng Fund Project under Grant CCF-NSFOCUS202207.
文摘Web application fingerprint recognition is an effective security technology designed to identify and classify web applications,thereby enhancing the detection of potential threats and attacks.Traditional fingerprint recognition methods,which rely on preannotated feature matching,face inherent limitations due to the ever-evolving nature and diverse landscape of web applications.In response to these challenges,this work proposes an innovative web application fingerprint recognition method founded on clustering techniques.The method involves extensive data collection from the Tranco List,employing adjusted feature selection built upon Wappalyzer and noise reduction through truncated SVD dimensionality reduction.The core of the methodology lies in the application of the unsupervised OPTICS clustering algorithm,eliminating the need for preannotated labels.By transforming web applications into feature vectors and leveraging clustering algorithms,our approach accurately categorizes diverse web applications,providing comprehensive and precise fingerprint recognition.The experimental results,which are obtained on a dataset featuring various web application types,affirm the efficacy of the method,demonstrating its ability to achieve high accuracy and broad coverage.This novel approach not only distinguishes between different web application types effectively but also demonstrates superiority in terms of classification accuracy and coverage,offering a robust solution to the challenges of web application fingerprint recognition.
基金support from the National Natural Science Foundation of China(Grant No.42175070)。
文摘While being successful in the detection and attribution of climate change,the optimal fingerprinting method(OFM)may have some limitations from a physics-and-dynamics-based viewpoint.Here,an analysis is made on the linearity,noninteraction,and stationary-variability assumptions adopted by OFM.It is suggested that furthering OFM needs a viewpoint beyond statistical science,and the method should be combined with theoretical tools in the dynamics and physics of the Earth system,so as to be applied for the detection and attribution of nonlinear climate change including tipping elements within the Earth system.
