Top Read Articles

    Published in last 1 year |  In last 2 years |  In last 3 years |  All
    Please wait a minute...
    For Selected: Toggle Thumbnails
    Sea-surface multi-target tracking method aided by target returns features
    ZHANG Yichen,SHUI Penglang,LIAO Mo
    Journal of Xidian University    2023, 50 (5): 1-10.   DOI: 10.19665/j.issn1001-2400.20230201
    Abstract449)   HTML96)    PDF(pc) (3631KB)(346)       Save

    Due to the complex marine environment and the dense sea-surface targets,radars often face the tough tracking scenarios with a high false alarm rate and high target density.The measurement points originating from clutter and multiple closely-spaced targets appear densely in the detection space.The traditional tracking methods only use the position information,which cannot distinguish the specific source of the measurement well,resulting in serious degradation of the tracking performance.Target returns features can be used to solve the problem without increasing the complexity of the algorithm,but the generalization ability of the features is low.It is necessary to select suitable features according to different radar systems,working scenes and requirements.In this paper,the test statistic and the target radial velocity measurement are used as the target returns features,and the tracking equation is reconstructed so that features can be fully applied in all aspects of tracking.In addition,this paper adopts a "two-level" tracking process,which divides track and candidate track according to track quality.Experimental results show that the proposed method can achieve robust target tracking in the complex multi-target scenarios on the sea surface.

    Table and Figures | Reference | Related Articles | Metrics
    Construction method of temporal correlation graph convolution network for traffic prediction
    ZHANG Kehan,LI Hongyan,LIU Wenhui,WANG Peng
    Journal of Xidian University    2023, 50 (5): 11-20.   DOI: 10.19665/j.issn1001-2400.20221103
    Abstract227)   HTML27)    PDF(pc) (4522KB)(186)       Save

    The existing traffic prediction methods in the virtual network of data centers characterize the correlation between links with difficulty,which leads to the difficulty in improving the accuracy of traffic prediction.Based on this,this paper proposes a Temporal Correlation Graph Convolutional neural Network (TC-GCN),which enables the representation of Temporal and spatial Correlation of the data center Network link traffic and improves the accuracy of traffic prediction.First,the graph convolutional neural network adjacency matrix with the time attribute is constructed to solve the problem of prediction deviation caused by traffic asynchronism between virtual network links,and to achieve accurate representation of link correlation.Second,a traffic prediction mechanism based on long/short window graph convolutional neural network weighting is designed,which adapts the smooth and fluctuating segments of the traffic sequence with a finite length long/short window,effectively avoids the vanishing gradient problem of the neural network,and improves the traffic prediction accuracy of the virtual network.Finally,an error weighting unit is designed to sum the prediction results of the long/short window graph convolutional neural network.The output of the network is the predicted value of link traffic.In order to ensure the practicability of the results,the simulation experiments of the proposed temporal correlation graph convolutional network are carried out based on the real data center network data.Experimental results show that the proposed method has a higher prediction accuracy than the traditional graph convolutional neural network traffic prediction method.

    Table and Figures | Reference | Related Articles | Metrics
    Algorithm for prediction of the 6G vehicle trajectory based on the GNN-LSTM-CNN network
    CAI Gouqing,LIU Ling,ZHANG Chong,ZHOU Yiqing
    Journal of Xidian University    2023, 50 (3): 50-60.   DOI: 10.19665/j.issn1001-2400.2023.03.005
    Abstract213)   HTML3)    PDF(pc) (1869KB)(64)       Save

    The 6G era will realize the interconnection of all things and establish a multi-layer and full-coverage seamless connection.The Internet of Vehicles will be developed and deployed with the help of the 6G technology as a key area for the integration and intersection of communication,transportation,automobile and other industries.Aiming at the insufficient accuracy of the prediction of vehicle trajectories in the 6G Internet of Vehicles,this paper proposes a three-channel neural network model with the method of deep learning.This model takes the impacts of vehicle interaction information,target vehicle trajectories and lane structure information on trajectories into consideration.The long short-term memory network (LSTM) is used to extract the vehicle track information features,graph neural network (GNN) to extract interaction features between different vehicles,and the convolution neural network (CNN) is used to extract lane structure features.The predicted trajectory of the target vehicle is obtained by calculating the weight of the three-channel feature vector and the model is trained and tested by the NGSIM data set.The test results show that the three-channel network prediction method considering multi-dimension information has advantages in prediction accuracy and long time domain prediction compared with other prediction models,and the prediction accuracy is improved by more than 20%.Reducing the data transmission volume of the 6G Internet of Vehicles system can improve the user’s privacy security of the Internet of Vehicles system.

    Table and Figures | Reference | Related Articles | Metrics
    Research on threat intelligence extraction and knowledge graph construction technology
    SHI Huiyang,WEI Jingxuan,CAI Xingye,WANG He,GAO Suixiang,ZHANG Yuqing
    Journal of Xidian University    2023, 50 (4): 65-75.   DOI: 10.19665/j.issn1001-2400.2023.04.007
    Abstract200)   HTML9)    PDF(pc) (1879KB)(76)       Save

    At present,the infrastructure used by attackers can adapt to more target environments.After successfully invading the target,the attackers use legitimate user credentials to gain trust,and continuously learn to exploit new vulnerabilities to achieve the purpose of attacks.In order to combat attacks and to improve the quality and utilization efficiency of the threat intelligence,this paper constructs a knowledge mapping framework of threat intelligence through the following four processes:intelligence collection,information extraction,ontology construction,and knowledge reasoning.The proposed framework can realize the search for and correlation of essential indicators in the intelligence.Then,an indicator of compromise (IOC) recognition extraction method based on the Bert+BISLTM+CRF is proposed and a regular matching mechanism is applied to limit the output for identifying and extracting IOC information from the text information,followed by performing the structured threat information expression (STIX) standard format conversion.The accuracy and recall rate of this extraction model for the text information extraction are higher through horizontal and vertical comparison.Finally,by taking the APT1 as an example,this paper constructs the entity-relationship diagram of threat intelligence.The attack behavior is transformed into a structured format combined with the adversarial tactics,techniques,and common knowledge (ATT & CK) framework.A knowledge map of ontology and atomic ontology is established which is used to analyze the potential associations between data through the knowledge map associations and to discover potential associated information and attack agents in threat intelligence with similarity and correlation.The correlation analysis of threat intelligence is carried out,which provides the basis for the formulation of defense strategy.

    Table and Figures | Reference | Related Articles | Metrics
    Differentially private federated learning framework with adaptive clipping
    WANG Fangwei,XIE Meiyun,LI Qingru,WANG Changguang
    Journal of Xidian University    2023, 50 (4): 111-120.   DOI: 10.19665/j.issn1001-2400.2023.04.011
    Abstract175)   HTML15)    PDF(pc) (1094KB)(69)       Save

    Federation learning allows the parties involved in training to achieve collaborative modeling without sharing their own data.Its data isolation strategy safeguards the privacy and security of user data to a certain extent and effectively alleviates the problem of data silos.However,the training process of federation learning involves a large number of parameter interactions among the participants and the server,and there is still a risk of privacy disclosure.So a differentially private federated learning framework ADP_FL based on adaptive cropping is proposed to address the privacy protection problem during data transmission.In this framework,each participant uses its own data to train the model by performing multiple iterations locally.The gradient is trimmed by adaptively selecting the trimming threshold in each iteration in order to limit the gradient to a reasonable range.Only dynamic Gaussian noise is added to the uploaded model parameters to mask the contribution of each participant.The server aggregates the received noise parameters to update the global model.The adaptive gradient clipping strategy can not only achieve a reasonable calibration of the gradient,but also control the noise scale by dynamically changing the sensitivity while considering the clipping threshold as a parameter in the sensitivity.The results of theoretical analysis and experiments show that the proposed framework can still achieve a great model accuracy under strong privacy constraints.

    Table and Figures | Reference | Related Articles | Metrics
    Research on a clustering-assisted intelligent spectrum allocation technique
    ZHAO Haoqin, YANG Zheng, SI Jiangbo, SHI Jia, YAN Shaohu, DUAN Guodong
    Journal of Xidian University    2023, 50 (6): 1-12.   DOI: 10.19665/j.issn1001-2400.20231006
    Abstract167)   HTML24)    PDF(pc) (3593KB)(149)       Save

    Aiming at the problem of low spectrum utilization of the traditional spectrum allocation scheme in a large-scale and high dynamic electromagnetic spectrum warfare system,intelligent spectrum allocation technology research is carried out.In this paper,first,we construct a complex and highly dynamic electromagnetic spectrum combat scenario,and under the coexistence conditions of multiple types of equipment such as radar,communication and jamming,we model the spectrum allocation of the complex electromagnetic environment as an optimization problem to maximize the number of access devices.Second,an intelligent spectrum allocation algorithm based on clustering assistance is proposed.Aiming at the centralized resource allocation algorithm facing the problem of exploding action space dimensions,a multi-DDQN network is used to characterize the decision-making information of each node.Then based on the elbow law and K-means++ algorithm,a multi-node collaborative approach is proposed,where nodes within a cluster make chained decisions by sharing action information and nodes between clusters make independent decisions,assisting the DDQN algorithm to intelligently allocate resources.By designing the state,action space and reward function,and adopting the variable learning rate to realize the fast convergence of the algorithm,the nodes are able to dynamically allocate the multidimensional resources such as frequency/energy according to the electromagnetic environment changes.Simulation results show that under the same electromagnetic environment,when the number of nodes is 20,the number of accessible devices of the proposed algorithm is increased by about 80% compared with the number by the greedy algorithm,and about 30% compared with that by the genetic algorithm,which is more suitable for the spectrum allocation of multi-devices under dynamic electromagnetic environment.

    Table and Figures | Reference | Related Articles | Metrics
    Adaptive secure stream encryption supporting pattern matching
    LI Yiming,LIU Shengli
    Journal of Xidian University    2023, 50 (4): 1-10.   DOI: 10.19665/j.issn1001-2400.2023.04.001
    Abstract163)   HTML30)    PDF(pc) (1249KB)(141)       Save

    The stream encryption supporting pattern matching(SEPM) is a primitive proposed to provide privacy protection while doing pattern matching.On the one hand,one can use the SEPM to perform pattern matching on some ciphertext to find out whether and where a keyword exists in its corresponding plaintext.On the other hand,the security of the SEPM guarantees that no information about the plaintext will be revealed except for the results of pattern matching.Up to now,there have been several constructions of the SEPM,but none of them achieves the adaptive security from non-interactive assumptions(especially post-quantum assumptions),and supports pattern matching with the wildcard simultaneously.In this paper,we propose a new generic construction of the SEPM from a functional encryption(FE),achieving the adaptive security and supporting pattern matching with the wildcard.Further,we instantiate the generic construction of the SEPM by existing learning with error(LWE)-based instantiations of the FE.Finally,an SEPM scheme is obtained which could achieve the adaptive security from a non-interactive and post-quantum assumption (the LWE assumption) and could support pattern matching with the wildcard simultaneously.

    Table and Figures | Reference | Related Articles | Metrics
    Review on polarimetric SAR terrain classification methods using deep learning
    XIE Wen,HUA Wenqiang,JIAO Licheng,WANG Ruonan
    Journal of Xidian University    2023, 50 (3): 151-170.   DOI: 10.19665/j.issn1001-2400.2023.03.015
    Abstract158)   HTML8)    PDF(pc) (5211KB)(44)       Save

    Polarimetric synthetic aperture radar (PolSAR) is one of the main sources of remote sensing data,because it can realize all-day and all-weather imaging.Terrain classification is an important research in the field of PolSAR data interpretation,which has become one of the hotspots in the research field and has been widely used in both military and civilian applications.In recent years,deep learning has achieved remarkable results in many research fields,some of which have been made in the field of PolSAR image processing.Compared with traditional image classification methods,the deep learning method has the advantages of automatic extracting deep features,strong generalization and high accuracy.In this paper,the existing terrain classification methods for the PolSAR image based on deep learning are reviewed.According to the different network models in deep learning,the research on PolSAR terrain classification is described in detail from three aspects,that is,deep belief network,sparse autoencoder network and convolutional neural network.Finally,the advantages and disadvantages of PolSAR terrain classification based deep learning are summarized in comparison with classical classification methods.Meanwhile,the development trend of PolSAR terrain classification is analyzed and discussed.

    Table and Figures | Reference | Related Articles | Metrics
    Spectrum compression based autofocus algorithm for the TOPS BP image
    ZHOU Shengwei, LI Ning, XING Mengdao
    Journal of Xidian University    2024, 51 (1): 1-10.   DOI: 10.19665/j.issn1001-2400.20230102
    Abstract141)   HTML20)    PDF(pc) (4183KB)(154)       Save

    In the high squint TOPS mode SAR imaging of the maneuvering platform,by using the BP imaging algorithm in the rectangular coordinate system of the ground plane,the wide swath SAR image without distortion in the ground plane can be obtained in a short time.However,how to quickly complete the motion error compensation and side lobe suppression of the BP image is still difficult in practical application.This paper proposes an improved spectral compression method,which can quickly realize the follow-up operations such as autofocus of the BP image of the ground plane in the high squint TOPS mode of the mobile platform.First,by considering that the traditional BP spectral compression method is only applicable to the spotlight imaging mode,combined with the virtual rotation center theory of high-squint TOPS SAR and the wavenumber spectrum analysis,an improved exact spectral compression function is derived,which can give rise to the unambiguous ground plane TOPS mode BP image spectrum through full-aperture compression,on the basis of which the phase gradient autofocus(PGA) can be used to quickly complete the full aperture motion error estimation and compensation.In addition,based on the unambiguous aligned BP image spectrum obtained by the improved spectral compression method proposed in this paper,the image sidelobe suppression can be realized by uniformly windowing in the azimuth frequency domain.Finally,the effectiveness of the proposed algorithm is verified by simulation data processing.

    Table and Figures | Reference | Related Articles | Metrics
    Efficient deep learning scheme with adaptive differential privacy
    WANG Yuhua,GAO Sheng,ZHU Jianming,HUANG Chen
    Journal of Xidian University    2023, 50 (4): 54-64.   DOI: 10.19665/j.issn1001-2400.2023.04.006
    Abstract135)   HTML10)    PDF(pc) (1184KB)(76)       Save

    While deep learning has achieved a great success in many fields,it has also gradually exposed a series of serious privacy security issues.As a lightweight privacy protection technology,differential privacy makes the output insensitive to any data in the dataset by adding noise to the model,which is more suitable for the privacy protection of individual users in reality.Aiming at the problems of the dependence of iterations on the privacy budget,low data availability and slow model convergence in most existing differential private deep learning schemes,an efficient deep learning scheme based on adaptive differential privacy is proposed.First,an adaptive differential privacy mechanism is designed based on the Shapley additive explanation model.By adding noise to the sample features,the number of iterations is independent of the privacy budget,and then the loss function is perturbed by the function mechanism,thus achieving the dual protection of original samples and labels while enhancing the utility of data.Second,the adaptive moment estimation algorithm is used to adjust the learning rate to accelerate the model convergence.Additionally,zero-centralized difference privacy is introduced as a statistical mechanism of privacy loss,which reduces the risk of privacy leakage caused by the privacy loss exceeding the privacy budget.Finally,a theoretical analysis of privacy is made,with the effectiveness of the proposed scheme verified by comparative experiments on the MNIST and Fashion-MNIST datasets.

    Table and Figures | Reference | Related Articles | Metrics
    Analysis of the spatial coverage area of linear distributed directional array beamforming
    DUAN Baiyu,YANG Jian,CHEN Cong,GUO Wenbo,LI Tong,SHAO Shihai
    Journal of Xidian University    2023, 50 (5): 32-43.   DOI: 10.19665/j.issn1001-2400.20230103
    Abstract133)   HTML8)    PDF(pc) (3773KB)(89)       Save

    The phased array antenna has been widely used in radar,communication and other fields because of its advantages of high gain,high reliability and controllability of the beam.Considering the limitations of the size,the deployment terrain and the power consumption of the phased array antenna,it is difficult for a single phased array antenna to meet the requirements in some complex scenes,especially in some scenarios such as the space-earth communication,reconnaissance and jamming,so it is necessary to deploy multiple phased array antennas in a distributed manner for cooperative beamforming to obtain a higher power gain than a single array antenna.The distributed directional array uses multiple distributed array nodes to realize a virtual antenna array,sending or receiving the same signal by adjusting the phase of each array element to form the directional beam.A calculation method is proposed based on the principle of array synthesis and spatial analytic geometry aiming at the problem of calculating the gain coverage area of the distributed directional array beam in a specific height plane.Analysis and simulation results show that the gain coverage area of the linear distributed directional array beam,including the main lobe and gate lobe beam gain coverage area,is strongly correlated with the elevation angle of the distributed array,the height of the target plane,the signal carrier frequency and the number of distributed nodes,while it is weakly correlated with the distance between the distributed nodes.The analytical value of the proposed method is consistent with the computer simulation value,which can provide a theoretical reference for the implementation of the long-distance high-power distributed array in engineering.

    Table and Figures | Reference | Related Articles | Metrics
    Federated learning scheme for privacy-preserving of medical data
    WANG Bo,LI Hongtao,WANG Jie,GUO Yina
    Journal of Xidian University    2023, 50 (5): 166-177.   DOI: 10.19665/j.issn1001-2400.20230202
    Abstract132)   HTML8)    PDF(pc) (4010KB)(83)       Save

    As an emerging training model with neural networks,federated learning has received widespread attention due to its ability to carry out model training on the premise of protecting user data privacy.However,since adversaries can track and derive participants’ privacy from the shared gradients,federated learning is still exposed to various security and privacy threats.Aiming at the privacy leakage problem of medical data in the process of federated learning,a secure and privacy-preserving medical data federated learning architecture is proposed based on Paillier homomorphic encryption technology (HEFLPS).First,the shared training model of the client is encrypted with Paillier homomorphic encryption technology to ensure the security and privacy of the training model,and a zero-knowledge proof identity authentication module is designed to ensure the credibility of the training members;second,the disconnected or unresponsive users are temporarily eliminated by constructing a message confirmation mechanism on the server side,which reduces the waiting time of the server and reduces the communication cost.Experimental results show that the proposed mechanism has high model accuracy,low communication delay and a certain scalability while achieving privacy protection.

    Table and Figures | Reference | Related Articles | Metrics
    Algorithm for recognition of lightweight intelligent modulation based on the CNN-transformer networks
    YANG Jingya,QI Yanli,ZHOU Yiqing,ZHAO Dengpan,WANG Shangquan,SHI Jinglin
    Journal of Xidian University    2023, 50 (3): 40-49.   DOI: 10.19665/j.issn1001-2400.2023.03.004
    Abstract131)   HTML5)    PDF(pc) (3561KB)(54)       Save

    Existing modulation recognition methods based on deep learning have the problems of low recognition accuracy under the influence of noise and uncertain channel interference,and are difficult to apply to mobile terminals due to a large number of parameters.This paper proposes a lightweight modulation recognition method based on the Convolutional Neural Network (CNN) and Transformer to solve the above problems.In order to improve the accuracy,the CNN is first used to extract the local features of the signal.Then,the CNN-based channel attention and Transformer-based temporal attention modules are used to focus on the features that are most conducive to recognition from the two dimensions of the signal channel and time domain,respectively,while reducing the impact of the channel,noise,etc.The proposed method can be applied to a variety of signal representations,such as raw IQ signals,amplitude-phase signals,and transform domain features.Simulation shows that on the RadioML2016.10b dataset,compared with the existing convolutional network methods,the average recognition accuracy of the proposed method is increased by 8%~12%.Compared with the methods based on the residual neural network and long-term memory network,the number of parameters is reduced by 90%~92%,and the amount of calculation is reduced by about 83%~93%.Experimental results show that the proposed method can improve the accuracy of model classification while effectively reducing the number of parameters and the amount of calculation.

    Table and Figures | Reference | Related Articles | Metrics
    Indoor pseudolite hybrid fingerprint positioning method
    LI Yaning,LI Hongsheng,YU Baoguo
    Journal of Xidian University    2023, 50 (5): 21-31.   DOI: 10.19665/j.issn1001-2400.20221102
    Abstract128)   HTML22)    PDF(pc) (4526KB)(99)       Save

    At present,the interaction mechanism between the complex indoor environment and pseudolite signals has not been fundamentally resolved,and the stability,continuity,and accuracy of indoor positioning are still technical bottlenecks.Existing fingerprint positioning methods face the limitation that the collection workload is proportional to the positioning accuracy and positioning range,and have the disadvantage that the positioning cannot be completed without actual collection.In order to solve the above shortcomings of the existing methods,by combining the advantages of actual measurement,mathematical simulation and the artificial neural network,an indoor pseudolite hybrid fingerprint location method based on actual fingerprints,simulation fingerprints and the artificial neural network is proposed.First,the actual environment and signal transceiver are modeled.Second,both the simulated fingerprints generated by ray tracing simulation after conversion and the measured fingerprints are added to the input of the neural network,which expands the sample characteristics of the input data set of the original single measured fingerprints.Finally,the artificial neural network positioning model is jointly trained by the mixed fingerprints and then used for online positioning.By taking an airport environment as an example,it is proved that the hybrid method can improve the positioning accuracy of the sparsely collected fingerprint region,and that the root mean square error is 0.485 0 m,which is 54.7% lower than that of the traditional fingerprint positioning method.Preliminary positioning can also be completed in areas where no fingerprints are collected,and the root mean square positioning error is 1.123 7 m,which breaks through the limitations of traditional fingerprint location methods.

    Table and Figures | Reference | Related Articles | Metrics
    Efficient federated learning privacy protection scheme
    SONG Cheng,CHENG Daochen,PENG Weiping
    Journal of Xidian University    2023, 50 (5): 178-187.   DOI: 10.19665/j.issn1001-2400.20230403
    Abstract126)   HTML7)    PDF(pc) (1908KB)(79)       Save

    Federated learning allows clients to jointly train models with only shared gradients,rather than directly feeding the training data to the server.Although federated learning avoids exposing data directly to third parties and plays a certain role in protecting data,research shows that the transmission gradient in federated learning scenarios will still lead to the disclosure of private information.However,the computing and communication overhead brought by the encryption scheme in the training process will affect the training efficiency,and it is difficult to apply to resource-constrained environments.Aiming at the security and efficiency problems of privacy protection schemes in current federated learning,a safe and efficient privacy protection scheme for federated learning is proposed by combining homomorphic encryption and compression techniques.The homomorphic encryption algorithm is optimized to ensure the security of the scheme,reduce the number of operations and improve the efficiency of operations.At the same time,a gradient filtering compression algorithm is designed to filter out the local updates that are not related to the convergence trend of the global model,and the update parameters are quantized by a computationally negligible compression operator,which ensures the accuracy of the model and increases the communication efficiency.The security analysis shows that the scheme satisfies the security characteristics such as indistinguishability,data privacy and model security.Experimental results show that the proposed scheme has not only higher model accuracy,but also obvious advantages over the existing schemes in terms of communication cost and calculation cost.

    Table and Figures | Reference | Related Articles | Metrics
    Adaptive score-level fusion for multi-modal biometric authentication
    JIANG Qi,ZHAO Xiaomin,ZHAO Guichuan,WANG Jinhua,LI Xinghua
    Journal of Xidian University    2023, 50 (4): 11-21.   DOI: 10.19665/j.issn1001-2400.2023.04.002
    Abstract125)   HTML13)    PDF(pc) (1111KB)(66)       Save

    In recent years,biometric-based authentication has played a vital role in our daily life.The multi-modal authentication method by fusing multiple biometrics to authenticate users can provide a higher security and authentication accuracy than single-modal authentication.However,most of the existing multi-modal authentication schemes adopt fusion strategies with fixed rules and parameters to achieve authentication,which cannot adapt to different authentication scenarios,thus resulting in a sub-optimal authentication performance.To solve the above problems,this paper proposes an Adaptive Particle Swarm Optimization based multi-modal authentication scheme that adaptively fuses multiple biometrics at the score level.First,the proposed scheme determines the security level required for the current authentication scenario according to the context information,and then adaptively selects rules and parameters of the fusion strategy to provide secure authentication and to ensure the best authentication performance of the system.Second,the collected multi-modal biometric data after preprocessing and feature extraction is fused using the selected optimal fusion strategy to achieve authentication.Finally,experimental analyses on the public dataset demonstrate that the proposed scheme is of feasibility and effectiveness by actual data,and can achieve a smaller global error rate than existing schemes under the same authentication security requirements.

    Table and Figures | Reference | Related Articles | Metrics
    Industrial control protocol reverse analysis based on active interactive learning
    FU Anmin,MAO An,HUANG Tao,HU Chao,LIU Ying,ZHANG Xiaoming,WANG Zhanfeng
    Journal of Xidian University    2023, 50 (4): 22-33.   DOI: 10.19665/j.issn1001-2400.2023.04.003
    Abstract120)   HTML12)    PDF(pc) (1680KB)(70)       Save

    As an important basis for information exchange in industrial control systems,the standardization and completeness of the design and implementation of industrial control protocols involve the security of the entire industrial control system.For the reverse of unknown industrial control protocols,although the protocol reverse method based on traffic samples has attracted more and more attention because it does not need to analyze the system firmware and other advantages,this type of method also has the disadvantage of relying too much on sample diversity.Especially,insufficient sample diversity can easily lead to problems such as field division errors,state identification errors,and only a subset of protocol specifications can be obtained from analysis.For this reason,this paper proposes an industrial control protocol reverse analysis method based on active interactive learning.On the basis of the reverse results of traffic samples,a data packet set is constructed according to the initial reverse results,and interactive learning is carried out with real devices to detect unknown protocol fields and state machines.Simulation experimental results of interactive learning with industrial control simulation software show that this method can effectively verify field semantics,expand field values,expand abnormal sample types,and solve the problem of pseudo-long static fields caused by insufficient sample diversity and that it can detect new states and state transitions,greatly improving the accuracy of unknown protocol reverse.

    Table and Figures | Reference | Related Articles | Metrics
    Work pattern recognition method based on feature fusion
    LIU Gaogao, HUANG Dongjie, XI Xin, LI Hao, CAO Xuyuan
    Journal of Xidian University    2023, 50 (6): 13-20.   DOI: 10.19665/j.issn1001-2400.20230705
    Abstract113)   HTML12)    PDF(pc) (810KB)(84)       Save

    Operational pattern recognition is one of the important means in the field of intelligence reconnaissance and electronic countermeasures,which is to determine the function and behavior of radar through signal processing and analysis.With the diversification of modern airborne radar functions,the corresponding signal styles are becoming more and more complex,and the increasingly complex reconnaissance environment also leads to the uneven quality of reconnaissance signals,which brings about great difficulties to the traditional operational pattern recognition methods.To solve this problem,based on the existing work pattern recognition methods,a new work pattern recognition method is proposed,which integrates parameter feature recognition and D-S evidence theory recognition.First,for the radiation source characteristic signals processed by each reconnaissance plane,the feature parameter recognition algorithm is used to quickly obtain the working mode information,and the recognition results are verified by the D-S evidence theory.Second,for the signal that can not be recognized by a single platform,the method of D-S evidence theory fusion recognition is used to distinguish the working mode.From the theoretical analysis,it can be concluded that the algorithm has the advantages of fast operation speed and simple structure,and that the new fusion recognition method can improve the recognition accuracy of the working mode.Finally,the feasibility of the method is verified by simulation.

    Table and Figures | Reference | Related Articles | Metrics
    Cause-effectgraph enhanced APT attack detection algorithm
    ZHU Guangming,LU Zijie,FENG Jiawei,ZHANG Xiangdong,ZHANG Fengjun,NIU Zuoyuan,ZHANG Liang
    Journal of Xidian University    2023, 50 (5): 107-117.   DOI: 10.19665/j.issn1001-2400.20221105
    Abstract108)   HTML10)    PDF(pc) (2814KB)(74)       Save

    With the development of information technology,the cyberspace also derives an increasing number of security risks and threats.There are more and more advanced cyberattacks,with the Advanced Persistent Threat(APT) attack being one of the most sophisticated attacks and commonly adopted by modern attackers.Traditional statistical or machine learning detection methods based on network flow are challenging in coping with complicated and persistent APT-style attacks.Aiming to overcome the difficulty in detecting APT attacks,a cause-effect graph enhanced APT attack detection algorithm is proposed to model the interaction process between network nodes at different times and identify malicious packets in the attack process in network flows.First,the causal-effect graph is used to model the network packet sequences,and the data flows between IP nodes in the network are associated to establish the context sequence of attack and non-attack behaviors.Then,the sequence data are normalized,and the deep learning model based on the long short-term memory network(LSTM) is used for sequence classification.Finally,based on the sequence classification results,the original packets are screened for malignancy.A new dataset is constructed based on the DAPT 2020 dataset,with the proposed algorithm’s ROC-AUC indicator on the test set reaching 0.948.Experimental results demonstrate that the attack detection algorithm based on causal-effect graph sequences has obvious advantages and is a feasible algorithm for detecting APT attack network flow.

    Table and Figures | Reference | Related Articles | Metrics
    Deterministic service of space-air-ground integrated networks: architecture,challenges and key technologies
    CAO Huan,CHEN Yan,ZHOU Yiqing,SU Yongtao,LIU Zifan,CHEN Daojin,DING Yashuai
    Journal of Xidian University    2023, 50 (3): 1-18.   DOI: 10.19665/j.issn1001-2400.2023.03.001
    Abstract108)   HTML3)    PDF(pc) (6645KB)(67)       Save

    It is one of the important development directions of future 6G communication to meet the extreme communication needs of users in the global vertical industry,cooperate with the terrestrial mobile communication network and the rapidly developing non-terrestrial network (NTN),and break the traditional rigid service mode of "doing your best" to provide users with global deterministic services.First,this paper summarizes the future space-air-ground integrated network architecture and the deterministic service connotation and scenario requirements under this architecture.In addition,it proposes a global network-oriented deterministic service management and a control technology framework.Then,it analyzes the three major challenges faced in the process of global deterministic service,including the difficulty in ensuring the business awareness of users in the global whole scene,orchestrating the end-to-end slicing network space-air-ground integration,and quickly coordinating and scheduling the global multidimensional resources in the slicing subnet.In response to the above challenges,three solutions are introduced respectively,namely,intelligent cloud-based full-domain,full-scene service sensing technology,satellite earth end-to-end intelligent slicing orchestration based on network topology prediction,and data and model-driven satellite earth resource intelligent allocation technology,which provide a reference for the development of space-air-ground integrated network extreme service technology.

    Table and Figures | Reference | Related Articles | Metrics
    Privacy preserving byzantine robust federated learning algorithm
    LI Haiyang,GUO Jingjing,LIU Jiuzun,LIU Zhiquan
    Journal of Xidian University    2023, 50 (4): 121-131.   DOI: 10.19665/j.issn1001-2400.2023.04.012
    Abstract106)   HTML7)    PDF(pc) (1782KB)(62)       Save

    Federated learning is a distributed machine learning paradigm,in which the original training sets of the nodes do not have to leave the local area and they collaborate to train machine learning models by sharing model updates.Most of the current privacy-preserving and Byzantine attack detection researches in the field of federated learning are carried out independently,and the existing Byzantine attack detection methods cannot be directly applied to the privacy-preserving environment,which does not meet the practical application requirements of federated learning.To address these problems,this paper proposes a federated learning algorithm for Byzantine robustness in a privacy-preserving environment with data non-independent and identically distributed.First,privacy protection is provided for model updates (local model gradient information) by differential privacy techniques; then the credibility is evaluated for the current state of nodes based on historical model updates uploaded by nodes; and finally,global model aggregation is performed based on the evaluation results.Simulation results show that in a federated learning environment with data non-independent and identically distributed,and with the privacy protection and Byzantine node ratio of 20%~80%,the proposed algorithm performs Byzantine node detection with both the miss detection rate and the false detection rate at 0%.Meanwhile,the time overhead of Byzantine node detection tends to linearly increase with the increase in the number of the nodes.Compared with the existing Byzantine node detection algorithms,the proposed algorithm can obtain a global model with a higher accuracy in the case of data being non-independent and identically distributed and model privacy protection.

    Table and Figures | Reference | Related Articles | Metrics
    Traffic flow prediction method for integrating longitudinal and horizontal spatiotemporal characteristics
    HOU Yue,ZHENG Xin,HAN Chengyan
    Journal of Xidian University    2023, 50 (5): 65-74.   DOI: 10.19665/j.issn1001-2400.20221101
    Abstract105)   HTML7)    PDF(pc) (4496KB)(78)       Save

    Aiming at the problems of insufficient mining of time delay characteristics and spatial flow characteristics of upstream and downstream traffic flow as well as insufficient consideration of spatiotemporal characteristics of lane-level traffic flow in existing urban road traffic flow prediction research,a traffic flow prediction method for integrating longitudinal and horizontal spatiotemporal characteristics is proposed.First,the method quantifies and eliminates the effect of spatial time lag between upstream and downstream traffic flow by calculating the delay time to enhance the spatiotemporal correlation of upstream and downstream traffic flow sequences.Then,the traffic flow with the elimination of spatial time lag is passed into the bidirectional long short-term memory network through the vector split data input method to capture the longitudinal transmission and backtracking bidirectional spatiotemporal relationship of upstream and downstream traffic flow.At the same time,the multiscale convolution group is used to mine the multi-time step horizontal spatiotemporal relationship between the traffic flows of each lane in the section to be predicted.Finally,the attention mechanism is used to dynamically fuse the longitudinal and horizontal spatiotemporal characteristics to obtain the predicted value.Experimental results show that by applying the proposed method in the single-step prediction experiment,the MAE and RMSE decrease by 15.26% and 13.83% respectively,and increase by 1.25% compared with conventional time series prediction model.In the medium and long-term multi-step prediction experiment,it is further proved that the proposed method can effectively mine the fine-grained spatiotemporal characteristics of longitudinal and horizontal traffic flow,and has a certain stability and universality.

    Table and Figures | Reference | Related Articles | Metrics
    Nuclear segmentation method for thyroid carcinoma pathologic images based on boundary weighting
    HAN Bing,GAO Lu,GAO Xinbo,CHEN Weiming
    Journal of Xidian University    2023, 50 (5): 75-86.   DOI: 10.19665/j.issn1001-2400.20230501
    Abstract100)   HTML6)    PDF(pc) (5221KB)(72)       Save

    Thyroid cancer is one of the most rapidly growing malignancies among all solid cancers.Pathological diagnosis is the gold standard for doctors to diagnose tumors,and nuclear segmentation is a key step in the automatic analysis of pathological images.Aiming at the low segmentation performance of existing segmentation methods on the nuclear boundary of the cell nucleus in the thyroid carcinoma pathological image,we propose an improved U-Net method based on boundary weighting for nuclear segmentation.This method uses the designed boundary weighting module,which can make the segmentation network pay more attention to the boundary of the nuclear.At the same time,in order to avoid the proposed network paying too much attention to the boundary and ignoring the main part of the nucleus,which leads to the failure for some lightly stained nuclei segmentation,we design a segmentation network to enhance the foreground area and suppresses the background area in the upsampling stage.In addition,we build a dataset for nuclear segmentation of thyroid carcinoma pathologic images named VIP-TCHis-Seg dataset.Our method achieves the Dice coefficient(Dice) of 85.26% and the pixel accuracy(PA) of 95.89% on self-built TCHis-Seg dataset,and achieves the Dice coefficient(Dice) of 81.03% and the pixel accuracy(PA) of 94.63% on common dataset MoNuSeg.Experimental results show that our method can achieve the best performance on both Dice and PA as well as effectively improve the segmentation accuracy of the network at the boundary compared with other methods.

    Table and Figures | Reference | Related Articles | Metrics
    Real-time power scheduling optimization strategy for 5G base stations considering energy sharing
    LIU Didi,YANG Yuhui,XIAO Jiawen,YANG Yifei,CHENG Pengpeng,ZHANG Quanjing
    Journal of Xidian University    2023, 50 (5): 44-53.   DOI: 10.19665/j.issn1001-2400.20230101
    Abstract100)   HTML9)    PDF(pc) (2647KB)(70)       Save

    To alleviate the pressure on society's power supply caused by the huge energy consumption of the 5th generation mobile communication (5G) base stations,a joint distributed renewables,energy sharing and energy storage model is proposed with the objective of minimizing the long-term power purchase cost for network operators.A low-complexity real-time scheduling algorithm for energy sharing based on the Lyapunov optimization theory is proposed,taking into account the fact that the a priori statistical information on renewable energy output,energy demand and time-varying tariffs in smart grids are unknown.A virtual queue is constructed for the flexible electricity demand of the base stations in optimization problem solving.The energy storage time coupling constraint is transformed in the energy scheduling problem into a virtual queue stability problem.The proposed algorithm schedules the renewable energy output,energy storage,energy use and energy sharing of the base stations in real time,and minimizes the long-term cost of network operators purchasing power from the external grid on the premise of meeting the electricity demand of each base station.Theoretical analysis shows that all the proposed algorithm needs is to make real-time decisions based on the current system state and that the optimization result is infinitely close to the optimal value.Finally,simulation results show that the proposed algorithm can effectively reduce the power purchase cost of the network operator by 43.1% compared to the baseline greedy Algorithm One.

    Table and Figures | Reference | Related Articles | Metrics
    Multi-scale object detection algorithm combined with super-resolution reconstruction technology
    WANG Juan,LIU Zishan,WU Minghu,CHEN Guanhai,GUO Liquan
    Journal of Xidian University    2023, 50 (3): 122-131.   DOI: 10.19665/j.issn1001-2400.2023.03.012
    Abstract100)   HTML7)    PDF(pc) (4169KB)(47)       Save

    At present,most object detection algorithms have poor performance because of the large span of scales,leading to errors and omissions.To address the above issues,a multi-scale object detection algorithm combined with the super-resolution technology is proposed in this paper.First,based on the one-stage YOLO framework,the super-resolution module is employed to the neck network during the process of multi-scale feature fusion,which avoids further loss of detailed features in deeper layers.Second,the attention module is integrated in the shallower layers to focus on the channel information on object contour features and to suppress irrelevant features,thus improving the superficial representational capacity.Finally,ablation and comparative experiments are carried out on PASCAL VOC 2007 and MS COCO 2017 public datasets.Experimental results show that the proposed module can improve the detection performance.Compared with the current contrast algorithms,not only can the average accuracy rate of small,medium and large objects be increased by 1.20%,1.20% and 1.30%,but also the average recall rate can be improved by 4.20%,3.50% and 4.20%,respectively.

    Table and Figures | Reference | Related Articles | Metrics
    Anti-occlusion PMBM tracking algorithm optimized by fuzzy inference
    LI Cuiyun,HENG Bowen,XIE Jinchi
    Journal of Xidian University    2023, 50 (5): 54-64.   DOI: 10.19665/j.issn1001-2400.20230401
    Abstract97)   HTML11)    PDF(pc) (6914KB)(77)       Save

    Target occlusion is a common problem in multiple extended target tracking.When the distance between targets is close or there are unknown obstacles within the scanning range of the sensor,the phenomenon of partial or complete occlusion of the target will occur,resulting in underestimation of the target quantity.Aiming at the problem that the existing Poisson multi-Bernoulli mixture(PMBM) filtering algorithms cannot perform stable tracking in occlusion scenarios,this paper proposes a GP-PMBM algorithm incorporating fuzzy inference.First,based on the random set target tracking framework,the corresponding extended target occlusion model is given according to different occlusion scenarios.On this basis,the state space of the GP-PMBM filter is expanded,and the influence of occlusion on the target state is taken into account in the filtering steps of the algorithm by adding variable detection probability.Finally,a fuzzy inference system that can estimate the target occlusion probability is constructed and combined with the GP-PMBM algorithm,and the accurate estimation of the target in occlusion scenarios is achieved with the help of the description ability of the fuzzy system and the good tracking performance of the PMBM filter.Simulation results show that the tracking performance of the proposed algorithm in target occlusion scenarios is better than that of the existing PMBM filtering algorithms.

    Table and Figures | Reference | Related Articles | Metrics
    Privacy-preserving federated learning with non-transfer learning
    XU Mengfan,LI Xinghua
    Journal of Xidian University    2023, 50 (4): 89-99.   DOI: 10.19665/j.issn1001-2400.2023.04.009
    Abstract96)   HTML8)    PDF(pc) (1612KB)(54)       Save

    The model stealing and gradient leakage attacks have increasingly become the bottlenecks that limit the broad application of federated learning.The existing authorization-based intellectual property protection schemes and privacy-preserving federated learning schemes have conducted a lot of research to solve the above challenges.However,there are still issues of authorization invalidation and high computational overhead.To solve the above problems,this paper proposes a model intellectual property and privacy-preserving method in federated learning.This method can protect the privacy of local gradients while ensuring that the aggregated model authorization is not invalidated.Specifically,a lightweight gradient aggregation method based on the blind factor is designed to significantly reduce the computational overhead of the encryption and decryption process by aggregating blinding factors.On this basis,an interactive co-training method based on anti-transfer learning is further proposed to ensure that the model can only be used by authorized users in authorized domains while protecting the privacy of local gradients,where the Shannon mutual information between the representation vector of the auxiliary domain data and the obstacle is increased.The security and correctness of the scheme are theoretically proved,and the system’s superiority is verified on the public data set.It is shown that the performance of the proposed method in the unauthorized domain is at least 47% lower than that of the existing schemes,and the computational complexity is reduced at the level of gradient dimension.

    Table and Figures | Reference | Related Articles | Metrics
    ResNet enabled joint channel estimation and signal detection for OTFS
    ZHOU Shuo,ZHOU Yiqing,ZHANG Chong,XING Wang
    Journal of Xidian University    2023, 50 (3): 19-30.   DOI: 10.19665/j.issn1001-2400.2023.03.002
    Abstract95)   HTML3)    PDF(pc) (3272KB)(32)       Save

    Orthogonal time frequency space (OTFS) modulation can realize reliable broadband communication at a high doppler frequency offset,which is one of the potential application technologies in the 6G communication-sensing-computing scenario.In order to solve the problems of high complexity and limited performance of the receiver in this system,a joint channel estimation and signal detection algorithm based on modified ResNet is proposed,with the transmission symbol information recovered directly without obtaining explicit channel information.According to the stability of the delay doppler domain channel,deep learning technology is introduced into the receiver design,and a lightweight residual neural network model that can fully extract the signal features is designed by using the embedded pilot data frame structure.It can directly fit the input-output relationship of delay doppler domain signals to achieve implicit channel estimation and complete signal detection.In the joint design,the optimal network model is obtained by off-line training with the data collected in the actual communication link,which can be used for on-line detection.Meanwhile,the joint optimization of channel estimation and signal detection is realized with the help of an error back propagation mechanism and gradient descent criterion,which effectively improves the communication performance.Simulation results show that the proposed scheme has better robustness and good generalization compared with the traditional receiver algorithm,which not only reduces the algorithm complexity,but also improves the BER performance by about 2dB.

    Table and Figures | Reference | Related Articles | Metrics
    Cloth-changing person re-identification paradigm based on domain augmentation and adaptation
    ZHANG Peixu,HU Guanyu,YANG Xinyu
    Journal of Xidian University    2023, 50 (5): 87-94.   DOI: 10.19665/j.issn1001-2400.20221106
    Abstract94)   HTML8)    PDF(pc) (1901KB)(76)       Save

    In order to solve the influence of the clothing change on the model’s recognition accuracy of the personal identity,a clothes-changing person re-identification paradigm based on domain augmentation and adaptation is proposed,which enables the model to learn general robust identity representation features in different domains.First,a clothing semantic-aware domain data enhancement method is designed based on the semantic information of the human body,which changes the color of sample clothes without changing the identity of the target person to fill the lack of domain diversity in the data; second,a multi-positive class domain adaptive loss function is designed,which assigns differential weights to the multi-positive class data losses according to the different contributions made by different domain data in the model training,forcing the model to focus on the learning of generic identity features of the samples.Experiments demonstrate that the method achieves 59.5%,60.0%,and 88.0%,84.5% of Rank-1 and mAP on two clothing change datasets,PRCC and CCVID,without affecting the accuracy of non-clothing person re-identification.Compared with other methods,this method has a higher accuracy and stronger robustness and significantly improves the model’s ability to recognize persons.

    Table and Figures | Reference | Related Articles | Metrics
    Double adaptive image watermarking algorithm based on regional edge features
    GUO Na,HUANG Ying,NIU Baoning,LAN Fangpeng,NIU Zhixian,GAO Zhuojie
    Journal of Xidian University    2023, 50 (5): 118-131.   DOI: 10.19665/j.issn1001-2400.20221107
    Abstract91)   HTML5)    PDF(pc) (2865KB)(69)       Save

    Local image watermarking is a hotspot technology which embeds watermark in a partial image and can resist cropping attacks.Existing local watermarking technology locates the embedded region by the feature points,which may be offset when attacks occur.Due to the obvious difference of the pixel near edges,if the region contains many edges,the offset will lead to excessive regional pixel error and fail the watermark extraction.To solve this problem,a double adaptive image watermarking algorithm based on regional edge features is proposed.First,a method to determine the embedding region is proposed,which uses a sliding window to choose embedding regions with few edges and good hiding ability by taking image features such as the edge,texture,etc.into account.Second,a double adaptive watermark embedding scheme is proposed,which is divided into blocks,with each block embedding 1-bit watermark information by modifying the pixel value.In the first coarse-grained adaptive scheme,the function between the embedding parameter and the number of edge pixels is established through linear regression analysis,and the embedding strength is adaptively adjusted by the function to enhance the robustness of blocks containing edges.In the second fine-grained adaptive scheme,the gaussian window is used to adaptively adjust the modifications of different pixels to improve the imperceptibility of the watermark.Experiments show that the proposed algorithm can effectively enhance the robustness of the watermark at the edge,and improve its imperceptibility.

    Table and Figures | Reference | Related Articles | Metrics
    Integrated sensing,communication,and computation via the over-the-air computing architecture
    HAN Kaifeng,ZHOU Ziqin,WANG Zhiqin,GONG Yi,LI Xiaoyang
    Journal of Xidian University    2023, 50 (3): 31-39.   DOI: 10.19665/j.issn1001-2400.2023.03.003
    Abstract89)   HTML1)    PDF(pc) (1802KB)(47)       Save

    Limited by equipment conditions and computer network hierarchy,traditional researches treat data sensing,communication,and computation as independent processes,which lack global consideration and seriously hinder the efficiency of data processing.To improve the efficiencies of data sensing and transmission,the sensing-communication fusion is proposed to design dual-functional signals for supporting both radar sensing and data communication.To improve the efficiencies of data communication and computation,over-the-air computing aims to exploit the waveform superposition property of signals in the multiple access channel to enable data computation during transmission.In order to achieve efficient data processing,the dual-functional signals in sensing-communication fusion and the waveform superposition property in over-the-air computing are utilized to realize the integrated sensing,communication,and computation over the air.The corresponding beamformers are designed to reduce channel noise and signal interference,so as to improve the accuracies of sensing,communication,and computation.This technology can be applied in various fields such as target detection,vehicle networking,and edge intelligence.Experimental results show that the technology can significantly improve the efficiency and accuracy of data processing in comparison with the traditional schemes.

    Table and Figures | Reference | Related Articles | Metrics
    The design and cryptanalysis of a large state lightweight cryptographic S-box
    FAN Ting,FENG Wei,WEI Yongzhuang
    Journal of Xidian University    2023, 50 (4): 170-179.   DOI: 10.19665/j.issn1001-2400.2023.04.017
    Abstract86)   HTML5)    PDF(pc) (945KB)(51)       Save

    Alzette is a 64 bit lightweight S-box based on the ARX structure proposed at the CRYPTO 2020.It has many advantages such as excellent hardware and software performance,strong diffusion and high security,so that it receives wide attention domestically and internationally.However,64-bit lightweight S-boxes with execllent performance and security are rare.Whether it is possible to design the large state lightweight S-box with better performance than Alzette is difficult in current research.In this paper,a large state lightweight cryptographic S-box based on the ARX structure with an excellent performance and security is designed.A “hierarchy filtering method” is proposed to determine the optimal rotation parameters by setting the best differential/linear characteristic bounds in advance,and the security evaluation for the new S-box is given.It is shown that the software and hardware implementation performance of the new S-box is equivalent to that of the Alzette.For the new S-box,the probability of 5-round best differential characteristic (linear approximation) up to 2-17(2-8),and the probability of 7-round best linear approximation reaches 2-17.But for the Alzette,the 5-round best differential characteristic (linear approximation) with probability of 2-10>2-17(2-5>2-8),and the 7-round best linear approximation with probability of 2-13>2-17.The new S-box shows a stronger resistance against differential cryptanalysis and linear cryptanalysis.

    Table and Figures | Reference | Related Articles | Metrics
    UAVs trajectory planning and power allocation based on the convergence of communication,sensing and computing
    WU Yihao,QI Yanli,ZHOU Yiqing,CAI Qing,LIU Ling,SHI Jinglin
    Journal of Xidian University    2023, 50 (3): 61-74.   DOI: 10.19665/j.issn1001-2400.2023.03.006
    Abstract85)   HTML5)    PDF(pc) (2245KB)(29)       Save

    Regional natural disasters often cause damage to ground-based communication facilities,and UAVs networks can act as aerial base stations to restore communications.Existing research has focused on how to provide efficient communication services to rescuers in static scenarios with a limited UAV spectrum and battery capacity.However,the location movement and service changes of communication rescuers in real scenarios lead to the failure of static schemes.To solve this problem,this paper proposes a collaborative UAVs scheduling algorithm through the convergence of communication,sensing and computing.First,we perform sensing the environmental information,i.e.,the rescuers' historical location information and service demand,in real-time to realize the prediction of the rescuers' future location and service demand and provide a priori information for the scheduling of UAVs.Second,an improved k-sums algorithm is proposed to deploy the UAVs' location concerning the UAV load constraint to achieve UAVs' load balancing.Furthermore,a reinforcement learning algorithm is used to optimize the UAVs' transmit power to ensure rescuers' communication service quality under a limited bandwidth.Compared to static scenarios where rescuer-UAV associations are established based on signal-to-noise ratios,the proposed UAV scheduling algorithm through the convergence of communication,sensing and computing in this paper can effectively improve network utility (network communication benefits minus communication costs) by 20%.The algorithm provides a guaranteed business experience for rescuers in emergency disaster relief scenarios.

    Table and Figures | Reference | Related Articles | Metrics
    COLLATE:towards the integrity of control-related data
    DENG Yingchuan,ZHANG Tong,LIU Weijie,WANG Lina
    Journal of Xidian University    2023, 50 (5): 199-211.   DOI: 10.19665/j.issn1001-2400.20230106
    Abstract84)   HTML5)    PDF(pc) (3156KB)(55)       Save

    Programs written in C/C++ may contain bugs that can be exploited to subvert the control flow.Existing control-flow hijacking mitigations validate the indirect control-flow transfer targets,or guarantee the integrity of code pointers.However,attackers can still overwrite the dependencies of function pointers,bending indirect control-flow trans-fers(ICTs) to valid but unexpected targets.We introduce the control-related data integrity(COLLATE) to guarantee the integrity of function pointers and their dependencies.The dependencies determine the potential data-flow between function pointers definition and ICTs.The COLLATE identifies function pointers,and collects their dependencies with the inter-procedure static taint analysis.Moreover,the COLLATE allocates control-related data on a hardware-protected memory domain MS to prevent unauthorized modifications.We evaluate the overhead of the COLLATE on SPEC CPU 2006 benchmarks and Nginx.Also,we evaluate its effectiveness on three real-world exploits and one test suite for vtable pointer overwrites.The evaluation results show that the COLLATE successfully detects all attacks,and introduces a 10.2% performance overhead on average for the C/C++ benchmark and 6.8% for Nginx,which is acceptable.Experiments prove that the COLLATE is effective and practical.

    Table and Figures | Reference | Related Articles | Metrics
    Random chunks attachment strategy based secure deduplication for cloud data
    LIN Genghao,ZHOU Ziji,TANG Xin,ZHOU Yiteng,ZHONG Yuqi,QI Tianyang
    Journal of Xidian University    2023, 50 (5): 212-228.   DOI: 10.19665/j.issn1001-2400.20230503
    Abstract84)   HTML9)    PDF(pc) (5198KB)(65)       Save

    Source based deduplication prevents subsequent users from uploading the same file by returning a deterministic response,which greatly saves the network bandwidth and storage overhead.However,the deterministic response inevitably introduces side channel attacks.Once the subsequent uploading is not needed,an attacker can easily steal the existent privacy of the target file in cloud storage.To resist side channel attacks,various kinds of defense schemes such as adding trusted gateways,setting trigger thresholds,confusing response values,and so on are proposed.However,these methods suffer from the problems of high deployment costs,high startup costs and the difficulty in resisting random chunks generation attack and learn remaining information attack.Thus,we propose a novel secure deduplication scheme,which utilizes the random chunks attachment strategy to achieve obfuscation in response.Specifically,we first add a certain number of chunks with the unknown existent status at the end of the request to blur the existent status of the original requested ones,and then reduce the probability of returning a lower boundary value in response by scrambling strategy.Finally,the deduplication response is generated with the help of the newly designed response table.Security analysis and experimental results show that,compared with the existing works,our scheme significantly improve the security at the expense of just a little extra overhead.

    Table and Figures | Reference | Related Articles | Metrics
    Anti-collusion attack image retrieval privacy protection scheme for ASPE
    CAI Ying,ZHANG Meng,LI Xin,ZHANG Yu,FAN Yanfang
    Journal of Xidian University    2023, 50 (5): 156-165.   DOI: 10.19665/j.issn1001-2400.20230406
    Abstract83)   HTML8)    PDF(pc) (1886KB)(61)       Save

    The existing algorithm based on Asymmetric Scalar-Product-Preserving Encryption (ASPE) realizes privacy protection in image retrieval under cloud computing.But due to untrustworthy cloud service providers and retrieval users during retrieval and the existence of an external adversary,it cannot resist the collusion attack of malicious users and cloud servers,which may lead to the leakage of image data containing sensitive information.Aiming at multi-user scenarios,an Anti-collusion attack image retrieval privacy protection scheme for ASPE is proposed.First,the scheme uses proxy re-encryption to solve the problem of image key leakage caused by transmitting private keys to untrusted users.Second,the feature key leakage problem between the cloud service provider and the retrieval user due to collusion attacks is solved by adding a diagonal matrix encryption at the client side.Finally,linear discriminant analysis is used to solve the problem of retrieval accuracy drop caused by dimensionality reduction when locality sensitive hashing is used to construct an index.The security analysis proves that the scheme is safe and effective and that it can not only resist collusion attacks from cloud service providers and untrusted users,ciphertext-only attacks,known background attacks and known plaintext attacks,but also realize protection of images and private keys during the process.Experimental results show that under the premise of protecting image privacy and ensuring retrieval efficiency,the retrieval accuracy of the proposed scheme in the ciphertext domain and that in the plaintext domain are only about 2% different.

    Table and Figures | Reference | Related Articles | Metrics
    Research on the interference combinational sequence generation algorithm for the intelligent countermeasure UAV
    MA Xiaomeng, GAO Meiguo, YU Mohan, LI Yunjie
    Journal of Xidian University    2023, 50 (6): 44-61.   DOI: 10.19665/j.issn1001-2400.20230903
    Abstract81)   HTML12)    PDF(pc) (8025KB)(59)       Save

    With the maturity and development of the autonomous navigation flight technology for the unmanned aerial vehicle(UAV),the phenomenon of the unauthorized UAV flying in controlled airspace appears,which brings a great hidden danger to personal safety and causes a certain degree of economic losses.The research of this paper is on improving the effectiveness of adaptive measurement and control and navigation interference in the unknown situation of UAV flight control on the basis of identifying the UAV flight status and real-time evaluation of countermeasure effectiveness,and finally realizing the intelligent countermeasure game between the non-intelligent UAV based on the combination of remote communication interference and navigation and positioning interference.In this paper,a game model of the anti-UAV system(AUS) and UAV confrontation is developed based on the original units of radar detection,GPS navigation positioning,UAV remote communication suppression jamming and GPS navigation suppression and spoofing.The mathematical model is constructed by using deep reinforcement learning and the Markov decision process.Meanwhile,the concept of situation assessment ring for the classification of the UVA flight status is proposed to provide basic information for network sensing jamming effectiveness.The near-end strategy optimization algorithm,maximum entropy optimization algorithm and actor-critic algorithm are respectively used to train the constructed intelligent AUS for many times,and finally the network parameters are generated to generate the intelligent interference combination sequence according to the UAV flight state and countermeasures efficiency.The intelligent interference combination sequences generated by various deep reinforcement learning algorithms in this paper all achieve the initial goal of deceiving UAVs,which verifies the effectiveness of the anti-UAVs system model.The comparison experiment shows that the proposed situation assessment loop is sufficient and effective in the aspect of AUS sensing interference effectiveness.

    Table and Figures | Reference | Related Articles | Metrics
    Detecting compromised email accounts via spatiotemporal login behavior analysis
    ZHAO Jianjun,WANG Xutong,CUI Xiang,LIU Qixu
    Journal of Xidian University    2023, 50 (4): 34-44.   DOI: 10.19665/j.issn1001-2400.2023.04.004
    Abstract79)   HTML8)    PDF(pc) (1178KB)(38)       Save

    Compromised email accounts detection faces various challenges in the system administration and attack forensics,such as the lack of threat intelligence,a large amount of data to be analyzed,and the difficulty with direct confirmation with the email owners.To address the above problems,this paper proposes a compromised email accounts detection method using only login logs without relying on any labeled samples.First,this paper summarizes the attack features and proposes an email accounts compromise model.Second,based on the email accounts compromise model,this paper characterizes the spatial similarity and temporal synchronization when invading the email accounts.When using the spatial similarity to detect the compromised email accounts,this paper uses graphs to construct the spatial distances between accounts;and then,the accounts with a similar spatial distance are grouped into the same community,and the possibility of accounts compromising is evaluated according to the community size.When using the temporal synchronization to detect the compromised email accounts,this paper proposes a metric to describe the abnormal login behaviors and evaluates the possibility of compromise by checking if other accounts have similar abnormal behaviors in the same period.Finally,a sorted list of email accounts is outputted to provide priority reference for analysts according to the possibility of compromise.Experimental results show that the method proposed in this paper can detect about 98% of the compromised email accounts with 70% workload reduced,and the detection effect is better than that of the similar studies.Additionally,the detection method can discover the unknown attackers and the undisclosed malicious IP addresses.

    Table and Figures | Reference | Related Articles | Metrics
    Generative adversarial model for radar intra-pulse signal denoising and recognition
    DU Mingyang, DU Meng, PAN Jifei, BI Daping
    Journal of Xidian University    2023, 50 (6): 133-147.   DOI: 10.19665/j.issn1001-2400.20230312
    Abstract76)   HTML9)    PDF(pc) (14404KB)(62)       Save

    While deep neural networks have achieved an impressive success in computer vision,the related research remains embryonic in radio frequency signal processing,i.e.,a vital task in modern wireless systems,for example,the electronic reconnaissance system.Noise corruption is a harmful but unavoidable factor causing severe performance degradation in the signal processing procedure,and thus has persistently been an intractable problem in the radio frequency domain.For example,a classifier trained on the high signal-to-noise ratio(SNR) data might experience a severe performance degradation when dealing with low SNR data.To address this problem,in this paper we leverage the powerful data representation capacity of deep learning and propose a Generative Adversarial Denoising and classification Network(GADNet) for radar signal restoration and a classification task.The proposed GADNet consists of a generator,a discriminator and a classifier fulfilling an end-to-end workflow.The encoder-decoder structure generator is trained to extract the high-level features and recover signals.Meanwhile,it fools the discriminator’s judges by bewildering the denoising results coming from the clean data.The classification loss from the classifier is adopted jointly to the training procedure.Extensive experiments demonstrate the benefit of the proposed technique in terms of high-quality restoration and accurate classification for radar signals with intense noise.Moreover,it also exhibits superior transferability in low SNR environments compared to the state-of-the-art methods.

    Table and Figures | Reference | Related Articles | Metrics
    Improved short-signature based cloud data audit scheme
    CUI Yuanyou,WANG Xu’an,LANG Xun,TU Zheng,SU Yunxuan
    Journal of Xidian University    2023, 50 (5): 132-141.   DOI: 10.19665/j.issn1001-2400.20230107
    Abstract74)   HTML5)    PDF(pc) (1741KB)(54)       Save

    With the development of the Internet of Things,Cloud storage has experienced an explosive growth.Effective verification of the integrity of data stored on the Cloud storage service providers(CSP) has become an important issue.In order to solve the problem that the existing data integrity audit scheme based on the BLS short signature is inefficient,ZHU et al.designed a data integrity audit scheme based on the ZSS short signature in 2019.However,this paper points out that the proof generated by ZHU et al.'s scheme in the challenge phase is incorrect and can be subjected to replay attacks or attacked by using a bilinear map,so as to pass the audit of a third party auditor(TPA).Then,this paper proposes an improved cloud audit scheme based on the short signature by improving the calculation method of proof in the challenge stage and optimizing the equations used by the third party auditor in the verification stage for verifying proof.This paper proves the correctness of the improved scheme,compensates for the shortcomings in the original scheme,and analyzes the security of the scheme.The improved scheme not only can make attackers including the third party auditor unable to recover users’ data,but also can resist replay attacks and forgery attacks of attackers including malicious cloud storage service providers.Through numerical analysis,it is found that the computational cost did not change much,and that the communication cost decreased,thus providing a better computational accuracy than the original scheme.

    Table and Figures | Reference | Related Articles | Metrics
    Point set registration optimization algorithm using spatial clustering and structural features
    HU Xin,XIANG Diyuan,QIN Hao,XIAO Jian
    Journal of Xidian University    2023, 50 (5): 95-106.   DOI: 10.19665/j.issn1001-2400.20230411
    Abstract70)   HTML6)    PDF(pc) (6289KB)(61)       Save

    The existence of noise,non-rigid deformation and mis-matching in point set registration results in the difficulty of solving nonlinear optimal space transformation.This paper introduces local constraints and proposes a point set registration optimization algorithm using spatial distance clustering and local structural features(PR-SDCLS).First,the motion consistency clustering subset and outlier clustering subset are constructed by using the point set space distance matrix;Then,the Gaussian mixture model is used to fit the motion consistency cluster subset,and the mixing coefficient considering global and local features is obtained by fusing the shape context feature descriptor and weighted spatial distance.Finally,the maximum expectation algorithm is used to complete the parameter estimation,and the non-rigid point set registration model of the Gaussian mixture model is realized.In order to improve the efficiency of the algorithm,the model transformation uses the reproducing kernel Hilbert space model,and uses the kernel approximation strategy.Experimental results show that the algorithm has a good registration effect and robustness in the face of a large number of outliers on non-rigid data sets involving different types of data degradation(deformation,noise,outliers,occlusion and rotation),and the mean value of registration average error is reduced by 42.053 8% on the basis of classic and advanced algorithms.

    Table and Figures | Reference | Related Articles | Metrics
    Privacy preserving multi-classification LR scheme for data quality
    CAO Laicheng,WU Wentao,FENG Tao,GUO Xian
    Journal of Xidian University    2023, 50 (5): 188-198.   DOI: 10.19665/j.issn1001-2400.20230601
    Abstract69)   HTML10)    PDF(pc) (1676KB)(61)       Save

    In order to protect the privacy of the multi-classification logistic regression model in machine learning,ensure the quality of training data,and reduce the computing and communication costs,a privacy preserving multi-classification logistic regressions cheme for data quality is proposed.First,based on the homomorphic encryption for arithmetic of approximate numbers technology,the batch processing technology and single-instruction multi-data mechanism are used to package multiple messages into one ciphertext,and the encrypted vector is safely shifted into the ciphertext corresponding to the plaintext vector.Second,the binary logistic regression model is extended to multiple classifications by training multiple classifiers using the "One vs Rest" disassembly strategy.Finally,the training data set is divided into several matrices of a fixed size,which still retain the complete data structure of the sample information.The fixed Hessian method is used to optimize the model parameters so that they can be used in any case and keep the parameters private.during model training.The scheme can reduce data sparsity and ensure data quality.The security analysis shows that the training model and user data information cannot be leaked in the whole process.Meanwhile,the experiment shows that the training accuracy of this scheme is greatly improved compared with the existing scheme and almost the same as that obtained by training unencrypted data,and that the scheme has a lower computing cost.

    Table and Figures | Reference | Related Articles | Metrics
    Algorithm for H.264/AVC adaptive watermarking
    WANG Yong,HUANG Junyu,CHEN Yifang,ZHANG Jun,CHEN Xiaozong
    Journal of Xidian University    2023, 50 (3): 95-104.   DOI: 10.19665/j.issn1001-2400.2023.03.009
    Abstract68)   HTML4)    PDF(pc) (2656KB)(25)       Save

    H.264/AVC video watermarking plays an important role in video copyright protection and information hiding.At present,the existing watermarking researches lack the consideration of the variable complexity of the video frame in the process of watermark embedding,which can easily lead to two problems:first,these watermark algorithms may not be able to fully embed the required watermark in frames with low picture complexity,and second,the redundancy of frames with high picture complexity cannot be fully utilized.Therefore,in this paper we analyze the factors of frame complexity in the process of video coding,and propose an adaptive watermarking algorithm based on the complexity of the video frame.In the proposed algorithm,the complexity degree of the current key frame is predicted using the total number of Intra_4×4 sub macroblocks of the previous key frame that satisfy the embedding conditions.The repetition time of embedding for each watermark bit is adjusted automatically according to the predicted complexity degree to achieve adaptiveness.A key frame with a prediction error is labelled an invalid frame.The watermark extraction is conducted by majority voting.Experimental results show that the proposed algorithm achieves watermark invisibility,and code stream stability as well,which outperforms the other recent algorithms.The algorithm is instructive to the research based on repeatedly embedding watermark bits.

    Table and Figures | Reference | Related Articles | Metrics
    Dark web author alignment based on attention augmented convolutional networks
    YANG Yanyan,DU Yanhui,LIU Hongmeng,ZHAO Jiapeng,SHI Jinqiao,WANG Xuebin
    Journal of Xidian University    2023, 50 (4): 206-214.   DOI: 10.19665/j.issn1001-2400.2023.04.020
    Abstract68)   HTML6)    PDF(pc) (2158KB)(42)       Save

    Dark network users engage in a large number of illegal and criminal activities in the underground market.The anonymity of the dark network brings great convenience to the communication between users of the dark network,but great difficulties to the police.In recent years,the deep neural network has been widely successful in various fields,and more and more researchers have begun to use the neural network to identify anonymous network text authors.In order to better align users in the dark web and find more different users with the same identity,we use the neural network method to identify and align users in the dark web.However,the existing methods focus mainly on the short text and are not good at dealing with the global and long sequence information.In this paper,we propose a self-attention mechanism to enhance the convolution operator and use long sequence information to strengthen the user representation,named DACN.DACN starts from the text content,and multiple account associations are carried out for anonymous dark web users to aggregate information from multiple anonymous accounts,proving mores clues for obtaining the users’true identity.Our recent analysis involves conducting a thorough assessment of two distinct dark web market forums,whereby we evaluate our methodology in comparison to the current state-of-the-art techniques.Experimental results show that our approach is remarkably effective,with a demonstrated average mean retrieval ranking (MRR) enhancement of 2.9% and 3.6%,as well as an improved Recall@10 of 2.3% and 3.0%.This evaluation offers robust evidence of the efficacy of our approach in dark web market forums.

    Table and Figures | Reference | Related Articles | Metrics
    Fast algorithm for intelligent optimization of the cross ambiguity function of passive radar
    CHE Jibin, WANG Changlong, JIA Yan, REN Zizheng, LIU Chunheng, ZHOU Feng
    Journal of Xidian University    2023, 50 (6): 21-33.   DOI: 10.19665/j.issn1001-2400.20231003
    Abstract68)   HTML9)    PDF(pc) (5554KB)(53)       Save

    The passive radar system realizes the target detection by receiving the direct wave signal from the emitter and the target echo signal.The cross ambiguity function is an important means to improve the coherent accumulation of the echo signal.However,the echo signal received by the passive radar is very weak,so it is necessary to increase the accumulation time to improve the estimation accuracy.When the target speed is fast,the frequency search range increases.In order to achieve a range of target detection requirements and take into account the real-time performance of data processing,it is of great significance to study the fast calculation method of the cross ambiguity function,and due to the objective requirements of long-time accumulation and large-scale time-frequency search,the computation of the cross ambiguity function is huge,which makes it difficult for the traditional accelerated calculation method based on ergodic search to meet the real-time requirements of system processing.In order to improve the efficiency of cross ambiguity function optimization,a time-frequency difference calculation method based on multi-group feature optimization is proposed in this paper.By deeply analyzing the characteristics of typical digital TV signals,a two-stage cross ambiguity intelligent optimization fast calculation method based on target characteristics is designed in the framework of particle swarm optimization theory.By designing an effective search strategy,this method introduces the multi-population iteration mechanism and shrinkage factor,which avoids the disadvantages of the traditional method of redundant computation.On the premise of ensuring the calculation accuracy,the time-frequency point calculation is greatly reduced,and the search efficiency of cross ambiguity function is improved.

    Table and Figures | Reference | Related Articles | Metrics
    Privacy-preserving internet of things data filtering scheme
    ZHOU Rang,ZHANG Xiaosong,WANG Xiaofen,LI Dongfen,CHEN Tao,ZHANG Xiaojun
    Journal of Xidian University    2023, 50 (4): 45-53.   DOI: 10.19665/j.issn1001-2400.2023.04.005
    Abstract67)   HTML4)    PDF(pc) (1123KB)(45)       Save

    With the development of industry 5.0,the operational data need to be collected and uploaded in real time in the practical Internet of Things (IoT).To describe and analyze the working state of the IoT more precisely,high accurate and real-time data are required.Then,in practical applications,many different types of IoT data are stored together without classifying,which could reduce the efficiency of data analysis.In order to improve the efficiency of data analysis in the hybrid data storage environment,it is necessary to use the method of data shunting in the process of data upload to realize the classified storage of data.However,the traditional data shunting method shunts the plaintext data according to its source identity,during which the source information on the plaintext data will leak the identity and privacy of the IoT devices.Therefore,how to realize the classified storage of these IoT data through the data shunting without revealing the privacy has become an urgent problem to be solved in the security management of the IoT data.In this paper,a new privacy-preserving IoT data filtering scheme is proposed.On the basis of maintaining the context and device identity privacy,each data filtering rule is set by a filtering trapdoor,which is computed from the identity of the data source device.Then,the data can be classified and routed by the relay nodes following the given rules in the data uploading phase,from which the heterologous data can be classified and the homologous data are stored together,which can help further data access control and data analysis.Experiment results show that our scheme is efficient and practical.

    Table and Figures | Reference | Related Articles | Metrics
    Verifiable traceable electronic license sharing deposit scheme
    WANG Lindong,TIAN Youliang,YANG Kedi,XIAO Man,XIONG Jinbo
    Journal of Xidian University    2023, 50 (5): 142-155.   DOI: 10.19665/j.issn1001-2400.20230408
    Abstract66)   HTML8)    PDF(pc) (4017KB)(61)       Save

    Verifiability and traceability are important challenges to the sharing and retention of electronic licenses.Traditional methods only ensure the verifiability of the issuer through electronic signature technology,but the verifiability of the holder and the depositor and the traceability of the license leakage are difficult to guarantee.Therefore,a verifiable and traceable electronic license sharing deposit scheme is proposed.First,aiming at the problem of unauthorized use of electronic licenses and the inability to trace after leakage,a model of the electronic license sharing and deposit system is constructed.Second,aiming at the problem of watermark information loss in the traditional strong robust watermarking algorithm,the existing strong robust watermarking algorithm is improved based on the BCH code,so as to realize the error correction of watermark information distortion.Finally,in order to realize the verifiability of the issuer,the holder and the depositor as well as the efficient traceability after the leakage of the electronic license,the verifiable and traceable electronic license model is constructed by combining the proposed robust watermark and reversible information hiding technology,on the basis of which the electronic license sharing and deposit protocol is designed to ensure the real authorized use of the license and the efficient traceability after the leakage.The analysis of security and efficiency shows that this scheme can achieve an efficient traceability after license leakage and has a good anti-collusion attack detection ability under the premise of ensuring the verifiability of the three parties,and that its execution time consumption is low enough to meet the needs of practical applications.

    Table and Figures | Reference | Related Articles | Metrics
    Resource optimization algorithm for unmanned aerial vehicle jammer assisted cognitive covert communications
    LIAO Xiaomin, HAN Shuangli, ZHU Xuan, LIN Chushan, WANG Haipeng
    Journal of Xidian University    2023, 50 (6): 75-83.   DOI: 10.19665/j.issn1001-2400.20230603
    Abstract64)   HTML4)    PDF(pc) (1909KB)(53)       Save

    Aiming at the covert communication scenario of an unmanned aerial vehicle(UAV) jammer assisted cognitive radio network,a transferred generative adversarial network based resource optimization algorithm is proposed for the UAV’s joint trajectory and transmit power optimization problem.First,based on the actual covert communication scenario,the UAV jammer assisted cognitive covert communication model is constructed.Then,a transferred generative adversarial network based resource allocation algorithm is designed,which introduces a transfer learning and generative adversarial network.The algorithm consists of a source domain generator,a target domain generator,and a discriminator,which extract the main resource allocation features of legitimate users not transmitting covert message by transfer learning,then transform the whole covert communication process into an interactive game between the legitimate users and the eavesdropping,alternatively train the target domain generator and discriminator in a competitive manner,and achieve the Nash equilibrium to obtain resource optimization solution for the covert communications.Numerical results show that the proposed algorithm can attain near-optimal resource optimization solution for the covert communication and achieve rapid convergence under the assumptions of knowing the channel distribution information and not knowing the detection threshold of the eavesdropper.

    Table and Figures | Reference | Related Articles | Metrics
    Multi-scale convolutional attention network for radar behavior recognition
    XIONG Jingwei, PAN Jifei, BI Daping, DU Mingyang
    Journal of Xidian University    2023, 50 (6): 62-74.   DOI: 10.19665/j.issn1001-2400.20231005
    Abstract63)   HTML9)    PDF(pc) (8738KB)(59)       Save

    A radar behavior mode recognition framework is proposed aiming at the problems of difficult feature extraction and low recognition stability of the radar signal under a low signal-to-noise ratio,which is based on depth-wise convolution,multi-scale convolution and the self-attention mechanism.It improves the recognition ability in complex environment without increasing the difficulty of training.This algorithm employs depth-wise convolution to segregate weakly correlated channels in the shallow network.Subsequently,it utilizes multi-scale convolution to replace conventional convolution for multi-dimensional feature extraction.Finally,it employs a self-attention mechanism to adjust and optimize the weights of different feature maps,thus suppressing the influence of low and negative correlations in both channels and the spatial domains.Comparative experiments demonstrate that the proposed MSCANet achieves an average recognition rate of 92.25% under conditions of 0~50% missing pulses and false pulses.Compared to baseline networks such as AlexNet,ConvNet,ResNet,and VGGNet,the accuracy has been improved by 5% to 20%.The model exhibits stable recognition of various radar patterns and demonstrates enhanced generalization and robustness.Simultaneously,ablation experiments confirm the effectiveness of deep grouped convolution,multi-scale convolution,and the self-attention mechanism for radar behavior recognition.

    Table and Figures | Reference | Related Articles | Metrics
    Active zero trust model against APT theft in the industrial internet
    FENG Jingyu,LI Jialun,ZHANG Baojun,HAN Gang,ZHANG Wenbo
    Journal of Xidian University    2023, 50 (4): 76-88.   DOI: 10.19665/j.issn1001-2400.2023.04.008
    Abstract62)   HTML4)    PDF(pc) (2683KB)(43)       Save

    The comprehensive and deep integration of the new generation of information technology and industrial systems that induces the advanced persistent threat (APT) theft has become a killer-level insider threat that leaks sensitive data in the industrial internet environment.The critical infrastructure in the industrial internet environment generates and maintains a large number of sensitive data with "ownership" characteristics,which will bring immeasurable economic losses to enterprises once they are leaked.Aiming at the lag of sensitive data protection in the current industrial internet,an active zero trust model against APT theft is proposed.Our model introduces the long short-term memory neural network to construct a feature extractor based on its advantages in processing temporal data,to train abstract sequence features from behavioral data,and to extract regular trust factors.The block creation is carried out for industrial internet terminals respectively.The forward sequential redundant block elimination algorithm is designed to evolve a scalable blockchain called the ZTE_chain so as to achieve tamper-proof and low-load trust factor security storage.To respond to the behavior changes of compromised terminals in time,the convolutional neural network is introduced to predict the mutation factor,which is used to dynamically adjust the trust value,on the basis of which an authentication algorithm is given to quickly identify the compromised terminals and to actively block their APT theft threat.Experimental results show that the model proposed in this paper has a good effect of identifying compromised terminals,which is helpful in combating the APT theft threat generated by compromised terminals in the industrial internet environment.

    Table and Figures | Reference | Related Articles | Metrics