Advanced Search
Submit & Review

2021 Vol.30 No.4

Published on 19 July 2021

Articles in press have been peer-reviewed and accepted, which are not yet assigned to volumes /issues, but are citable by Digital Object Identifier (DOI).
Display Method:
Marginal Attacks of Generating Adversarial Examples for Spam Filtering
GU Zhaoquan, XIE Yushun, HU Weixiong, et al.
2021, 30(4): 595-602.   doi: 10.1049/cje.2021.05.001
Abstract(40) PDF(16)
Digit information has been used in many areas and has been widely spread in the Internet era because of its convenience. However, many ill-disposed attackers, such as spammers take advantage of such convenience to send unsolicited information, such as advertisements, frauds, and pornographic messages to mislead users and this might cause severe consequences. Although many spam filters have been proposed in detecting spams, they are vulnerable and could be misled by some carefully crafted adversarial examples. In this paper, we propose the marginal attack methods of generating such adversarial examples to fool a naive Bayesian spam filter. Specifically, we propose three methods to select sensitive words from a sentence and add them at the end of the sentence. Through extensive experiments, we show that the generated adversarial examples could largely reduce the filter’s detecting accuracy, e.g. by adding only one word, the accuracy could be reduced from 93.6% to 55.8%. Furthermore, we evaluate the transferability of the generated adversarial examples against other traditional filters such as logic regression, decision tree and linear support vector machine based filters. The evaluation results show that these filters’ accuracy is also reduced dramatically; especially, the decision tree based filter’s accuracy drops from 100% to 1.51% by inserting only one word.
Unsupervised, Supervised and Semi-supervised Dimensionality Reduction by Low-Rank Regression Analysis
TANG Kewei, ZHANG Jun, ZHANG Changsheng, et al.
2021, 30(4): 603-610.   doi: 10.1049/cje.2021.05.002
Abstract(18) PDF(8)
Techniques for dimensionality reduction have attracted much attention in computer vision and pattern recognition. However, for the supervised or unsupervised case, the methods combining regression analysis and spectral graph analysis do not consider the global structure of the subspace; For semi-supervised case, how to use the unlabeled samples more effectively is still an open problem. In this paper, we propose the methods by Low-rank regression analysis (LRRA) to deal with these problems. For supervised or unsupervised dimensionality reduction, combining spectral graph analysis and LRRA can make a global constraint on the subspace. For semi-supervised dimensionality reduction, the proposed method incorporating LRRA can exploit the unlabeled samples more effectively. The experimental results show the effectiveness of our methods.
Semi-supervised Robust Feature Selection with ℓq-Norm Graph for Multiclass Classification
LUO Hui, HAN Jiqing
2021, 30(4): 611-622.   doi: 10.1049/cje.2021.05.003
Abstract(17) PDF(1)
Flexible manifold embedding (FME) is a semi-supervised dimension reduction framework. It has been extended into feature selection by using different loss functions and sparse regularization methods. However, these kind of methods used the quadratic form of graph embedding, thus the results are sensitive to noise and outliers. In this paper, we propose a general semisupervised feature selection model that optimizes an ℓq-norm of FME to decrease the noise sensitivity. Compare to the fixed parameter model, the ℓq-norm graph brings flexibility to balance the manifold smoothness and the sensitivity to noise by tuning its parameter. We present an efficient iterative algorithm to solve the proposed ℓq-norm graph embedding based semi-supervised feature selection problem, and offer a rigorous convergence analysis. Experiments performed on typical image and speech emotion datasets demonstrate that our method is effective for the multiclass classification task, and outperforms the related state-of-the-art methods.
Improving Knowledge Graph Completion Using Soft Rules and Adversarial Learning
TANG Caifang, RAO Yuan, YU Hualei, et al.
2021, 30(4): 623-633.   doi: 10.1049/cje.2021.05.004
Abstract(17) PDF(4)
Knowledge graph is a useful resources and tools for describing entities and relationships in natural language processing tasks. However, the existing knowledge graph are incomplete. Therefore, knowledge graph completion technology has become a research hotspot in the field of artificial intelligence, but the traditional knowledge graph embedding method does not fully take into account the role of logic rules and the effect of false negative samples on knowledge embedding. Based on the logic rules of knowledge and the role of adversarial learning in knowledge embedding, we proposes a model to improve the completion of knowledge graph: soft Rules and graph adversarial learning (RUGA). Firstly, the traditional knowledge graph embedding model is trained as generator and discriminator by using adversarial learning method, and high-quality negative samples are obtained. Then these negative samples and the existing positive samples together constitute the label triple in the injection rule model. The whole model will benefit from both high-quality samples and logical rules. In addition, we evaluated the performance of link prediction task and triple classification task on Freebase and Yago datasets respectively. Finally, the experimental results show that the model can effectively improve the effect of knowledge graph completion.
Smoothing Neural Network for Non-Lipschitz Optimization with Linear Inequality Constraints
YU Xin, WU Lingzhen, XIE Mian, et al.
2021, 30(4): 634-643.   doi: 10.1049/cje.2021.05.005
Abstract(20) PDF(2)
This paper presents a smoothing neural network to solve a class of non-Lipschitz optimization problem with linear inequality constraints. The proposed neural network is modelled with a differential inclusion equation, which introduces the smoothing approximate techniques. Under certain conditions, we prove that the trajectory of neural network reaches the feasible region in finite time and stays there thereafter, and that any accumulation point of the solution is a stationary point of the original optimization problem. Furthermore, if all stationary points of the optimization problem are isolated, then the trajectory converges to a stationary point of the optimization problem. Two typical numerical examples are given to verify the effectiveness of the proposed neural network.
A Bionic Optimization Technique with Cockroach Biological Behavior
CHENG Le, CHANG Lyu, SONG Yanhong, et al.
2021, 30(4): 644-651.   doi: 10.1049/cje.2021.05.006
Abstract(12) PDF(456)
Many practical engineering problems can be abstracted as corresponding function optimization problems. During the last few decades, many bionic algorithms have been proposed for this problem. However, when optimizing for large scale problems, such as 1000 dimensions, many existing search techniques may no longer perform well. Inspired by the social model of cockroaches, this paper presents a novel search technique called Cooperation cockroach colony optimization (CCCO). In the CCCO algorithm, two kinds of special biological behavior of cockroach, wall-following and nest-leaving, are simulated and the whole population is divided into wall-following and nest-leaving populations. By the collaboration of the two populations, CCCO accomplishes the computation of global optimization. The crucial parameters of CCCO are set by the self-adaptive method. Moreover, a discussion on group model design is provided in this paper. The CCCO algorithm is evaluated with shifted test functions (1000 dimensions). Three state-of-the-art cockroach-inspired algorithms are used for the comparative experiments. Furthermore, CCCO is applied to a real-world optimization problem concerning spread spectrum radar poly-phase. Experiment results show that the CCCO algorithm can be applied to optimize large-scale problems with the good performance.
Differential Fault Attack on Camellia
ZHOU Yongbin, WU Wenling, XU Nannan, FENG Dengguo
2009, 18(1): 13-19.  
[Abstract](654) [PDF 423KB](32)
Camellia is the final winner of 128-bit blockcipher in NESSIE project, and is also certified as the international IETF standard cipher for SSL/TLS cipher suites.In this study, we present an effcient differential fault attack on Camellia. Ideally, by using our techniques, on average, the complete key of Camellia-128 is recovered with64 faulty ciphertexts while the full keys of Camellia-192and Camellia-256 are retrieved with 96 faulty ciphertexts.Our attack is applicable to generic block ciphers with overall Fiestel structure using a SPN round function.All theseattacks have been successfully put into experimental simulations on a personal computer.
An Ultra Low Steady-State Current Power-on- Reset Circuit in 65nm CMOS Technology
SHAN Weiwei, WANG Xuexiang, LIU Xinning, SUN Huafang
2014, 23(4): 678-681.  
[Abstract](921) [PDF 832KB](751)
A novel Power-on-reset (POR) circuit is proposed with ultra-low steady-state current consumption. A band-gap voltage comparator is used to generate a stable pull-up voltage. To eliminate the large current consumptions of the analog part, a power switch is adopted to cut the supply of band-gap voltage comparator, which gained ultra-low current consumption in steady-state after the POR rest process completed. The state of POR circuit is maintained through a state latch circuit. The whole circuit was designed and implemented in 65nm CMOS technology with an active area of 120μm*160μm. Experimental results show that it has a steady pull-up voltage of 0.69V and a brown-out voltage of 0.49V under a 1.2V supply voltage rising from 0V, plus its steady-state current is only 9nA. The proposed circuit is suitable to be integrated in system on chip to provide a reliable POR signal.
Face Liveness Detection Based on the Improved CNN with Context and Texture Information
GAO Chenqiang, LI Xindou, ZHOU Fengshun, MU Song
2019, 28(6): 1092-1098.   doi: 10.1049/cje.2019.07.012
[Abstract](661) [PDF 3162KB](47)
Face liveness detection, as a key module of real face recognition systems, is to distinguish a fake face from a real one. In this paper, we propose an improved Convolutional neural network (CNN) architecture with two bypass connections to simultaneously utilize low-level detailed information and high-level semantic information. Considering the importance of the texture information for describing face images, texture features are also adopted under the conventional recognition framework of Support vector machine (SVM). The improved CNN and the texture feature based SVM are fused. Context information which is usually neglected by existing methods is well utilized in this paper. Two widely used datasets are used to test the proposed method. Extensive experiments show that our method outperforms the state-of-the-art methods.
Identity Based Encryption and Biometric Authentication Scheme for Secure Data Access in Cloud Computing
CHENG Hongbing, RONG Chunming, TAN Zhenghua, ZENG Qingkai
2012, 21(2): 254-259.  
[Abstract](1093) [PDF 273KB](62)
Cloud computing will be a main information infrastructure in the future; it consists of many large datacenters which are usually geographically distributed and heterogeneous. How to design a secure data access for cloud computing platform is a big challenge. In this paper, we propose a secure data access scheme based on identity-based encryption and biometric authentication for cloud computing. Firstly, we describe the security concern of cloud computing and then propose an integrated data access scheme for cloud computing, the procedure of the proposed scheme include parameter setup, key distribution, feature template creation, cloud data processing and secure data access control. Finally, we compare the proposed scheme with other schemes through comprehensive analysis and simulation. The results show that the proposed data access scheme is feasible and secure for cloud computing.
A Global K-modes Algorithm for Clustering Categorical Data
BAI Tian, C.A. Kulikowski, GONG Leiguang, YANG Bin, HUANG Lan, ZHOU Chunguang
2012, 21(3): 460-465.  
[Abstract](457) [PDF 334KB](29)
In this paper, a new Global k-modes (GKM) algorithm is proposed for clustering categorical data. The new method randomly selects a sufficiently large number of initial modes to account for the global distribution of the data set, and then progressively eliminates the redundant modes using an iterative optimization process with an elimination criterion function. Systematic experiments were carried out with data from the UCI Machine learning repository. The results and a comparative evaluation show a high performance and consistency of the proposed method, which achieves significant improvement compared to other well-known k-modes-type algorithms in terms of clustering accuracy.
Large Spaceborne Deployable Antennas (LSDAs)-A Comprehensive Summary
DUAN Baoyan
2020, 29(1): 1-15.   doi: 10.1049/cje.2019.09.001
[Abstract](481) [PDF 4261KB](166)
This paper provides a survey of research activities of Large spaceborne deployable antennas (LSDAs) in the past, present and future. Firstly, three main kinds of spaceborne antennas, such as solid reflector, inflatable reflector and mesh reflector, are issued by showing the strengths and weaknesses. Secondly, a detailed research situation of LSDAs with mesh is discussed, for majority of the in-orbit large diameter and high frequency antennas are made in this type of structures. Thirdly, new conception of antenna is proposed as it does have both advantages of large aperture (high gain) and high precision (high frequency). Fourthly, the design theory and approach of LSDAs are concerned. It includes thermal-electromechanical multidisciplinary optimization, shaped beam design technique, performance testing technology and evaluation method, passive intermodulation of mesh, and application of new materials. Finally, the ultra large spaceborne deployable antennas of the next generation are presented, such as the deployable frame and inflatable reflector antennas, space-assembled ultra large antennas, smart array antennas and so on.

Review Articles

A Review of FPGA-Based Custom Computing Architecture for Convolutional Neural Network Inference
PENG Xiyuan, YU Jinxiang, YAO Bowen, et al.
2021, 30(1): 1-17.   doi: 10.1049/cje.2020.11.002
Abstract(307) HTML(149) PDF(77)
Deep Learning and Its Application in Diabetic Retinopathy Screening
ZOU Beiji, SHAN Xi, ZHU Chengzhang, et al.
2020, 29(6): 992-1000.   doi: 10.1049/cje.2020.09.001
Abstract(89) PDF(336)
Research and Application of Machine Learning in Automatic Program Generation
ZHANG Xiaojiang, JIANG Ying
2020, 29(6): 1001-1015.   doi: 10.1049/cje.2020.10.006
Abstract(94) PDF(244)
Transparent Computing: Development and Current Status
ZHANG Yaoxue, DUAN Sijing, ZHANG Deyu, et al.
2020, 29(5): 793-811.   doi: 10.1049/cje.2020.07.001
Abstract(310) PDF(465)
The Principle and Progress of Dynamically Reconfigurable Computing Technologies
WEI Shaojun, LU Yan'an
2020, 29(4): 595-607.   doi: 10.1049/cje.2020.05.002
Abstract(224) PDF(545)
Spatial Sound—History, Principle, Progress and Challenge
XIE Bosun
2020, 29(3): 397-416.   doi: 10.1049/cje.2020.02.016
Abstract(146) PDF(484)
Review on the Technological Development and Application of UAV Systems
FAN Bangkui, LI Yun, ZHANG Ruiyu, et al.
2020, 29(2): 199-207.   doi: 10.1049/cje.2019.12.006
Abstract(90) PDF(554)
Survey of Performance Evaluation Standardization and Research Methods on GNSS-Based Localization for Railways
CAI Baigen, WU Boqian, LU Debiao
2020, 29(1): 22-33.   doi: 10.1049/cje.2019.09.003
Abstract(42) PDF(858)