2021 Vol. 30, No. 4

COMPUTERS AND MICROELECTRONICS
Marginal Attacks of Generating Adversarial Examples for Spam Filtering
GU Zhaoquan, XIE Yushun, HU Weixiong, YIN Lihua, HAN Yi, TIAN Zhihong
2021, 30(4): 595-602. doi: 10.1049/cje.2021.05.001
Abstract(850) PDF(98)
Abstract:
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, WANG Lijun, ZHAI Yun, JIANG Wei
2021, 30(4): 603-610. doi: 10.1049/cje.2021.05.002
Abstract(697) PDF(51)
Abstract:
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(603) PDF(58)
Abstract:
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, SUN Ling, CHENG Jiamin, WANG Yutian
2021, 30(4): 623-633. doi: 10.1049/cje.2021.05.004
Abstract(661) PDF(39)
Abstract:
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, WANG Yanlin, XU Liuming, LU Huixia, XU Chenhua
2021, 30(4): 634-643. doi: 10.1049/cje.2021.05.005
Abstract(689) PDF(42)
Abstract:
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, WANG Haibo, XU Yihan, BIAN Yuetang
2021, 30(4): 644-651. doi: 10.1049/cje.2021.05.006
Abstract(659) PDF(484)
Abstract:
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.
Keyword Extraction from Scientific Research Projects Based on SRP-TF-IDF
WANG Zhuohao, WANG Dong, LI Qing
2021, 30(4): 652-657. doi: 10.1049/cje.2021.05.007
Abstract(658) PDF(23)
Abstract:
Keyword extraction by Term frequency-Inverse document frequency (TF-IDF) is used for text information retrieval and mining in many domains, such as news text, social contact text, and medical text. However, keyword extraction in special domains still needs to be improved and optimized, particularly in the scientific research field. The traditional TF-IDF algorithm considers only the word frequency in documents, but not the domain characteristics. Therefore, we propose the Scientific research project TF-IDF (SRP-TF-IDF) model, which combines TF-IDF with a weight balance algorithm designed to recalculate candidate keywords. We have implemented the SRP-TF-IDF model and verified that our method has better precision, recall, and F1 score than the traditional TF-IDF and TextRank methods. In addition, we investigated the parameter of our weight balance algorithm to find an optimal value for keyword extraction from scientific research projects.
A Multi-Granularity Density Peak Clustering Algorithm Based on Variational Mode Decomposition
GU Ziwen, LI Peng, LANG Xun, YU Yixuan, SHEN Xin, CAO Min
2021, 30(4): 658-668. doi: 10.1049/cje.2021.03.001
Abstract(626) PDF(22)
Abstract:

Density peak clustering (DPC) can identify cluster centers quickly, without any prior knowledge. It is supposed that the cluster centers have a high density and large distance. However, some real datasets have a hierarchical structure, which will result in local cluster centers having a high density but a smaller distance. DPC is a flat clustering algorithm that searches for cluster centers globally, without considering local differences. To address this issue, a Multi-granularity DPC (MG-DPC) algorithm based on Variational mode decomposition (VMD) is proposed. MG-DPC can find global cluster centers in the coarse-grained space, as well as local cluster centers in the fine-grained space. In addition, the density is difficult to calculate when the dataset has a high dimension. Neighborhood preserving embedding (NPE) algorithm can maintain the neighborhood relationship between samples while reducing the dimensionality. Moreover, DPC requires human experience in selecting cluster centers. This paper proposes a method for automatically selecting cluster centers based on Chebyshev’s inequality. MG-DPC is implemented on the dataset of load-data to realize load classification. The clustering performance is evaluated using five validity indices compared with four typical clustering methods. The experimental results demonstrate that MG-DPC outperforms other comparison methods.

Differential Fault Attack on GIFT
XIE Min, TIAN Feng, LI Jiaqi
2021, 30(4): 669-675. doi: 10.1049/cje.2021.05.008
Abstract(690) PDF(28)
Abstract:

GIFT, a lightweight block cipher proposed at CHES2017, has been widely cryptanalyzed this years. This paper studies the differential diffusion characteristics of round function of GIFT at first, and proposes a random nibble-based differential fault attack. The key recovery scheme is developed on the statistical properties we found for the differential distribution table of the S-box. A lot of experiments had been done and experimental results show that one round key can be retrieved with an average of 20.24 and 44.96 fault injections for GIFT-64 and GIFT-128 respectively. Further analysis shows that a certain number of fault injections recover most key bits. So we demonstrate an improved fault attack combined with the method of exhaustive search, which shows that the master key can be recovered by performing 216 and 217 computations and injecting 31 and 32 faults on an average for GIFT-64 and GIFT-128 respectively.

A Software Safety Requirements Elicitation Approach Based on Hazardous Control Action Tree Modelling
ZHU Danjiang, YAO Shuzhen, TAN Huobin
2021, 30(4): 676-685. doi: 10.1049/cje.2021.05.009
Abstract(491) PDF(13)
Abstract:

Software safety requirements are crucial for safety assurance of safety-critical software systems. A novel accident causality model, Systems-theoretic accident modeling and processes (STAMP), has been proposed to overcome the limitations of traditional safety techniques in software safety requirements elicitation. However, the STAMP-based method is ad-hoc with no rigorous procedure to elicit software safety requirements effectively. Furthermore, the time-related safety requirements, which are important to software safety, have been paid little attention in STAMP-based method. With the purpose of overcoming these limitations, this paper strives to find a systematic approach to elicit software safety requirements with STAMP, especially the time-related safety requirements. Firstly, a new process model of STAMP is proposed to model all the system varilables and the ralationship of them in control processes. Then based on the process model, an approach HCAT-SSRA is proposed to elicit the software safety requirements by building Hazardous control action tree (HCAT) for each control action in system control processes. Additionally, several rules are proposed to guide time-related software safety requirements analysis. Finally, a case study is given to illustrate the availability and feasibility of the proposed method.

TAKAP: A Lightweight Three-Party Authenticated Key Agreement Protocol with User Anonymity
QIAO Hui, DONG Xuewen, SHEN Yulong, LIU Haozhe, YANG Lingxiao
2021, 30(4): 686-696. doi: 10.1049/cje.2021.05.010
Abstract(477) PDF(9)
Abstract:

The three-party authenticated key agreement protocol is a significant cryptographic mechanism for secure communication, which encourages two entities to authenticate each other and generate a shared session key with the assistance of a trusted party (remote server) via a public channel. Recently, Wang et al. put forward a three-party key agreement protocol with user anonymity and alleged that their protocol is able to resist all kinds of attacks and provide multifarious security features in Computer Engineering & Science, No.3, 2018. Unfortunately, we show that Wang et al.’s protocol is vulnerable to the password guessing attack and fails to satisfy user anonymity and perfect secrecy. To solve the aforementioned problems, a lightweight chaotic map-based Three-party authenticated key agreement protocol (short for TAKAP) is proposed, which not only could provide privacy protection but also resist a wide variety of security attacks. Furthermore, it is formally proved under Burrows-Abadi-Needham (BAN) logic. Simultaneously, the performance analysis in this paper demonstrates that the proposed TAKAP protocol is more secure and efficient compared with other relevant protocols.

Message Authentication Codes Against Related-Key Attacks Under LPN and LWE
CUI Nan, LIU Shengli
2021, 30(4): 697-703. doi: 10.1049/cje.2021.05.011
Abstract(525) PDF(12)
Abstract:

Message authentication code (MAC) guarantees the authenticity of messages and is one of the most important primitives in cryptography. We study related-key attacks with which the adversary is able to choose function f and observe the behavior of the MAC under the modified authenticated key f(k), and consider unforgeability of MAC under (selectively) chosen message attack with f(k). We focus on MAC schemes from the Learning parity with noise (LPN) and the Learning with errors (LWE) problem by Kiltz et al. in EUROCRYPT 2011. We first prove that the MAC schemes from LPN/ LWE can resist key-shift attacks and enlarge the key-shift function set to support a subclass of affine functions.

Essential Protein Prediction Based on Shuffled Frog-Leaping Algorithm
YANG Xiaoqin, LEI Xiujuan, ZHAO Jie
2021, 30(4): 704-711. doi: 10.1049/cje.2021.05.012
Abstract(418) PDF(19)
Abstract:

Essential proteins are integral parts of living organisms. The prediction of essential proteins facilitates to discover disease genes and drug targets. The prediction precision and robustness of most of existing identification methods are not satisfactory. In this paper, we propose a novel essential proteins prediction method (EPSFLA), which applies Shuffled frog-leaping algorithm (SFLA), and integrates several biological information with network topological structure to identify essential proteins. Specifically, the topological property and several biological properties (function annotation, subcellular localization, protein complex, and orthology) are integrated and utilized to weight protein-protein interaction networks. Then the position of a frog is encoded and denotes a candidate essential protein set. The frog population continuously evolve by means of local exploration and global exploration until termination criteria for algorithm are satisfied. Finally, those proteins contained in the best frog are regarded as predicted essential proteins. The experimental results show that EPSFLA outperforms some well-known prediction methods in terms of various criteria. The proposed method aims to provide a new perspective for essential protein prediction.

Verifiable Multi-Party Universal Blind Quantum Computing in Distributed Networks
SHAN Ruiting, CHEN Xiubo, XU Gang, DOU Zhao, LI Zongpeng, MIAO Lihua, YANG Yixian
2021, 30(4): 712-718. doi: 10.1049/cje.2021.05.013
Abstract(466) PDF(17)
Abstract:

Blind quantum computing (BQC) ensures that a classical client could delegate complex computing tasks to a remote quantum server safely. In order to detect the dishonest behavior of the participants, we present a verifiable multi-party universal BQC protocol in distributed networks. By using the stabilizer formalism, we propose an honesty check method to test the correctness of the graph states generated by the servers. The honesty of both the clients and the servers can be judged fairly with the help of the arbitrator. Moreover, a load balancer is introduced to control the possible breakdown of servers in the network. No-signaling principle ensures the unconditional security of the protocol. Through the use of universal resource states, our protocol can be applied in more multi-party verifiable universal BQC protocols. The failure management and workload balancing of the protocol also provide a promising prospect for the future BQC network.

SIGNAL PROCESSING
Detection and Classification of Small Traffic Signs Based on Cascade Network
ZHANG Shufang, WANG Qinyu, ZHU Tong, LIU Yuhong
2021, 30(4): 719-726. doi: 10.1049/cje.2021.05.014
Abstract(501) PDF(21)
Abstract:

Research on the traffic sign detection is significant for driverless technology, which provides useful navigation information. Existing object detection methods are only applicable to large-size objects or small-scale specific types of traffic signs, and the performance of detecting traffic signs in street views is not adequate. In this regard, we propose a method to detect and classify small traffic signs by constructing a cascaded network. Specifically, the RetinaNet network is adopted firstly to integrate multi-layer information to identify small traffic signs in traffic scene images. The focal loss function is used to balance the biased distribution of traffic sign categories. Then, a two-class network is cascaded after the RetinaNet, which helps identify valid traffic signs from the first-stage prediction results. Experiments show that our cascaded network structure could achieve the balance of different categories of predictions and an improvement in precision and recall.

Segment-Adaptive Spread Spectrum Audio Watermarking Using NSGA-II
SU Zhaopin, ZHANG Guofu, ZHOU Xianxian, LI Wangwang
2021, 30(4): 727-735. doi: 10.1049/cje.2021.05.015
Abstract(401) PDF(20)
Abstract:

In the field of robust audio watermarking, how to seek a good trade-off between robustness and imperceptibility is challenging. The existing studies use the same embedding parameter for each part of the audio signal, which ignores that different parts may have different requirements for embedding parameters. In this work, the constraints on imperceptibility are first analysed. Then, we present a segment multi-objective optimization model of the scaling parameter under the constrained Signal-to-noise ratio (SNR) in Spread spectrum (SS) audio watermarking. Additionally, we adopt the Nondominated sorting genetic algorithm II (NSGA-II) to solve the proposed model. Finally, we compare our algorithm (called SS-SNR-NSGA-II) with the existing methods. The experimental results show that the proposed SS-SNRNSGA-II not only provides flexible choices for different application demands but also achieves more and better trade-offs between imperceptibility and robustness.

A Reframed Watermark Extraction Approach Using the ABC Algorithm
KRICHA Anis, KRICHA Zied, SAKLY Anis
2021, 30(4): 736-742. doi: 10.1049/cje.2021.05.016
Abstract(729) PDF(557)
Abstract:

The fast development and the rapid spread of information technology have raised the issue of digital-image copyright protection, which was mostly addressed through robust watermarking. To ensure the survival of the watermark after attacks, most works have been making use of an additional step at the embedding phase, and by so making the scheme unsuitable for real-time implementation. However, to be effective and secure, it is recommended to combine the watermarking algorithm and the capture sensor in one device. In this paper, we brought the extra stage to the detection side. We firstly studied the effect of some signal processing operations on a watermarked image, then we proposed a general model for watermark extraction whose parameters were determined using the artificial bee colony. The efficacy of the proposed method was validated by comparing its robustness with two state-of-art schemes focusing on the watermark extraction.

TELECOMMUNICATIONS
Objective Information Theory Exemplified in Air Traffic Control System
XU Jianfeng, WANG Shuliang, LIU Zhenyu, WANG Yingfei
2021, 30(4): 743-751. doi: 10.1049/cje.2021.05.017
Abstract(581) PDF(19)
Abstract:

Information is the core of Air traffic control system (ATCS). In this paper, objective information theory is extended to depict, model and measure the information exemplified in ATCS. The sextuple model is presented with information ontology, state occurrence time, state set, carrier, reflection time, and reflection set. The metric system is given to quantitively measure the information with extensity, detailedness, continuity, richness, containability, delay, pervasiveness, authenticity, and adaptability. The results show that the proposed method is potential to find out actual indexes in ATCS for flight safety.

Distribution of Reference Stations: Initial Explore on the Optimal Selection Strategy
BAO Junjie, LI Rui, SHAO Bo, HUANG Zhigang
2021, 30(4): 752-758. doi: 10.1049/cje.2021.05.018
Abstract(559) PDF(17)
Abstract:

The Satellite-based augmentation system (SBAS) is intended to provide real-time differential global navigation satellite system corrections with the high accuracy, availability, and integrity required for aviation applications. Since the performance of Satellite clock and ephemeris (SCE) corrections and Ionospheric range delay (IRD) corrections can vary dramatically depending on satellites and Ground reference stations (GRSs) geometry, therefore, we present a GRSs distribution optimized criteria and process to improve SBAS corrections performance. The present step-by-step optimized scheme using the average satellite surveillance dilution of precision and relative centroid metric availability of grid points as fitness values to determine the appropriate GRSs distribution to sufficiently meet the corrections requirements. The results show that the statistical mean RCM availability can reach more than 0.5518 for all IGPs and the coverage depth of GRSs in China and its surrounding areas is more than 25, which fully satisfies the requirement for solving SCE and IRD corrections.

Secure Localization Technology Based on Dynamic Trust Management in Wireless Sensor Networks
LI Peng, YU Xiaotian, XU He, WANG Ruchuan
2021, 30(4): 759-768. doi: 10.1049/cje.2021.05.019
Abstract(602) PDF(40)
Abstract:

Wireless sensor networks have critical applications in various fields, and the algorithm of their secure localization has become a vital technology to support a network. In the light of the self-organization, random deployment and dynamic topology, the localization process is vulnerable to various kinds of malicious attacks. The model of dynamic trust management for a given node is proposed to deal with security concerns in wireless sensor networks. The trust computation is divided into three stages, which are the stage of trust initialization, trust establishment, and trust evolution. The initial value of a global trust relationship is established through a corresponding global trust relation graph in the initial stage of trust. The trust value of each node is calculated by the attribute value in the stage of trust establishment. In the evolution of trust, the iterative process of trust value is accelerated via the finite state machine. Compared with the existing wireless sensor networks, simulation results show that the proposed security localization technology method can resist many kinds of attacks with low communication and time consumption.

A Multicarrier Phase-Coded Waveform Design Scheme for Joint Radar and Communication System
DU Zhen, ZHANG Zenghui, YU Wenxian
2021, 30(4): 769-780. doi: 10.1049/cje.2021.05.020
Abstract(452) PDF(22)
Abstract:

A Multicarrier phase-coded (MCPC) waveform design scheme with two steps for Joint radar and communication (JRC) system is developed. Firstly, an integrated MCPC waveform design method is addressed by simultaneously maximizing the Signal-to-clutter-to-noise ratio (SCNR) and the Shannon capacity, subject to both the Integrated sidelobe level ratio (ISLR) constraint and the energy constraint. This model is theoretically proved to be a convex optimization problem with respect to the absolute squares of the transmit weights corresponding to different subcarriers, of which the analytical result is also discussed. Subsequently, by further optimizing the phases of the transmit weights, minimizing the Peak to average power ratio (PAPR) for JRC system is recast as a Semidefinite programming (SDP) problem, which can be effectively solved with the Semidefinite relaxation (SDR) technique via Eigenvalue decomposition (EVD) or Complex Gaussian randomization (CGR). Numerical examples are provided to verify the effectiveness of the proposed scheme.

Coherent Integration Algorithm for Frequency-Agile and PRF-Jittering Signals in Passive Localization
JIANG Hongzhi, ZHAO Chuang, ZHAO Yongjun
2021, 30(4): 781-792. doi: 10.1049/cje.2021.06.001
Abstract(637) PDF(23)
Abstract:

Increasing pulses Coherent processing interval (CPI) can effectively improve the location parameters estimation performance in passive localization. However, for a moving emitter transmitting pulses with Frequency agile and Pulse repetition frequency jittering (FA-PRFJ) in a CPI, there will exist random phase, uneven sampling and Range migration (RM), which deteriorates the estimation performance of location parameters. Aiming at long-time coherent localization parameters estimation for the above emitter, this paper proposes a joint Range difference (RD) and Range rate difference (RRD) estimation algorithm. Firstly, the signal model of a moving emitter transmitting FA-PRFJ signal is constructed, and the influence of the FA and PRFJ on coherent integration is analyzed. Secondly, the random phase induced by FA is eliminated by frequency symmetric autocorrelation function operation. Then, RD and RRD can be coherently estimated after RM correction via the modified scaled non-uniform fast Fourier transform. This method can be efficiently implemented by Fast Fourier transform (FFT), inverse FFT and FFT-based chirp-z transform without any searching operation. Simulation results demonstrate that the proposed method has a better antinoise capability with a much lower computational complexity compared with several representative methods.