Loading...
Office

Table of Content

    20 August 2019 Volume 46 Issue 4
      
    Efficient alarm analysis approach for communication networks
    QI Xiaogang,HU Qiuqiu,YAO Xuqing,LIU Lifang
    Journal of Xidian University. 2019, 46(4):  1-8.  doi:10.19665/j.issn1001-2400.2019.04.001
    Abstract ( 877 )   HTML ( 102 )   PDF (1455KB) ( 343 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Considering the irrationality of this hypothesis that previous most alarm analysis studies were based on the premise that all alarms in the communication network are equal, a weighted alarm analysis method is proposed. First, according to the significance of the alarm in the network, the entropy method is used to assign different weight values for different alarms, and these values are transformed into a sequence dataset suitable for data mining. Then, a weighted alarm sequential pattern mining algorithm is proposed which uses an original pruning strategy to reduce the size of dataset needing to be mined to improve the efficiency of the algorithm. Finally, the algorithm is used to mine sequential relationship in alarm data. Experimental results show that this weighted alarm analysis approach has a good performance in pruning effectiveness, mining of the important alarm sequential pattern and execution efficiency.

    Pilot contamination suppression in 3D massive MIMO systems
    LIU Can,LI Yongzhao,ZHANG Rui,LI Tao
    Journal of Xidian University. 2019, 46(4):  9-15.  doi:10.19665/j.issn1001-2400.2019.04.002
    Abstract ( 459 )   HTML ( 54 )   PDF (1007KB) ( 133 )   Save
    Figures and Tables | References | Related Articles | Metrics

    To tackle the pilot contamination (PC) problem in three-dimensional (3D) massive multiple-input multiple-output (MIMO) systems, a soft pilot reuse (SPR) based pilot contamination suppression scheme is proposed. By exploiting the elevation degree of freedom (DoF), the users with separable or inseparable elevation channel state information (CSI) among the adjacent cells are regarded as center users or edge users, respectively. The edge users in the adjacent cells are assigned orthogonal pilot sequences to eliminate the PC, while the center users in all cells reuse the pilot sequences and suppress the PC by the improved channel estimate method. Thus, the channel estimation accuracy of all the users in the cell can be improved. Simulation results demonstrate that, compared to the conventional pilot allocation schemes, the proposed scheme can decontaminate pilots more effectively and achieve a better system performance with a low complexity.

    New two-step semidefinite relaxation method for acoustic energy-based localization
    TIAN Qiang,FENG Dazheng,LI Jin,HU Haoshuang
    Journal of Xidian University. 2019, 46(4):  16-21.  doi:10.19665/j.issn1001-2400.2019.04.003
    Abstract ( 533 )   HTML ( 40 )   PDF (713KB) ( 122 )   Save
    Figures and Tables | References | Related Articles | Metrics

    A new two-step semidefinite relaxation method is proposed to deal with the nonlinear and non-convex problem of acoustic energy-based localization in wireless sensor networks. The proposed algorithm transforms the nonlinear positioning equations into a weighted least squares estimation problem of the unknown source location and signal transmit power, which is then solved in two steps. First, the signal transmit power is eliminated from the cost function by expressing it as a function of the source position in the least square sense. In the second step, the weighted least squares formulation is converted into a semidefinite programming(SDP) optimization problem by using a new convex relaxation technique. The tightness of the semidefinite relaxation method is theoretically proved. Simulation results indicate that compared with the previous methods, the proposed algorithm has a higher localization accuracy, especially when the measurement error is relatively large.

    Improved algorithm for hyperspectral endmember extraction and its FPGA implementation
    ZHANG Jintao,LEI Jie,WU Lingyun,HUANG Biying,LI Yunsong
    Journal of Xidian University. 2019, 46(4):  22-27.  doi:10.19665/j.issn1001-2400.2019.04.004
    Abstract ( 419 )   HTML ( 30 )   PDF (753KB) ( 105 )   Save
    Figures and Tables | References | Related Articles | Metrics

    The automatic target generation process (ATGP) is unsuitable for being deployed for high speed hardware implementation due to enormously complex inverse operations and the increasing scale of iterative operation problems. On the basis of an intensive study of the ATGP algorithm, a novel implementation framework is proposed in this paper, which takes advantage of simple regular matrix operations instead of increasingly complicated matrix inversion to update the orthogonal projection operator matrix. Furthermore, an accelerated design of the algorithm is implemented on FPGA by using high-level synthesis (HLS) for the first time. Experimental results demonstrate that the processing time achieved by the FPGA implementation is strictly real-time while retaining the same high detection accuracy.

    Analytical solution and verification of the workspace of the planar cable-driven mechanism
    HE Shikai,DUAN Qingjuan,LI Hao,ZHAO Ping
    Journal of Xidian University. 2019, 46(4):  28-34.  doi:10.19665/j.issn1001-2400.2019.04.005
    Abstract ( 308 )   HTML ( 17 )   PDF (1760KB) ( 65 )   Save
    Figures and Tables | References | Related Articles | Metrics

    In order to find the wrench closed workspace boundary of the Planar Multibody Cable-Driven Mechanism, a matrix null space method is proposed. By analyzing the cable force Jacobian matrix, the condition for guaranteeing non-negative cable forces are derived, and the boundary of wrench closed workspace and wrench feasible workspace are solved according to the condition. The Monte Carlo method is used to optimize the the position of the motor and cable connection point to maximize the wrench feasible workspace and ensure that the trajectory of the mechanism is within the wrench feasible workspace. Parameter optimization and experimental verification of a configuration of the planar serial two-link mechanism show that the optimal configuration guarantees the desired trajectory in the wrench feasible workspace, and that the analytical method can accurately derive the boundary of the wrench closure workspace.

    Method for bridge crack detection based on the U-Net convolutional networks
    ZHU Suya,DU Jianchao,LI Yunsong,WANG Xiaopeng
    Journal of Xidian University. 2019, 46(4):  35-42.  doi:10.19665/j.issn1001-2400.2019.04.006
    Abstract ( 688 )   HTML ( 45 )   PDF (2660KB) ( 139 )   Save
    Figures and Tables | References | Related Articles | Metrics

    In order to improve the accuracy of bridge crack detection, retain details, and get information on the crack width, the paper proposes a pixel-wise and small sample crack detection method by using U-Net convolutional neural networks. The method uses a U-Net network to extract crack features automatically by using multi-layer convolutions, and uses the superposition of the shallow network and deep network to realize the fusion of local features and abstract features of cracks. This method can retain the details of cracks and greatly improve the accuracy of detection. In order to refine the detection results, the paper presents the threshold method and an improved Dijkstra minimum spanning tree algorithm for eliminating noise and pseudo cracks. Finally, an eight-direction searching method is applied to measure the width of cracks in pixels. Experiments prove that the proposed method can accurately and completely detect bridge cracks and measure the width, which can meet the application requirements.

    Method for design and analysis of the cable net structure for a hoop/column antenna
    LIU Yang,LI Tuanjie,CHEN Congcong,TANG Yaqiong
    Journal of Xidian University. 2019, 46(4):  43-48.  doi:10.19665/j.issn1001-2400.2019.04.007
    Abstract ( 579 )   HTML ( 24 )   PDF (2274KB) ( 106 )   Save
    Figures and Tables | References | Related Articles | Metrics

    In order to solve the difficult problem of the design of the complex structure of the hoop/column antenna cable net and the fpretension, the geometric configuration of the hoop/column antenna cable net is calculated according to the structure of the hoop/column antenna. Considering the complexity of the design problem with too many design variables, the form-finding design of cable net structure is divided into two parts, one for the front net and the other for the rear net. First, the design variables of the front net are merged according to the symmetry of the cable net structure. Then, the pretension of the front cable net is calculated using the force density method and some nodal coordinates of the cable net are updated. Next, the pretension of the remaining cables is calculated according to the nodal force balance equations and the previously obtained pretensions. Design results are verified by finite element software. Simulation results indicate that the mesh reflector designed by the method this paper describes has a high precision and that the cable net structure design method is effective.

    Ciphertext-policy attribute-based encryption scheme with verifiability on authority
    YU Jinxia,HE Xu,YAN Xixi
    Journal of Xidian University. 2019, 46(4):  49-57.  doi:10.19665/j.issn1001-2400.2019.04.008
    Abstract ( 348 )   HTML ( 26 )   PDF (1249KB) ( 59 )   Save
    Figures and Tables | References | Related Articles | Metrics

    To address the problem of how to authenticate a valid authority in the real life, a new ciphertext-policy attribute-based encryption scheme with verifiability on authority is proposed. By employing ring signature technology, multiple authorities are organized into a valid organization, and the value of ring signature which can be used to prove the legitimacy of each authority is reasonably embedded in the user's private key, so that users are able to verify the validity of the authority. In addition, the linear secret sharing scheme is adopted to support any monotone access structures in our scheme, and it is proved to be full security based on the difficult assumptions on the composite order group under the standard model. Finally, compared with some related attribute-based encryption schemes, the proposed scheme has certain performance advantages in terms of ciphertext size, user's private key size, bilinear pairing calculation and so on. Therefore, it is more suitable for cloud environment and resource-limited mobile devices.

    IoT Node trust comprehensive evaluation model
    XIE Lixia,WEI Ruixin
    Journal of Xidian University. 2019, 46(4):  58-65.  doi:10.19665/j.issn1001-2400.2019.04.009
    Abstract ( 583 )   HTML ( 50 )   PDF (1557KB) ( 113 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Aiming at the low accuracy and capability to deal with malicious nodes of the Internet of Things(IoT) trust evaluation model, an IoT node comprehensive trust evaluation model (INCTEM) is proposed. First, the evaluation index of the same quality service strength is introduced to reduce the non-intrusive impact on trust evaluation. Second, the recommended node’s reliability is evaluated from three aspects: node similarity, difference and recommended node’s trust value. Then the recommended node’s reliability is used as the recommendation trust weight. Finally, the adaptive weights of direct and recommended trust are calculated to compute the comprehensive trust value. Experimental results show that the INCTEM has advantages in dealing with malicious services and malicious recommendation behaviors, and can reduce the transmission energy consumption while ensuring the validity of trust evaluation.

    Numerical models of short-circuit failure for field-effect transistors
    ZHOU Yuming, JIANG Baoguo, CHEN Zhaoquan, WANG Bing
    Journal of Xidian University. 2019, 46(4):  66-73.  doi:10.19665/j.issn1001-2400.2019.04.010
    Abstract ( 382 )   HTML ( 15 )   PDF (2078KB) ( 104 )   Save
    Figures and Tables | References | Related Articles | Metrics

    To analyze the failure mechanisms of the silicon carbide field-effect transistor (SiC MOSFET) and silicon field-effect transistor (Si MOSFET) under the short-circuit condition, numerical models for the silicon carbide field-effect transistor and silicon field-effect transistor have been built by technology computer aided design (TCAD), which can replicate the short-circuit failure of the silicon carbide field-effect transistor and silicon field-effect transistor. In the numerical models, the self-heating effect has been introduced to simulate the temperature change and thermal transmission inside the device, and the Fowler-Nordheim tunneling effect and Poole-Frenkel emission simulate the leakage current of the gate oxide. Experimental results have verified the numerical models. With the numerical models, the change of the gate driving voltage, the distribution of current flowlines and the temperature inside the two devices have been compared. The results have shown that the short-circuit failure of the silicon carbide field-effect transistor originates from the metal melting of the gate surface electrode and serious oxide degradation. By contrast, the short-circuit failure for the silicon field-effect transistor comes from the triggering of the parasitic bipolar transistor. As a result, the leakage current inside the silicon field-effect transistor is out of control, which leads to the catastrophic damage.

    Method for diagnosis of data-driven GMC sparse enhancement
    CHEN Baojia,HE Wangpeng,HU Jie,WANG Geng,GUO Baolong
    Journal of Xidian University. 2019, 46(4):  74-79.  doi:10.19665/j.issn1001-2400.2019.04.011
    Abstract ( 375 )   HTML ( 13 )   PDF (1291KB) ( 53 )   Save
    Figures and Tables | References | Related Articles | Metrics

    In mechanical fault diagnosis, to address the problem that the weak fault features extracted by traditional methods are easily disturbed by strong background noise and have a low accuracy, a data-driven sparse features extraction method using the generalized minimax-concave penalty is developed. This method constructs an effective sparse optimization objective function for mechanical fault diagnosis in order to improve the accuracy of fault feature extraction. It is also proved that the non-convex controllable parameters can guarantee the overall convexity of the objective function under certain constraints. The proximal algorithm is used to solve the unconstrained optimization problem. In addition, the data-driven regularization parameter setting criteria are studied to ensure that the proposed sparse feature extraction method has parameter adaptability. Finally, simulation results and practical fault experiment verify the effectiveness of the proposed method in the machinery fault diagnosis.

    Low-complexity multi-user detection algorithm for an SCMA system
    ZHU Cuitao,WU Bei
    Journal of Xidian University. 2019, 46(4):  80-86.  doi:10.19665/j.issn1001-2400.2019.04.012
    Abstract ( 361 )   HTML ( 26 )   PDF (821KB) ( 59 )   Save
    Figures and Tables | References | Related Articles | Metrics

    To further reduce the complexity of multi-user detection algorithm for a sparse code multiple access system, a multi-user detection algorithm applied Gaussian approximation based on partially resource nodes is proposed. We first compare the superiority levels of resource nodes, then selectresource nodes with higher superiority and apply the weighted message passing algorithm to those nodes. For the remaining resource nodes, the Gaussian approximation is performed. Meanwhile, we develop a method for combining the superiority levels of resource nodes and users, and directly decode and remove the users with a higher superiority level after each iteration, so the complexity of subsequent iteration is decreased gradually. Simulation results show that, by rationally selecting the resource nodes, the proposed algorithm can guarantee the detection performance and effectively reduce the detection complexity, thus achieving a better tradeoff between the decoding performance and complexity.

    Resource scheduling strategy in hierarchical software defined wireless sensor networks
    ZHEN Yan,ZHAO Hu
    Journal of Xidian University. 2019, 46(4):  87-98.  doi:10.19665/j.issn1001-2400.2019.04.013
    Abstract ( 320 )   HTML ( 17 )   PDF (2111KB) ( 56 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Aiming at the problem of low resource utilization, poor energy efficiency and inability to adapt to the dynamic changes of network topology in task-specific traditional wireless sensor networks(WSN), this paper proposes a hierarchical software-defined wireless sensor networks (SDWSN) resource scheduling strategy for the multi-task concurrent scenario. In this strategy, software-defined networking(SDN) is utilized to implement the decoupling between control layer and data layer, multiple tasks can be accomplished simultaneously by flexible network resource scheduling , and the network total energy consumption can be minimized under the premise of ensuring the sensing quality. Besides, the software-defined master node can obtain the changes of network topology in time, and then an intra-cluster resource scheduling strategy is implemented by cluster heads, which improves optimization efficiency and reduces energy consumption. The results show that the proposed global network resource scheduling strategy can improve energy efficiency and network resource utilization. Moreover, the intra-cluster resource scheduling strategy can reduce the control overhead of the master node while solving the network dynamic events efficiently.

    Multi-view fuzzy clustering algorithm using FCS
    LIU Yongli,GUO Chengyi,LIU Jing,WU Yan
    Journal of Xidian University. 2019, 46(4):  99-106.  doi:10.19665/j.issn1001-2400.2019.04.014
    Abstract ( 406 )   HTML ( 91 )   PDF (1457KB) ( 60 )   Save
    Figures and Tables | References | Related Articles | Metrics

    By synthesizing different representations of data, multi-view fuzzy clustering can produce more comprehensive and macroscopic clustering results. However it is vulnerable to noise. In order to improve the ability to resist noise, a multi-view fuzzy clustering algorithm is proposed which, inheriting the advantages of multi-view clustering and fuzzy compactness and separation clustering, can collaborate clustering according to the importance of different views and enhance robustness. In order to validate the effectiveness of this algorithm, four multi-view data sets are selected to carry out experiments. Experimental results show that this algorithm can not only achieve high clustering accuracy, but also effectively reduce the impact of noise data on clustering results.

    Deep learning algorithm for the segmentation of the interested region of an infrared thermal image
    ZHU Li,ZHAO Jun,FU Yingkai,ZHANG Jing,SHEN Hui,ZHANG Shoufeng
    Journal of Xidian University. 2019, 46(4):  107-114.  doi:10.19665/j.issn1001-2400.2019.04.015
    Abstract ( 698 )   HTML ( 33 )   PDF (2054KB) ( 219 )   Save
    Figures and Tables | References | Related Articles | Metrics

    To tackle difficulties of the segmentation of the interested region in complex background, a deep-learning segmentation algorithm based on the fully convolutional network and the dense conditional random field is proposed. First, the fully convolutional network is leveraged for pixel-level feature extraction to obtain the coarse segmentation result. Then, the dense conditional random field which is used to optimize the context information is performed for detailed segmentation. Five-fold cross-validation experiments are carried out on an actually acquired infrared thermal image of the solar panel. Experimental results show that the proposed algorithm has an average precision rate of 89.96%, an average recall rate of 94.55%, an average F1 index of 0.9118 and an average J index of 0.8687. At the same time, the algorithm achieves the best maximum precision rate of 93.35%, a maximum recall rate of 97.59%, a maximum F1 index of 0.9562 and a maximum J index of 0.9125 compared with those by main existing algorithms. Moreover, this method takes less time and requires less manual interference. In conclusion, the proposed algorithm is capable of the segmentation of the interested region in the infrared thermal image effectively in the complex background.

    Representation method for the compressed full-attitude quaternion
    LIU Gongxu,YU Baoguo,SHI Lingfeng,LIU Hong,HE Rui,DONG Yajun
    Journal of Xidian University. 2019, 46(4):  115-121.  doi:10.19665/j.issn1001-2400.2019.04.016
    Abstract ( 348 )   HTML ( 15 )   PDF (2651KB) ( 33 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Due to the double spreadability of the quaternion for attitude representation or the non-one-to-one-mapping between the attitude and the quaternion, an attitude distraction issue may occur. A method for representing the compressed full-attitude quaternion is proposed in this paper to solve the issue mentioned above. First, two concepts of compressed attitude-space and compressed quaternion-space are introduced, and then the one-to-one mapping relationship between them is derived based on the East-North-Day coordinate system. Finally, the effectiveness of the proposed method is verified by numerical simulation experiments. The simulation results show that the proposed method not only inherits the non-singularity advantages of the quaternion for attitude representation, but also avoids the problem of attitude distraction caused by double spreadability.

    Method for estimation of the two-dimensional direction of wave arrival using the rectangular array
    WANG Jianshu,FAN Yangyu,DU Rui,LV Guoyun
    Journal of Xidian University. 2019, 46(4):  122-129.  doi:10.19665/j.issn1001-2400.2019.04.017
    Abstract ( 410 )   HTML ( 81 )   PDF (1607KB) ( 82 )   Save
    Figures and Tables | References | Related Articles | Metrics

    To improve the performance of existing two-dimensional (2-D) grid-less irection of arrival(DOA) estimation methods using the uniform rectangular array(URA) or sparse rectangular array(SRA), a novel 2-D grid-less DOA estimation method based on doubly Toeplitz matrix reconstruction and 2-D ESPRIT is proposed. First, using URA or SRA, the doubly Toeplitz structure of the associated covariance matrix is established. Second, by applying the log-det sparse metric and semi-definite positive constraints, the constrained optimization problem is presented and solved by the majorization-minimization (MM) algorithm. Finally, the azimuth angles and elevation angles are estimated by the 2-D ESPRIT method. The proposed method needs to solve semi-definite programming (SDP) problems repeatedly, which results in a high complexity, while it always provides a superior performance of DOA estimation. In simulations, the proposed method has a very small root-mean-square error (RMSE) in the case of URA and SRA, which can approach the Crammer-Rao bound. Simulation results prove the good performance of the proposed method.

    Speech enhancement method based on the time-frequency smoothing deep neural network
    YUAN Wenhao,LIANG Chunyan,LOU Yingxi,FANG Chao,WANG Zhiqiang
    Journal of Xidian University. 2019, 46(4):  130-136.  doi:10.19665/j.issn1001-2400.2019.04.018
    Abstract ( 379 )   HTML ( 20 )   PDF (984KB) ( 56 )   Save
    Figures and Tables | References | Related Articles | Metrics

    In the existing speech enhancement methods based on the deep neural network, the characteristics of speech enhancement problem are not fully considered in the design of the network structure. In view of this problem, based on the different characteristics of speech enhancement in time and frequency, inspired by the feature calculation method in the traditional speech enhancement methods, a time-frequency smoothing network with different processings in time and frequency is designed. In this network, a gated recurrent unit is used to express the correlation of noisy speech with time, and a convolutional neural network is used to express the correlation of the noisy speech with frequency, which realizes a time-frequency smoothing process similar to that of the traditional speech enhancement methods. Experimental results show that the proposed time-frequency smoothing network can significantly improve the speech enhancement performance compared with other networks under the premise of ensuring the causality of the speech enhancement system and that the enhanced speech has a better speech quality and intelligibility.

    Extraction of features of the urban high-rise building from high resolution InSAR data
    GUO Rui,ZANG Bo,PENG Shuming,XING Mengdao
    Journal of Xidian University. 2019, 46(4):  137-143.  doi:10.19665/j.issn1001-2400.2019.04.019
    Abstract ( 413 )   HTML ( 87 )   PDF (2327KB) ( 65 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Under decimeter resolution, the projection of an individual high-rise building shows its distinctive features in interferometric synthetic aperture radar (InSAR) images, which can help extract more accurate detail information and supply prior information for advanced interferometric techniques such as Tomography SAR (TomoSAR). In this paper, a top-to-down coarse-to fine approach is proposed to extract high-rise building features, which works at four different levels, namely, image-, region of interest (ROI)-, building- and pixel-wise processing, and which considers the analysis of high-rise building characteristics in very high resolution InSAR data pairs. Finally, the features of the high-rise building including the building orientation and pixels at the same height with the similar orientation are defined and estimated. The proposed approach is validated by using the single-pass interferogram from the TanDEM-X system over high-rise urban areas.

    Design and verification of a film bulk acoustic resonator
    SHEN Hongxia,OU We1
    Journal of Xidian University. 2019, 46(4):  144-149.  doi:10.19665/j.issn1001-2400.2019.04.020
    Abstract ( 477 )   HTML ( 24 )   PDF (1768KB) ( 70 )   Save
    Figures and Tables | References | Related Articles | Metrics

    In order to simplify the design of the film thickness of the film bulk acoustic resonator, a design method for film bulk acoustic resonator film thickness is proposed. The simulation structure consists of the induced layer and the upper electrode-piezoelectric-electrode sandwich structure. The initial film of the film bulk acoustic resonator is designed by the optimal effective electromechanical coupling coefficient, and then the thickness of the induced layer and the corresponding frequency offset value are determined. The parallel resonant frequency is compensated with the frequency offset, and then the film thickness of the electrode and the piezoelectric are recalculated. Finally, the structure is simulated by COMSOL. When the parallel resonant frequency is 3.60 GHz, the frequency offset value of the 100 nm AlN is 0.20 GHz. The effective electromechanical coupling coefficient of the AlN is 5.907%. After frequency compensation, the series resonant frequency and parallel resonant frequency of the AlN are 3.48 GHz and 3.60 GHz, respectively. The design method is verified. The induced layer effectively optimizes the C-axis characteristics of the piezoelectric layer and reduces energy loss.

    Visual tracking method using discriminant dictionary learning
    WANG Hongyan,QIU Helei,PEI Tengda
    Journal of Xidian University. 2019, 46(4):  150-158.  doi:10.19665/j.issn1001-2400.2019.04.021
    Abstract ( 266 )   HTML ( 15 )   PDF (1725KB) ( 33 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Focusing on the issue of the great decrease in object tracking performance induced by complex background and occlusion, a visual tracking method is proposed. The object and background samples are first obtained according to the local correlation of the object in the temporal-spatial domain. In what follows, a dictionary learning model is established: the outliers generated by occlusion are captured by error terms, and the sparse encoding matrix and error matrix are punished by nonconvex minimax concave plus functions. In addition, inconsistent constraints are imposed on the dictionaries to improve the robustness and discriminability of dictionaries. Concerning the established nonconvex dictionary learning optimization issue, the majorization-minimization (MM) optimization method can be exploited to get better convergence. Finally, the reconstruction errors of the candidate object are computed from the learned discriminative dictionary to construct the object observation model, and after that, the object tracking is realized accurately based on the Bayesian inference framework. As compared to the existing state-of-the-art algorithms, simulation results show that the proposed algorithm can improve the accuracy and robustness of the object tracking significantly in complex environments.

    Robust detection of weak transient signals with changing background
    WU Yong,ZHENG Wei,NIU Wenlong,YANG Zhen
    Journal of Xidian University. 2019, 46(4):  159-166.  doi:10.19665/j.issn1001-2400.2019.04.022
    Abstract ( 325 )   HTML ( 12 )   PDF (1769KB) ( 42 )   Save
    Figures and Tables | References | Related Articles | Metrics

    In the detection of a weak transient signal, the signal is completely submerged in the noise, so the detection performance of traditional methods will be dramatically deteriorated when the background changes. To solve this problem, a new weak transient signal based on the kernel function is proposed, which is robust to the continuous background change. The signal is mapped to a higher dimensional feature space by use of the kernel function to solve the problem that there is linear inseparability in the original space. The kernel function is constrained according to characteristics of the signal and noise, so that the detection method is robust to the continuous background change while detecting transient signals effectively. The proposed method is applied to weak moving point target detection based on the high frame rate image sequence. The experiment shows that the proposed method can obtain better detection results in both simulation data and real world data, and that the proposed method is more superior when the background is changing continuously.

    Hardware trojan attack methods and security analysis under split manufacturing
    YANG Yajun,CHEN Zhang
    Journal of Xidian University. 2019, 46(4):  167-175.  doi:10.19665/j.issn1001-2400.2019.04.023
    Abstract ( 344 )   HTML ( 14 )   PDF (1238KB) ( 29 )   Save
    Figures and Tables | References | Related Articles | Metrics

    The Hardware Trojan (HT) threats the security of chips greatly, and in order to explore the effectiveness of split manufacturing (SM) securing circuits from HT insertion, an HT threat model is first proposed under SM. Then we propose two possible attack methods: one is the improved proximity-based attack and the other is the genetic algorithm based attack, which leverage different physical heuristics revealed from the physical design process. Experimental results show that both the proposed methods are effective on inserting HTs under SM. Especially the genetic algorithm based attack method is shown to be the most effective with the attack accuracy of up to 78.62%, thus demonstrating that SM is not secure inherently and that more efforts need to be made in order to make SM more secure.

    Monte Carlo simulation of uplink laser transmission across media
    LIU Li,YUE Peng,CUI Zongmin
    Journal of Xidian University. 2019, 46(4):  176-181.  doi:10.19665/j.issn1001-2400.2019.04.024
    Abstract ( 329 )   HTML ( 71 )   PDF (1215KB) ( 44 )   Save
    Figures and Tables | References | Related Articles | Metrics

    In order to study the distribution and loss characteristics of the cross-media uplink laser during transmission, the existing laser transmission model is improved based on the Monte Carlo method. Assuming that the underwater laser source is a collection of photons and that the light intensity obeys the Gaussian distribution, by tracking the transmission process of a large number of photons, statistical results are obtained to analyze the effects of the wind speed and underwater transmission distance on the photon distribution and weight on the receiving surface. Results show that when the sea surface wind speed is constant, the underwater transmission distance is larger, and the photon distribution is more divergent, with the weight of the center location sharply decreasing. When the underwater transmission distance is constant, the lower wind speed on the sea surface has a little effect on the photon distribution. As the wind speed increases, the trend of divergence is obvious for the photon distribution. Therefore, both a large wind speed on the sea surface and a long distance of underwater transmission have a great influence on the laser transmission across the media.

    Analysis and detection of TDDB degradation for DRAM in 3D-ICs
    JIA Dingcheng,WANG Leilei,GAO Wei
    Journal of Xidian University. 2019, 46(4):  182-189.  doi:10.19665/j.issn1001-2400.2019.04.025
    Abstract ( 319 )   HTML ( 17 )   PDF (2212KB) ( 45 )   Save
    Figures and Tables | References | Related Articles | Metrics

    3D multicore systems with stacked DRAM are capable of boosting system performance significantly, but accompanied with the key problem of the effect of heat density and heat dissipation on circuit reliability. Aiming to study the TDDB (Time Dependent Dielectric Breakdown) effect in DRAM of 3D-ICs, we adopt a physical-based SPICE model and analyze the statistical TDDB degradation induced by the gate leakage current in peripheral circuits of DRAM. Meanwhile, a TDDB detection design is proposed based on the 45nm process, which is suitable for large scale integration of the memory circuit. And the operation of the detection circuit is analyzed based on the BTI (Bias Temperature Instability) effect. Experimental results show that sense amplifiers are more susceptible to time dependent dielectric breakdown than word-line drivers in DRAM. The proposed TDDB detection design can completely meet the maximum fault coverage rate with good robustness to BTI, and it will send out an alarm signal when TDDB happens in the sense amplifier.

    Vehicle re-identification by multi-cameras for public security surveillance
    WANG Yanfen,ZHU Xuran,YUN Xiao,SUN Yanjing,SHI Yunkai,WANG Sainan
    Journal of Xidian University. 2019, 46(4):  190-196.  doi:10.19665/j.issn1001-2400.2019.04.026
    Abstract ( 335 )   HTML ( 21 )   PDF (2048KB) ( 70 )   Save
    Figures and Tables | References | Related Articles | Metrics

    The existing vehicle re-identification (Re-ID) methods mostly perform Re-ID between images marked with vehicle bounding boxes, but there are no vehicle bounding boxes in the real scene; at the same time, the complexity of environment and the similarity and diversity in appearance among vehicles can also cause a low accuracy of Re-ID. Therefore, this paper proposes a multi-camera vehicle Re-ID method combining vehicle detection and recognition for the unmarked original video in the field of public safety surveillance. First, the Binary-Single Shot MultiBox vehicle detection network is proposed to obtain the vehicle bounding boxes and generate candidate database online. Second, a multi-task Siamese vehicle recognition network is designed to improve the Re-ID accuracy. Finally, the “VeRi-1501” vehicle dataset is established, which expands the number of vehicle IDs and balances the number of images for each ID in the case of different cameras. The proposed method has achieved good results in the VeRi-1501 dataset and the actual traffic scene.