Please wait a minute...
, No 11
■ Cover Page (PDF 4745 KB) ■ Editorial Board (PDF 145 KB) ■ Table of Contents, November 2018 (PDF 91 KB)
  • Communication Protocols for Distributed Wireless Sensor Network based on Viterbi Algorithm
    Binbin Yu
    2018, 14(11): 2553-2560.  doi:10.23940/ijpe.18.11.p1.25532560
    Abstract    PDF (219KB)   
    References | Related Articles
    Routing is one of the most important technologies in wireless sensor networks. The energy efficiency of the sensor network is realized by combining other technologies with routing protocols in this paper. First, through the analysis of the energy model of the clustering routing algorithm, the local signal to noise ratio (SNR) is used to purposefully select the nodes in the cluster to transmit data. In order to avoid the energy consumption of a large number of nodes to transmit data, a simple architecture is proposed to combine the distributed source code with the cluster routing protocol. The method uses the data on the cluster head as the edge information to decode the compressed information transmitted from the cluster nodes. The simulation result shows that the energy efficiency of the network can be achieved through this combination.
    Short-Range Wireless Network Communication and Application based on Decision Tree Algorithm
    Xie Wei, Xu Huoxi, Peng Liping, and Lan Zhigao
    2018, 14(11): 2561-2573.  doi:10.23940/ijpe.18.11.p2.25612573
    Abstract    PDF (1449KB)   
    References | Related Articles
    Due to the rapid development of computer technology, various technologies have become very advanced and have already played an important role in people's lives and work. The research development speed based on the decision tree algorithm in short-range wireless network communication technology is slow. The decision tree algorithm was used to optimize short-range wireless network communication technology and its application, which can better help related systems improve work efficiency. Taking one Cangshan community of Fuzhou City as an example, the decision tree algorithm and wireless network communication model were used to design and implement a short-range wireless communication network covering the whole community, providing necessary theoretical support for related types of research.
    Architectural Design Model based on BIM Management System Model and Data Mining
    Tiandong Shao and Chunming Zhang
    2018, 14(11): 2574-2580.  doi:10.23940/ijpe.18.11.p3.25742580
    Abstract    PDF (404KB)   
    References | Related Articles
    An empirical study of the architectural design model based on the BIM management system model and data mining is carried out. Combined with the current code for the design of building structures, a two-level hybrid optimization algorithm is proposed. The optimal design of the structure is divided into two levels: unit optimization and overall stiffness optimization. First, the result of the overall stiffness optimization is taken as the lower limit. The genetic algorithm is used to complete the structural strength optimization design, and then the result of the unit optimization is used as the lower limit. The above optimization algorithm is implemented on large finite meta-software and special software for building structure design. The tall building structure of the two-story is optimized, and the result shows that the proposed method is effective and operational.
    Elderly Health Care Interventions under the Mode of Smart Sports Rehabilitation and the Background of Big Data
    Yanping Jiang
    2018, 14(11): 2581-2589.  doi:10.23940/ijpe.18.11.p4.25812588
    Abstract    PDF (343KB)   
    References | Related Articles
    In recent years, there has been a tremendous transformation in the way of health care for the elderly. The in-depth research on the smart sports rehabilitation model has promoted medical innovation. China's health and medical interventions for the elderly have also shown a further deepening trend. Therefore, it is necessary to study elderly health care interventions under the smart physical rehabilitation mode in the context of big data. By constructing a smart sports rehabilitation model, all data can be collected and compared, and a targeted rehabilitation selection mode based on computer algorithms can be realized. The experiments have shown that the use of smart medical care is more convenient and effective for elderly health care interventions.
    A Data Glove-based KEM Dynamic Gesture Recognition Algorithm
    Rui Han, Zhiquan Feng, Changsheng Ai, Wei Xie, and Kang Wang
    2018, 14(11): 2590-2600.  doi:10.23940/ijpe.18.11.p5.25892600
    Abstract    PDF (758KB)   
    References | Related Articles
    Data gloves-based gesture recognition plays a very important role in the virtual reality interaction system. A new dynamic gesture recognition method, that is, K-means clustering dimensionality reduction and Euclidean metric template matching algorithm based on data glove (KEM algorithm), is proposed in this paper. First, high-dimensional data is clustered in the K-means clustering algorithm to achieve dimensionality reduction. Then, the low-dimensional data is put into the template matching method based on Euclidean metric to get the distance that matches all the templates. Finally, the corresponding gesture is identified according to the template matching. The main innovations of the proposed KEM algorithm are as follows: (a) K-means clustering is applied to dynamic gesture recognition for the first time to achieve real-time recognition, (b) the classical K-means method is optimized, and (c) the template matching process is more reasonable. Experiments show that the proposed KEM method can achieve 99.42% in recognition rate. The validity of the KEM method has been verified in a 3D Intelligent Teaching System.
    Locality Preserving Hashing based on Random Rotation and Offsets of PCA in Image Retrieval
    Shan Zhao and Yongsi Li
    2018, 14(11): 2601-2611.  doi:10.23940/ijpe.18.11.p6.26012611
    Abstract    PDF (1198KB)   
    References | Related Articles
    Manifold-based subspace feature extraction methods have recently been deeply studied in data dimensionality reduction. Inspired by PCA Hashing (PCAH), if the Locality Preserving Projection (LPP) is directly used in the hash image retrieval, it is prone to shortcomings such as being inefficient and time-consuming. In order to address these deficiencies, this paper mainly combines Principal Component Analysis (PCA) and manifold subspace feature extraction method LPP, and we present a RLPH framework using random rotation. Among them, PCA processing solves the eigenvalue problem encountered in the calculation of LPP, thereby improving the recognition effect of the algorithm. The PCA projection needs to ensure that the variance of the sample points after projection is as large as possible. However, projections of small variance may produce unnecessary redundancy and noise. Therefore, in the subspace after the PCA projection, we only extract the eigenvectors that contain most of the information at the top of the PCA projections. Then, we utilize a random orthogonal matrix to randomly rotate and shifts the eigenvectors and the reduced-dimensional sample obtained after the top eigenvectors of the PCA projection is subjected to LPP mapping. Random rotation produces many thin projection matrices blocks that are then concatenated into one final projection matrix. Random rotation is a key step in this paper that minimizes the quantization error for codes. The proposed method greatly improves the retrieval efficiency, and extensive experiments demonstrate its effectiveness.
    An Improved Influential Cover Set Mining Algorithm
    Jia Liu, Wei Chen, Ziyang Chen, and Huijuan Liu
    2018, 14(11): 2612-2623.  doi:10.23940/ijpe.18.11.p7.26122623
    Abstract    PDF (671KB)   
    References | Related Articles
    The influential cover set (ICS) problem is a hot research issue in social networks and road networks. Considering that existing methods cannot take into account both the efficiency and the accuracy of results, we propose a partition-based influential cover set mining algorithm with the index. Firstly, we create the inverted index for each attribute to be queried, thus avoiding traversing all nodes while querying the node covered attributes and reducing the query time with high accuracy. Then, we design the pruning strategy according to the upper bound of cover-group for filtering, which reduces the number of the partition combination in the treatment on each tuple in the linked list of partitions, reduces the overall computation, and improves the processing speed and efficiency. Our experimental results on 15 real datasets verify the efficiency of our method in terms of different metrics, including indexing time, accuracy of results, influence of results, and query processing time.
    Contrast Enhancement of Illumination Layer Image using Optimized Subsection-based Histogram Equalization
    Yongxin Wang, Ming Diao, and Haibin Wu
    2018, 14(11): 2624-2632.  doi:10.23940/ijpe.18.11.p8.26242632
    Abstract    PDF (314KB)   
    References | Related Articles
    A key problem of underwater image sharpening is to improve image contrast while retaining image detail. The retinex model is used to obtain the illumination and detail layer images. The histogram of the illumination layer image is divided into under-exposure subsection and over-exposure subsection by using the maximum interclass variance method, and the histogram subsections are equalized separately. The above process of histogram dividing and equalization is repeated until the difference between adjacent thresholds for dividing histogram subsections reaches its optimal value. This enhances the contrast of the illumination layer image. As a result, our contrast enhancement method of illumination layer image using optimized histogram subsection based histogram equalization is formed. Furthermore, by multiplying the enhanced contrast of illumination layer image with the original detail layer image, the contrast of underwater image is enhanced and its original details are retained. Some evaluations, e.g. information entropy and mean structure similarity, are examined to show that the underwater image quality is improved appropriately.
    Defensive Strategy Selection based on Attack-Defense Game Model in Network Security
    Ningbin Zhang
    2018, 14(11): 2633-2642.  doi:10.23940/ijpe.18.11.p9.26332642
    Abstract    PDF (570KB)   
    References | Related Articles
    Security analysis and attack-defense modeling are effective methods to identify the vulnerabilities of information systems for proactive defense. The attack graph model reflects only attack actions and system state changes, without considering the perspective of the defenders. To assess the network information system and comprehensively show attack and defense strategies and their cost, a defense graph model is proposed. Compared with the attack graph, the model makes some improvements. The defense graph will be mapped to the attack and defense game model, in order to provide a basis for active defense policy decision. Moreover, a generation algorithm of defense graph is proposed. A representative example is provided to illustrate our models and demonstrate the high efficiency of the algorithm.
    An Attribute Key Distribution Scheme based on Low Dependence Trusted Authorities
    Xuewang Zhang, Wei Wang, and Jinzhao Lin
    2018, 14(11): 2643-2651.  doi:10.23940/ijpe.18.11.p10.26432651
    Abstract    PDF (742KB)   
    References | Related Articles
    Aiming at the problem of “authority deception” and attribute restoration on the key distribution center, an attribute key distribution scheme under the environment of low reliance on trusted authorities is proposed in this paper. In this scheme, the users’ attributes are generated by the Cloud Storage Service Provider (CSSP), and then the attributes are divided into several attribute blocks and distributed to users. The user exchanges attribute blocks with other users and restore the attributes through the Chinese remainder theorem after receiving the attribute blocks. The multiple Key Management Centers (KMC) generate the attribute parameters according to the user attributes, and the last user calculates the attribute keys through the received attribute parameters. This scheme avoids the problem of “authority deception” and has high computational efficiency. Finally, the experimental results show that the program is safe and efficient.
    A Top-r k Influential Community Search Algorithm
    Wei Chen, Jia Liu, Ziyang Chen, and Jianqi Chen
    2018, 14(11): 2652-2662.  doi:10.23940/ijpe.18.11.p11.26522662
    Abstract    PDF (531KB)   
    References | Related Articles
    Top-r k influential community search is one of the hot topics in social network research, the solution of which usually adapts the “index + query” strategy. Aiming at the problems of low index efficiency and unreasonable metric of the influence, we first propose a new index construction method that not only improves the efficiency of constructing index but also reduces the index size. In the community search, the metric of the influence on the community is redefined and the search algorithm is proposed on this basis to make the search results more practical. Finally, according to experiments on 12 datasets, we verify the high efficiency of the method proposed in this paper compared with the existing methods from the following aspects including the index construction time, the index size, and the search time.
    Aerial Image Matching based on NSST and Quaternion Exponential Moment
    Huan Wang, Zhenhua Jia, and Yunfeng Zhang
    2018, 14(11): 2663-2673.  doi:10.23940/ijpe.18.11.p12.26632673
    Abstract    PDF (492KB)   
    References | Related Articles
    In this paper, we propose an aerial image matching algorithm based on NSST and quaternion exponential moment. Firstly, we use non-subsampled shearlet transform (NSST) to decompose the reference image and the to-be-matched image, and the scale invariant feature with error resilience (SIFER) operator is used to extract stable feature points from NSST low-frequency sub-bands and construct local feature areas respectively. Subsequently, local features of each feature area are solved by quaternion exponential moment to constitute feature vectors of such feature points for pre-matching. In the end, mismatching point pairs are removed by the random sample consensus (RANSAC) algorithm. Finally, experimental results show that compared with the SIFT and SURF algorithms, the algorithm proposed in this paper makes faster operations, has higher matching precision, and is significantly better than the other two methods in resisting rotation, noise, brightness change, and integrated disturbance.
    An Improved Location Algorithm for Wireless Sensor Networks
    Qiang Zhang
    2018, 14(11): 2674-2682.  doi:10.23940/ijpe.18.11.p13.26742682
    Abstract    PDF (960KB)   
    References | Related Articles
    The ranging error of WSN (wireless sensor network) is usually large in complex environments. We find that the elements of the coordinate inner product matrix may fluctuate in a certain range with the changing ranging error. Therefore, we present a maximum likelihood estimation (MLE) location algorithm based on the coordinate inner product matrix for determining the relative locations of sensor nodes in complex environments with large ranging error. Based on the global topological structure and the connectivity of WSNs, the geodesic distance between each node and the coordinate inner product matrix are obtained. Using the maximum likelihood estimator for a coordinate inner product matrix, we can finally estimate the sensor node coordinates by finding the global optimal solution. The experimental results show that the algorithm has good noise resistance for ranging noise; therefore, it is suitable for WSN node locating with large range noise. When the node distance error is large, it can also achieve high location accuracy.
    SRD: Static Data Race Detection for Concurrent Programs
    Yang Zhang, Yanan Liang, and Dongwen Zhang
    2018, 14(11): 2683-2691.  doi:10.23940/ijpe.18.11.p14.26832691
    Abstract    PDF (660KB)   
    References | Related Articles
    Data race probably occurs when many threads concurrently access the same memory location and at least one is a write thread. Data race detection suffers from false negatives and false positives. How to detect data race and avoid false negatives and false positives has become a hot topic. This paper proposes a static data race detection methodology to eliminate false negatives and false positives. We use Soot to conduct intra-thread and inter-thread analysis. Our data race detection focuses on variable access events that are collected from call graphs. Several program analysis technologies, such as alias variable analysis, alias lock analysis, happens-before analysis, constraint graph, and slicing analysis, are used to improve the coverage and precision of the detection results. In the experimentation, several benchmarks, such as raytracer, sor, and mergesort, have been selected to evaluate our methodology. Experimental results show that SRD can not only eliminate fake races but also identify potential races. Furthermore, SRD can detect more positive races than the existing tool RVPredict.
    Performance Analysis of Software Aging Prediction
    Yongquan Yan
    2018, 14(11): 2692-2701.  doi:10.23940/ijpe.18.11.p15.26922701
    Abstract    PDF (335KB)   
    References | Related Articles
    Software aging is a problem that was discovered two decades ago. Since then, many research studies have investigated how to manage aging problems caused by memory leakage and accumulated round-off error through resource consumption prediction or state forecasting. When applying state prediction, the performances of various aging classification algorithms are compared by the prediction error. Since forecasting error is not a precise measure and must be estimated, the forecast error variance needs to be analyzed. In this work, we carefully analyze the forecast error variance by three steps. In the first step, we propose a method to decompose the variance by considering the influence of the data sampling process and data partition procedure. In the second step, we use an enhanced Friedman test and the Nemenyi post hoc test to analyze the influence of the data sampling process on the data partitioning procedure. In the last step, a corrected t-test is proposed to compare the performance of two off-the-shelf classification algorithms. The software comparison experiment is based on a real-time web environment. We end this work by proposing a set of feasible suggestions.
    A Multi-Target Detection Algorithm for Infrared Image based on Retinex and LeNet5 Neural Network
    Lijun Yun, Tao Chen, Zaiqing Chen, and Kun Wang
    2018, 14(11): 2702-2710.  doi:10.23940/ijpe.18.11.p16.27022710
    Abstract    PDF (762KB)   
    References | Related Articles
    Objectdetection in infrared video images is an important and challenging work. Due to low resolution, poor contrast, and low visual quality, target detection in infrared images is inefficient and prone to having higher false positive and lower precision rates. To improve detection efficiency, according to the characteristics of infrared images, we proposed a multi-target detection algorithm based on image enhancement and the LeNet5 deep neural network. In our method, we used the Retinex image enhance algorithm to protrude the edge contour and contrast, highlight the detailed features, and enhance the overall visibility of infrared images. In particular, the LeNet5 convolution neural network and CVC vehicle-assisted driving database were used to train the interesting target in the infrared image to generate the target data model, and the selective search algorithm was used to segment the candidate detect object regions in the image. The separated candidate regions were sent to the trained data model to classify the type and locate the position of objects in the image. The simulation results in CVC infrared image subset datasets show that our algorithm has higher detection speed and accuracy than the traditional HOG-based and LBP-based detection algorithms.
    Degradation and Reliability Modeling of Two-Component System with Degradation Rate Interaction
    Zhiyuan Yang, Jianmin Zhao, Chiming Guo, and Liying Li
    2018, 14(11): 2711-2722.  doi:10.23940/ijpe.18.11.p17.27112722
    Abstract    PDF (620KB)   
    References | Related Articles
    The stochastic dependency, which has a significant impact on degradation and reliability modeling, exists in many complex systems. The stochastic dependency is analyzed from a degradation processes interaction perspective in this paper. For a two-component system, a degradation rate interaction model is developed to describe the dependency between the degradations of two components with a nonlinear Wiener process. This is achieved by considering that the degradation rate of a component is affected by the degradation state of the other component in the proposed system. Then, the parameter estimation approach is given, and the reliability models of the components and system are derived. Finally, a numerical example about fatigue crack development is presented to validate the developed models. Moreover, a comparison study with some present models is taken to show the performance of the developed degradation and reliability models.
    Data Aggregation in WSN based on Deep Self-Encoder
    Lishuang Zhao
    2018, 14(11): 2723-2730.  doi:10.23940/ijpe.18.11.p18.27232730
    Abstract    PDF (634KB)   
    References | Related Articles
    In order to reduce the energy consumption of data transmission in the limited resources of wireless sensor networks, a WSN data fusion algorithm based on AEDA) is proposed. Firstly, the deep self-encoder (DESAE) is constructed and the training is completed at the sink node, and the trained parameters are passed to the corresponding sensor nodes. The algorithm proposes two kinds of data fusion models, which can extract the raw data through the network model to obtain a small amount of feature data and send it to the sink node, reducing the amount of data transmission. The simulation results show that compared with the LEACH algorithm, this algorithm can significantly reduce the energy consumption, extend the network life cycle, and is more suitable for large-scale networks.
    A Synergy Metric for Educational Emergency Governance based on Information Entropy
    Mingjing Tang, Li Liang, and Wei Zhong
    2018, 14(11): 2731-2741.  doi:10.23940/ijpe.18.11.p19.27312741
    Abstract    PDF (528KB)   
    References | Related Articles
    Educational emergency governance is a complex work that requires the participation of multiple subjects. Studying the coordination of educational emergency governance can improve coordination efficiency, reduce losses, and achieve optimal governance effect. In this paper, collaborative entropy is introduced to establish an evaluation model of coordination in educational emergency governance. Collaborative matrices, collaborative degree, and collaborative efficiency between governing organizations and governance works are determined based on this model. Finally, case data is used as an example to evaluate collaborative degree and collaborative efficiency for educational emergency governance.
    Hotspots, Theme Structure, and Development Trends on Maker Education: A Quantitative and Co-Word Analysis
    Haiyan Xie, Jianhou Gan, Bo Zhao, and Ying Xiong
    2018, 14(11): 2742-2751.  doi:10.23940/ijpe.18.11.p20.27422751
    Abstract    PDF (893KB)   
    References | Related Articles
    The cultivation of innovative talents is the focus of the education field in every country. In recent years, Maker Education (ME), as one of the ways to promote innovation of education, has become a key research area. By means of quantitative and qualitative analysis of the research literatures in recent years, it is possible that scholars have a relatively more comprehensive command over the latest subjects, and this can provide some directions for scholars when launching other studies to promote the sustainable development of ME. In this paper, the research hotspots, theme structure, and development trends of ME were surveyed based on literatures in CNKI (China National Knowledge Infrastructure). Firstly, among all the extracted keywords, 33 high-frequency keywords representing research hotspots were identified. Secondly, the status and relationship of the research hotspots were calculated and described through social network analysis (SNA). Finally, three research themes were classified by cluster analysis, and their distribution was depicted in a two-dimensional map. In conclusion, the results show that the current research direction of ME in China mainly revolves around two main lines and three research theme areas. Although China has made some achievements, the development of ME is still immature and unbalanced, and four development trends will be the research focus in the future.
    Performance Analysis of Beam Error Parasitic Loop of Phased Array Radar Seeker
    Jing Yu, Zhiyi Lu, Xiangping Li, Qi Chen, and Xiaohai Zou
    2018, 14(11): 2752-2759.  doi:10.23940/ijpe.18.11.p21.27522759
    Abstract    PDF (372KB)   
    References | Related Articles
    For the parasitic loop problem caused by beam errors of phased array radar seekers, a beam error parasitic loop model is established. The concept of beam error slope is put forward, and thus the corresponding closed-loop transfer function of parasitic loop model is deduced. Through the Routh criterion, the stability domain of a parasitic loop is determined. The performance of the parasitic loop is analyzed from the time domain and frequency domain, and the guidance parameters that influence the stability of parasitic loops are given. It shows that the stability of parasitic loops is influenced by missile flight attitude and guidance time. Additionally, it was found that the influence of dynamic link order in parasitic loop positive feedback is much smaller than that in parasitic loop negative feedback, but it is more sensitive to the amplitude of beam error slope and more easily unstable. This provides a theoretical basis for industrial design in the future and plays an important role in improving the performance of missiles.
    Formal Verification of Double Two out of Two Computer Systems
    Haonan Feng, Xiaojiao Ma, Wei Fu, and Ming Pan
    2018, 14(11): 2760-2768.  doi:10.23940/ijpe.18.11.p22.27602768
    Abstract    PDF (579KB)   
    References | Related Articles
    The double two out of two safety computer system is widely used in China’s rail transit. To enhance the safety integrity level of such a system, safety related logic is described and modelled by FSP (finite state process) language in a simple and explicit manner. A new method based on LTS (labelled transition system) model checking is proposed for verifying the system safety properties. The LTS method is adapted to model system behaviors by means of LTSA (labelled transition system analyzer) software. It visualizes overall activity traces and is easy for analysis and safety verification by developers. Simulation and verification results indicate that the LTS method provides great assistance for designers to develop more efficient and reliable complex systems.
    Evaluation of Creative Talents in Cultural Industry based on BP Neural Network
    Xiaolan Chang and Wenjun Li
    2018, 14(11): 2769-2776.  doi:10.23940/ijpe.18.11.p23.27692776
    Abstract    PDF (203KB)   
    References | Related Articles
    Since the creative talents evaluation is a basic link of decision-making in the cultural creative industry, this study establishes an evaluation indicator system for creative talents in the cultural industry, examines common evaluation methods and the back propagation (BP) neural network evaluation method, builds an evaluation model for creative talents in the cultural industry based on the BP neural network, and evaluates the evaluation indicator system of creative talents in the cultural industry by using common methods, which provide a sample set for the training and testing of the BP neural network model. Furthermore, this article adopts the unique nonlinear mapping capability, self-learning, and strong fault-tolerant abilities of the BP neural network to construct an evaluation model of creative talents in the cultural industry based on the BP neural network and carries out case analysis and verification, which show that the evaluation model based on the BP neural network is appropriate for the evaluation of cultural creative talents. Compared with the conventional evaluation methods, the BP neural network can simulate the experts to conduct a quantitative evaluation through repeated learning and training, so as to effectively avoid human error in the evaluation process. The structure and algorithm of the BP neural network are simple, and computers can simulate the evaluation process, thus reducing the manpower for calculation.
    A Novel Image Inpainting Method for Object Removal based on Structure Sparsity
    Lei Zhang and Minhui Chang
    2018, 14(11): 2777-2788.  doi:10.23940/ijpe.18.11.p24.27772788
    Abstract    PDF (872KB)   
    References | Related Articles
    In the traditional image inpainting method for object removal, for each target patch, the entire source region must be traversed to search for the exemplar patch, which may make the restoration process time-consuming and affect the restoration efficiency. Even worse, the target patch may be replaced by an inappropriate exemplar patch during the process, which will introduce some unexpected objects in the restored image and make the result unable to meet the requirements of visual consistency. In view of these problems, we propose a novel image inpainting method for object removal based on structure sparsity. First, we calculate the structure sparsity of the target patch, and then identify the local characteristics of the region where the target patch is located. Then, we set different search regions for the target patches according to different regional characteristics. Finally, we find the exemplar patch in the search region and restore the target patch. Experiments on a number of natural images show that the proposed method can reduce the restoration time and improve the restoration efficiency. Additionally, it can prevent the mismatch to some extent and improve the restoration effect.
    Multiple Signals Estimation for Overlapping Nyquist Folding Receiver
    Wan Zhu, Shuang Zhao, Lei Chen, and Kang Chen
    2018, 14(11): 2789-2797.  doi:10.23940/ijpe.18.11.p25.27892797
    Abstract    PDF (467KB)   
    References | Related Articles
    The estimation of multiple signals across an extremely wide radio frequency bandwidth is a problem that is relevant to a large number of fields. In this paper, an overlapping division method of Nyquist Folding Receiver (NYFR) is proposed. A typical NYFR allows multiple Nyquist zones to be directly undersampled and then folded into a continuous time analog interpolation filter. The folding is achieved by undersampling the RF spectrum with a stream of signals that have different characteristics. However, the signals after modulation have a certain bandwidth that may be eliminated by the baseband filter. The overlapping structure proposed in this paper can avoid this situation. It achieves good and stable detection performance over the whole frequency band of the receiver. The NYFR folds all the signals of different Nyquist zones into the same band. The signals that belong to different Nyquist zones may cover each other. To solve this problem, the searching algorithm is presented. In the numerical results we show that, in the proposed framework, the overlapping NYFR has excellent detection performance.
    Kinematics Analysis and Optimization Design of Multi-Link High-Speed Precision Press
    Menglei Li, Hao Liu, Fuxing Li, and Maohua Xiao
    2018, 14(11): 2798-2807.  doi:10.23940/ijpe.18.11.p26.27982807
    Abstract    PDF (804KB)   
    References | Related Articles
    In this paper, taking the multi-link driving mechanism of a high-speed precision press as the research object, the corresponding kinematics equations are deduced. At the same time, kinematics analysis is performed by using MATLAB software to obtain the motion law of the slider, and the advantage of this mechanism is proposed in comparison with the crank slider mechanism. Then, based on ADAMS software, the parametric model of the driving mechanism is established and simulated. By comparing the simulation data of being imported into MATLAB with the derived formula, the correctness of the parametric model is verified. In addition, by analyzing the influence of the length change of each rod on the stamping slider and the mechanical pressure angle, the multi-objective optimization of the mechanism is carried out, which provides reference for the optimization design of homogeneous mechanisms.
    Prediction of Daily Pollen Concentration using Support Vector Machine and Particle Swarm Optimization Algorithm
    Wenfang Zhao, Jingli Wang, Dongchang Yu, and Ge Zhang
    2018, 14(11): 2808-2819.  doi:10.23940/ijpe.18.11.p27.28082819
    Abstract    PDF (414KB)   
    References | Related Articles
    In this paper, a support vector regression model for daily pollen concentration forecasting combined with the particle swarm optimization algorithm was proposed. Firstly, feature vector extraction was carried out by using the correlation analysis technique from meteorological data such as temperature, wind, relative humidity, precipitation, sunshine hours, and atmospheric pressure. Secondly, a support vector regression prediction model based on these vectors and pollen concentration observation data were established. Based on the Spark framework, a parallel particle swarm optimization algorithm was designed to optimize the parameters in the support vector regression algorithm, and then the optimal parameters were used to construct the daily pollen concentration prediction model. Finally, daily prediction of pollen concentration was made by using the optimized support vector regression model. The comparison among the accuracy of this optimized support vector regression model, the multiple linear regression (MLR) model, and the back propagation neural network (BPNN) model is performed to evaluate their performance. The results show that the proposed support vector regression model performs better than the MLR and BPNN models. Meanwhile, they also indicate that SVM provides promising results for prediction of daily pollen concentration.
    Remote Sensing Identification of Black Cotton Soil based on Deep Belief Network
    Lingling Wang, Wenyin Gong, and Xiang Li
    2018, 14(11): 2820-2830.  doi:10.23940/ijpe.18.11.p28.28202830
    Abstract    PDF (533KB)   
    References | Related Articles
    As a type of expansive soil, black cotton soil swells when absorbing water and shrinks when dehydrated, and the cycle of swelling-shrinking movements can readily occur repeatedly. These characteristics result in serious consequences both to land surfaces and to surface buildings such as ground fracturing, building settling, and road buckling and cracking, having extreme adverse effects on the quality and safety of road transportation. With Kitui, Kenya as the research area and a GF-1 remote sensing image as the vector, this study focuses on in-depth exploration of the application of a deep belief network to identify and classify black cotton soil based on the characteristics of the local black cotton soil in the remote sensing image. The results indicate that given the sample database available to this study, when the network depth was 3, the number of nodes in each hidden layer was 60, the learning rate was 0.01, the number of iterations was 20, and the number of samples was 2,000,000. The best classification result could be achieved with a precision of about 90% per the evaluation criteria proposed in this study, indicating a significant advantage of the deep belief network in remote sensing identification of black cotton soil.
    A Strongly Secure and Efficient Certificateless Authenticated Asymmetric Group Key Agreement Protocol
    Haiyan Sun, Zengyu Cai, Jianwei Zhang, Ling Zhang, and Yong Gan
    2018, 14(11): 2831-2841.  doi:10.23940/ijpe.18.11.p29.28312841
    Abstract    PDF (492KB)   
    References | Related Articles
    In Eurocrypt’2009, Wu et al. (2009) presented an important primitive named the asymmetric group key agreement (AGKA) protocol. In such a primitive, a group of users generate a common public encryption key, and each user only holds his own secret decryption key. Authenticated asymmetric group key agreement (AAGKA) protocols are a kind of AGKA protocol that can be secure against active attacks. AAGKA protocols in certificateless public key cryptography (CL-PKC) have some preponderance than those in identity-based cryptography and PKI cryptography. However, existing AAGKA protocols in CL-PKC only consider security against normal type adversaries, the weakest adversaries considered in CL-PKC literature. To solve this problem, an improved security model that considers security against super adversaries and a provably secure certificateless AAGKA protocol under the improved security model are proposed. Efficiency comparison shows that the proposed protocol is more efficient.
    Spatio-Textual Query: Review and Opportunities
    Lianyin Jia, Binglin Shen, Mengjuan Li, Jing Zhang, and Jiaman Ding
    2018, 14(11): 2842-2851.  doi:10.23940/ijpe.18.11.p30.28422851
    Abstract    PDF (348KB)   
    References | Related Articles
    Recent years have seen a rapid development of location-based services. As a consequence, spatio-textual query is becoming ubiquitous in real life. A great number of works have been conducted in this area over the last decade to efficiently support a variety of different queries. Unfortunately, scarce literatures have been witnessed to systematically categorize these works and to comprehensively compare them. To tackle this issue, in this paper, we provide a detailed survey in this field. Firstly, to capture the main differences among different algorithms, we divide spatio-textual queries into two categories from the textual perspective: spatio-textual containment query and spatio-textual similarity query. Secondly, the existing indexes in each category are compared and analyzed in detail. Finally, the paper points out the challenges and opportunities in this area.
    Recognition and Classification of High Resolution Remote Sensing Image based on Convolutional Neural Network
    Guanyu Chen, Zhihua Cai, and Xiang Li
    2018, 14(11): 2852-2863.  doi:10.23940/ijpe.18.11.p31.28522863
    Abstract    PDF (1198KB)   
    References | Related Articles
    High resolution remote sensing image data is veritable big data. It is not only massive, multi-source, and heterogeneous, but also high-dimensional, multi-scale, and non-stationary. In order to overcome the reduction of classification accuracy and redundancy of spatial data when dealing with high resolution remote sensing images using traditional classification methods, this paper improves the traditional Convolution Neural Network (CNN) from the aspects of both the network structure and the training method, and the improved CNN is used in the classification and recognition of high resolution remote sensing images. The experiments show that the classification accuracy of the improved CNN is better than that of the traditional CNN. Furthermore, the classification accuracy of the improved CNN is better than the Deep Belief Network (DBN), Support Vector Machine (SVM), and traditional BP.
    An Online HDP Mixture Model for Video Mining
    Lin Tang, Lin Liu, Mingjing Tang, and Yu Sun
    2018, 14(11): 2864-2876.  doi:10.23940/ijpe.18.11.p32.28642876
    Abstract    PDF (654KB)   
    References | Related Articles
    In this paper, we address two problems in video mining: real-time inference and the automatic decision of the number of activities in videos. To solve these problems, we present a real-time Bayesian non-parametric model that is able to discover activities and interactions of videos in real-time. In this model, there are two layers modeled in each scene, which are activities and interactions. An activity is represented as the distribution over visual words, and an interaction is represented as the distribution over activities. Then, the Hierarchical Dirichlet Process (HDP) model connects these two layers of video and automatically decides the number of clusters. Moreover, we developed a hybrid stochastic variational Gibbs sampling algorithm for inferring the parameters of the HDP mixture model. This online inference algorithm has the capacity to process the massive video stream dataset. Finally, the detailed experimental results in a crowded traffic scene and a simulated dataset are described and reveal that our online HDP mixture model achieves superior performance in real-time anomaly activity detection.
    Evaluation of Teaching Effectiveness based on Classroom Micro-Expression Recognition
    Xiaoxu Guo, Juxiang Zhou, and Tianwei Xu
    2018, 14(11): 2877-2885.  doi:10.23940/ijpe.18.11.p33.28772885
    Abstract    PDF (237KB)   
    References | Related Articles
    The improvement of teaching quality has been a persistent theme in education. To improve the quality of teaching in the classroom, teachers need to interact with students, pay attention to each student’s emotional changes, and closely follow each student’s changes in learning status, so as to make effective adjustments for teaching content. However, students’ responses often cannot be captured in time due to the limitations of the teacher in the classroom. Advances in computer and Internet technology as well as the development and maturation of image processing and artificial intelligence have provided technical support for the evaluation system of facial expression recognition in intelligent classrooms. In this paper, we propose an effective method to evaluate teaching effectiveness based on facial micro-expression recognition. An evaluation system is also designed and realized based on analyzing the change of classroom micro-expressions and the concentration of students. In such an evaluation system, face detection, tracking, and micro-expression recognition technology are applied to analyze the emotional changes during the learning process. Then students’ attention in class will be timely fed back to teachers, which can help teachers adjust teaching methods and strategies in a timely manner to improve teaching quality. In an informational teaching environment with general monitoring equipment, our proposed system can automatically track and analyze the degree of student’s concentration in the teaching process. Furthermore, it can also track the specified objects and analyze the change of their learning status in a certain period of time, which can help teachers conduct expediently multi-dimensional evaluation and guidance.
    Online Learning Behavior Evaluation Modeling and Application
    Yu Sun, Yanlong Sun, Bin Wen, and Lin Tang
    2018, 14(11): 2886-2896.  doi:10.23940/ijpe.18.11.p34.28862896
    Abstract    PDF (343KB)   
    References | Related Articles
    To motivate learners’ online learning behavior, this paper attempts to inspire online learners’ effective learning behavior in a data-driven way through the learning performance of online learners. Firstly, a theoretical model of online learning behavior evaluation with knowledge acquisition dimension, collaborative communication dimension, and learning attitude dimension was proposed and discussed in detail. Secondly, to realize the model, data from 1000 learners (20 general education courses) on the “Erya Online Classroom” learning platform of Yunnan Normal University was collected, and SPSS 22.0 was used to study the correlation between each evaluation index and the final exam results. Finally, 1000 online learners’ learning behavior data was taken in order to verify the validity of the model. The authors considered two models and compared the final evaluation results with the learner’s final exam results to verify the validity of the model.
    Removing Streak Interference from a Single Image based on Joint Priors
    Ao Li, Xin Liu, Deyun Chen, Kezheng Lin, Guanglu Sun, and Qidi Wu
    2018, 14(11): 2897-2904.  doi:10.23940/ijpe.18.11.p35.28972904
    Abstract    PDF (276KB)   
    References | Related Articles
    Streaks due to weather, such as rain or snow, degrade image quality and affect the performance of subsequential high-level vision tasks by the generated undesired artifacts. Hence, removing streak interference is an ongoing and challenging issue for many applications in real-time mobile surveillance systems. In this paper, streak interference removal from a single image is the focus. To sufficiently extract streak interference from an observed image, the image was firstly filtered with the nonsubsampled contourlet transform. Then, the residual part between the original and filtered image was decomposed into the streak component and detail component of background. Based on the additive layer model, we designed two specific priors that constrain the detail and streak interference respectively and established a model with joint priors for residual image decomposition. As a result, the resulting image can be synthesized with the filtered image and detail component. Experimental results show that our proposed method outperforms existing methods both qualitatively and quantitatively.
ISSN 0973-1318