Author Login Editor-in-Chief Peer Review Editor Work Office Work

05 April 2009, Volume 35 Issue 7
    

  • Select all
    |
    Degree Paper
  • JIAN Qiang; GUI Chun-mei; GONG Zheng-hu; LIU Xiang-hui
    Computer Engineering. 2009, 35(7): 1-4. https://doi.org/10.3969/j.issn.1000-3428.2009.07.001
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper studies Minimal Spatial Reuse Link Scheduling(MSRLS) model in wireless sensor networks and presents a distributed algorithm named MSRLS-D to solve MSRLS problem in normal networks. Because of considering both link bandwidth requirement distribution and link interference in wireless communication, MSRLS-D algorithm is more suitable for real physical environment. Analysis and comparison with central algorithm and graph-based algorithm validates the performance of MSRLS-D.
  • YUAN Zhi-jian; CHEN Ying-wen; MIAO Jia-jia; JIA Yan; YANG Shu-qiang
    Computer Engineering. 2009, 35(7): 5-7. https://doi.org/10.3969/j.issn.1000-3428.2009.07.002
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Bloom Filter(BF) is a space-efficient randomized data structure for representing data set with a small false positive probability. This paper compares and analyzes the similarities and differences among Counting Bloom Filter(CBF), Spectral Bloom Filter(SBF) and Dynamic Counting Filter(DCF), gives some examples and applications in data stream which include duplicate items detecting and frequent items mining. It summarizes the challenges in data stream brought by BF, including element burst problem and number of distinguish element changing problem.
  • FENG Hao-ran; ZHAO Qian-chuan; MU Chun-di
    Computer Engineering. 2009, 35(7): 8-10,1. https://doi.org/10.3969/j.issn.1000-3428.2009.07.003
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    According to the problem that competing node number estimation algorithm of the optimal enhanced IEEE 802.11 Distributed Coordination Function(DCF) protocol is complex and imprecise, this paper proposes a Self-Adaptive Contention Window Adjustment(SACWA) scheme. It is proved that the saturation throughput is almost independent of the competing node number when the collision probability keeps invariable, which is the theoretic foundation of this protocol. Simulation results demonstrate that SACWA can achieve nearly optimal saturation throughput no matter the node number keeps stable or changes rapidly.
  • NA Wen-wu; KE Jian; MENG Xiao-xuan; ZHU Xu-dong; XU Lu
    Computer Engineering. 2009, 35(7): 11-13,2. https://doi.org/10.3969/j.issn.1000-3428.2009.07.004
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper presents an optimal RAID6 algorithm, called XP-Code, to achieve small-write operations for double disk failure correction in RAID system. XP-Code partitions the erasure correcting group in the way of diagonal or anti-diagonal. There are one parity block and N-2 data blocks in each erasure correcting group. Parity blocks in diagonal group are stored on main anti-diagonal while parity blocks in anti-diagonal group are stored on main diagonal. The formula based on XOR operations is simple and easy to implement. By analyzing the algorithm and comparing with other RAID6 algorithms, the efficiency of calculating parity blocks, recovering missing blocks and small-write operations in XP-Code is proved to be optimal.
  • LING Ping; ZHOU Chun-guang; WANG Zhe
    Computer Engineering. 2009, 35(7): 14-16,3. https://doi.org/10.3969/j.issn.1000-3428.2009.07.005
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the bottleneck of poor performance and expensive consumption of traditional Support Vector Clustering(SVC), this paper proposes Least-Square Support Vector Clustering(LSSVC) model, and designs self-adaptive parameterization strategies. The model includes a new cluster labeling algorithm and fast training approach. The cluster labeling algorithm clusters Support Vectors(SVs) and non-SVs respectively. The fast training approach is implemented in incremental learning process, and after each data’s increment, a double-way learning procedure is conducted to adjust clustering model. Experiments demonstrate the improvement of LSSVC over its counterparts in efficiency and its competitive performance. And when the size of incremental data is 10% of the working set, it can balance cost and clustering accuracy well.
  • ZHANG Zhong-ping; ZHAO Hai-liang; ZHANG Zhi-hui
    Computer Engineering. 2009, 35(7): 17-19. https://doi.org/10.3969/j.issn.1000-3428.2009.07.006
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Concept similarity computation is a hot issue in information retrieval. With the wide applications of ontology in the fields of information retrieval and artificial intelligence, it brings new methods for concept similarity computation. Using ontology, a new method for computing concept similarity is proposed. This method combines the semantic distance and statistical properties of the ontology. The assistant influences of the depth and semantic superposition and intensity between concepts are taken into account. Experimental result shows this method is feasible for computing the concept similarity, it can be applied in the field of Web information retrieval.
  • DONG Wen-juan; WANG Hong-yi; HAN Jian-tao; CHEN Zeng-ping
    Computer Engineering. 2009, 35(7): 20-22. https://doi.org/10.3969/j.issn.1000-3428.2009.07.007
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to solve real-time performance of large field, high resolution optoelectronic processing system, this paper analyzes the characteristic of task of detecting optoelectronic small weak targets and the primary principle of design of real-time parallel processor through three factors such as processor nodes, memory and the network connecting the processor nodes. It puts forward a parallel processor consists of double DSP + FPGA based on SRIO technology. The low level algorithm is realized with FPGA and the high level algorithm is realized by DSPs. A region-decomposing parallel algorithm which is fit for digital image processing is adopted for dividing image-processing task. Experimental result shows the feasibility of the design.
  • YE Qian-yu; LIU Yu; ZHANG Lin
    Computer Engineering. 2009, 35(7): 23-25. https://doi.org/10.3969/j.issn.1000-3428.2009.07.008
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In Wireless Sensor Networks(WSN), it is not energy efficient to transfer the redundant information. This paper distributes source coding which is utilized to compress the information, but the method needs the suitable network structure. The cluster based network and distributed source coding method are combined to meet the requirement of energy efficiency of WSN. Simulation results show that there is about 25%-30% less energy consumption than the conventional routing method.
  • MAO Yan-e; ZHOU Zhen-hui; WANG Hong-wei ; JING Yuan-wei
    Computer Engineering. 2009, 35(7): 26-28. https://doi.org/10.3969/j.issn.1000-3428.2009.07.009
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    A robust Active Queue Management(AQM) congestion control algorithm is proposed on the basis of the observer to solve the problem of dynamic network congestion. The algorithm ensures asymptotic stability of time-varying delay uncertainties network systems with input saturation. Through Lyapunov stability analysis, the system can be stabilized with the obtained observer and the state feedback control law by using Linear Matrix Inequality(LMI). Simulation results demonstrate that it can lead to the convergence of the queue length to the set value quickly and maintain the oscillation small, and also shows that the scheme is very roubust against disturbance and burst flows under various network conditions.
  • ZHOU Yan; ZHAO Hai; WANG Shi-juan; LI Hai-cheng
    Computer Engineering. 2009, 35(7): 29-31. https://doi.org/10.3969/j.issn.1000-3428.2009.07.010
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Due to the noise interference, distance errors of unknown node can be occurred in the 3-D smart space. The errors due to positioning an regional are studied, and some conclutions are deduced that errors of uknown node are smallest when positioning reference nodes and unknown node meet certain conditions. Thus, the layout model theorem of space positioning refrence nodes is proposed. Simulation results indicate that the positioning reference point placement theorems can meet the requirement of samrt terminal’s real-time positioning and possesses the preferable positioning precision.
  • Software Technology and Database
  • HU Zhi-gang; FU Yi; XIAO Peng; HU Zhou-jun
    Computer Engineering. 2009, 35(7): 32-34. https://doi.org/10.3969/j.issn.1000-3428.2009.07.011
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To provide trust-enhanced service selection, a Bayesian network-based approach is presented to evaluate the trustworthiness of grid service. The approach employs performance history to construct an evaluation model, which ranks the services in terms of their trustworthiness of Quality of Service(QoS) information and considers degree of consistency in QoS provision. Experimental result shows that the proposed approach can effectively enhance trustworthiness of service information and improve feasibility of service selection.
  • SHI Ye-ling; LI Jian-hui
    Computer Engineering. 2009, 35(7): 35-38,5. https://doi.org/10.3969/j.issn.1000-3428.2009.07.012
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to overcome the shortcomings of automatic-schema-translation method in heterogeneous relational databases integration, the paper investigates the relational model, XML Schema model and the mappings generated by the automatic schema translation, and proposes RtoX, a universal model of mappings, up on the formal definitions of relational Schema and XML Schema. Formal definition of RtoX is given in detail, as well as the model-validating algorithm. A new method based on this universal model to integrate relational databases is described in detail along with a simple application.
  • CHEN Yu; LI Zhi-shu; JIN Hu; HE Jiang
    Computer Engineering. 2009, 35(7): 39-41. https://doi.org/10.3969/j.issn.1000-3428.2009.07.013
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To solve the key problem of how to calculate the effective subset of the control flow paths of the tested program, this paper proposes an algorithm for working out the subset of the whole control flow paths of the program with the help of the LCC compiler in software coverage testing. The control flow paths in the subset are based on the strategy that the loop statements in the program execute less than twice. Adjoining matrix is introduced by the algorithm to work out the subset of the control flow paths with the help of customized stack structure. The correctness of the algorithm is verified on a sample program, which indicates that the subset of the whole control flow paths of the tested program can be calculated efficiently and accurately with this method.
  • LI Dong; PENG Yu-hui; YIN Jiang-long
    Computer Engineering. 2009, 35(7): 42-45,4. https://doi.org/10.3969/j.issn.1000-3428.2009.07.014
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Traditional index structures do not work well on moving objects because of the need of frequently updating the index which leads to the poor performance. This paper presents a novel indexing structure based on Quadtree and Hash table, namely the QH-index which can index the past, present and future positions of moving object and can support moving object’s range queries and point queries which include the object identifier based query. Merging timely the corresponding nodes to degrade the depth of the tree can guarantee the query efficiency. Experiments show that the QH-index gains much better performance in updating and in querying by the object identifier than those of the TPR-tree, and the efficiency of range query is no less than that of TPR-tree.
  • CHEN Yong-chao; LIU Gui-quan
    Computer Engineering. 2009, 35(7): 46-48. https://doi.org/10.3969/j.issn.1000-3428.2009.07.015
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    A new way of clustering the search results based on named entities——NEC is introduced. This paper proposes an algorithm which improves the readability of the cluster labels. Named entities all have certain meanings with themselves, can indicate themes of the documents which they are in, and more readable for the users. It uses the named entities in the documents as cluster labels, and gets the final result after label-selecting and cluster-merging strategies. Experiments show that it is a feasible way for search results clustering.
  • FANG Yuan-kang; HU Xue-gang; XIA Qi-shou
    Computer Engineering. 2009, 35(7): 49-51. https://doi.org/10.3969/j.issn.1000-3428.2009.07.016
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Session identification is an important step in data preprocessing of Web log mining. This paper proposes an improved session identification algorithm. After identifying users, effective Web pages in experiment are reduced greatly by filtering frame pages, and the access time threshold is adjusted by the Web contents and site’s structure on this condition. Compared with the traditional method that defines a uniform threshold for all Web pages experimentally, the approach can decide the access time threshold more accurately. Experiments proves that the algorithm enhances the quality of transaction session.
  • GONG Hong-cui; YU Jiong; HOU Yong; LIU Hong-wei
    Computer Engineering. 2009, 35(7): 52-54,5. https://doi.org/10.3969/j.issn.1000-3428.2009.07.017
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    According to dynamic characteristic of grids, especially to the dynamic QoS request of the users, this paper describes multiple QoS attributes as utility functions. Meanwhile, considering the system indexes such as makespan, load balancing, the notion of system utility is introduced, and the load balance degree is defined to guide and evaluate the scheduling. A user QoS and system index guided task scheduling algorithm, UQSI, is presented to schedule independent tasks with multiple QoS. Simulation results reveal that the algorithm can satisfy the user’s demands and improve the system’s performance, and it is more suitable for the complex grid environments.
  • XIN Min-jie; GAO Jian-hua
    Computer Engineering. 2009, 35(7): 55-58. https://doi.org/10.3969/j.issn.1000-3428.2009.07.018
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In response to the limitation in Graphic User Interface(GUI) automated test, this paper proposes an improved framework for Daily Automated Regression Tester(DART). DART process and event-flow graph is studied. And event-interaction graph which can simplify the event-flow graph is also included. In addition, DART is improved as an efficient and brief automated smoking test framework by weight selection.
  • LIU Qun; LENG Ping; SUN Ling-yu
    Computer Engineering. 2009, 35(7): 59-60,6. https://doi.org/10.3969/j.issn.1000-3428.2009.07.019
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    By analyzing the human concept formation, this paper proposes an incremental concept formation approach based on alternate object and attribute. This approach starts from the empty concept lattice, inserts alternately object and attribute of the formal context, and constructs the concept lattice incrementally. The approach can construct incrementally the concept lattice, and avoid reconstructing the concept lattice in view of the attribute and the object alternately increasing.
  • LI Xin-ke; ZHANG Lei-lei
    Computer Engineering. 2009, 35(7): 61-63. https://doi.org/10.3969/j.issn.1000-3428.2009.07.020
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The effectiveness of user session test techniques depend on the set of collected user session data, the wider this set, the greater the capability of the approach, but the wider the user session data set, the greater the cost of collecting, storing and analyzing data. For solving this problem, this paper develops a technique for reducing user session data sets in Web application test. It uses concept analysis for clustering user sessions to test suite reduction, and exploits incremental concept analysis to update data sets. It constructs an automated test prototype framework that enables to evaluate effectiveness of reduced user session data sets.
  • XU Wen-jie; CHEN Qing-kui
    Computer Engineering. 2009, 35(7): 64-67. https://doi.org/10.3969/j.issn.1000-3428.2009.07.021
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes a parallel cluster crawling model to improve the mining efficiency of massive data on Web. For fully using of the ability of parallel nodes in computer cluster, a vector measurement technology is introduced to solve the matching problem between crawling task and computer node. After giving the definitions of crawling task vector and computer node vector, cosine vector similarity formula is described, and the parallel crawling algorithms is designed. Experimental results show that the system is effective in distribution adaptability and load balance.
  • CHENG Jiang; YI Yun-fei; LIN Jian-hui; YU Qi-gang
    Computer Engineering. 2009, 35(7): 68-69,7. https://doi.org/10.3969/j.issn.1000-3428.2009.07.022
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In view of that the boolean association rules can not express the association of fuzzy data, this paper gives a series of definitions of fuzzy association rules and proposes a mining algorithm based on prefix tree for fuzzy association rules. The algorithm can store fuzzy pattern candidate sets and frequent sets compressibly by constructing prefix tree, which effectively saves the memory cost, besides it only scans database twice. The efficiency of the algorithm is verified by the experiment.
  • DENG Xin-guo; LIN Zi-yu; XIAO Ru-liang
    Computer Engineering. 2009, 35(7): 70-72. https://doi.org/10.3969/j.issn.1000-3428.2009.07.023
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The model-to-code transformation is necessary in development environments for model-driven business process based on some standards. This paper presents a novel solution method based on structured theorem, which shifts the unstructured programs to structured ones for Business Process Modeling Notation(BPMN). The feasiblilty of the method is tested through a practical case. Experimental results show this method is effective.
  • YANG Xi-quan; CHEN Xue-feng
    Computer Engineering. 2009, 35(7): 73-75. https://doi.org/10.3969/j.issn.1000-3428.2009.07.024
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The existent XML signature technique has two problems that it is hard to search subtle granular information and slowly to sign. In order to solve these problems, this paper proposes a transformation algorithm of abstract information based on the semantic path, in the light of constructed XML index tree. With using transformation algorithm, speed of searching information is quick. Through signing the semantic path and the information, the efficiency and flexibility of signature can be enhanced, date security and good formation of document can be guaranteed.
  • CAI Liang; HUANG Hao
    Computer Engineering. 2009, 35(7): 76-77,8. https://doi.org/10.3969/j.issn.1000-3428.2009.07.025
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    By using the way of P2P, a P2P storage system is built. The nodes involved in the system are grouped by the number of the subject of resources and the predicted network distance. The overlay network maintained by super nodes is formed, which implements effective management of nodes. Moreover, by using the multiplex naming mechanism of nodes, the efficiency of data storage in system is promoted. The performance of this storage system is testified on the basis of simulation experiment.
  • FANG Gang
    Computer Engineering. 2009, 35(7): 78-81. https://doi.org/10.3969/j.issn.1000-3428.2009.07.026
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    An Algorithm of Constrained Association Rules Mining Based on Binary(ACARMB) is presented, which uses digital section to ascertain rang of candidate frequent items that can crosswise generate candidate items by the methods about digital ascending and descending, computes support by binary logic operation and uses digital character to reduce the number of scanned transactions, and extracts association rules satisfied with Constrained Condition(CC). This algorithm is suitable for mining any length frequent item sets, and has higher efficiency for calculation.
  • ZHANG Yu-yan; FENG Wen-tang; HUANG Guo-dong; HOU Jin-kui
    Computer Engineering. 2009, 35(7): 82-84. https://doi.org/10.3969/j.issn.1000-3428.2009.07.027
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    According to the theory of Model-driven Architecture(MDA) and following the idea of the separation between application descriptions and UI designs, a development model for Web report systems is proposed systematically based on the high-level abstract description of system function and user interface. ASP.NET is used as a target platform in the experiment which shows that this approach follows the essence, process and requirements of MDA. Great improvement of quality and efficiency of Web report system development can be obtained, for the model can be well combined together with application development models.
  • Networks and Communications
  • LI Hong-fei; ZHAO Ming
    Computer Engineering. 2009, 35(7): 85-86,8. https://doi.org/10.3969/j.issn.1000-3428.2009.07.028
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    A network processing structure based on multiple PicoBlaze soft-core processors is introduced. The structure suits for parallel process and hierarchical network processing. Using this structure, it is easy to develop some complex network protocols. As an example, a complete remote boot server is implemented on FPGA.
  • ZHOU Si-qing; CHEN Rui-biao
    Computer Engineering. 2009, 35(7): 87-89. https://doi.org/10.3969/j.issn.1000-3428.2009.07.029
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problems of lower location coverage rate for APIT location algorithm in Wireless Sensor Network(WSN), an improved algorithm called IAPIT is proposed. The performances of these two algorithms are compared in terms of different anchor ratios, different communication radius of nodes and the same anchor ratio node. Simulation experimental results show this improved algorithm has higher location coverage rate.
  • LIU Hong-bo; CAI Hui; HAN Guo-dong
    Computer Engineering. 2009, 35(7): 90-92. https://doi.org/10.3969/j.issn.1000-3428.2009.07.030
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problem of no optimizing path in Mobile IP(MIP), this paper proposes an approach based on the active network technology that makes these nodes deal with the active packets in MOBEE which is an execution environment in the MAC. It is proved that MOBEE in ACR can make mobile IP come true. And there is the better performance of MOBEE than Execution Environment(EE) based on the language of C or Java.
  • LI Wei; WEN Li
    Computer Engineering. 2009, 35(7): 93-95. https://doi.org/10.3969/j.issn.1000-3428.2009.07.031
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Due to large amount of backbone bandwidth consumed by Peer-to-Peer(P2P) connections in BitTorrent(BT) system, centralized management and peer-node clustering to globally manage P2P traffic is proposed. Through the network distance measurement, all nodes with their neighboring nodes are put in one group. The experimental result shows that the strategy can greatly increase proportion of local area network traffic while download time of BT is not apparently affected, and restrictions on the data flow within the local area network is a good way to reduce backbone load.
  • ZHANG Wen-bo; FENG Yong-xin; JIANG Yue-qiu; XU Ye
    Computer Engineering. 2009, 35(7): 96-98. https://doi.org/10.3969/j.issn.1000-3428.2009.07.032
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In the integrated satellites information network, in terms of the characteristic that satellite network node is far from terrestrial station, the transmission delay is high and the mobility is intensive. Based on that characteristic, a new architecture is built for the network management architecture of the integrated satellites information network. The method that disposes a Local Policy Decision Point(LPDP) between the traditional Policy Decision Point(PDP) and Policy Enforcement Point(PEP) is proposed, and the LPDP can exchange the policy information with the PEP by disposing the LPDP on the satellite agent which has the function of management. The simulation results show that the new architecture can reduce the response delay of the network management.
  • ZHANG Xiao-xiao; TANG Lun; CHEN Qian-bin; WU Xiao-ping
    Computer Engineering. 2009, 35(7): 99-100,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.033
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes Call Admission Control(CAC) algorithm based on VCG mechanism in wireless networks. It uses VCG mechanism regulating user needs, controls the allocation of resources, achieves the rational use of resources. The algorithm is fit for multi-business services and multi-priority services in wireless networks. Simulation results show that the performances verify the considerable improvement achieved by the integration of VCG mechanism with CAC in wireless networks, and it can improve the link level QoS performance.
  • WANG Jian-wei; YANG Zhong-xiu; SONG Jia-tao; REN Xiao-bo; YU Wei-ying
    Computer Engineering. 2009, 35(7): 101-103. https://doi.org/10.3969/j.issn.1000-3428.2009.07.034
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    P2P networks always use the Chord algorithm which uses the uniform Hash function for separately assigning a m identifier to the key word and the node, and obtains a ring-like structure which is loaded with balance. This paper introduces connectirity choice into Chord algorithm. According to the received and dispatched information analysis of the node, it establishes the key value of information connects, adjusts the sequence of the node and the key word, and reduces the inquiry and transmission of the low effect information. Simulation experiment proves that the improved algorithm has a lower scotoma rate, higher effect and higher search ability.
  • MENG Xiang-zhong; SONG Bao-ye
    Computer Engineering. 2009, 35(7): 104-106. https://doi.org/10.3969/j.issn.1000-3428.2009.07.035
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper presents a kind of centroid locating algorithm based on hop weighted. This algorithm is based exclusively on connectivity and uses hop information as weight factor to calculate the estimated position of unknown nodes, whose hop influences the result of polygon centroid calculation. Simulations demonstrate that this algorithm can reduce the average locating errors and it is fit for mass Wireless Sensor Network(WSN) application.
  • ZHANG Shuo; ZHAO Rong-cai; AN Ke
    Computer Engineering. 2009, 35(7): 107-109. https://doi.org/10.3969/j.issn.1000-3428.2009.07.036
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper presents a generating technology of self-similar test traffic. By comparing typical self-similar traffic models, it analyzes the multifractal wavelet model, brings out an algorithm to generate self-similar test traffic, and applies it on the network testing system based on IXP2400 programming multi-core processor. With the network testing system, it presents the test results of the self-similar test traffic according to the value of Hurst (0.5, 1.0) and proves the correctness and validity of the paper’s technology.
  • CHEN Wen-bin; LI Xiao-hong; WANG Dong
    Computer Engineering. 2009, 35(7): 110-112. https://doi.org/10.3969/j.issn.1000-3428.2009.07.037
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Topology control tries to reduce radio interference, lower energy consumption and enhance network capacity by adjusting transmission power of nodes. This paper studies some classical topology control algorithms and finds that the resulted topology has several nice properties such as connectivity, less node degree and transmission range. Simulation results show that when the delivery rate of packets in the network is greater than a certain value, which affects the effectiveness of topology control. The network’s throughput can be significantly improved.
  • WANG Xin-ying; WU Zhao
    Computer Engineering. 2009, 35(7): 113-115. https://doi.org/10.3969/j.issn.1000-3428.2009.07.038
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes an optimized routing protocol based on AODV for Mobile Ad Hoc Network(MANET), optimizes the RREQ dissemination mechanism, HELLO mechanism and link repair mechanism, provides many pathways not related, which reduces network routing cost, improves the data transfer ratio and balances network load. The performance of simulation shows that it is better than AOMDV in packet delivery ratio, average delay and routing cost.
  • SUN Pei; KONG Fan-ting
    Computer Engineering. 2009, 35(7): 116-118,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.039
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To select the best transmission rate based on dynamic channel quality in IEEE 802.11 wireless networks, this paper presents weighted frame-loss-ratio based adaptive threshold rate adaptation algorithm. It calculates weighted frame loss ratio and decides upscaling and downscaling thresholds. The mechanism avoids wrong choice because of random frame loss and rate fluctuation. Simulation results show that it is robust for Rayleigh and Ricean channel, the throughput is better than other rate adaptation solutions.
  • HUANG Wen-ming; CAI Yang; LEI Ling-hong; ZHU Ying
    Computer Engineering. 2009, 35(7): 119-121. https://doi.org/10.3969/j.issn.1000-3428.2009.07.040
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The characteristics of power law of non-scale network are applied to improve the structured P2P overlay. BAChord model based on power law is presented, which makes full use of the information of nodes including connection degree and physical adjacency to help the overlay match the physical topology at some extent, thus increasing the probability of discovery of short-chain link in the routing process. Experiments show that this model has the advantage of routing performance over the Chord model by increasing a slight routing overhead.
  • WANG Xin-sheng; CUI Xiao-wei; DONG Lei-gang; LI Chun-feng
    Computer Engineering. 2009, 35(7): 122-124. https://doi.org/10.3969/j.issn.1000-3428.2009.07.041
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Resource search is an important problem of research for unstructured P2P system. In order to improve the efficiency of the data search, it proposes an algorithm for intelligent search based on skyline query technology. Combining the similarity of messages and the hit rate of nodes, a best set of neighbor nodes which send messages is obtained by skyline query technology. The experiment shows that it can obviously reduce the number of messages and advance the efficiency of query.
  • Security Technology
  • WAN Xiao-yu; YANG Bo; FAN Zi-fu
    Computer Engineering. 2009, 35(7): 125-126,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.042
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Pattern matching algorithm is an important method in intrusion detection. The matching efficiency is the bottleneck of intrusion detection, which indirectly affects the real-time performance of the system. By analyzing four pattern matching algorithms, this paper proposes an improved Sunday algorithm. It finds the character-word in the pattern string before matching, and then compares the character-word and the last word of the pattern string with the text. If the matching fails, the character-word will be moved as far as possible. Experiment of result proves that its match efficiency is better than Sunday algorithm.
  • HONG Zhi-guo; WANG Yong-bin; SHI Min-yong
    Computer Engineering. 2009, 35(7): 127-129. https://doi.org/10.3969/j.issn.1000-3428.2009.07.043
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    On the basis of analyzing the communication process and geometric characteristics, this paper constructs a Stochastic Petri Net(SPN) model for Low Earth Orbit(LEO) satellite network under full duplex mode. Contraposing to two types of security protocols: IKE and ISAKMP’s aggressive mode and Space Communication Protocol Specification-Security Protocol(SCPS-SP), average time delay of satellite networks is analyzed and model-based simulation is conducted with Stochastic Petri Net Package(SPNP) 6.0.
  • TANG Yang; LAN Ju-long
    Computer Engineering. 2009, 35(7): 130-131,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.044
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The efficiency of re-keying is a main factor that influences the performance of secure multicast. This paper presents a new key management algorithm to resolve the problem of high computation cost. The algorithm based on the membership of group is believable and choose a group leader to make a group key for the secure multicast, which introduces the key management algorithm into key agreement algorithm. The algorithm can reduce the computational complexity of group key generation, and improve the scalability of group key management algorithm.
  • TIAN Bo; CHEN Yi-yun; WANG Wei; LI Zhao-peng; WANG Zhi-fang
    Computer Engineering. 2009, 35(7): 132-135. https://doi.org/10.3969/j.issn.1000-3428.2009.07.045
    Abstract ( )   Knowledge map   Save
    This paper introduces the design and implementation of a certifying complier back end for a C-like language, PointerC. The back end adopts the strongest post-conditions calculation for both integer assertion and pointer assertion synchronously, and proofs the verification conditions involving integers and pointers. It generates the proofs of pointer safety at the assembly level full-automatically, and solves the problem of the non-uniform alias analysis in commonly used data structures. The proof checker, which checks the integrity of proof-carrying code, is included in the back end.
  • LIANG Zhong-yin; YANG Xiao-yuan; ZHOU Xuan-wu; SHI Guo-wei
    Computer Engineering. 2009, 35(7): 136-138,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.046
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper shows that Lu’s proxy signature is insecure against a key replacement attack and malicious-but-passive(KGC) attack. The attack shows that anyone who replaces both the original signer and the proxy signer’s public key can forge valid signatures which seems to be generated by the proxy signer on behalf of this adversary himself, and KGC who can’t replace the public key can get the signers’ private keys and forge valid signatures. To thwart these attacks, this paper improves the scheme. The improved scheme is secure against the key replcement attack and the malicious-but-passive KGC attack.
  • SANG Jun; XIANG Hong; YE Chun-xiao; HU Hai-bo
    Computer Engineering. 2009, 35(7): 139-141. https://doi.org/10.3969/j.issn.1000-3428.2009.07.047
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper presents a kind of zero-watermark technique based on neural network in wavelet domain. Different from the traditional watermarking techniques, it extracts the feature from the host image and uses the feature as the watermark for copyright proof, which can eliminate the image distortion of watermark embedding. The host image is transformed via Discrete Wavelet Transform(DWT) and some coefficients at the low frequency sub-band are selected randomly as the extracted feature. A neural network is used to manage, store and detect the selected feature coefficients, which improves the practicability of the algorithm.
  • SHI Ming-ruo; JIANG Zhong-hua
    Computer Engineering. 2009, 35(7): 142-143,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.048
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper puts forward a new authenticated key agreement protocol for Wireless Local Area Network(WLAN), which provides secure mutual authentication, key establishment and key confirmation. It can resist dictionary attacks by either passive or active networks intruder, man-in-the middle attack, and impersonate attack. It also offers perfect forward secrecy. The protocol combines techniques of challenge response protocols with symmetric key agreement protocols. Both computational and communication performance are analyzed.
  • HAN Ji-hong; ZHOU Zhi-yong; GUO Yuan-bo; WANG Ya-di
    Computer Engineering. 2009, 35(7): 144-146,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.049
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper introduces two attacks on security protocols which have cipher homomorphism, defines term representation and security protocol model based on role’s action sequences, proposes a method of transforming role instances interleaving to sequence of constraints, extends the Dolev-Yao intruder model with equational theory, putforwards a intruder deduction system in first-order logic, models and verifies NSL protocol using constraint solving method, discovers the attack based on cipher homomorphism which can not be found under perpect cryptography assumptions.
  • WU Xiang; YANG Xiao-yuan; ZHANG Min-qing; ZHOU Hong-fei
    Computer Engineering. 2009, 35(7): 147-149. https://doi.org/10.3969/j.issn.1000-3428.2009.07.050
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper presents an audio zero-digital watermark algorithm based on statistical feature of approximation signal and the technology of restoration of the linear scaling. The original watermark is error corrected coded with repetition code. In order to enhance the correct probability of detecting, the recoded watermark is modulated with PN sequence P which is produced by the key through Pseudo-random sequences generator, with the same length of the recoded code. During the process of detecting, the technology of restoration of the linear scaling is used to eliminate the influence brought by linear scaling on time axis, and a blind detecting is realized. Experiment confirms the expected high robustness to many kinds of attacks especially to cutting-attack and MP3-attack.
  • QI Mei; GUO Ya-jun; YAN Hui-fang
    Computer Engineering. 2009, 35(7): 150-152. https://doi.org/10.3969/j.issn.1000-3428.2009.07.051
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes a hybrid anti-pollution mechanism based on Peer reputation and object reputation where the reputation of selected object is determined by the reputation of voting Peers. Severe penalty strategy and polling incentive mechanism are introduced in computing Peer reputation, which can isolate the polluters effectively, stimulate users’ awareness for file pollution, and avoid pollution further spreading. Simulation results show that compared with object reputation system, hybrid reputation mechanism converges faster, and has better performance of anti-pollution.
  • TIAN Hong; WANG Ya-wei; WANG Xiu-kun
    Computer Engineering. 2009, 35(7): 153-155. https://doi.org/10.3969/j.issn.1000-3428.2009.07.052
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    With the growing concern over data privacy-preserving problem, how to discover association rules from distributed databases becomes one of the hot topics of this field. This paper is devoted to study the problem of discovering global frequent itemsets from distributed vertically partitioned databases with the goal of preserving the confidentiality of each database. All sites are worked together to find global frequent itemsets without revealing private data, each one holds some attributes of global database. The paper presents an invertible matrix-based encryption protocol based on the research of commodity server model, which protocol is of great confidentiality, effectiveness and correctness for distributed vertically partitioned databases.
  • LI Guo-min; HE Da-ke;
    Computer Engineering. 2009, 35(7): 156-157. https://doi.org/10.3969/j.issn.1000-3428.2009.07.053
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Authenticated Group Key Agreement(AGKA) protocol allows a group of users in a public network to share a session key which may be used to achieve desirable secure communication. The protocol compiler proposed by Abdalla et al transforms any provably secure authenticated 2-party key establishment into a provably secure AGKE with two more rounds of communication. This paper points out an impersonation attack against the protocol compiler that two malicious users can impersonate an entity to agree some session keys in a new group if these two malicious users have the previous commitment transcripts of the entity. The improvement to eliminate the security leaks is proposed.
  • HU Yan-zhi; MA Da-wei; TIAN Zeng-shan; TIAN Tian
    Computer Engineering. 2009, 35(7): 158-160. https://doi.org/10.3969/j.issn.1000-3428.2009.07.054
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    According to the feature of wireless group communication, a new group key distribution protocol in wireless network is proposed based on homomorphic encryption scheme. The scheme can realize member node authentication, anonymous identity, forward privacy and backward privacy. Calculation in advance introduced in initialization phase reduces the calculation cost of terminal in the process of execution, and satisfies real time request of wireless group communication. Moreover, performance of the scheme is analyzed. Result shows that the scheme has higher security, lower complexity, which satisfies the requirements of wireless network.
  • KANG Xiao-wen; YANG Ying-jie; WANG Bao-yun
    Computer Engineering. 2009, 35(7): 161-163. https://doi.org/10.3969/j.issn.1000-3428.2009.07.055
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper makes a brief analysis about the existing secure file system, and in the view of the problems of these secure file system, this paper designs a kind of disaster-oriented strong secure file system. The system is based on Linux, and it introduces the technology of virtual storage. According to the rationale of stackable file system, it addes the layers of authentication, audit, access control, transparent encryption and disaster tolerant to implement virtual file system which integrates disaster tolerant and strong security.
  • ZHANG Yu-li; CAI Qing-jun
    Computer Engineering. 2009, 35(7): 164-165. https://doi.org/10.3969/j.issn.1000-3428.2009.07.056
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The algebraic immunity is a criteria on evaluating the property of the Boolean function recently. This paper studies the relation of the nonlinearity and algebraic immunity of the functions having odd weight and gives the lower bound of the nonlinearity when the algebraic is given, further shows that this bound is tight. The weight distribution has a range when the algebraic immunity is given, i.e. d, and shows that the range is also tight, that is for any integer in that range, there exists a Boolean function that its weight is t and the algebraic immunity is greater than d.
  • Artificial Intelligence and Recognition Technology
  • LIU Li-mei; CAI Zi-xing; PAN Wei
    Computer Engineering. 2009, 35(7): 166-167,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.057
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Map building is very important to accomplish autonomous navigation. This paper presents the research on the problem of uncertainty sensor information processing in map building of mobile robots, and especially focuses on the scattering and specular reflection of sonar. In this paper, an improved method based on probabilistic grid is proposed. This method introduces range confidence factor in ultrasonic model and is applied to autonomous map building of mobile robot. So uncertainty in sonar sensor responses can be successfully reduced. Experimental results in robots platform indicate that the model improves the performance of sonar sensors.
  • YANG Bin; MA Rui-sheng; HE Li; YANG Lei; SUN Han; LIU Zhi-ping
    Computer Engineering. 2009, 35(7): 168-169. https://doi.org/10.3969/j.issn.1000-3428.2009.07.058
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Focus on forest fire detection by Unmanned Aerial Vehicle(UAV) remote sensing, this paper proposes a new algorithm of identifying fume based on the color features of its spectrum. Due to the require of real-time processing, it uses linear de-noising method to remove the noise. It develops the program using this method by interactive data language. Experimental results show the identification rate with the method is higher than 77.71%.
  • CAI Zhao-quan; HUANG Han
    Computer Engineering. 2009, 35(7): 170-171,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.059
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Classical Particle Swarm Optimization(PSO) algorithm has bad diversity and is easy to converge locally. This paper puts forward a smallest- variation-first mutation to design an improved CLPSO algorithm named as CLPSO-M algorithm. The experimental result of solving the benchmark problems indicates that CLPSO-M performs better and more steadily than CLPSO.
  • YUAN Ning; WU Xiao-jun; WANG Shi-tong; YANG Jing-yu; Josef Kittler
    Computer Engineering. 2009, 35(7): 172-174,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.060
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    An improved face verification algorithm is proposed based on the modular 2DPCA and Client Specific Kernel Discriminant Analysis (CSKDA) because of the disadvantage of CSKDA. CSKDA first transforms an image matrix to a vector which caused high dimensionality and computational complexity and not considered the local feature. While the method first extracts the local features with the original images which are divided into modular sub-images, CSKDA is utilized. The discriminant information obtained from the between-class and the within-class scatter matrix are developed. Moreover, client specific subspace can describe the diversity of different faces. Experimental results obtained on XM2VTS and ORL show the effectiveness of the method.
  • WANG Xiao-bin; WEN Chun; SHI Zhao-xiang
    Computer Engineering. 2009, 35(7): 175-176,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.061
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at organizing hidden Web databases according to their topic domains, this paper proposes an Independent Component Analysis(ICA) based algorithm for hidden Web domain clustering. Text is extracted from search interface pages as common Web pages, and TF-IDF formula is applied to weight terms. After selecting the top N-highest weight terms to construct VSM, the algorithm performs a singular value decomposition to implement features reconstruction. It applies ICA decomposition to obtain the cluster information. The main idea is utilizing the co-occurrence analysis and noise eliminating ability of Latent Semantic Index(LSI) to improve cluster performance. Experiment shows that the average precision is higher than 90 percent.
  • XIE Jian-bin; LIU Tong; CHEN Zhang-yong; CHENG Yong-mao
    Computer Engineering. 2009, 35(7): 177-179. https://doi.org/10.3969/j.issn.1000-3428.2009.07.062
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to solve the problem of characteristic extracting for bill’s fibrin, a new characteristic extracting method for bill’s fibrin based on extremum filtering and OTSU is proposed in this paper. Extremum filtering is used to enhance characteristic of fibrin and integrate the complex background. And the optimal OTSU algorithm is proposed to partition the image of bill and extract characteristic of fibrin. The experiment shows the good effect and fast speed of this method, and its consistency performance is above 91%.
  • TANG Hui; LI Bi-cheng; QU Dan; ZHANG Lian-hai
    Computer Engineering. 2009, 35(7): 180-182. https://doi.org/10.3969/j.issn.1000-3428.2009.07.063
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper presents compressed speaker recognition approach for VoIP(Voice over IP) which is based on the micro-clustering algorithm. It designs a framework based on the micro-clustering algorithm and performs speaker recognition on the feature vector which is directly extracted from G.729, G723.1(6.3 Kb/s), G723.1(5.3 Kb/s) compressed stream. Experimental result shows that the new method is more accurate and efficient than the widely used Gaussian Mixture Model(GMM) which uses the same feature vector.
  • WANG Xiao-bin; WEN Chun; SHI Zhao-xiang
    Computer Engineering. 2009, 35(7): 183-185. https://doi.org/10.3969/j.issn.1000-3428.2009.07.064
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    There are many applications that can benefit from topic identification and keyword extraction. The traditional way of choosing the topic number depends on human labeling or automatic clustering which is immeasurable. This paper utilizes the Bayes Information Criteria(BIC) based model selection theory to evaluate the probability of each topic numbers taking. After the topic number is acquired, the paper implements the Independent Component Analysis(ICA) decomposition of term-document, then calculates the weight and extracts the keyword according to the ICA separating matrix. Experiments show this method extracts the keyword in a meaningful way.
  • GE Hong-wei; LIU Lin-ju
    Computer Engineering. 2009, 35(7): 186-188. https://doi.org/10.3969/j.issn.1000-3428.2009.07.065
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To solve the rectangle packing problem, one of the NP hard problems, a bi-group Particle Swarm Optimization(PSO) algorithm with mutation operator is presented, by searching the two sub-groups which are parallely performed, it can achieve a better location in both overall and local situations. Information is exchanged when sub-groups are reorganized. Meanwhile, mutation operator is adopted to search in the neighborhood of local optimal solution. The effectiveness of this algorithm is demonstrated for rectangle-packing problem through a number of typical computational experiments.
  • WANG Xin-sheng; LI Xue; JIA Dong-yan
    Computer Engineering. 2009, 35(7): 189-190,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.066
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Resources search is one of research hotspots in the filed of P2P technique. Aiming at the problems of existing P2P resources search algorithm, such as huge message packets, low search efficiency, this paper proposes resources search mechanism based on Ant Colony Optimization(ACO) for unstructured P2P. This mechanism directs the query routing effectively according to the positive feedback principle of the ant pheromone. It sends the query messages to the area where may store the requested resources. Simulation experimental results show that this mechanism can achieve better search performance by increasing the resource hit ratio and reducing the query message packets of redundancy.
  • CHEN Chao-xiang; DING Jian-long; CHEN You-rong
    Computer Engineering. 2009, 35(7): 191-194. https://doi.org/10.3969/j.issn.1000-3428.2009.07.067
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Based on comprehensively research on the SMS technology and employment, this paper proposes the SMS employment services solution on Web site based on ontology and Agent technology. It establishes the ontology of employment filed, proposes compatible matching algorithm, constructs the tripartite Agent model of job-seekers, recruitment and intermediary and realizes intelligent search and two-way push. And taking Agent for example, the realization process of SMS employment platform is described.
  • WU Li-sheng; QUAN Long; YANG Fu-sheng
    Computer Engineering. 2009, 35(7): 195-197. https://doi.org/10.3969/j.issn.1000-3428.2009.07.068
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    An improved Ant Colony Algorithm(ACA) is presented that can be used to search shortest grinding path for the grinding machine and the grinding path optimization mathematical model is given. The attractiveness normalization is introduced and adopted to decrease the optimization result’s dependence on the parameters selected in the basic model ant colony algorithm. Simulation experiment and application show that the location path is shortened greatly after the path is optimized by ant colony algorithm.
  • WU Jian-jie; WANG Yong-xin; QIU De-hong; FANG Shao-hong
    Computer Engineering. 2009, 35(7): 198-199,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.069
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To solve the problem of high sensitivity to position of component that exists in the algorithms widely used in Automatic Optical Inspection(AOI) for chip component, an automatic optical inspection algorithm is presented. The gray-level change of component image in local feature region is used as a detection criterion. By introducing local principal wave probability, the similarity between sample image and inspected image is described by intrinsic characteristic of the principal wave. Thus it can be decided whether the component is a defective one. The type of defect can be concluded. Experimental results show that the algorithm is effective.
  • Graph and Image Processing
  • GUO Yao; ZHANG Min-qing; YANG Xiao-yuan; LIU Jia
    Computer Engineering. 2009, 35(7): 200-202. https://doi.org/10.3969/j.issn.1000-3428.2009.07.070
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    For a lot of images to be retrieved, after cut step, not only the big areas contribute to the whole image and human vision, but also the small ones do the same effect. This paper proposes to take these small areas into account in order to describe the whole image more completely and more accurately. Abstract the distribute characteristics of the small areas using mean value and variance methods, and then combines them with the big areas’ characteristics, and start retrieving last. The experiment shows that this retrieval method’s performance does better than that considering the big areas’ characteristics only.
  • HU Wei-qiang; ZHANG Cong-pin; LIU Chao; CHEN Zhi-fang
    Computer Engineering. 2009, 35(7): 203-205. https://doi.org/10.3969/j.issn.1000-3428.2009.07.071
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The fast classification of image based on content is the groundwork for real-time searching and filtration of Web image. After the analysis of the distributing characteristics for image feature, a fast image classification algorithm based on local feature is proposed. Compared with the algorithms at present, the features like color, texture, shape are got just in local area of images. And then, images can be fast and automatically classified using Bayesian classifier. In this paper, the algorithm is implemented and the result is improved that the classification is fast and efficient by contrastive experiments.
  • LI Jing-hui; ZHANG Xiao-feng; MA Yan
    Computer Engineering. 2009, 35(7): 206-208. https://doi.org/10.3969/j.issn.1000-3428.2009.07.072
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Image restoration is the main research area in the domain of digital image processing. This paper introduces the Criminisi algorithm in the texture synthesis, and aiming at the problems existed in the process of making use of the priority values on the boundary to adjust inpainting order, an improved algorithm is proposed. This novel algorithm not only improves the effect of image inpainting, but also implements the replacement of big forth scenery wih digital image background. Experimental results show this improved algorithm is effective.
  • OU Xiao-zhe; GENG Guo-hua; FENG Jun; XU Xiang
    Computer Engineering. 2009, 35(7): 209-211. https://doi.org/10.3969/j.issn.1000-3428.2009.07.073
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    A biomedical data restoration algorithm based on statistical deformable model is proposed. Specifically, the statistical model is partitioned into known and unknown parts, and the unknown data are estimated by the prior knowledge constructed by the statistical model and the known data. Experiments on missing liver edge points demonstrate that the restoration error can be controlled at less than 1% under 40% unknown data points with proper resolution.
  • Engineer Application Technology and Realization
  • ZHU Ying; BAI Guo-qiang; CHEN Hong-yi
    Computer Engineering. 2009, 35(7): 212-214. https://doi.org/10.3969/j.issn.1000-3428.2009.07.074
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The method to design PCI-X interface for high speed network security co-processor is introduced, which uses two protocols such as IPSec and SSL/TLS protocols to optimize the system, and deploys different kinds of algorithm engines. The co-processor uses PCI-X bus interface and SoC chip with higher performance, which meets the requirement of both PCI-X bus protocol and the internal data transfers of co-processor. Experimental results show this method is feasible.
  • LIU Lei-ming; WU Yan-kun; TONG Chao-nan
    Computer Engineering. 2009, 35(7): 215-216,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.075
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    For the dynamic output feedback control problem over networks, delay quantization and augmented plant vector method is applied and a discrete time Markov jump system model is established, the stabilization controller design method is presented, the simulation on a cart and inverted pendulum is made. Since the delay quantization method is applied in the system, the model and the design method are also fit for solving dynamic output feedback control problem with known distribution stochastic delays.
  • LI Jing-feng; WANG Xue-jing; YE Fan; REN Jun-yan
    Computer Engineering. 2009, 35(7): 217-219. https://doi.org/10.3969/j.issn.1000-3428.2009.07.076
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    A novel method to implement low-complexity Field Programmable Gate Array(FPGA) for (de)interleaver is proposed, which uses dual-proted memory of Xilinx FPGA. The consumption for resources to FPGA is reduced and the read-width is not necessarily equal to write-width. It is fitting to Multi-Bandwidth Orthogonal Frequency Division Multiplex(MB-OFDM) Ultra-Wide Bandwidth(UWB) systems. Experimental results show the numbers of occupied slices are reduced by 46% and 78% for interleaver and deinterleaver respectively.
  • XIAO Ye-wei; GUO Xue-feng; HUANG Hui-xian; PAN Hong-bin
    Computer Engineering. 2009, 35(7): 220-222. https://doi.org/10.3969/j.issn.1000-3428.2009.07.077
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problem that it is difficult to implement the dynamic optimization for traffic signal in traffic signal machine, a self-adaptive intelligent traffic signal machine system based on dual Multi Controller Unit(MCU) is proposed, and the structure and operating principle of which is introduced clearly. This paper also expatiates the design of the software and hardware of the system. The optimal object of the system is to minimize average vehicle delay to realize self-adaptive control with real-time fuzzy methods. Experimental results show this system has higher integration, powerful functions, reliable performance, and it is convenient to achieve the traffic network coordinated control.
  • LIU Yuan-yuan; TENG Dong-xing; MA Cui-xia; YANG Hai-yan; DAI Guo-zhong;
    Computer Engineering. 2009, 35(7): 223-225. https://doi.org/10.3969/j.issn.1000-3428.2009.07.078
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problems existed in switchover between Computer Aided Design(CAD) and traditional pen-and-paper model, a knitwear craft design system based on pen-and-paper model is proposed. The design workflow for knitwear industry in agile manufacturing environment is researched, and uses sketch recommendation technologies and curve editing technique to implement this system. Experimental results show this system reduces the complexity of use, increases the design productivity, and has gained better effect for application in a factory.
  • WANG Jue; SUN Ya-xing; LIU Rong; XU Li-bing
    Computer Engineering. 2009, 35(7): 226-228,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.079
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to meet the requirement of high speed, stability and reliability of the data transmission for large industrial Computerized Tomography(CT) system, this paper studies the feasibility of the scheme of data transmission based on the combination of switch Ethernet and the TCP/IP protocol. Based on analyzing transmission delay of switch Ethernet, the model of data transmission system for large industrial CT is founded, and test is done by using Socket mechanism. The result and application reflect that this scheme can be applied to data transmission system of large industrial CT, and the satisfying results are procured.
  • CHENG Ling-fei; MA Xing-he
    Computer Engineering. 2009, 35(7): 229-231. https://doi.org/10.3969/j.issn.1000-3428.2009.07.080
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    There are some differences between mine ADSL and common ADSL, the transmission line is the one of the differences. Compared with the common twisted pair, the mine twisted pair has larger diameter, so the pair can carry more data and information. This paper analyzes the characteristics of mine ADSL transmission technology, introduces the procedure for calculating channel capacity of mine twisted pair based on the iterative water-filling algorithm, and on basis of which, the relationship among channel capacity, loop distances and frequency bandwidths of mine twisted pairs.
  • ZHANG Fei; BAI Rui-lin; LU Lin
    Computer Engineering. 2009, 35(7): 232-234. https://doi.org/10.3969/j.issn.1000-3428.2009.07.081
    Abstract ( )   Knowledge map   Save
    This paper proposes a design and implementation of WinCE 5.0 Bootloader based on ARM9. By analyzing the structure of Bootloader, based on the development of WinCE 5.0 BSP of touch-screen controller, it designs and implements WinCE 5.0 Bootloader based on AT91SAM9261. It describes the development process of boot code and main code, and gives the detailed hardware initialization steps of main code. Test result indicates that the Bootloader meets the design objectives and is used in touch-screen controller.
  • Developmental Research
  • KONG Zhong-zhe; WANG Ying-jie; YU Zhuo-yuan
    Computer Engineering. 2009, 35(7): 235-238,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.082
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper discusses the key theory and technology for design of information management and geo-visualization system for population (VisualMap) based on domestic spatial database, named uBase. It proposes system design framework and design method of functional modules. The pilot system is constructed in Shandong province using China fifth census data. It proves VisualMap system achieves a completely home-made, component-based, easy to expand data management and visualization platform.
  • LIN Jia-wen; XIE Xiao-gang; TAO Ying-xue; CAI Jun
    Computer Engineering. 2009, 35(7): 239-241,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.083
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to improve adaptive optics simulation and modeling abilities to SciSimu which is developed independent, Code for Adaptive Optics System needs to be added to SciSimu as an expansion. The design principle of expandsion using interface library, automatic management and inside compiler is presented. A lot of simulation tests are implemented, which testifies the design principle can make SciSimu have ability of adaptive optics simulation, and be more user-friendly and expandable.
  • WANG Xuan-qiang; WANG Xiang-qian; ZHANG Yun-quan;
    Computer Engineering. 2009, 35(7): 242-244. https://doi.org/10.3969/j.issn.1000-3428.2009.07.084
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper describes a comparative performance test on IBM JS21 BladeCenter using High Performance Computing Challenge (HPCC) tests. It applies the hierarchic analytic model AHPCC to analyze the test results of HPCC. The previous HPL tests get poor results and few reasons can be found. However, through AHPCC model, HPCC can obtain more detailed benchmark information and find performance bottlenecks of the system.
  • LI Kai; ZHONG Er-shun
    Computer Engineering. 2009, 35(7): 245-246,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.085
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    It’s a fundament feature of the next generation dynamic navigation system to support Real-Time Traffic Information(RTTI). Based on the existing hierarchal road model of navigation electronical map, this paper proposes an uniform method for multilayer road coding and a Dynamic Hierarchal Road Model(DyHiRD) to improve the efficiency and correctness of RTTI storage. Experimental results show that this model can satisfy the requirements of dynamic navigation application in large cities.
  • SHUI Wu-yang; ZHOU Ming-quan; GENG Guo-hua
    Computer Engineering. 2009, 35(7): 247-249. https://doi.org/10.3969/j.issn.1000-3428.2009.07.086
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper researches the reconstruction of cerebral cortex based on a group of continual brain slices. Volume data field and the processes of 3D reconstruction system are introduced. An effective segmentation algorithm for the gray matter is proposed and the image of gray matter is gained through image preprocessing and threshold segmentation. Cerebral cortex surface is reconstruction based on the improved Marching Cubes algorithm, avoiding extracting the contour lines of cerebral cortex. Experimental results verify that this algorithm is feasible and usable.
  • WANG Kui-sheng; ZHA Li; QU Zhan
    Computer Engineering. 2009, 35(7): 250-252. https://doi.org/10.3969/j.issn.1000-3428.2009.07.087
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper uses the real-time data of drilling operation as parameters to query the history drilling operating datas on the same structure or layer. These datas are feed backed to the well site to increase the efficiency of drilling operation. The drilling history information is stored in distribution and heterogeneous environment. It gives the architecture and implementation method of drilling operation guide system based on real-time data.
  • CHEN Dao-da; CAO Dong-lin; LI Shao-zi
    Computer Engineering. 2009, 35(7): 253-255,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.088
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    JGraph is an open-source and Swing compatible graph component written in Java. JGraph provides a range of graph API and good extensibility. This paper implements a cross-platform graph configuration system of substation integrated automation based on JGraph, which is more automatic than the existing system. The feature of higher automation is proved by the function of automatic equipment recognition and automatic diagram design. It simplifies the diagram design of the tradition substation graph system.
  • LIU Dong
    Computer Engineering. 2009, 35(7): 256-258. https://doi.org/10.3969/j.issn.1000-3428.2009.07.089
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Most of embedded system simulators only support simulation of CPU. It’s difficult to change the architecture of simulator to fit for the users’ specific hardware environments. To resolve this problem, this paper introduces a new sort of general pure mathematics simulation system which supports transparently forwarding data among components and scheduling the running components in real time. It’s easy and flexible to adjust the components which involve CPU and peripheral device modules according to different hardware designs of embedded systems.
  • HAN Ming-feng; LI Xiao-bin; ZHENG Yong-zhi
    Computer Engineering. 2009, 35(7): 259-260,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.090
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper analyzes the waiting-timeout mechanism of uC/OS-II kernel from source code. It indicates waiting-timeout of uC/OS-II is not correst in some cases. The kernel can return success while it is time out. System efficiency is low if kernel waiting-timeout mechanism has additional suspension operation. The reason why waiting-timeout mechanism of uC/OS-II kernel is not perfect is analyzed. A method is advanced to resolve these problems.
  • LU Jun; ZHANG Bao-ming; HUANG Wei; LI Er-sen
    Computer Engineering. 2009, 35(7): 261-263. https://doi.org/10.3969/j.issn.1000-3428.2009.07.091
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper designs a remote sensing image data fusion algorithm based on Graphics Processing Unit(GPU) using the programmability of GPU which is a parallel vector processor, and both the IHS and the inverse IHS transform computation are mapped into GPU. It realizes parallel rendering and output of the three components of the IHS and the inverse IHS transform useing Render To Texture(RTT) and Multiple Render Targets(MTR) method to speed up the computation. Result shows that this algorithm on GPU is much faster than the previous CPU-based algorithm for large IHS transform.
  • GAO Jia-li; LIAO Xiao-feng
    Computer Engineering. 2009, 35(7): 264-266. https://doi.org/10.3969/j.issn.1000-3428.2009.07.092
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Bloom filter based filtering mechanisms have one major drawback, that is, after a query item passes through the filter, the entire key set has to be searched ever for false positives. The search cost is therefore bound to be higher than necessary. This paper developes a new search filter which can lower the possibly of false positives, and significantly reduce the overall search cost, since a large amount of unnecessary search operations can be avoided.
  • ZHU Zhen; YAO Qi-fu
    Computer Engineering. 2009, 35(7): 267-269,. https://doi.org/10.3969/j.issn.1000-3428.2009.07.093
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper analyzes the deficiency of traditional digital campus solution and puts forward a new solution of digital campus system based on Service-Oriented Architecture(SOA). After analyzing the principle of SOA, a service oriented digital campus frame is proposed. This paper expatiates methods and steps of integration based on SOA and proves the feasibility and superiority developed in digital campus system integration.
  • PI Yang; LIU Chang-jiang
    Computer Engineering. 2009, 35(7): 270-272. https://doi.org/10.3969/j.issn.1000-3428.2009.07.094
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Slotted ALOHA is commonly used protocol in anti-collision of RFID system. It uses TDMA mode to exchange their data. Tags can only translate its information in a slot. This paper introduces a dynamic frame distributed slots algorithm, which can reduce the rate of collision tags, and the slots can be changed according to the quantity of collision immediate. Simulation results show that the rate of identifying is improved in evidence.
  • LI Yuan-yuan; MENG Xiang-ru; ZHANG Li; ZHUANG Ling-yi
    Computer Engineering. 2009, 35(7): 273-276. https://doi.org/10.3969/j.issn.1000-3428.2009.07.095
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Correlative transmitting of network faults may bring lots of redundancy information included in the network fault datas, which will affect the precision and efficiency of diagnosis. According to the characteristic of fault datas, numerical attribute reduction algorithms based on neighborhood rough approximation are adopted to carry out fast and highly efficient faults diagnosis by uniting rough set with Support Vector Machine(SVM). The discrete error in the classical RS is conquered, the memory space of data is curtailed, the complexity of SVM training model is reduced greatly, and the speed of training is put up. The well generalization of this method is analyzed and validated by ROC performance curve.
  • LIU Zhao-bin; LI Zhi-min; WANG Bing-zhi
    Computer Engineering. 2009, 35(7): 277-279. https://doi.org/10.3969/j.issn.1000-3428.2009.07.096
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Storage resource in large-scale ocean storage system has more complicated architecture, protocol and I/O pattern than traditional storage system. However, current methods lack of thorough and intelligent theory model to deal with these complicated storage environments. In this paper, based on the characteristic of network storage, a self-adaptive storage service Agent model is introduced to match individual applications to most appropriate storage systems and storage services. The system architecture and essential modules are presented. The parameter matrices regarding the service-aware and storage-environment-aware are thoroughly studied. Self-adaptive module and service scheduling algorithm of multi-Agent collaboration are proposed. An experiment based on Swarm is deployed. This methodology can automatically match the service requirements and appropriate storage system so as to improve the storage QoS.
  • CAI Song-song; LIU Qi; WANG Jian; LIU Jin-gang
    Computer Engineering. 2009, 35(7): 280-282. https://doi.org/10.3969/j.issn.1000-3428.2009.07.097
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Binary translation is the most important approach to achieve system migration, but the one based on general architectural and software suffers from poor performance. This paper introduces an optimization method of QEMU binary translation based on GODSON2F CPU. This optimization mainly contains two aspects: tool chain and QEMU binary translator itself. Aiming at the latter one, the methods are direct mapping of registers and the optimization to Multimedia Extensions(MMX) instructions. Experimental results show that the OS obtains speed-up of 1.45 through direct mapping of registers, and the MMX programs attains 80% performance of the host machine through the optimization to MMX instructions.
  • ZHONG Yan-ru; HONG Lei
    Computer Engineering. 2009, 35(7): 283-封三. https://doi.org/10.3969/j.issn.1000-3428.2009.07.098
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the shortcoming of PLC programmer, this paper presents an abstract model for PLC universal programmer, based on which the downloadable PLC program is generated using intermediate codes and configuration files. Intermediate codes are transformed from the specific Signal Interpreted Petri Net(SIPN) control system model. To different kinds of PLC systems, it can extract one’s own systematic characteristic information, and produce the disposition file of XML type in the same schema. According to the selection of the machine, PLC universal programmer extracts the corresponding configuration files, links with the intermediate codes, recombines codes and generates the executable program adapting to the PLC system.