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

20 May 2010, Volume 36 Issue 10
    

  • Select all
    |
    Networks and Communications
  • GUO Min-yi
    Computer Engineering. 2010, 36(10): 1-7. https://doi.org/10.3969/j.issn.1000-3428.2010.10.001
    Abstract ( ) Download PDF ( )   Knowledge map   Save

    Green computing is a new kind of computing mode to promote the scientific and technological improvement and sustainable social development. It is the focus of international competition and concerns national politics, economics and social safety. It involves system architecture, system software, parallel and distributed computing and computer network. Under the premise of guaranteeing computing efficiency, reliability and providing pervasive service, it aims to realize low power. The new kind of computing mode, which is new computer system architecture-oriented and cloud computing-included, through building energy awareness computing system, Internetworking environment, and computing service system, provides low power supporting environment for personalized, diversified information service mode.

  • Degree Paper
  • WAN Ya-ping; FENG Dan; LIU Li; SHEN Hong-jian
    Computer Engineering. 2010, 36(10): 8-10. https://doi.org/10.3969/j.issn.1000-3428.2010.10.002
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to solve the problems of data loss and corruption caused by failure of the single controller, this paper designs a dual-controller Redundant Array of Independent Disk(RAID) based on internet Small Computer System Interface(iSCSI). Stochastic Petri Net(SPN) model is established to evaluate the performance impact on controller caused by the design. Function test results prove that the average utilization rate of string controller in the dual-controller model is enhanced, and the dual-controller RAID has better fault-tolerance, higher availability and lower cost.
  • GAO Lin; LIU Ji-ming; ZHANG Shi-wu; YANG Jie
    Computer Engineering. 2010, 36(10): 11-13. https://doi.org/10.3969/j.issn.1000-3428.2010.10.003
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problem that the properties of nodes and links in real social networks change dynamically with time, this paper proposes a Social Memory Network(SMN) model based on preference attachment theory by analyzing the dynamic properties of nodes and links in real social networks, so that the model can describe the evolution process of social networks more clearly, and explain the metabolism phenomena. Effectiveness of the model is validated by analyzing the model and data from real networks. The influence of model parameters on the topological structures of networks are concluded.

  • YANG Jing-yu; ZHANG Yong-sheng; LIU Zhao-hua
    Computer Engineering. 2010, 36(10): 14-15. https://doi.org/10.3969/j.issn.1000-3428.2010.10.004
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problems of traditional remote images pyramid unitary store model in multi-user parallel access application, this paper proposes a separable store model suitable for multi-user parallel access, which makes use of distributed data store and processing task decomposition and transferring to improve I/O performance and response efficiency of data servers. Experimental result proves that the effectiveness of the model, and the model system has good expansibility and stability.
  • XIANG Ming; WU Xiao-pei; LIU Ming-sheng
    Computer Engineering. 2010, 36(10): 16-18. https://doi.org/10.3969/j.issn.1000-3428.2010.10.005
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the shortage of classification unbalanced in traditional fingerprint classification algorithm, this paper presents a secondary fingerprint classification algorithm based on independent component analysis. It extracts a group of characteristic fingerprint image in terms of high level statistics relevance, uses this group of characteristic as base images, the fingerprint can be projected into the feature space. Combining coefficient vector with Henry classification mode to set up two level index which classifies input fingerprints into eleven kinds of category. Application results show that this algorithm can save operation time and reduce complexity.
  • CHEN Xiao-long; TIAN Yi-qiang; ZHANG Yun
    Computer Engineering. 2010, 36(10): 19-21. https://doi.org/10.3969/j.issn.1000-3428.2010.10.006
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Transmission Control Protocol(TCP) self-synchronization can improve the stability of network, existing TCP/Random Early Detection (TCP/RED) congestion control model ignores TCP self-synchronization, aiming at this problem, this paper proposes an improved TCP/RED congestion control model including TCP self-synchronization. It applies Nyquist stability criterion to prove that the system is stable if the system model in a half-plane that does not include -1+j0 in the complex plane, receives a more relaxed RED parameter setting scope. NS2 simulation is used to validate the effectiveness of stability condition.

  • GUO Jian-wen; YU De-jie; LIU Jian; ZENG Wei
    Computer Engineering. 2010, 36(10): 22-24. https://doi.org/10.3969/j.issn.1000-3428.2010.10.007
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To make knowledge in E-maintenance to support maintenance task, this paper proposes a business-oriented knowledge supply model. Maintenance knowledge concept ontology is built to descript concept of task, maintenance object and knowledge function. Knowledge object ontology is built to integrate knowledge resource. Integrating requirement and permission of knowledge supply into business process, a knowledge supply information model is built, and knowledge matching method is also introduced to control knowledge supply. Application results on E-maintenance platform in certain petrochemical enterprise show that this model can enhance enterprise maintenance level.
  • ZHOU Yi; ZHENG Xue-feng; YU Yi-ke
    Computer Engineering. 2010, 36(10): 25-28. https://doi.org/10.3969/j.issn.1000-3428.2010.10.008
    Abstract ( ) Download PDF ( )   Knowledge map   Save

    Aiming at the definitionproblem of that the trust definition and the current model of trust has not been establishedve no common definitionsrepresentations currently ,The general description to the definition the and model of the trust must be needed, therefore we this paper presents a trusted computing platform model based on existing computer network, whichby makes full use of using the versatility of existing computer and network platform and it’s versatility. Through By putting the existing computing layered platforms into different logical trust layers, not onlyit reduces the difficulty of the implementation to theing trusted computing platform models but also , and improves its flexibility, scalability and efficiency of the models. At the same time, tAhe model building the credibility of a complete trusted chain is established based on in the starting phase to of the client system, thereby ensuring to ensure the security of trusted computingthe platform security. The Implementation results of the implementation to the prototype system showed that:t in the system, all the computers and their networks in the system constitute a unified platform for trusted computing, which mademakes the transfer of trust from the the user to the the user come true, and then realizeand realizes the true user experience for the trusted computing between users.

  • DENG Shao-jiang; LI Yan-tao; ZHANG Dai-gu; XIAO Di
    Computer Engineering. 2010, 36(10): 29-31. https://doi.org/10.3969/j.issn.1000-3428.2010.10.009
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes a one-way Hash function construction algorithm based on chaotic look-up table. It maps the message with the corresponding data in chaotic look-up table which is updated by control characters, and generates Hash value with length of 128 bit according to the rules. Theoretical analysis and simulation results indicate that the algorithm has characteristics of one way, confusion and diffusivity, and collision-resistance, and it can satisfy various performance requirements of Hash function.
  • SUN Li-bo; LIU Yan; SUN Ji-zhou; NI Hui
    Computer Engineering. 2010, 36(10): 32-34. https://doi.org/10.3969/j.issn.1000-3428.2010.10.010
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problem of real-time obstacle avoidance for autonomous virtual humans in dynamic unknown environment, this paper presents a path planning model based on the mixed perception information. The model is composed of a global planner and a local planner. The global planner computes an optimal path according to the known environment information, and the local planner establishes the assessment function of motion cost for local dynamic planning according to the analysis of the human planning behavior, realizes the plausible intelligent planning behaviors for virtual humans in dynamic unknown situation. Experimental results show that this model can run in real-time, it can simulate the planning behavior well fitting for human characteristic.
  • Software Technology and Database
  • WEI Guang-bo; ZHANG Ping; HUANG Guo-rui
    Computer Engineering. 2010, 36(10): 35-37. https://doi.org/10.3969/j.issn.1000-3428.2010.10.011
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Program parallelization is an available method to exert the multi-core process’s performance sufficiently. Because the existing programming patterns are limited by synchronization mode, such as lock mechanism and pipeline, the parallelization of them is very difficult to improve. Aiming at the above problem, this paper presents a conflict manager approach based on Rochester Software Transaction Memory(RSTM) system to supply the APIs for the existing program language, which improves the extent of program parallelization by the approach of transaction, and solve the problem of how to arbitrate the conflict of two transactions by the method of PRI. It can cancel some unnecessary consistency validation at some time and reduce the system cost.
  • SUN Yan-qing; YIN Shu-hua; WANG Ji
    Computer Engineering. 2010, 36(10): 38-40. https://doi.org/10.3969/j.issn.1000-3428.2010.10.012
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes an algorithm based on improved rough set and a pretreatment method for decision table. The algorithm does the attribute reduction of decision table based on the core attribute after pretreating the decision table, and uses the method of improved weighted mean attribute significance to select the non-core attribute. It adds the converse eliminate action to ensure that it can get a Pawlak reduct, so as to improve the efficiency of case searching. Using Matlab to implement the algorithm, the running result shows the validity and veracity of the method by analyzing and comparing algorithm through an example.

  • JI Chun-lei; TAN Ye-yong; SONG Guo-xin
    Computer Engineering. 2010, 36(10): 41-43. https://doi.org/10.3969/j.issn.1000-3428.2010.10.013
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the existing problem of strong subjectivity and weak repeatability in software estimation by introducing Unified Modeling Language(UML) into software estimation, this paper analyses the semantics of UML artifacts(use case diagram, class diagram and sequence diagram) and primary elements of COSMIC-FFP(functional user, layer, boundary, data group and functional process) and establishes the mapping from UML artifacts to COSMIC-FFP elements. As a result, 15 mapping rules are proposed. This paper proposes a UML-based measurement process of COSMIC-FFP and presents an example for that. Experimental result shows that it is effective and objective to the software estimation.
  • GAO Bao-qiang; HAO Ke-gang
    Computer Engineering. 2010, 36(10): 44-46. https://doi.org/10.3969/j.issn.1000-3428.2010.10.014
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Simulation of workflow process based on extended XinPai driven workflow model consists of two parts which are static inspection and dynamic simulation. This paper studies several key problems in process simulation——getting performance index and structure-function index, analyzing simulation results. It employs white-box and black-box methods to test workflow process with high coverage rate, gives expression to type, distribution and trend of process defects. And it outputs simulation reports.

  • WANG Miao; HAO Zhong-xiao;
    Computer Engineering. 2010, 36(10): 47-49. https://doi.org/10.3969/j.issn.1000-3428.2010.10.015
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    A lot of reverse nearest-neighbor query on uncertain objects can not answer whether the uncertain object is reverse nearest-neighbor of query object. Aiming at this problem, this paper proposes the notion of probabilistic reverse nearest neighbor query, devises a probabilistic reverse nearest neighbor query index and offers a reverse nearest neighbor query algorithm on uncertain objects based on this index.
  • ZHAN Qin
    Computer Engineering. 2010, 36(10): 50-51. https://doi.org/10.3969/j.issn.1000-3428.2010.10.016
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    For how to utilize the huge original data to analyse the current situation and predict the future effectively, based on the antibodies selection strategy, this paper brings forward a clonal selection mining algorithm. The algorithm evaluates the degree of support, confidence and affinity, and thus achieves effective association rules. Experiments confirm that this algorithm gets the understandable rules faster and better in terms of the average precision rate.
  • LIU Shuang-shuang; RAO Tian-gui; SUN Jian-hua
    Computer Engineering. 2010, 36(10): 52-55. https://doi.org/10.3969/j.issn.1000-3428.2010.10.017
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    After introducing the basic theory of concept lattice, this paper brings forward an improved construction algorithm named Godin algorithm, adds the index for pruning, which can be helpful to pruned concept lattice, combines quantified concept lattice, pruned concept lattice with the improved algorithm, and gives non-redundant rules extraction algorithm. Theoretic analysis and experiment show that the algorithm and the non-redundant rules extraction algorithm are correct and effective.
  • QI Xin
    Computer Engineering. 2010, 36(10): 56-57. https://doi.org/10.3969/j.issn.1000-3428.2010.10.018
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Object-Oriented Programming(OOP) standard is used to implement traditional Chain of Responsibility(CoR) design pattern, which puts too much emphasis on the expansion process, without taking into account the complicated nature of space and time problem. The process causes waste of time and space greatly and reduces the treatment of complicated procedures. Aiming at the problems, a CoR creating process is presented, which uses delay initialization technology to reduce unnecessary requests for the creation of the treatment process. This method is not only a responsibility to maintain the chai’s core applications and standards for the realization of the ideological model, but also can request in accordance with the actual code, to deal with the creation of specific examples, which greatly improves the performance of the procedure.
  • TIAN Sheng-wen; YANG Hong-yong; LI A-li; WANG Yi-lei
    Computer Engineering. 2010, 36(10): 58-60. https://doi.org/10.3969/j.issn.1000-3428.2010.10.019
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the disadvantage of low clustering coefficient in initial attractive model proposed by Dorogovtsev and its extended model, a complex network evolving model based on node attraction with clustering effect(CALW model) is proposed. According to the local-world property of preferential attachment existed in real networks, CALW model constructs local-world by referencing on the idea of forest fire, and defines node attraction as a dynamic function with the change. Simulation results show that the network generated by CALW model follows power-law degree distribution.
  • HUANG Ya-jing; GAO Jian-hua
    Computer Engineering. 2010, 36(10): 61-63. https://doi.org/10.3969/j.issn.1000-3428.2010.10.020
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to study the complexity of the large-scale open source software, this paper models the packages in the open source software as vertices and the dependency relationships among these packages as directed edges. It uses random graph measure and structure entropy to propose a new method of measuring structural complexity of open source software. It uses the method by the fact to investigate that as the release of the open source software evolves, the coupling and cohesion grow from lower to higher.
  • WAN Jia; TANG Chun; TANG Sheng-qun; QU Zhen-xin
    Computer Engineering. 2010, 36(10): 64-66. https://doi.org/10.3969/j.issn.1000-3428.2010.10.021
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Under the background of Visualized Ontology Editor(VOEditor), a prototype system of semantic query integration platform is proposed with the related ontology knowledge, the Oracle 11g semantic technique and some inference engines’ semantic technique, which can query with inference of different extent and support large-scale semantic application. This paper explores on the architecture, query mechanism, implementation technique and preliminary application.

  • HE Wei; HE Jian-zhong
    Computer Engineering. 2010, 36(10): 67-69. https://doi.org/10.3969/j.issn.1000-3428.2010.10.022
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The uncertainty of memory allocation time and low memory utilization are shortages of Free Real Time Operating System(FreeRTOS), which also lacks good support to dynamic memory allocation. To solve such problems, this paper analyzes the memory management mechanism of FreeRTOS and compares several typical dynamic memory management algorithms and try to improve its performance through the transplantation of modified Two Level Segregated Fit(TLSF) algorithm. The smaller memory is divided into fixed-size memory blocks and organized by bitmap index, while the larger memory is organized by TLSF. Experimental results show that this method can improve the memory allocation speed and memory utilization.
  • ZUO Wei; FENG Jin-fu; XU Ling-quan
    Computer Engineering. 2010, 36(10): 70-72. https://doi.org/10.3969/j.issn.1000-3428.2010.10.023
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In allusion to weapon control system that typical concurrent and real-time application and according to the development course of COMET, the requirement model, analysis model and design model are developed, and the software architecture based on concurrent tasks and connector is constructed. The modeling course of each model is described explicitly and the graphic presentation of each model is presented. The schedulability of the concurrent tasks is analyzed by using real-time schedule theory. The result shows that the designed tasks are schedulable.
  • ZHOU Zhong-hua; YANG Xiao-hu
    Computer Engineering. 2010, 36(10): 73-75. https://doi.org/10.3969/j.issn.1000-3428.2010.10.024
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Dedicated to the lack of efficient developing approach of message-oriented application, a kind of Message-oriented Component(MoC) is proposed. This paper introduces conception and characteristic of the component, and analyzes the internal structure of MoC. An example is given to show how to use MoC for fast development in practice. The results indicate MoC approach increases the speed and the efficiency of developing message-oriented application.
  • LV Zheng; CHEN Hao; LV Yi
    Computer Engineering. 2010, 36(10): 76-79. https://doi.org/10.3969/j.issn.1000-3428.2010.10.025
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Cluster file system sharing semantics is one of the key issues that address functional correctness, performance and usability of cluster file system. The concept of file sharing semantics test is proposed and a file sharing semantics test system(FSbench) is implemented. FSbench uses random test and partion test to verifiy whether can satisfy semantic characteristics. Test shows that the method can get satisfied results.
  • ZHU Shuai-shuai; YANG Xiao-yuan; ZHANG Wei
    Computer Engineering. 2010, 36(10): 80-82. https://doi.org/10.3969/j.issn.1000-3428.2010.10.026
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper analyzes the relevancy quantitatively in Distributed Storage Network(DSN) in the view of attributes relativity of data share. Based on the analysis, it presents a new mean-weight clustering algorithm in distributed storage networks based on the idea of dispersing correlative data share on different storage nodes. According to the simulative test, the algorithm can greatly lower compromising probability after taking the data relativity into consideration.

  • MA Deng-wu; DENG Li; WAN Bing
    Computer Engineering. 2010, 36(10): 83-84. https://doi.org/10.3969/j.issn.1000-3428.2010.10.027
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The FPF code format is widespread saved by the imported battle plane objective check system parameter information currently. Aim at the problem that the FPF coding format is unknown causes the parameter information makes use of insufficiency definitely, the method that unites WinHex and C++Builder to analyze and check FPF codes is putted forward. That method realizes FPF encoding, which adopts bit computation and stream technique. The method is useful is proved by comparing with the imported analytical software. The CDT document developed easily is acquired.
  • Networks and Communications
  • ZHANG Lei; DUAN Li-li; QIAN Zi-juan; HUANG Guang-ming
    Computer Engineering. 2010, 36(10): 85-87. https://doi.org/10.3969/j.issn.1000-3428.2010.10.028
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    A novel localization algorithm based on Genetic Algorithm(GA) is presented. The algorithm is divided into two stages. The sampling method is used to obtain initial location and the initial location is refined based on GA. Simulation experimental results show this algorithm is effective for Wireless Sensor Network(WSN). Its location accuracy is better.
  • WANG Shi-guo; LIU Xian-ru
    Computer Engineering. 2010, 36(10): 88-90. https://doi.org/10.3969/j.issn.1000-3428.2010.10.029
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In the construction of Mobile Ad hoc Peer-to-Peer Networks(MANET P2P), cross-layer design idea is utilized, and through modifying the traditional protocol named Chord, a novel construction method for P2P MANET is presented, in which the received signal power and node finger table are used to estimate the distance between two nodes and local node distribution density respectively. A node ID is allocated to realize matching between logical network and physical network, as well as load balance. Simulation results show that the MANET P2P constructed through the proposed method shortens the resource-locating path, minimizes the latency and drop-package rate, and load capability is improved.
  • LIN Lin; WANG An-ping; ZHOU Xian-wei; LIU Zhen-zhen
    Computer Engineering. 2010, 36(10): 91-93. https://doi.org/10.3969/j.issn.1000-3428.2010.10.030
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at delay-constrained Steiner tree problem in Ad hoc network, this paper designs a distributed rapid heuristic algorithm DCST. The algorithm builds a delay-constrained Steiner tree through labeling on the nodes and changing the adjacent relationship between nodes based on the labels. Under the premise of time synchronization of nodes in Ad hoc network, the time complexity of the algorithm is O(n). Compared with some traditional Steiner tree algorithms, the algorithm has distinct advantages.

  • XIANG Min; WANG Ping; LUO Zhi-yong
    Computer Engineering. 2010, 36(10): 97-99. https://doi.org/10.3969/j.issn.1000-3428.2010.10.032
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Node management of cluster is very important to the performance of Wireless Sensor Networks(WSN), so a method based on the integrative support degree is presented. The key of method is that the integrative support degree of sensing data is obtained with the error function and fuzzy correlation function, and sensor nodes and their sensing data are divided into two types of reliable and unreliable through their integrative support degrees and increments of integrative support degrees, which provide the feasibility for exact controlling the redundant nodes and unreliable nodes. Simulation results demonstrate that the method is efficient to realize sensor node classification.
  • HAN Dong-xue; ZHANG Rui-hua; LIU Dan-hua
    Computer Engineering. 2010, 36(10): 100-102. https://doi.org/10.3969/j.issn.1000-3428.2010.10.033
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The clustering algorithm for Wireless Sensor Network(WSN) is optimized by Particle Swarm Optimization(PSO) algorithm. The master cluster-head and the vice cluster-head are considered according to the information of both normal nodes and the cluster-heads. The master cluster- head is used for date collecting and date integration. The integrated dates are sent to the vice one. The vice cluster-head charges for the communication with base station. This algorithm can balance the energy consumption, so it can extend the network lifetime effectively. Simulation results show the lifetime of the algorithm is extended for 50% contrast with LEACH.
  • XIANG You-huan; XIONG Yan; MIAO Fu-you
    Computer Engineering. 2010, 36(10): 103-104. https://doi.org/10.3969/j.issn.1000-3428.2010.10.034
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problems that peers have a strong ability of moving and the topology of network changes frequently in the mobile Peer-to-Peer(P2P) network, which make low efficiency and fallibility of the system, this paper proposes an algorithm of super-peer selection in mobile P2P network, it selects a peer with more powerful and longer lifetime as a super-peer, and uses the technique of backup super-peer, which can improve the efficiency of a system in the mobile P2P network system. Experimental results show that the algorithm can reduce the failure rate of super-peer and shorten the query delay.
  • WANG Xin-sheng; LIANG Ping; ZHANG Yun-chao; WANG Wei-jie; DING Xue-yong
    Computer Engineering. 2010, 36(10): 105-107. https://doi.org/10.3969/j.issn.1000-3428.2010.10.035
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to locate the resource key with high efficiency and improve the system’s capability in P2P system, a new Chord algorithm by extending the finger table on basis of deep analysis of Chord is presented. This algorithm has lower latency and shorter search path than Chord. Simulation results indicate this algorithm inherits the advantages of Chord, such as simplicity, efficiency, reliability, load balance and low cost. So it gets higher search efficiency than original Chord.
  • FAN Jun-qing; ZHAO Li-hua; JIANG La-lin; ZHANG Si-fa
    Computer Engineering. 2010, 36(10): 108-109. https://doi.org/10.3969/j.issn.1000-3428.2010.10.036
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    For the disadvantage of some nodes energy consumption nonequilibrium nowadays in the Wireless Sensor Network(WSN), this paper presents the energy aware routing protocol based on the grid 2D cluster. Through the process of cluster, the cluster’s size can be confirmed by nodes’ position. Network partition is equilibrium with the cluster node. By the theoretic analysis and emulator experiments, performance of the protocol is verified and compared with LEACH, result of which shows that this protocol can equally balance the energy consumption of all the nodes, and the lifecycle of wireless network can be prolonged.
  • XU Jing-dong; ZHAO Wen-yao; LI Miao; HE Liang
    Computer Engineering. 2010, 36(10): 110-112. https://doi.org/10.3969/j.issn.1000-3428.2010.10.037
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Based on Chip CC2430 of TI and ZigBee protocol stack, a simple prototype of wireless temperature sensor network is proposed and implemented. It discusses details on hardware and software of the prototype, including chip selection of sensor node, topology discovery and data query protocol, GUI and programming logic of sink node. The correctness and effectiveness of the prototype are verified by the practical experiment.
  • WANG Xin-sheng; PAN Shu-qin; LIU Yang; ZHANG Yun-chao
    Computer Engineering. 2010, 36(10): 113-114. https://doi.org/10.3969/j.issn.1000-3428.2010.10.038
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    According to the question how to choose multipath routes in Ad hoc networks, this paper improves Backup Source Routing(BSR) protocol. It considers link service time and transmission delay of the network in routing discovery, chooses a primary route according to a factor and a backup route which has little correlation with the primary route according to the improved cost function. It transmits data packet by using the route pairs with the primary route and the backup route. Experimental results demonstrate that the improved protocol can decrease the count of routing discovery and improve the routing reliability and network performance.
  • JIANG Ming; BIAN Hao; CHEN Qin
    Computer Engineering. 2010, 36(10): 115-117. https://doi.org/10.3969/j.issn.1000-3428.2010.10.039
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    HCHOKe controlls and punishes non-adaptive flows, makes adaptive flows share the bandwidth stably. HCHOKe combines CHOKe’s hit and uses congestion self-adaptive while picking packets, gives a way of dropping packets. It analyzes HCHOKe and other main schemes about CHOKe, then uses NS-2 and shows that HCHOKe performs better than CHOKe and other main schemes in distributing the bandwidth in network fairly.
  • HE Xuan; SONG Yong; HAO Qun
    Computer Engineering. 2010, 36(10): 118-120. https://doi.org/10.3969/j.issn.1000-3428.2010.10.040
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes an event driven clustering algorithm for Wireless Sensor Network(WSN) via choosing the cluster-head with a head-power-tradeoff algorithm and joining the event driven mechanism. The event driven algorithm can rebuild clusters in the random event areas when sudden event happens, and can reduce data redundancy and improve network efficiency. It builds a simulation model for analyzing the trace and dynamic clustering situation of nodes. Experimental results show that the algorithm can require the demands of the event driven clustering, it consumes balanced energy compared with traditional algorithms.
  • YUAN Hui-yong; ZHOU Na-qin; YI Ye-qing;
    Computer Engineering. 2010, 36(10): 121-123. https://doi.org/10.3969/j.issn.1000-3428.2010.10.041
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problem of “hotspots” in Wireless Sensor Networks(WSNs), a grid-based power control algorithm which divides a network monitor area into different widths of grid is proposed. To achieve the uniform distribution of energy consumption in sensor networks, the widths of the grid and node transmission radius are computed by analyzing node energy consumption. Simulation results show the algorithm is superior to others in energy-balancing.
  • MA Xiao-qiang; ZHANG Chun-ye; ZHANG Bo
    Computer Engineering. 2010, 36(10): 124-126. https://doi.org/10.3969/j.issn.1000-3428.2010.10.042
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Analyzing the QoS mechanism of IEEE 802.16 and according to the situation that this protocol does not stipulate admission control algorithm detailedly, this paper proposes an improved priority-based dynamic bandwidth allocation admission control mechanism with an agreement in the QoS features of service types and the signaling interactive mechanism. Simulation results show that compared with the fixed bandwidth allocation mechanism in which the bandwidth is reserved, the system handoff connection dropping probability, new connection blocking probability and bandwidth utility can be improved by using this admission control mechanism.
  • ZHAO Xian-lin; PANG Bin; SHEN Ming-xia
    Computer Engineering. 2010, 36(10): 127-128. https://doi.org/10.3969/j.issn.1000-3428.2010.10.043
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The paper introduces an associated simulation tool——TrueTime tool box, using the tool box builds a simulation model of DC motor in Wireless Network Control System(WNCS). The farthest signal transmission distance is obtained through the investigation of path loss function. Simulation experimental result shows that the node power control of WNCS can be realized by reducing the transmit power within the farthest signal transmission distance.
  • ZHANG Hui-juan; WANG Ke-te
    Computer Engineering. 2010, 36(10): 129-132. https://doi.org/10.3969/j.issn.1000-3428.2010.10.044
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Due to the increase of nodes quantity and communications, the central node of the Ad hoc network may lead to forward too much data, which causes network congestion, and the whole network can be completely paralyzed. In this paper, a new network routing algorithm is designed based on the improvement of AODV algorithm in order to solve the problem of unbalanced load. Experimental results show that the improved AODV algorithm can alleviate the load of the network center areas and increase the throughput of the network.

  • XUE Jun-han; ZHANG Xiao-min; CHEN Gao-ming
    Computer Engineering. 2010, 36(10): 133-135. https://doi.org/10.3969/j.issn.1000-3428.2010.10.045
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper corrects the service time of G/G/∞ queuing model and gets the accurate result of the transmission probability. Based on the corrected model, the saturation throughput of IEEE 802.11 Distributed Coordinated Function(DCF) WLAN basic access mechanism and RTS/CTS access mechanism is analyzed, and the average packet delay for a lossy channel is derived. The correctness of this method is verified by simulation with NS.
  • XUE Jian-sheng; LIU Jian; REN Xiu-li; SONG Bao-yan
    Computer Engineering. 2010, 36(10): 136-139. https://doi.org/10.3969/j.issn.1000-3428.2010.10.046
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problem of energy loss result from non-uniform cluster generated by LEACH protocol, a routing protocol, LEACH-EP, is proposed, which is based on the node’s remaining energy and the geographical position, for hierarchical symmetrical clustering in administrative levels and the establishment of the spare cluster to reduce the frequency of repeated clustering. The simulation results show that under the same conditions of initial energy in the nodes and network size, the energy consumption is far less than the LEACH protocol, and the nodes survive longer, more suitable for large-scale low-power wireless sensor networks routing protocols.
  • MU Peng-zhi; LI Fei; LUO Chuan-jun
    Computer Engineering. 2010, 36(10): 140-142. https://doi.org/10.3969/j.issn.1000-3428.2010.10.047
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problems of topology discovery inaccurate and topology display not timely in Network Management System(NMS), a kind of three-level network topology discovery method based on the SNMP agents is studied. The strategy of “many kinds of ways, discover accurately” is presented to ensure equipment discovery accurately and efficiently. A kind of service procedure of “background scanning” is designed to realize the real-time updates. A model of network topology display based on SVG technology is presented to display changing topology real-timely. Experimental results show that the NMS adopting the above data update strategy can meet the requirements in the accuracy of topology discovery and real-time of topology display.
  • ZHANG Wen-bo; XU Ye
    Computer Engineering. 2010, 36(10): 143-145. https://doi.org/10.3969/j.issn.1000-3428.2010.10.048
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In terms of the time-varying feature of the satellite network topology and partition policy of the management field in satellite network, the fault detection algorithm of the satellite network is presented. When the satellite can respond to the network management instruction, the network management technique is used to detect the interfaces in the satellite. When the satellite can not respond to the network management, the intra-domain cooperation or inter-domain cooperation is activated. The suspected fault satellite is detected through cooperation among the faulty detection agents in the satellite. Simulation results show that in the circumstance of the low faulty frequency, the algorithm is effectively used in satellite network with short cooperative time and low throughput.
  • Security Technology
  • OUYANG Qian; LIU Yi-guang; SONG Shuang; GAO Qiang
    Computer Engineering. 2010, 36(10): 146-148. https://doi.org/10.3969/j.issn.1000-3428.2010.10.049
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Most of the existing Digital Rights Management(DRM) system can not meet the requirement of security and flexibility in digital media application. Aiming at this problem, this paper proposes a DRM system based on PKCS#12. It designs the system architecture and the security protocol of user registration, certificate issuing, encrypted digital content distribution, authorized license delivery, authentication and decryption, etc. With the security of PKCS#12 and the security of the designed protocols, the proposed system provides the secure storage and transfer of certificate and private key, lets users participate in system applications through different devices, and prevents digital rights from illegal sharing.
  • MO Jie; WANG Xiao-ming
    Computer Engineering. 2010, 36(10): 149-151. https://doi.org/10.3969/j.issn.1000-3428.2010.10.050
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper presents a distributed certificate revocation scheme for Ad Hoc network. It uses two tables to describe creditability and relationship of network nodes, and one-way Hash chain to authenticate accuse info. It reflects the intensity of complaints accurately and provides protection against certificates being wrongfully revoked through malicious accusations by dynamic updating creditability of the nodes. By introducing subspending certificates, it provides protection against the certificates of good nodes being revoked for accidental misbehavior.
  • ZHOU Hai-qing; CHEN Qi-mai; LIU Hai
    Computer Engineering. 2010, 36(10): 152-154. https://doi.org/10.3969/j.issn.1000-3428.2010.10.051
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problem of unauthorized access to data warehouses and indirect inferences of sensitive data in data warehouses and OLAP system, this paper builds a three-layer inference control architecture for OLAP based on statistical databases security architecture, and proposes a data cube-based inference control method implemented on the basis of the three-layer inference control architecture. This method prevents multi-dimensional inferences and then eliminates one-dimensional inferences, which simplifies the process of detecting multi-dimensional inference.

  • ZENG Ying-luo; PAN Xue-zeng; CHEN Jian
    Computer Engineering. 2010, 36(10): 155-157. https://doi.org/10.3969/j.issn.1000-3428.2010.10.052
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes a dynamic security access control model based on system reliability, so as to improve the flexibility of combination among multiple security access control strategies, and to ensure system security and practicality. The model can choose the proper kind of combination among the security access control strategies dynamically in order to deal with different system states and improve the whole system performance. Experimental results prove that this model makes the multiple strategies control more flexible, ensures the system security, and improves practicality of the system.

  • LIANG Hai-hua; LI Ke-qing; PAN Li-na
    Computer Engineering. 2010, 36(10): 158-159. https://doi.org/10.3969/j.issn.1000-3428.2010.10.053
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The measures of Intrusion Detection System(IDS) in special threshold-value are derived from the contingency table. ROC curve facilitates to evaluate general performance of IDS. In practice, ROC Convex Hull(ROCCH) method which combines techniques from computational geometry is used. In order to low computational complexity, the algorithm of MLE-ROC which depending on monotonic regression under maximum likelihood estimation of value of Area-Under-Curve(AUC) is proposed. Experimental results show that the proposed algorithm is computationally efficient while having better AUC approximation level.
  • ZHANG Jian-zhong; LAN Jian-qing; GUO Zhen
    Computer Engineering. 2010, 36(10): 160-162. https://doi.org/10.3969/j.issn.1000-3428.2010.10.054
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the defect of existing dynamic generalized threshold secret sharing scheme among weighted participants, this paper proposes a threshold secret sharing scheme among participants with different weight based on Chinese remainder theorem. Participant’s sub-secret is selected and kept by the participant himself and the sub-secret of every participant can be used many times. A secure channel between the dealer and participants is unnecessary. A participant can be freely accepted or fired and many secrets can be recovered in one time. Analysis results show that the scheme is secure and practical.
  • LIAO Qi-nan
    Computer Engineering. 2010, 36(10): 163-164. https://doi.org/10.3969/j.issn.1000-3428.2010.10.055
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper studies the mechanism of saving image object data and method of obtaining their intact data in Word and PowerPoint files, sets the image of the files as the direct carrier of secret message and the file as secondary cover medium, and presents an image information hiding algorithm based on Least Significant Bit(LSB). Experimental results show that this algorithm has big information content, strong imperceptibility and anti-interference performance.

  • LI Qin; ZENG Qing-kai;
    Computer Engineering. 2010, 36(10): 165-167. https://doi.org/10.3969/j.issn.1000-3428.2010.10.056
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Anonymous on Demand Routing(ANODR) protocol is designed for avoiding node identity from being leaked by other nodes during communication in MANET and insuring the communication route not be discovered. The objectives of the protocol include sender anonymity, reception anonymity and route anonymity. The protocol is decomposed into three components, each of which implements one of three objectives respectively. Cord logic with extension is used to verify the failure of the component for route anonymity that destroys the anonymity service providing by the protocol as an entirety.
  • ZHANG Jian-zhong; WEI Chun-yan;
    Computer Engineering. 2010, 36(10): 168-169. https://doi.org/10.3969/j.issn.1000-3428.2010.10.057
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Different from traditional public key cryptosystem, certificateless public key cryptosystem eliminates the extra spending of certificate management and the key escrow problem in identity based public key cryptosystem. This paper proposes a certificateless proxy signature scheme with the requirement of proxy signature unforgeal be property, non-repudiation. It can avoid abuse of signature power and has high efficiency.
  • WANG Ze; LIU Wen-ju; KE Yong-zhen
    Computer Engineering. 2010, 36(10): 170-172. https://doi.org/10.3969/j.issn.1000-3428.2010.10.058
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    EAP-AKA protocol can realize bidirectional authentication between 3G users and 3G network, but it lacks authentication for Wireless Local Area Networks(WLANs) and updating scheme for user master key, and the cost of roaming among WLANs is relatively higher. Aiming at such problems, this paper designs a fast authentication roaming protocol without participation of 3G network, in which ID-based encryption and ticket technology is utilized to implement master key updating and access network authentication.
  • ZHU Hui; SHEN Ming-xing; LI Shan-ping
    Computer Engineering. 2010, 36(10): 173-175. https://doi.org/10.3969/j.issn.1000-3428.2010.10.059
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper studies the code injection vulnerabilities of Web application, modifies and expands the definition of this kind of vulnerabilities with summarizing and analyzing the features of them, and transforms the causes of vulnerabilities into two kinds of coding errors to present a new test method based on testing the two kinds of coding errors. Experimental result shows that the test method can test all the code injection vulnerabilities of Web application effectively with less test workload.

  • GAO Fei; GAO Yong-ren
    Computer Engineering. 2010, 36(10): 176-178. https://doi.org/10.3969/j.issn.1000-3428.2010.10.060
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problem of the key service security situation for the quantitative analysis, this paper designs and implements a novel quantitative evaluation model for key service security situation. This model makes survivability and service performance indicator as internal and external factors for reflecting service change, and performance indicator is taken into account to validate outside change in fastness survivability. It uses Domain Name System(DNS) to verify this model, results show that the model can accomplish comprehensively the quantitative analysis and situation generating, effectively detect the deviation and the failure of service, and help security administrators to make correct decisions.
  • LIU Yi-ning; CAO Jian-yu
    Computer Engineering. 2010, 36(10): 179-180. https://doi.org/10.3969/j.issn.1000-3428.2010.10.061
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Random number is always chosen by the participant in information security protocol, which can not be verified by other participants, so that the fairness of protocol is affected. Aiming at the problem, this paper proposes a scheme for generating verifiable random number. It does not need Trusted Third Party(TTP). The task of calculation center is only to perform calculation, not to play a role as TTP. The scheme allows every participant to verify the Randomness of the random number, which ensures the fairness of the protocol.
  • ZHOU Jian-ling; WANG Xiu-yan
    Computer Engineering. 2010, 36(10): 181-183. https://doi.org/10.3969/j.issn.1000-3428.2010.10.062
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Considering the importance of dynamic r matching algorithm for Intrusion Detection System(IDS), this paper researches the mechanism of endocrine system, proposes a dynamic balance model of endocrine hormone concentration and a dynamic matching algorithm based on hormone concentration adjustment. Simulation experimental results show that the matching algorithm can adjust matching preferences adaptively with the change of network condition, and has good performance in matching effect.
  • Networks and Communications
  • YANG Yong, MA Zhi-Meng, XU Chun
    Computer Engineering. 2010, 36(10): 184-186. https://doi.org/10.3969/j.issn.1000-3428.2010.10.063
    Abstract ( ) Download PDF ( )   Knowledge map   Save

    The Chan-Vese(C-V) active contour model utilizes global region information of images, so it is difficult to handle images with intensity inhomogeneity. A Local region-based C-V(LCV) model based on image local region information is proposed, which utilizes the weighted average intensity inside a local window to replace the global average intensity of C-V model. Moreover, the distance penalized energy function is incorporated into it, which makes the expensive re-initialization unnecessary. Experimental results of medical image segmentation show it has a distinctive advantage over C-V model for images with intensity inhomogeneity, and it is more efficient than LBF.

  • LIN Yun-Chi, DIAO Dun-Gong, SHU Hua-Feng, HU Yong-Jian
    Computer Engineering. 2010, 36(10): 187-189. https://doi.org/10.3969/j.issn.1000-3428.2010.10.064
    Abstract ( ) Download PDF ( )   Knowledge map   Save

    Aiming at the drawbacks of Total Variation(TV) model for image impainting, this paper proposes an improved inpainting model based on image decomposition. Using image decomposition technology, it extracts the structure information and texture information from the image. The improved TV model is applied to structure part of image to effectively avoid step effect in TV model on smooth region. On the process of iterative, it respectively analyzes the feature points and non-feature points to avoid the fuzziness of feature points during the inpainting. Image texture information uses improved exemplar-based inpainting technology. Matlab simulation experimental results show that the inpainting effect and the results of peak signal to noise ratio are better than that of original algorithm.

  • TUN Jia-Feng, YANG Zhao-Shua, HAN Dong, BAI Zhuo-Fu, SU Yo-Ting
    Computer Engineering. 2010, 36(10): 190-192. https://doi.org/10.3969/j.issn.1000-3428.2010.10.065
    Abstract ( ) Download PDF ( )   Knowledge map   Save

    In the situation of unbalancing illumination or too complex background, global threshold can not binarize the 2D barcode image correctly, and local threshold can cause the block effect. In order to resolve this problem, a binarization method based on wavelet analysis and the Otsu method is proposed. This method estimates the approximate illumination distribution using wavelet analysis, and eliminates the effect of unbalancing illumination with the acquired illumination distribution. After that, the method binarizes the processed image using the Otsu method. This method not only eliminates the effect of unbalancing illumination, but also avoids the block effect encountered in local threshold method, by selecting appropriate wavelet function to make the approximate illumination distribution smooth.

  • CHEN Shen, HOU Yu-Jing, YANG Xu-Lang, HAN Min
    Computer Engineering. 2010, 36(10): 193-195. https://doi.org/10.3969/j.issn.1000-3428.2010.10.066
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes a mixed image denoising method combining wavelet transform with Partial Differential Equation(PDE), the method is denoised by the windows improved Wiener filtering, and is processed by the PDE combined model. Based on the characteristics of the P-M equation, the directed diffusion equation and LLT high-order equation, two weight functions are used to build the PDE combined denoising model. Experimental result shows that the mixed denoising method not only can overcome the drawbacks of three PDE models and Wiener filtering, but also can improve image denoising capacity and edge protection, and it can obtain a better visual effect and enhance efficiency of the algorithm after denoising.
  • SHU Qiang-Sheng, HU Yuan-Cui, CENG Lian-Qiu, FANG Fan, LV Zhong-Hua
    Computer Engineering. 2010, 36(10): 196-198. https://doi.org/10.3969/j.issn.1000-3428.2010.10.067
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To solve the problem of the growth deformation of the plant organs, this paper presents an approach of organs growth modeling based on L system and 3D Morphing. The approach describes the plant topological structure with L system, simulates the deformation process of the plant leaves and fruits with 3D morphing, and introduces the modeling method to implement the opening process of flowers. This paper brings forth a modeling process of cayenne. Experimental result shows that the approach solves the problem of the plant organs deformation, improves the modeling efficiency.
  • LIU Hai-Chao, YUAN Chang-An, LIU Hai-Long, XUE Lin, LI Gui-Lai
    Computer Engineering. 2010, 36(10): 199-200. https://doi.org/10.3969/j.issn.1000-3428.2010.10.068
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The accuracy of traditional classification method based on remote sensing image is difficult to achieve practical requirements because of remote sensing information uncertainty and the existence of mixed pixel. So Fuzzy C-Means clustering method is analyzed and realized. The researches show the behavior of the FCM clustering depends on the quality of the initialization of the parameters strongly. Remote Sensing Digital Image Fuzzy Clustering based on Gene Expression Programming(RSDIFC-GEP) is proposed and realized. By incorporating the local and global search and taking the clustering result of GEP as the initialized value of the FCM, the algorithm eliminates FCM trapped local optimum and is sensitive to initial value effectively.
  • SU Zhi-Xun, HU Jian-Beng, LIU Xiu-Beng, CAO Dun-Jie
    Computer Engineering. 2010, 36(10): 201-202. https://doi.org/10.3969/j.issn.1000-3428.2010.10.069
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper gives a remeshing algorithm according to an arbitrary genus and boundary 3D mesh model. It improves triangle quality and vertex position distribution by performing a series of adaptive local modification operations on the mesh. In order to reduce the accumulation of the error during optimization process, an algorithm based on distance field is developed to remain the optimized vertices on the original mesh. The implementation of the algorithm is simple and it does not need the complicated global parameterization operation. Experiment results show that the algorithm is effective, fast and robust.
  • CAO Jian, WANG Wu-Jun, HAN Fei, LIU Yu-Shu
    Computer Engineering. 2010, 36(10): 203-205. https://doi.org/10.3969/j.issn.1000-3428.2010.10.070
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the shortcoming of target recognition technology based on global features, this paper introduces target recognition technology based on local feature. The main component parts and the merits of the technology are introduced briefly. And the local features are divided into three classes and some approaches in common use are explored. From the point of applications, the difficulties of target recognition are discussed. The development trend of the area is forecasted.
  • WANG Jia-Yi, ZHANG Li-Qing
    Computer Engineering. 2010, 36(10): 206-208. https://doi.org/10.3969/j.issn.1000-3428.2010.10.071
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Recognition performances of traditional feature extraction methods degenerate dramatically in noisy environment. In this paper, a new approach called sparse discriminant analysis is developed and its algorithm derived by gradient descent method is given. It combines Linear Discriminant Analysis(LDA) with sparse constraint. Experimental results demonstrate that the method improves the speaker recognition performance in noisy environment.
  • CHENG Hui-Jie, ZHANG Guo-Yi, HE Ying
    Computer Engineering. 2010, 36(10): 209-211. https://doi.org/10.3969/j.issn.1000-3428.2010.10.072
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Due to the characteristic of small sample numbers and high dimensionality in tumor gene expression profile, an ensemble algorithm of particle swarm neural network is proposed to classify tumor subtypes. The genes irrelevant to classification are eliminated by different correlation functions and candidate feature subsets are formed. BP neural network based on sensitivity analysis is used as base classifier to learn the subsets and redundant genes are further removed. The parameters and thresholds of classifiers are optimized by improved Particle Swarm Optimization(PSO) algorithm. Experimental results show that the proposed method can obtain better recognition rates in tumor classification and only 54 feature genes in the feature set.
  • TANG Liang-Rong, JIANG Zhen, XU Wei-Hong, LI Ying
    Computer Engineering. 2010, 36(10): 212-214. https://doi.org/10.3969/j.issn.1000-3428.2010.10.073
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Based on fuzzy composition of maximum operation and a t-norm Tξ with a parameter ξ, a parameterized general fuzzy associative memory network Max-Tξ FAM and its effective learning algorithm are presented. By adjusting parameter ξ, the Max-Tξ FAM has good adaptability and flexibility in practice. It is proved theoretically that, using the mentioned above learning algorithm, Max-Tξ holds weak robustness to perturbations of training pattern pairs for any [0.1]. Experiment about image association validates this conclusion.
  • HE Ge-Jia
    Computer Engineering. 2010, 36(10): 215-217. https://doi.org/10.3969/j.issn.1000-3428.2010.10.074
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Dynamic Power Management(DPM) aims at minimization of power consumption of electronic systems by dynamically switching the power state of power manageable components according to the variations of workloads. Probability-based Adaptive Learning Tree(ALT) is provided against the defect of adaptive learning tree model. By characterizing the device activity in probability, probability-based ALT has higher hit ratio and can optimize the balance between performance and power. It is one kind of prediction strategy, with collection of prediction, control and feedback. Experimental result indicates that the probability-based ALT forecast strategy has the very good stability, comparied with other prediction strategies, it may further reduce system’s power loss.
  • Artificial Intelligence and Recognition Technology
  • HU Yi-fan; LIN Xin; DING Yong-sheng; WU Yi-zhi
    Computer Engineering. 2010, 36(10): 218-220. https://doi.org/10.3969/j.issn.1000-3428.2010.10.075
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper designs a wearable posture recognition system based on kernel-based Support Vector Machines(SVM). With the characteristic parameters extraction of tilt sensors embedded in garments, it completes posture classification using two kinds of classification algorithms. Experiment which is subjected to five testers shows that using kernel-based SVM can give the best performance, and the designed system can make a good recognition rate when classifying everyday posture, which has a considerable value in the remote monitoring field.
  • Networks and Communications
  • ZHANG Li-Qun
    Computer Engineering. 2010, 36(10): 221-222. https://doi.org/10.3969/j.issn.1000-3428.2010.10.076
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In accordance with folk Wudao chess game, a Wudao chess computer game program is designed and realized. The function and design goal of Wudao chess computer game program is introduced. The key problems processing method and software realizing method of Wudao chess computer game program are presented. It is proved that the Wudao chess computer game program runs fast and reliably. The design and realization of Wudao chess computer game program has some referential value to accomplish the computer game programs of other sorts of folk chess.
  • LIU Fu-Liang, HU Re-Ning, GAO Ying, SU Rui-Juan
    Computer Engineering. 2010, 36(10): 223-225. https://doi.org/10.3969/j.issn.1000-3428.2010.10.077
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    o solve the problem of converging to local optima, two methods are introduced to modify the inertia weight in parallel. When particles’ fitness values are worse than the average, the dynamic P chaotic map formula is devised to modify the inertia weight, which can make particles break away from local best and search global best dynamically. The introduced dynamic nonlinear equation is used to modify the inertia weight, which can make particles retain favorable conditions and converge to global best continually. Two methods coordinate dynamically. Experimental results demonstrate that the introduced algorithm outperforms other improved Particle Swarm Optimization(PSO) algorithms on many well-known benchmark problems.
  • CHEN Liang, SUN Min
    Computer Engineering. 2010, 36(10): 226-227. https://doi.org/10.3969/j.issn.1000-3428.2010.10.078
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In view of Web services composition method with QoS, this paper proposes an Immune Genetic Algorithm(IGA). The global performance of algorithm is improved by introducing immune theory into Genetic Algorithm(GA). This is mainly manifested in two aspects: one is that immune selection can be effective in preventing premature, the other is that the strategy based on immunological memory sub-group information exchange can be faster convergence. Because GA has the insufficiency in the Web services selection, and IGA can choose effectively from the alternative Web service and compose the Web service composition, which improves the service combination quality and the convergence speed. Simulation results indicate the feasibility of the algorithm by comparing IGA algorithm with GA algorithm.
  • HONG Ri-Beng, LIU Si-An, NI Tian-Quan-
    Computer Engineering. 2010, 36(10): 228-230. https://doi.org/10.3969/j.issn.1000-3428.2010.10.079
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In higher dense multi-return environment, reference target for the radar system to track the location of information may occur a poor association accuracy. Aiming at this problem, this paper adds Doppler frequency parameter to improve the accuracy of target tracking. In the process of data fusion, aiming at the difference of radar error, it gives a dynamic weight value allocation strategy. The weight value is sent to radar data fusion center for calculation and integration. Simulation results indicate that the method improves the tracking precision of radar system.
  • LIU Ji-Ban, SONG Mao-Jiang, DIAO Fang, YANG Zhi-Yu
    Computer Engineering. 2010, 36(10): 231-232. https://doi.org/10.3969/j.issn.1000-3428.2010.10.080
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To better represent the influence of the context to the part of speech of the current word, this paper proposes a second-order hidden Markov model based on the traditional hidden Markov model and applies it to part-of-speech tagging in Chinese. In the improved statistical model, sparse data problem occurs due to the shortage of training data. To solve this problem, an improved smoothing algorithm based on index linear interpolation is proposed, which provides effective smoothing. Experiments show that the second-order Hidden Markov Model(HMM) based on the context has higher correct rate and disambiguation rate of part-of-speech tagging than the traditional hidden Markov model.
  • YI Wen-Zhou, ZHANG Chao-Yang, WANG  Jiang, HU E-Mei, ZHOU Jin-Ling
    Computer Engineering. 2010, 36(10): 233-235. https://doi.org/10.3969/j.issn.1000-3428.2010.10.081
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In accordance with the respective advantages and disadvantages of Particle Swarm Optimization(PSO) algorithm and Differential Evolution(DE) algorithm, a novel hybrid algorithm is achieved through the improvement of Particle Swarm Optimization(PSO) algorithm and mixture with Differential Evolution(DE) algorithm. Massive experiments of five standard benchmark functions in five different dimensions suggest that this novel hybrid algorithm effectively overcomes the respective disadvantages of PSO algorithm and DE algorithm. It produces a conspicuous effect, which results in satisfactory outcome in experiments especially in high dimension.
  • SUN Pan-Jing, LI Yan, JIAN Jian-Sheng, MA Xiao-Dong
    Computer Engineering. 2010, 36(10): 236-238. https://doi.org/10.3969/j.issn.1000-3428.2010.10.082
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    According to the different demands of periodic and aperiodic data, the regulation of node transfer is presented based on non-destroyed priority bit-by-bit arbitrate regulation. The probability of data collision is minished effectually, and a limited transmission time is ensured. The maximum delay time is made adjustable and the real-time of the data communication is improved. The real-time protocol is described detailedly and the method of exception is given. View of the new protocol, the calculating methods of maximum transmitting delay is discussed and analyzed.
  • CENG Hui-Yan, YAN Ping, HE Xiao-Hui, JI Wei-Long
    Computer Engineering. 2010, 36(10): 239-241. https://doi.org/10.3969/j.issn.1000-3428.2010.10.083
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problems of high costs, low efficiency and poor universality in monitoring software at present, a development method of configurable intelligent monitoring software based on embedded platform with the thoughts of configuration is presented. The method can reconstruct the Human Machine Interface(HMI) of monitoring software with simple configuration way. The working principle and key technologies of the software are described, and the application example is given. Experimental results show the monitoring software developed with the method can configure HMI according to different environment. It realizes rapid reconstruction function of HMI.
  • LIU Dun, TAN Jian-Jun, SHAO Chang-Gao
    Computer Engineering. 2010, 36(10): 242-244. https://doi.org/10.3969/j.issn.1000-3428.2010.10.084
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In view of disadvantages of poor interactive, slow response and low scalability in traditional WebGIS application, this paper proposes a rich Internet WebGIS framework based on Flex after studying Flex technology. It elaborates on the hierarchical structure, composition and reciprocity of this framework and gives implementation. After applied in a public information service system, it shows that this framework can effectively enhance user’s experience, improve operating efficiency and provide stronger technical support for the release of spatial information.
  • XIANG Hui, WANG Hua-Bao
    Computer Engineering. 2010, 36(10): 245-247. https://doi.org/10.3969/j.issn.1000-3428.2010.10.085
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    A basic principle of PEB20534 is presented. The structure of multilevel FIFO, DMA controller and the interrupt system are discussed. It emphasizes on the discussion of software design, including the initialization of chain structure and interrupt processing. The technology is used for asynchronous serial port driving software based on WDM mode. It not only meets the requirements of serial port communication, but also ensures the correctness of data packet transmission with serial port.
  • LIU Yu, WANG Yu-Yan
    Computer Engineering. 2010, 36(10): 248-250. https://doi.org/10.3969/j.issn.1000-3428.2010.10.086
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To realize switch and control, it is necessary to select suitable data buffer structure for Ethernet switch and control chip. This paper uses pagination management mode of packet buffer space, assignment and release method of idle buffer space, management technology of output port queue. It increases usage efficiency of switch and control chip buffer space and enhances chip performance through the design method for descriptor of packet buffer space and the relative analysis of destination port structure.
  • ZHOU Xiao, LI Meng, BO Jia-Dun, YANG Zhi
    Computer Engineering. 2010, 36(10): 251-253. https://doi.org/10.3969/j.issn.1000-3428.2010.10.087
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes a remote medical surveillance system based on mobile equipment which uses General Packet Radio Service(GPRS) communication and encryption technology to combine with the development of General Packet Radio Service(GPRS). It considers over service client, network transmission, mobile terminal and data collection terminal synthetically to achieve real-time monitor, diagnose patient’s heartthrob and blood pressure and receive feedback information from service client. The principles and the overall program of the system is introduced. Data collection terminal design, control center software design and diagnostic algorithms are expatiated.
  • ZHANG Shi-Kun, DAN Chong-Shan
    Computer Engineering. 2010, 36(10): 256-258. https://doi.org/10.3969/j.issn.1000-3428.2010.10.089
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes an anti-pattern detection model for the anti-pattern “fine grain remote call” and uses this model to introduce an anti-pattern based J2EE middleware optimization model. Without interruption of the application system, this optimization model can support the online refactoring and overcome the disadvantage of the amend of the off-line client and non-automatic refactoring. It is used at Enterprise Assets Management(EAM) system, and the results prove that it realizes automatic online refactoring and reduces the workload of the developers while making other properties remain the same.
  • CHEN Jing-Bei, WANG Yong-Gui
    Computer Engineering. 2010, 36(10): 259-261. https://doi.org/10.3969/j.issn.1000-3428.2010.10.090
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problems of report system, such as model difficult to modify and poor reusability, this paper presents a semantic-based report system model to realize flexible separation and binding report data and report structure. It establishes the mapping relations between report structure and semantics, as well as report semantics and data. The report system developed based on this model is successfully run in the oil field, and meets the needs of complex report.
  • WANG Hui, GAO Cheng-Yang, LIU Ning
    Computer Engineering. 2010, 36(10): 262-264. https://doi.org/10.3969/j.issn.1000-3428.2010.10.091
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    According to the shortage of client-side Mashup, a development platform on server-side is proposed. It is designed with object-oriented method based on interface, which makes it have good extensibility. The object pooling is used for platform performance and a data bus is designed to reduce the coupling in platform and improve the flexibility of platform. Practice proves that the platform can meet the requirements of Mashup development on server-side.
  • ZHOU Gang, WEI Zheng-Xi
    Computer Engineering. 2010, 36(10): 265-266. https://doi.org/10.3969/j.issn.1000-3428.2010.10.092
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Combined with the mobile Agent technology and the characteristics of the distributed information system, this paper introduces a Cooperative Information System Oriented Middleware(CISOM) model based on the mobile Agent. It expatiates the designing ideas, system structure of the CISOM. It realizes the information communication between cooperative entities depending on migration and communication of mobile Agent. It analyzes the system security of the Agent, network transmission and resource access.
  • LEI Jing-Beng, SONG Yang-Xiong, GU Feng
    Computer Engineering. 2010, 36(10): 267-269. https://doi.org/10.3969/j.issn.1000-3428.2010.10.093
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper presents a design project of network behavior identification system based on PowerPC processor and embedded Linux. Data plane of the system is optimized to meet the needs of working under the high-speed network environment. NAPI network device driver and efficient buffer storage structure are designed. Data processing flow of Linux network protocol stack is shortened to promote the performance of Linux. Experimental result shows the throughput of the optimized system is improved compared with the original system.
  • Developmental Research
  • WANG Zhi; XIA Shi-xiong; NIU Qiang; LI Lian-xi
    Computer Engineering. 2010, 36(10): 270-272. https://doi.org/10.3969/j.issn.1000-3428.2010.10.094
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To make up the insufficiency of traditional motor fault knowledge representative methods, related theories of knowledge engineering and ontology are adopted to build domain ontology of mine-motor fault knowledge by analyzing mine-motor fault knowledge, and a knowledge base model is built based on ontology and rule-based reasoning. Tests results show that the method concluded in this paper can effectively represent the semantic relationships among motor fault knowledge and improve the motor fault knowledge reusability.
  • Networks and Communications
  • ZHENG Ji-Meng, WANG Jin-Song
    Computer Engineering. 2010, 36(10): 273-275. https://doi.org/10.3969/j.issn.1000-3428.2010.10.095
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Current pitch period detection algorithms can not work well under high noise environment. Aiming at this problem, this paper utilizes multi-scale continuous wavelet to pre-filtering the signal of voice. Based on the short-time autocorrelation function and the average magnitude difference function, a method for voice pitch period detection is proposed. This method needs little computation expense and can obtain the right pitch period. Smoothing algorithm is exploited to remove the errors of multiple frequency point, half frequency point and random point, and realize pitch curve smoothing.
  • LI Kun, LIU Jia
    Computer Engineering. 2010, 36(10): 276-278. https://doi.org/10.3969/j.issn.1000-3428.2010.10.096
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To effectively extract the pitch period, an improved wavelet transformation method is presented. On basis of wavelet filter, the frequency multiplication suppressed coefficients are added to remove most of the frequency multiplication errors. According to the voice, the system combines the linear prediction, which has a suitable order adjusting to the speech, to remove the format interference, while avoiding excessive damage to the pitch. Experimental results show the accuracy rate of the method is 97.8% at quite environment, and when the SNR is 0 dB, the one of it is 91.1%. The system has a good accuracy, robustness, and noise immunity.
  • Developmental Research
  • ZHANG Yu-jia; WU Hua-rui; ZHU Hua-ji; SU Zhong-bin; TAO Yong
    Computer Engineering. 2010, 36(10): 279-281. https://doi.org/10.3969/j.issn.1000-3428.2010.10.097
    Abstract ( ) Download PDF ( )   Knowledge map   Save

    Aiming at complexity of village-town land circulation business and non-uniformity of approving flow standard, this paper proposes a dynamic data management method based on Object Exchange Model(OEM) to build individual dynamic customization for approving flow data structure. Village-town land approval dealing system is developed after mapping OEM representation of approval flow data into eXtensible Markup Language(XML) file. Application results show that this system can implement digital treatment and auxiliary decision-making analysis of planning program in the process of land approval.

  • Networks and Communications
  • WANG Zheng-Hua, GUO Wei, WEI Ji-Ceng
    Computer Engineering. 2010, 36(10): 282-284. https://doi.org/10.3969/j.issn.1000-3428.2010.10.098
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    This paper proposes a genetic search algorithm model based on minimum delay to solve the instruction scheduling problems such as pipeline scheduling conflicts, deadlock scheduling and resource conflicts. It gives a genetic search algorithm model based on minimum delay. The software bypass optimization and dynamic allocation of resources are integrated into this model. Experimental results show that the algorithm can produce higher quality parallel code. Compared with the list scheduling algorithm, instruction level parallelism of more than 90% testbench is higher by using this algorithm.
  • TUN Zhen-Hai, LIU Fu-Yan
    Computer Engineering. 2010, 36(10): 285-287. https://doi.org/10.3969/j.issn.1000-3428.2010.10.099
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    On traditional x86 processors, address space switch usually requires flushing Translation Look-aside Buffer(TLB) and cache, which consumes much kernel time. With Fast Context Switch Extension(FCSE) mechanism on ARM920T embedded processors, the lower 32 MB of address space is redirected to the segment specified by process identifier. These segments are never overlapped, so that address information stored in TLB and cache keeps valid during process switching. Unnecessary flushing of TLB and cache is eliminated, and embedded system performance is improved.
  • ZHANG Li-Ping, CHEN Wei
    Computer Engineering. 2010, 36(10): 288-290. https://doi.org/10.3969/j.issn.1000-3428.2010.10.100
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Based on the mathematical model of customer-oriented catalog segmentation problem, through the risk assessment of customers, this paper introduces the factor and threshold of risk and makes the catalog segmentation problem into the interaction between customers and product. It gives the priority value of product a new calculation rules and presents a policyholder-oriented catalog segmentation model, makes it more suitable for marketing life insurance companies. The model is applied to an insurance software company Customer Relationship Management(CRM) system.
  • CHEN Bai-Lin, ZHONG Chun-Bei, TIAN Lian-Fang, MO Hong-Jiang, ZHENG Ze-An
    Computer Engineering. 2010, 36(10): 291-292. https://doi.org/10.3969/j.issn.1000-3428.2010.10.101
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    According to the actual monitoring requirement of cough patients in hospital wards, this paper presents a portable real-time cough monitor system. The system is developed based on DSP chip TMS320VC5509A. Overall framework is given, and the design of each module, the implementation elements and the software flow chart, including the process of the system, are introduced. Experiment in actual ward environment verifies the effectiveness and stability of the system.