Loading...
  • 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

Current Issue

    • 论文
      Topology Awareness of the Hierarchical PeertoPeer Networks Based on Network Coding
      YANG Jun,DAI Bin,HUANG Benxiong,HUANG Chen
      2011, 33(2): 1-6. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 602 )   PDF (659KB) ( 484 )     

      Network coding is a promising technology to improve the throughput and reliability of networks. However, its benefit is affected by network topology, especially in peertopeer systems. System employed network coding can improve its performance in terms of throughput and reliability by precise topology awareness. In this paper, we propose a TANCF (Topology Awareness based on Network Coding with Feedback) scheme which is based on the feedback from the intermediate nodes and the receiver nodes to detect the path in hierarchical P2P networks. It makes use of network coding with feedback to reveal the network connectivity and construct the network topology. According to the close relationship between the adjacency matrix and path matrix, it can be used to construct the network topology in order to optimize the parameters of network coding. Through a theoretical analysis, we obtain the probability equation and give the conditions under which our scheme can work. Moreover, by simulation experiments, we prove that the TANCF scheme can obtain the path matrix correctly.

      Development of a PriceBased Congestion Control Model
      PENG Lihong,ZHANG Heying,JIANG Xinwen,DOU Wenhua
      2011, 33(2): 7-11. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 541 )   PDF (389KB) ( 406 )     

      The congestion control problem in communication networks has been abstracted as an aggregate utility maximization problem in pricebased congestion control model. Pricebased model has become the general framework to design and analyze congestion control mechanisms. However, a pricebased model is different from real network environments since it is a simplified linear model. In recent years, research efforts are made to extend and improve the pricebased model. In order to systemicaly understand the development of this research area, we first briefly review the concept of pricebased model, and analyzes its limitation, then summarizes the research efforts. Finally, we provide some further interesting research directions.

      Design and Implementation of a Visual Network Simulation System Based on NS2
      REN Ju,DENG Xiaoheng,CHEN Zhigang
      2011, 33(2): 12-16. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 539 )   PDF (878KB) ( 768 )     

      NS(Network Simulator) has always been widely used in network simulation. However, some of its shortcomings such as difficulty in use, lack of longrange simulation support have limited its development. This paper proposes a remote visualization simulation system based on NS2 and gives the implementing method. We use Flex and the ActionScript programming language to develop the client program. In addition, we use component technology to develop the library of scenarios and a distributed component autonomy model to realize the automatic generation of the script. Then the system conveys the script to the server and the NS2 child process executes this script by remote calling. Meanwhile, the thirdparty software library Gawk and Pnuplot are integrated to analyze the simulation results and realize the curves description and the animation show of the results, which ensures that protocol researchers can have an intuitive and indepth understanding of the complex network behavior. It provides a good platform for the users who are learning computer network remotely.

      District,Changsha,Hunan 410073,P.R.China Network Analysis Based on the Dioid Theory
      FAN Baohua,ZHANG Heying,DOU Wenhua
      2011, 33(2): 17-22. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 453 )   PDF (447KB) ( 436 )     

      We study the application of the dioid theory especially maxplus algebra in computer network analysis. The IO characteristics of a discrete token bucket is analyzed under the maxplus algebra. Moreover we obtain the (b,ρ) token bucket’s transfer matrix and state space equations under the maxplus algebra, we also propose a framework of maxplus network calculus and give the definition of the maxplus arrival curve and the maxplus service curve. On the basis of these two concepts, we prove the theorems about the delay and the burst of output flow. And finally, we give the application of maxplus network calculus in network analyisis, we use the nonpremptive priority node and the Guaranteed rate server as our analyzing objects and obtain their maxplus service curves respectively. We also compare our results to the minplus network calculus and show the advantages of maxplus based network analysis.

      An Ontology Based Testing Case Reuse Model for Penetration Testing
      LOU Fang,LI Liang,HE Zhiqiang
      2011, 33(2): 23-26. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 452 )   PDF (546KB) ( 377 )     

      To improve the capability and efficiency of penetration testing teams, a test case reuse model for penetration testing is built, which is based on the ontology for penetration test knowledge. Meanwhile, this paper gives the methods to describe, organize and query the knowledge in a formal way. It gives the solution for knowledge management in the penetration testing field, which can be used to improve the knowledge sharing in testing teams and different projects.

      Design and Implementation of the DM9000 Network Device Driver Based on ARM and Linux
      YUAN Anfu,XIA Shengfeng
      2011, 33(2): 27-31. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 478 )   PDF (744KB) ( 424 )     

      DM9000 is a widely used Ethernet chip in the arena of embedded systems.This paper makes an indepth study of embedded systems baesd on the use of ARM920Tchips S3C2440 and network chip DM9000 interface circuits, analyses and gives the network device driver functionality abstraction of design templates in the Linux operating system, and realizes the DM9000 network interface’s driver finally,which gives a good example to network device driver development.

      An AnonymityLevel Selectable  Hierarchical Anonymous Communication Framework
      CHEN Xin1,WANG Wei2,LIU Bo1,XIAO Fengtao1,HUANG Zunguo1
      2011, 33(2): 32-36. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 468 )   PDF (824KB) ( 350 )     

      With  more emphasis paid on network security and privacy, the research on anonymous communication systems has developed quickly. In the past, researchers mainly focused on how to improve the anonymity, but few focused on the usability. In this paper, an anonymitylevel selectalbe  hierarchical anonymous communication framework (ALHACF) is presented. In this framework, firstly the anonymity level is selected by the user according to his requirement, then the path length and set of relay routers will be determined by the anonymity level. The anonymity and communication performance of ALHACF are analyzed. The analysis indicates that ALHACF inherits the merits of HLLACF, and it can prevent some kinds of attacks, keep good anonymity and increase the communication performance markedly.

      Research on the Evolution of Complex Networks and Their Application in Internet Load Balancing
      YANG Zhongming1,QIN Yong1,HUANG han2,CAI Zhaoquan3,WU Yugang1,4
      2011, 33(2): 37-41. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 455 )   PDF (532KB) ( 575 )     

      Some research results of the structural characteristics of complex networks are summarized and the future application of structural property research in Internet load balancing is prospected. The evolution of the Internet structures of complex networks is summarized. ASlevel selfgovernment domain networks usually have only one or some exports, and longrange connectivity decides the shortest path routing from each group to the Internet, which means that the heaviest load is on the largest range. The problem to be resolved of longrange connectivity load balancing mechanism modeling about the Internet local world is presented, which explores the research of range and longrange connectivity load balancing in networks and makes an analysis of the drawbacks of the problem. An algorithm of multipath traffic load balancing of networks with local world property and its application are studied.

      Improvement of the Precise Time Synchronization Algorithm Based on IEEE1588
      LI Xueqiao,CHEN Yuan,LIANG Shuang
      2011, 33(2): 42-45. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 431 )   PDF (474KB) ( 346 )     

      In the field of industrial control, clock synchronization precision is an important factor that  affects the realtime performance of the industrial Ethernet. In order to satisfy the request for precise clock synchronization in the industrial Ethernet, this paper conducts the research to the IEEE1588 highprecision time synchronization protocol, carries on the elaboration to the principle of highprecision synchronization, proposes the corresponding improvement algorithm for asymmetric communication paths. Finally, the algorithm is tested in a selfdesigned system. The results of test show that the improved algorithm can effectively increase the clock synchronization accuracy under asymmetric communication paths.

      Research on the Software Cache of Cell Processors
      YANG Canqun,WANG Feng,DU Yunfei
      2011, 33(2): 46-50. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 431 )   PDF (784KB) ( 259 )     

      For improving the performance of references to the system memory on a Cell processor, this paper implements a software cache supporting the released memory model. The experimental results show that the software cache can reduce the overhead of the references to the system memory, and can improve the performance of OpenMP programs.

      A Survey of the NonUniform Cache Architecture
      WU Junjie,YANG Xuejun
      2011, 33(2): 51-60. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 501 )   PDF (960KB) ( 456 )     

      Because of the memory wall problem, cache memory technologies have been very important. In the future, growing wire delays will become one of the key factors restricting the design of large caches. To provide a uniform access delay, the traditional cache design methodology has to accommodate cache access time to the farthest cache bank from processors. Thus, researchers proposed a kind of NonUniform Cache Architecture (NUCA), which is almost the inevitable outcome in intending processors. If a processor hits the data  stored in the nearby banks of NUCA, the cache access time will be short. Otherwise, the access time will be long. This paper summarizes the origin, development and typical cases of NUCA. It is important that two memory technologies of multiprocessors, from which the research of NUCA may borrow ideas, are introduced. Finally, the key problems of the NUCA research as well as solutions are proposed in the authors' opinion.

      Design of a 2~5GHz 0.18μm CMOS  Wideband Low Noise Amplifier
      HE Xiaowei, ZHANG Minxuan
      2011, 33(2): 61-64. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 452 )   PDF (854KB) ( 356 )     

      A twostage 25GHz CMOS Low Noise Amplifier(LNA) for the lower side of UltraWideBand(UWB) applications(3.15GHz) is presented. This LNA consists of two stages. The first stage is a commongate structure maintaining good linearity with better input matching; the second stage is a commonsource stage stacked with a current mirror, which obtains low Noise Figure(NF) and reduces the LNA’s power consumption. By introducing commongate and commonsource stages to compensate gain mutually, the proposed LNA has achieved nearly flat power gain and NF. This LNA has been implemented by a 0.18μm CMOS process. The successive simulation results indicate that the power gain(S21) keeps 11.5dB over the wide frequency band of 25GHz with NF 5.1dB and input return loss(S11) below 22dB in the entire band. The simulated inputreferred thirdorder intercept point(IIP3) at 4GHz is 10dBm. The LNA consumes about 11mW under the 1.8V supply.

      Performance Analysis of the Nonlinear Asynchronous Pipeline Based on Queuing Networks
      JIN Gang,WANG Lei,WANG Zhiying
      2011, 33(2): 65-69. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 401 )   PDF (437KB) ( 373 )     

      It is very important to analyse the performance of the asynchronous pipeline during the design of the asynchronous circuit. But there is no very efficient way to evaluate the performance of the complex nonlinear asynchronous pipeline. The queuing network is a very intuitive tool to model the asynchronous pipeline, and a performance model of the asynchronous pipeline based on the forkjoin queuing network with blocks has been introduced in this paper, then an equivalent stochastic marked graph has also been introduced, and the upper and lower bounds of the average cycle time are presented in the end of this paper.

      Effect of Quantum Capacitance in DualGateMaterial MOSlike Carbon Nanotube Field Effect Transistors
      ZHOU Hailiang,ZHANG Minxuan,FANG Liang
      2011, 33(2): 70-74. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 393 )   PDF (944KB) ( 370 )     

      The dualgatematerial MOSlike Carbon Nanotube Field Effect Transistors (CCNFETs) based device design, proposed by our research group, can not only increase the ONOFF current ratio, tune the subthreshold voltage, decrease the device power, but also eliminate the ambipolar transport property of the traditional MOSlike CNFETs effectively. However, as a typical lowdimensional system, such CCNFETs design suffers from the effect of quantum capacitance inevitably. In this paper, based on a study of the working principle of this novel device design, we make a first research into the effect of quantum capacitance on the transport property of it. The results show that quantum capacitance can not only result in a great increase of the OFF current and a corresponding decrease of the ONOFF current ratio, but also destroy the unipolar transport property of this novel device design, which would have a negative impact on its application in circuits. In order to decrease the impacts of quantum capacitance as greatly as possible, several suggestions are given at the end of the paper.

      Research on SingleEvent Transients in VoltageControlled Oscillators of PhaseLocked Loops
      QIN Junrui,CHEN Jihua,ZHAO Zhenyu,LIANG Bin,LIU Zheng
      2011, 33(2): 75-79. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 425 )   PDF (1067KB) ( 316 )     

      Based on process calibration, single event transient in VoltageControlled Oscillators of high frequency PhaseLocked Loops in a 0.18μm bulk process is studied by the TCAD Mixedmode simulation. The impact of voltage, LET and temperature on SET are studied. Our simulation results demonstrate that the max phase difference caused by the particles on VCO is weak when NMOS works in the cutoff region. The pulse error number caused by the particles on VCO increases linearly as LET increases, and when the device temperature increases, the max phase difference on VCO also generally increases.

      Memristor and Its Application
      ZHANG Rongfen,DENG Chaoyong,FU Xinghua
      2011, 33(2): 80-85. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 565 )   PDF (624KB) ( 1226 )     

      Memristor is the contraction for memory and resistor, means memorable resistor, which can store information without the need of a power source. It was predicted in 1971 by Leon Chua as a “Missing Circuit Element” and was identified by R. Stanley Williams as a possible nanoscale device. In this paper, we review the concept and the mathematical definition of memristor, introduce the HP’s sandwich model of Pt/TiO2/Pt, and provide a fundamental analysis of the memristor behavior , such as pinched hysteretic loops. Finally, we discuss its application and research fields as a nanoscale device in the future.

      Approximate GradientBased Big Bang Search Algorithm
      CAO Ju,JIANG Xingwen
      2011, 33(2): 86-91. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 538 )   PDF (426KB) ( 307 )     

      A new intelligence optimization,Big Bang Search (BBS),is proposed by simulating the big bang process. Inspired by the classical optimization method, the concept of “Approximate Gradient” is defined and the “Approximate Gradient Explosion” (AGE) operator is created,and an improved method called Approximate Gradientbased Big Bang Search (AGBBS) is proposed. AGBBS keeps down the excellent feature of BBS, the nice combination of uniformity and randomness of distributed candidate solutions; it fully uses the information of explosive pieces, which enhances the algorithm’s search ability. By improving some heuristic operators, the convergence of the algorithm and the accuracy of solutions are improved. The testing of 12 standard benchmark functions and a comparative analysis demonstrate the effectiveness of the new algorithm and the robustness of the AGBBS.

      Study on Infrequent Itemsets Mining AlgorithmsBased on Vector Inner Product
      LIU Caihong1,LIU Qiang2,LI Aiping3
      2011, 33(2): 92-96. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 488 )   PDF (597KB) ( 345 )     

      Aiming at how to produce infrequent itemsets in the negative association rules, this paper introduces vector inner product to this field. By converting the transaction database to the Boolean Vector Matrix, and by allotting a equitable data storage structure, we put forward a new algorithm to produce infrequent itemsets effectively. First of all, we convert a database to a Boolean Vector Matrix; and then calculate the inner vector in the matrix, and finally produce infrequent itemsets and frequent itemsets with the restriction of the 2LS model according to the idea of incremental change layer after layer ,which makes sure that infrequent itemsets not only can be produced by the joint of frequent itemsets , but also can be produced by the joint between infrequent itemsets and frequent itemsets, and between infrequent itemsets and infrequent itemsets .The experimental results show that this method not only scans the database only once, and also has the virtues such as dynamic pruning, without saving mid items, saving lots of memories, and without losing infrequent itemsets, which has an important meaning to the negative association rule mining and all kinds of itemsets with the characteristics of low frequent appearance, strong correlation in databases.

      A Novel Evolution Strategy Algorithm for Solving the Standard Eigenvalue of the Symmetric Interval Matrix
      XIA Huiming1,ZHOU Yongquan2
      2011, 33(2): 97-101. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 490 )   PDF (525KB) ( 341 )     

      For the eigenvalue problem of real symmetric interval matrixes, the uncertainties of the interval are thought as one perturbed quantity around the midpoint of the interval. In this paper an Evolution Strategy Algorithm is presented for solving the eigenvalue problem of interval matrixes based on the expansion of the interval. Taking the center point of the interval matrix as the balancing point and the interval indefinite quantity as the perturbation quantity, the solving the minimum and maximum eigenvalues of the interval matrix by perturbation formulas, the solution of the eigenvalue problem is obtained. The example which is given in the article has illustrated that the algorithm is effective. The main feature of the algorithm is that the result region is very accurate and the convergence speed is fast.

      A New Image Fusion Algorithm Based on Wavelet Transform and PCNN
      ZOU Beiji,HU Yiling,XIN Guojiang
      2011, 33(2): 102-107. doi: 10.3969/j.issn.1007130X.2011.
      Abstract ( 510 )   PDF (766KB) ( 444 )     

      With the development of fusion technology and the growth of the wavelet theory,wavelet transform uses its excellent time and frequency localization,excels in the area of fusion.On the basis of the wavelet theory,we propose a new algorithm which combines wavelet transform with the pulse coupled neural networks.Firstly,we perform a wavelet multiscale decomposition of two original images which have been registered.Secondly, we propose a novel fusion rule based on the improved PCNN.The key point in this paper is that in allusion to the feature of the high frequency domain and the low frequency one,using various methods to each frequency domain,respectively.Finally,it can obtain a fused image by taking inverse wavelet transform to reconstruct images.The results of simulation and quantifying evaluation show that this algorirhm can preserve more useful information from the original images effectively,and enhance the quality of the fused image.

      论文