Current Issue

• 2011, No. 11　Published:25 November 2011 Last issue    Next issue
  Columns in this issue: 论文

• 论文

  An Edge-to-Edge Network Architecture for Mass Streaming Media Transmission

  WANG Hui,SUN Zhigang,GONG Zhenghu,HE Junfeng,HE Ningwu

  2011, 33(11): 1-9. doi:

  Abstract ( 377 )   PDF (675KB) ( 379 )   　　

  With the progress of network technology, the Internet has shifted from barely file data exchang to application services. There are plenty of defects in the traditional EndtoEnd network architecture, such as deficiency of resource controllability, incapability of guarantee QoS, etc. So it is unsuitable to transfer the mass streaming media data. Although the P2P applications alleviate the load of servers, they damage the profit of ISP. In this paper, we present an EdgetoEdge network service model for mass streaming media transmission, which isolates the edge network from the core network, adds storage and calculation functions at the edge to increase network performance.We analyse the key techniques in this new network architecture, and network simulation results show it can decrease the stratup latency of users effectively and lessen the traffic flow of the core netwok remarkably.

  A Method of Network Traffic Identification Based on Improved Clustering Algorithms

  WANG Yuke1,LI Wenwei2,SU Xin2

  2011, 33(11): 10-14. doi:

  Abstract ( 316 )   PDF (805KB) ( 440 )   　　

  The automatic detection of applications associated with network traffic is very important for network security and traffic management. Unfortunately, because of some of the applications like P2P, VOIP applications using dynamic port numbers, masquerading techniques, and encryption, it is difficult using simple portbased analysis to classify packet payloads in order to identify these applications. And many research works have proposed using the clustering algorithms to identify network traffic, but these algorithms have some defects in how to choose the cluster center and the number of clusters. In this paper, we first use the Weighting D2 algorithm to improve the selection of the initialized cluster centers, and use the value of NMI(Normalize Mutual Information)to ascertain the number of clusters, and then get an improved clustering algorithm, and finally propose a application level identification method based on this algorithm. The experimental results show that this method  reaches 90% accuracy or more, and gets lower False Positive Rate and False Rejection Rate.

  An Algorithm of Mobile Node Localization Based on Weighted Centroid for Wireless Sensor Networks

  ZHU Xiaojuan1,MENG Xiangrui2

  2011, 33(11): 15-19. doi:

  Abstract ( 357 )   PDF (445KB) ( 604 )   　　

  In this paper we present an algorithm of mobile node localization based on weighted Centroid for wireless sensor networks. The proposed method splits the original sampling period of the Centroid algorithm into temporal windows in order to maintain a record of past information during movement. The selection of the anchor nodes is based on the weights of the temporal window. The RSSI ratios received from the anchor nodes are used as weights to determine the position of an unknown node. The results from simulation show that the proposed algorithm has low computing complexity, low cost and higher accuracy characteristics.

  A NFC Authentication Model Based on Trusted Computing

  WU Junjun，FANG Mingwei,ZHANG Xinfang

  2011, 33(11): 20-26. doi:

  Abstract ( 344 )   PDF (665KB) ( 355 )   　　

  By considering the privacy issues occurred in mobile payment with the NFC (Near Field Communication) devices, a NFC authentication model is proposed. This model can implement mutual authentication between the mobile platform and the service provide by using the anonymous authentication scheme. By issuing the qualification certificate and certificating the service provider, the proposed model can prevent the attacks from malicious service providers. Meanwhile, this model can achieve trusted authentication without leaking the configuration information and the identity information of the platform. Moreover, this scheme can identify the user’s account with the help of the issuer in case of the user’s anonymity.

  A Secure Protocol for Digital Copyright Without the ThirdParty Based on Dynamic License

  GUO Wei1,LIU Kun1,XIAO Zhenjiu2

  2011, 33(11): 27-31. doi:

  Abstract ( 357 )   PDF (494KB) ( 390 )   　　

  A secure protocol for digital copyright protection without thirdparty is presented. This protocol makes digital content and license each package separately, and distribute independently, which gets rid of a thirdparty  in the traditional DRM model, thus it can improve efficiency for both the client and the server. License is registered and checked by an online server, implementing the license dynamically allocated. Considering the integrity and security of the dynamic license, cryptography algorithms are used for data encryption and digital signature respectively. Besides, the used license will have been deleted while the online checking is over. The license database will be updated when the licenses are  exhausted, and it is effective to protect copyright from the cloned attack and the collusion attack.

  Analysis and Improvement of an Authenticated Key Agreement Protocol

  LU Shouke，SHI Guochuan

  2011, 33(11): 32-35. doi:

  Abstract ( 316 )   PDF (403KB) ( 394 )   　　

  The authenticated key agreement protocol proposed by Chen et al. does not have the property of KeyCompromise Impersonation (KCI) resilience. This paper conducts a detailed analysis on the flaw．To avoid this shortcoming，an improvement of the protocol is proposed based on the original scheme．This scheme permits two communicating sides to authenticate each other’s identity and agree the session key fairly. This protocol has better security characteristics and efficiency.

  A General Model of Multisite Synergetic Service and Its Application

  2011, 33(11): 36-39. doi:

  Abstract ( 341 )   PDF (604KB) ( 303 )   　　

  To solve the crossorganizational, crosssectoral and interregional synergetic application requirement based on the Internet environment, a general model of multisite synergetic service is proposed first. Then we design and develop a corresponding supporting infrastructure based on the model. This paper also introduces some related principles and conventions for the establishment of multisite synergetic service environment and its structure. An instance shows that the proposed model and its supporting infrastructure can be effective in realizing the cross multisite business connection and information sharing problems in the process of building EGovernment.

  Hybrid Custom Hardware Acceleration for Coarsegrained Dataflow Network Processor

  LI Tao,SUN Zhigang

  2011, 33(11): 40-47. doi:

  Abstract ( 307 )   PDF (789KB) ( 390 )   　　

  Aimed at the limitation of ILP exploitation and the fixed topology of controlflow NP, a novel scheme of coarsegrained dataflow NP architectureDynaNP is presented in this paper. DynaNP not only improves the programmability of the NP by the controlflow structure of Processing Elements (PEs), but also effectively exploits the tasklevel parallelism by introducing a dataflow model into the packet processing. Moreover，to further improve the system throughput of DynaNP, a mechanism of hybrid custom hardware acceleration is proposed taking consideration of the multicore and dataflow characteristics of DynaNP. Moreover, some key techniques of implementing the hybrid custom hardware accelerating unit are also addressed. The mechanism provides a unified interface for two kinds of hardware acceleration of custom instructions and coprocessors.

  AOP Based Methodology for Hardware Design Space Exploration

  LIU Feng,TAN Qingping

  2011, 33(11): 48-53. doi:

  Abstract ( 286 )   PDF (757KB) ( 319 )   　　

  As a new paradigm that provides new modularity constructs on top of objectoriented languages, Aspectoriented programming (AOP) was mainly applied in software domain previously. In this paper, we present a novel and synergistic approach to efficiently create systemlevel hardware prototypes based on AspectC++ and SystemC. This approach can be used in the hardware design space exploration. A case study on a 32bit adder is performed to demonstrate the efficiency of our methodology.

  国家自然科学基金资助项目（60921062，61003087）

  JIA Jia

  2011, 33(11): 54-59. doi:

  Abstract ( 331 )   PDF (526KB) ( 336 )   　　

  The applicationlevel checkpointing technique is one of the most commonly used and well matured faulttolerance techniques in homogenous systems. However, It is on its infant phase in heterogeneous systems and there are not accurate and reasonable solutions or approaches with respect to architectures and fault models of heterogeneous systems. Motivated by this observation, based on the architecture and programming model of the CUDA heterogeneous system, this paper analyzes the execution mode of CUDA programs running on CPUs and GPUs and proposes an asynchronous execution mechanism for the applicationlevel checkpointing technique in heterogeneous systems. With this mechanism, we explore a solution of optimal placement of checkpoints in heterogeneous systems. Finally, three experimental cases in the CUDA platform are used to evaluate our technique’s performance, feasibility and viability. The results demonstrate the effectiveness of our asynchronous execution mechanism for the applicationlevel checkpointing technique in heterogeneous systems. Compared with the synchronous execution mechanism, our mechanism is more flexible and has broader optimization space to explore. Moreover, Our solution of optimal placement of checkpoints can efficiently reduce checkpointing overhead and hence obtain higher performance.

  A Static Routing Algorithm Based on the 3D Mesh Structure

  ZHANG Linlin,LI Dongsheng,YONG Aixia

  2011, 33(11): 60-64. doi:

  Abstract ( 418 )   PDF (563KB) ( 305 )   　　

  3D NoC architectures offer better performance than 2D NoC in multiprocessor systems. The paper studies the 3D Mesh structure on the basis of the ideal topological structure of the average latency and the ideal throughput, and puts forward an evaluation based on the 3D Mesh new static routing algorithms. Finally we use the NS2 network simulation software to simulate  and compare the algorithms. The experimental results show that the new routing algorithm can improve the throughput, and largescale data throughput, shows better performance.

  A Study of the Two Helmholtz Solvers in the GRAPES Model Using GCR and GMRES

  SONG Junqiang1,WU Xiangjun1，2,ZHANG Lilun1,CHEN Dehui2,JIN Zhiyan2,HU Jiang

  2011, 33(11): 65-70. doi:

  Abstract ( 452 )   PDF (1034KB) ( 434 )   　　

  GRAPES（Global and Regional Assimilation and PrEdiction System）is a new generation of NWP model in CMA (China Meteorological Administration) for the operational implementation. After the discretization of computing equations for GRAPES's model, the first calculation becomes the solution of the Helmholtz equations which is the kernel computing of the dynamic framework. The running time for solving the Helmholtz equations is more than onethird of the total cost for GRAPESglobal mode at 0.5ox0.5o horizontal resolution with 38 vertical levels, and for the higher resolution model, the timecost is an exponential growth. The generalized conjugate residual method is employed to solve the 3D Helmholtz equation in the version of the GRAPES mode currently, as a contrast, another method which is based on GMRES（generalized minimal residual method）of PETSc(Portable, Extensible Toolkit for Scientific computation) is used here. The computation shows that the GMRES method with high precision can improve the forecast accuracy and has much better scalability for largescale parallel computing.

  Application of Metabolic Computing in Molecule Stability Analysis

  HU Yang,NIAN Xiaohong

  2011, 33(11): 71-74. doi:

  Abstract ( 296 )   PDF (336KB) ( 363 )   　　

  This paper summaries the main application fields in the artificial metabolic algorithm and discusses the development trend of information technology in molecule simulation. Based on the coding design mode in metabolic computing, the paper describes an analysis process about molecule stabilization. As some common molecules for examples, molecule stability can be discussed and compared by the metabolic computing method. The result shows the  feasibility and effectiveness of the method in molecule stabilization analysis.

  The Public Transport Dispatching Problem and Its ParthenoGenetic Algorithm

  YAO Chun,LI Maojun

  2011, 33(11): 75-79. doi:

  Abstract ( 315 )   PDF (568KB) ( 498 )   　　

  The bus scheduling problem is a typical combinatorial optimization problem. In this paper, combining with China’s urban public transport scheduling model and considering both the interests of company and passengers, the bus service frequency optimization model is established with company income satisfaction and passenger waiting satisfaction as the goals. Also a parthenogenetic algorithm for this model is designed. It introduces the random mutation step in individual reproduction, adding migration, reinsertion and other operations in intermediate subgroups. Finally, this algorithm is verified by a concrete examples. The simulation results show the superiority of the algorithm.

  Research of the Bidirectional MasterSlave Chord Resource Search Algorithm

  XU Zhihong,ZHANG Zhulou,GUO Hao,LI Xueyuan

  2011, 33(11): 80-83. doi:

  Abstract ( 300 )   PDF (424KB) ( 343 )   　　

  Chord is a structured P2P network model, it has the advantages of high speed, needing no center control, extensibility, load balance, fault tolerance, etc. But, as Chord is a oneway search algorithm, when the destination node is far from the current node, the routing delay increases because of multihops. And, low ability nodes with responsibility for large numbers of search and download tasks, and the nodes join in or leave the system momentarily and frequently, which can decrease the network efficiency obviously. An improved algorithm called the bidirectional masterslave chord algorithm supports twoway search, and all the network nodes are divided into super nodes and common nodes, and the super nodes with higher evaluation results construct the main ring of Chord. The experimental results show that the improved algorithm can reduce routing hops and network delay effectively.

  A Bee Colony Optimization Algorithm for ClusteringConstraint Scheduling

  ZHAO Lianghui,WANG Tianqing

  2011, 33(11): 84-88. doi:

  Abstract ( 305 )   PDF (517KB) ( 331 )   　　

  A beecolonyoptimization based scheduling algorithm is proposed to meet the clustering constraint which often appears in real production fields. The algorithm performs its searching operation by means of imitating the bee colony’s “foraging” and “waggle dance”, including endowing bees with different “believes” to keep the searching diversification of the colony, merging the clustering constraint into bees’ behavior as “social rule” to meet the user’s individual demands, and sharing information among the bees through “waggle dance”, i.e. making bees display their paths and choose a preferred path from others at the same time. The computational experiments on a series of JSP benchmark problems and the comparison with other algorithms validate the effectiveness of the proposed algorithm.

  A Survey of Human Pose Estimation from Single Images

  TANG Zesheng,WANG Zhaozhong

  2011, 33(11): 89-97. doi:

  Abstract ( 416 )   PDF (1321KB) ( 684 )   　　

  Pose estimation from images is to detect the position where the human body is located in the image and computes the orientation and scale for each body part, the results of pose estimation is always classified as 2D and 3D, and the approaches to these results can be divided as modelbased approaches and modelfree approaches. In this paper we first give a simple introduction to the background and applications of pose estimation, then conceptions relevant to pose estimations are discussed, and the different representations of the results produced are compared, the remaining part of this paper focuses on the human body detection and human pose estimation, which are discussed in detail, all the different methods are discussed theoretically by applicability, at the end of this paper.The  problems remaining to be located are analyzed.

  An Image Enhancement Method Based on DualTree Complex Wavelet Transform

  ZHANG Jing,LI Yibing,LI Ao

  2011, 33(11): 98-102. doi:

  Abstract ( 371 )   PDF (683KB) ( 427 )   　　

  In traditional image enhancement methods, such as histogram equalization, wavelet transform, etc, these methods achieve good performance, however, there still exist some drawbacks. The noise is amplified during the image enhancement processing. In this paper, an image enhancement method based on the 2D dualtree wavelet transform is proposed, aiming at the exsiting problem in traditional image enhancement methods.The 2D dualtree complex wavelet transform can describe the directional information of images better.The enhancement algorithm divides coefficients into three classes: strong, weak edge and noise point.The enhancement function will process the coefficients of each class respectively. The enhancement method adopted in this paper can suppress noise effectively. In this paper, the effect of parameters on the enhancement function is discussed taking PSNR as the evaluation standard, in order to optimize parameter selection.The experimental results show that the proposed method receives better performance.

  A Wavelet Transform Image Fusion Algorithm Based on Related Coefficients

  YAN Liwei,REN Xiaokang

  2011, 33(11): 103-107. doi:

  Abstract ( 321 )   PDF (809KB) ( 522 )   　　

  The key problem in multifocus image fusion is how to keep the detail and edge in the original image.Therefore,a new image fusion algorithm is developed based on the related coefficients of wavelet transform.The schemes of choosing high frequency and low frequency coefficients are two approaches designed. While choosing high frequency coefficients,we select these that have the maximal absolute values and combination of averages with selection.When choosing low frequency coefficients,we select two principles based combination of average with selection and coefficients related.The algorithm is used to fuse two multifocused images. In addition, the performance of the fusion methods is appraised and analysed with objective evaluation index. The experiments show that this algorithm can preserve all the useful information from the primitive images and all the targets in the fused image are very clear. Each fusion evaluation index achieves improvement. The algorithm is of certain effective and parctial.

  A Segmentation Method for the Occluded Vehicles Based on Perspective Points and Contours

  LI Yucheng,OU Xiaodan,TIAN Zhen

  2011, 33(11): 108-112. doi:

  Abstract ( 354 )   PDF (757KB) ( 406 )   　　

  The paper proposes a new method of combining perspective points and vehicle contours to segment the occluded vehicles in the image for resolving the problem of multivehicle occlusion in vehicle detecting and tracking. Perspective points are detected by analyzing the lane lines in the highway scene, and the cuboids of vehicle models are built with the perspective points. The ratio of the cuboid’s height to width and the size of the vacant area inside the cuboid models are used to detect the occluded cars. Different searching methods are chosen to find the cutpoints, depending on the condition of occlusion. Then, the occluded vehicles are detected and segmented in sequence from front to back. The experiments show the accuracy and adaptability of the proposed method in the segmentation of the occluded vehicles on a straight road.