Current Issue

• 2012, No. 10　Published:25 October 2012 Last issue    Next issue
  Columns in this issue: 论文

• 论文

  OverSim：A Scalable ApplicationLayer Multicast Network Simulation Framework

  2012, 34(10): 1-5. doi:

  Abstract ( 298 )   PDF (630KB) ( 354 )   　　

  As a viable alternative of IP multicast, applicationlayer multicast is increasingly becoming a hot research topic of Internet applications. How to simulate largescale, dynamic applicationlayer multicast overlay network has become a common problem faced by the application layer multicast scheme researchers. Based on the analysis of several mainstream network simulators, this paper proposes the OMNet++ based overlay network simulation framework OverSim in order to simulate the largescale and dynamic application layer multicast overlay network, and further analyzes the solution of scalable application layer multicast. At last, we simulate the applicationlayer multicast protocol Scribe on the OverSim and analyze the simulation results. The simulation results show that the OverSim can steadily simulate overlay network with thousands of nodes under the limited hardware environment. Besides, the simulation consumes low system memory that is linearly related with the overlay network scale.

  A QoS Multicast Routing Scheduling Algorithm in Cognitive Radio Networks

  YANG Wenzhong,ZHANG Zhenyu,WU Xiangqian

  2012, 34(10): 6-10. doi:

  Abstract ( 237 )   PDF (558KB) ( 295 )   　　

  The multicast information is usually transmitted through multiple intermediate nodes to reach the final destination node in multihop cognitive radio networks. There exist a lot of multicast routing protocols, while these based on traditional wireless networks are not suitable for cognitive radio networks. The question to be solved is in this paper: in multihop wireless networks, given a multicast routing with QoS requirements, how to create a multicast routing path and schedule node transmission so as to achieve the least bandwidth consumption in the transmission process. In this paper, a distributed multicast routing protocol is proposed to resolve the issue, It not only completes the establishment of the routing process, but also conquers the reasonable scheduling of nodes in the routing transfer process. Experimental results show that the transmission scheduling policy can effectively reduce network bandwidth consumption, and increase the success rate of the response of the multicast request.

  Research on QoS Mechanism for Mobile P2P Live Streaming System

  JIANG Wenbin1,YU Linchen1,2,LIAO Guangxian1,LIAO Xiaofei1,WANG Pengcheng1,C

  2012, 34(10): 11-16. doi:

  Abstract ( 260 )   PDF (869KB) ( 300 )   　　

  With the development of mobile networks and smart mobile terminals, researches on mobile P2P live streaming systems become popular. However, extending the P2P live streaming system faces two obstacles: one is the differentiated services and the other is the Quality of Service (QoS) during the domainacross movement of mobile terminals . In the paper, an adaptive multirate streaming service method is proposed to deal with the differences between mobile terminals and PCs. Multiple substreams of one stream with different rates are distributed in one P2P distribution topology. Clients buffer all or part of substreams according to their situations and abilities. They can adaptively change their buffer and the playing substream according to the variation of the network environment, and provide differentiated services for their neighbors. Meanwhile, a crossdomain mobile QoS mechanism containing data prebuffering and temporarystoring is proposed. Experimental results demonstrate the efficiency of the ways presented for the QoS in the Mobile P2P Live Streaming System.

  An Interactive Assistant Communication Scheme of VANET

  YAO Hong,HUANG Huawei,CONG Lei

  2012, 34(10): 17-21. doi:

  Abstract ( 276 )   PDF (1451KB) ( 315 )   　　

  In this paper, we study such a scenario: in the network consisting of both V2I and V2V communication, a vehicle is driving on highway, and it is trapped because of various reasons. In order to assist this trapped vehicle to maintain its network communication as possible, we propose an interactive network model and relevant scheme, which expropriates part of communication capabilities of other passing vehicles to assist the trapped vehicle to propagate data in an interactive way. We simulate the MVA model and scheme in NS2 and compare the proposed MVA scheme with DSDV, AODV, DSR routings in the blind area of VANET. The simulation results show that the AODV and DSR routings perform not well in the blind area of VANET. And the proposed MVA scheme achieves higher improvement in the average amount of network flow performance than DSDV routing.

  Study of Realname Authentication Method for WiFi Networks and Its Implementation

  ZHANG Jing

  2012, 34(10): 22-27. doi:

  Abstract ( 258 )   PDF (947KB) ( 332 )   　　

  With the development of mobile internet technology and the widespread application of smart phones, the demands of WiFibased mobile internet business are rising. Wireless WiFi networks are built in a number of industries and public places, and for network security management, realname authentication for wireless WiFi networks must be adopted. User name and phone number authentication are two basic means for it. This paper discusses authentication ways that are independent from specific protocols and vendors, as well as the methods and processes of implementing realname authentication.

  An Efficient Homomorphic Timed Commitment

  RUAN Ou1,2,ZHOU Jing2,ZHENG Minghui3

  2012, 34(10): 28-31. doi:

  Abstract ( 258 )   PDF (521KB) ( 307 )   　　

  At Crypto 2000，Boneh and Naor first introduced  the timed commitment scheme , whose commitments are secure against parallel attacks while ensuring the possibility of forced opening. There are many important applications for the timed commitment scheme, such as fair exchange protocol, fair multiparty secure computation and fair multiparty coinflipping. However, the plain BonehNaor's construction is very expensive, it needs a lot of computation and bandwidth in each commitment.  In the paper, based on Pedersen commitment scheme, generalized BlumBlumShub assumption and master timelines techniques, we construct an efficient timed commitment scheme. The master timeline of new scheme is generated in the setup step, and can be reused. Compared to BonehNaor’s construction, we lower the session costs of existing timed applications. Furthermore, our commitment scheme has another important property: homomorphism.

  User Interest Tracking Method Based on Paging Cache Model

  LI Zhihao 1,NIE Wenhui 1,CHENG Peng 2 ,ZHANG Yubo2,YANG Zhimin2

  2012, 34(10): 32-37. doi:

  Abstract ( 241 )   PDF (614KB) ( 352 )   　　

  Aiming at tracking user’s interest in intelligent recommender systems, traditional methods such as Time Window, Forgotten Function, consider little in topic relevance when characterizing user’s interest model and could hardly make full use of historic data, leading to inaccurate in tracking user’s interest migration. Therefore, the paper proposes a new characteristic model based on paging cache, which forms user’s multiple interest domains classified by topics. This paper proposes the corresponding SIM algorithm which introduces sequence entropy difference to characterize the integral features of interest migration. Experimental results show that the proposed method provides lower mean absolute error of interest and precisely characterize user’s interest migration, so as to enhance the quality and efficiency in services.

  A TwoParty Authenticated Key Exchange Protocol Based on TTP

  ZHOU Huihua1,2

  2012, 34(10): 38-42. doi:

  Abstract ( 241 )   PDF (687KB) ( 242 )   　　

  In this paper, a TTPbased twoparty authenticated key exchange (TTPTPAKE) protocol is proposed. It is suitable for largescale clienttoclient communication environments. This protocol can be completed in five steps and four rounds, and communicating entities can authenticate each other and establish a session key through a trusted thirdparty. A formal proof is presented to demonstrate the AKE security of the proposed TTPTPAKE protocol in the ideal cipher model and random oracle model.

  Local Outlier Detection Algorithm in Sensor Network

  WU Zhongbo,YUAN Lei,WANG Min

  2012, 34(10): 43-47. doi:

  Abstract ( 252 )   PDF (1239KB) ( 327 )   　　

  Finding outlier data is very important for sensor network. In this paper we put forward the concept of local outlier by an example. Different with traditional outlier, whether a sensor node becomes a local outlier only depends on its neighbor nodes. We give its formula definition and put forward a Nave algorithm that consume excessive energy. Then we devise a distributed algorithm named DLODA. The DLODA divides the network into many areas and selects a cluster head for each area. The cluster head can compute part results innetwork and filter the data which will not become the results to save energy. Finally we evaluate the DLODA in terms of energy consumption and response time. Experiments show that the DLODA has good performance.

  Research on Specification Language and Policy Conflict of Access Control Policy

  MA Xiaopu1,LI Zhengyan1,LU Jianfeng2

  2012, 34(10): 48-52. doi:

  Abstract ( 266 )   PDF (461KB) ( 368 )   　　

  Access control is a key technology for network security, which can ensure the information confidentiality and integrality. With the rapid development of computer, internet and wireless communication technologies and the growing problem of network security, the access control policy has become a hot research topic in computer science. Firstly, five major access control policy description languages are described. Their characteristics and application environments are analyzed. Secondly, the causation, classification, detection and resolution of the access control policy conflict are summarized. Finally some promising research orientations for access control policy are also presented in conclusion.

  Study on Energy Saving for 6LoWPAN under Enabled Node Sleep State

  CHENG Hongbin,WANG Xiaonan,LIANG Wei,NIE Panhong,SUN Xia

  2012, 34(10): 53-57. doi:

  Abstract ( 238 )   PDF (634KB) ( 243 )   　　

  In order to reduce the energy consumption of a 6LoWPAN node, a channel contention access mechanism based on enabled node sleep state is proposed. In the mechanism, nodes can appropriately sleep during both the superframe backoff period time and the superframe dormant period. Then a Markov chain model is established for the mechanism, and the corresponding mathematical derivation is carried out. Based on this model, the mathematical analysis on the average energy consumption, the steadystate probability of sending packets and the packet service time is performed, and moreover the influence of the protocol parameters NB and minBE on the network performance is also analyzed. The mathematical analysis shows that the model describes the channel contention access mechanism of 6LoWPAN very well, and the mechanism effectively reduces the average consumption and greatly improves the protocol performance.

  Research on Admission Control Mechanism for Distributed Computing Environments

  WU Libing1,2，DANG Ping1，NIE Lei1，LI Fei1，QIU Xin1

  2012, 34(10): 58-63. doi:

  Abstract ( 225 )   PDF (875KB) ( 247 )   　　

  Admission control algorithm is the core of resource advance reservation mechanisms for distributed computing environments. This paper summarizes the existing flexible reservation admission control algorithms and proposes a novel algorithmadmission control algorithm with transferable reservation in advance. The algorithm can transfer the scalable requests that have been reserved to reserve enough resources for fixed requests, increasing the acceptance probability of fixed requests. Data structure that stores the reservation information acts as an important role in resource reservation. Reasonable data structure can significantly improve the overall performance of admission control. Therefore, this paper further summarizes the existing data structure, including the slot array, the slot segment tree, the tree of resources and the binary search tree. The comparison experiments with three kinds of transferable reservation admission control algorithm show that the transferable reservation in advance algorithm can improve the acceptance rate of fixed requests and the total request accepted rate.

  Research on Parallel Execution Platform for MultiAgentBased Crowd Simulation

  XIONG Muzhou,LUO Wenzhen

  2012, 34(10): 64-67. doi:

  Abstract ( 233 )   PDF (586KB) ( 276 )   　　

  This paper proposes a multiagent based parallel platform for simulating virtual crowds. It applies MPI as the communication software amongst nodes in masterslave mode. Using the kmeans algorithm, the master node ensures the load balance amongst slave nodes and distributes and schedules agent model execution tasks to slave modes. It also contains modules of simulation result retrieval and monitoring. Slave nodes execute agent model execution tasks according to the task distribution by the master node. It also reports status information to the master node and synchronizes the execution process with other slave nodes. Experimental results on the parallel platform indicate that it is able to apply various agentbased model to the platform with efficient execution cost and good scalability.

  Research on Monitoring Method of Cloud Computing Cluster Server System

  DONG Bo,SHEN Qing,XIAO Debao

  2012, 34(10): 68-72. doi:

  Abstract ( 250 )   PDF (986KB) ( 597 )   　　

  Nowadays, with cloud computing technology more and more applied to various areas of information industry, there is a growing demand for monitoring and managing cloud computing cluster server system. The cloud computing cluster server system is composed of  a series of server clusters which are based on distributed architecture, and contains tens of thousands servers. An effective monitoring system is needed to monitor and control a cloud computing cluster server system with such a scale if we want to manage it well and ensure its highperformance operation. However, there are some defects in traditional cluster monitoring systems. In the paper, a highperformance monitoring and dispatching scheme for cloud computing cluster system is analyzed from monitoring system’s architecture, collection of monitor data, load balance scheduling and so on. We construct a cloud system solution which can ensure the highperformance of the cloud computing cluster system.

  Design and Implementation of Cloud Monitoring System Based on Server Clusters

  SHEN Qing,DONG Bo,XIAO Debao

  2012, 34(10): 73-77. doi:

  Abstract ( 254 )   PDF (919KB) ( 580 )   　　

  In order to solve the problems of data delay and distributed IT resources, more and more companies put the server and the application from the remote branches to the data center in order to do centralized management. This leads to more and more IT resources in the data center, and the scale of the data center tends to be larger. The data center manages not only the traditional minicomputers, servers, databases and various applications, but also storage equipments, blades servers, virtualization and cloud computing. In this way, the traditional network management and monitoring faces great challenges. This paper describes the comprehensive monitoring and efficient management of server clusters in complex network environments, and focuses on the research of monitoring largescale and multitype data in the data center. The cloud monitoring and management system is made up with data collection and massive data monitoring, It can achieve effective management and comprehensive monitoring of 1 000~10 000 servers. Currently, it can monitor five categories of hosts, for instance, Windows, Linux, Unix, IBM, and Solaris, and seven kinds of applications, such as mySQL, SQLServer, Oracle, Aapache, Tomcat, IIS, site monitoring, etc.

  Optimization Scheme of Disk Write in Multiple Virtual Machines

  YU Linchen 1, 2,LIAO Xiaofei2

  2012, 34(10): 78-82. doi:

  Abstract ( 185 )   PDF (729KB) ( 273 )   　　

   In virtualization environment, how to develop a novel data transfer method, which not only ensures the stability but also guarantees the performance of the system based on writethrough and writeback schemes, is an attractive topic. Based on the platform of computing system virtualization XEN, this paper proposes a new scheme to deal with dirty memory data effectively and reliably with the help of virtualized environment. This method is equipped with the advantages of both writethrough and writeback method in traditional architectures. It not only ensures the integrity and reliability of user data in the situation of system crash, but also provides satisfying performance with applications. Experiments are carried out to evaluate the system. The results show that the new scheme is able to work efficiently and can well protect the integrity of user data even if the operating system crashes unexpectedly.

  Multichannel FCM Segmentation Algorithm of  Satellite Cloud Image

  WANG Wei1，LIU Juan1，MENG Zhibin1，GUO Juntian2

  2012, 34(10): 83-87. doi:

  Abstract ( 247 )   PDF (713KB) ( 367 )   　　

  The satellite cloud images are important materials in the remote sensing measurement of the earth’s atmosphere. Cloud identification, tracking and prediction need segmentation and edge extracted operator on satellite cloud images. In this paper, we improve the fuzzy Cmeans (FCM) algorithm for cloud image segmentation. The experimental result demonstrates that the improved algorithm features efficiency and accuracy. In the experiment, we employ the RGB multichannel satellite cloud images in June 2011, cluster the cloud image on the layering of image, and extract multichannel clustering intersection areas, etc. The result achieves the fast and accurate segmentation to satellite cloud images.

  An Improved Face Recognition Method Based on SingularValuePerturbed with Single Image

  LIU Song1,2,LI Shidong1,ZHENG Minghui1

  2012, 34(10): 88-91. doi:

  Abstract ( 207 )   PDF (416KB) ( 295 )   　　

  In view of the poor performance of face recognition，an improved face recognition method based on singularvalueperturbed is proposed in this paper. Firstly, the singularvalueperturbed is applied to the single image so as to obtain expanded image set. Secondly, the wavelet decomposition is used as the preprocessing method, the lowfrequency face image is chosen as a subimage，and the highorder features are extracted by kernel principal component analysis. Finally, the nearest neighbor classifier is used for identification. The experiment results on ORL and Yale face databases show that the proposed method improves the recognition performance in comparison with the comparative approach.

  Survey of Visual Objects Tracking

  Lv Ze hua1,LIANG Hu2,TANG He1,WANG Guangwei2

  2012, 34(10): 92-97. doi:

  Abstract ( 255 )   PDF (561KB) ( 518 )   　　

  Detecting and tracking objects are the most prevalent and challenging tasks,which will be wildly used in video monitoring,human and computer interaction as well as military fields. Realtime,accurately and steadily expressing the objects are the key problems,whereas abrupt object motion,changing appearance patterns of both the object and the scene may lead to low accuracy.In this paper, tracking problem is categorized into two parts:object detection and tracking. Besides,the scene understanding is considered in visual tracking.In the end,the paper points out the drawbacks of the existing visual tracking systems,and prospects the future researches.

  A Novel Temperature Normalization Method of Infrared Face Images in Fourier Domain

  XIE Zhihua,ZENG Jie

  2012, 34(10): 98-103. doi:

  Abstract ( 223 )   PDF (617KB) ( 313 )   　　

  This paper proposes a novel temperature normalization method in Fourier domain, which can lessen the effect on infrared face recognition from ambient temperature. Firstly, the infrared face images in different ambient temperatures are transformed to Fourier domain. Secondly, based on statistical theory, the variances of phase spectrum and amplitude spectrum of the infrared face are used to describe the extent affected by the ambient temperature. Then, to achieve the robust information, those parts with big variances in the phase spectrum and amplitude spectrum are discarded and replaced by corresponding mean parts in training database. The main idea of this process is that one can set a suitable threshold for the variance of phase spectrum and amplitude spectrum and find those characteristic points that should be replaced. Finally, in order to verify the effectiveness of our temperature normalization method, the normalized infrared face is applied to the traditional face recognition system based on classic PCA method. Experimental results show that our normalization method can obtain stable information in infrared face and improve the performance of the infrared face recognition system.