Current Issue

• 2010, No. 2　Published:25 January 2010 Last issue    Next issue
  Columns in this issue: 论文

• 论文

  A Flexible QoS Unicast Routing Scheme Based on Utility and QGA

  WANG Xin-Wei, JU Rong-Zhu, HUANG Min

  2010, 32(2): 1-3. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 2127 )   PDF (362KB) ( 838 )   　　

  Based on the knowledge of fuzzy mathematics, microeconomics and swarm intelligence, this paper proposes a flexible QoS unicast routing scheme with the QoS(Quality of Service) satisfaction degree and utility. By using QGA(Quantum Genetic Algorithm), the scheme searches for a QoS unicast path which maximizes the network provider utility and the user utility while trying to strike a right balance between the two. Simulation results show that the proposed scheme is both feasible and effective.

  esearch on the Efficiency and Fairness Issues in Bandwidth Allocation

  ZHANG Yang, HE Li-Yuan, XIE Ying-Ze

  2010, 32(2): 4-6. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 610 )   PDF (493KB) ( 1344 )   　　

  This paper presents new definitions to evaluate fairness and efficiency in bandwidth allocation. Based on these definitions, we compare the fairness and efficiency among different bandwidth allocation algorithms. The bandwidth allocation problem can be treated as solving a utility maximization problem. We propose a new utility function and compare the bandwidth allocation algorithm using this function with other fairness algorithms,and establish the relationship between them.

  A HashBased Locator/ID Mapping Mechanism

  CHU Rui, SU Jin-Shu

  2010, 32(2): 7-10. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 596 )   PDF (541KB) ( 480 )   　　

  With the development of the Internet, the original TCP/IP architecture is facing some critical challenges from its scalability, mobility, security and so on,and the flaws of which have hampered the development of some new network technologies such as multihoming and traffic engineering. One of the most important causes for this problem is the IP address semantic overloading. To address the issue, researchers commonly suggest that we should redesign the naming and addressing architecture of the next generation Internet. “Locator/ID Split” is one of the most important techniques in tackling the problem. At the core of “Locator/ID Split” is the scalable flatlabels based mapping service. This paper proposes HLIM, which uses hashbased routing to provide a deterministic mapping resolution for the edge network through distributed mapping servers. HLIM satisfies the scalability of the flatlabels “Identifier” mapping service, and can adapt to the dynamic joining and leaving of the mapping nodes to some extent.

  Research on the KConnected Cluster Topology Control Algorithm for Wireless Sensor Networks

  MENG Zhong-Lou-1, WANG Shu-1, WANG Qi-1, DIAO Feng-2

  2010, 32(2): 11-14. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 953 )   PDF (1078KB) ( 768 )   　　

  A Kconnected cluster topology control algorithm is proposed in this paper. The main content of the algorithm is the selection of the backbone nodes and the redundant backbone nodes during the procedure of the cluster forming and sleeping scheduling of the cluster nodes. In the selection of the backbone nodes, energy balancing, channel quality, node faults and mobility are taken into account. The sleeping scheduling gives different scheduling mechanisms for different types of nodes. The simulation results show that the algorithm has a good performance in network topology stability and lifetime with some fault tolerance.

  A Survey on the Neighbor Based Broadcast Protocols in Ad Hoc Networks

  CHEN Wei-Xu, SUN Wei-Wei

  2010, 32(2): 15-19. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 518 )   PDF (433KB) ( 586 )   　　

  Nowadays，ad hoc networks become very promising with their development.Since broadcast algorithms in ad hoc networks are crucial to route discovery, they attract increasing attentions from the researchers. Neighbor based broadcast algorithms are the most popular broadcast ones in ad hoc networks. In this paper, we demonstrate the detailed categories of these algorithms. We also analyze and present the strengths and weaknesses of different broadcast algorithms in detail, and give the comparisons of them. In the end, we conclude the whole paper and discuss some future trends of the research in broadcast algorithms in ad hoc networks.

  An Analysis of the Location Management Strategies for the WiMAX Networks Under Different Topologies

  ZHANG Jian-Wu, DIAO Qi, JU Jing-Yuan

  2010, 32(2): 20-22. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 752 )   PDF (692KB) ( 417 )   　　

  Supporting the mobility of the IEEE802.16e［1］ standard on fixed WirelessMANs is supposed to be a promising solution for mobile broadband wireless access (BWA). By integrating HMIPv6［2］ with the IEEE802.16e system,the performance of the system can be enhanced greatly. The mesh topology in the IEEE802.16 standard can achieve wide market coverage as well as low investments of the facility. In this paper, the cost and delay of the traditional location management schemes in Wireless Mesh Networks are analyzed theoretically and compared with that of the typical cellular networks. The results show that a more effective location management scheme which is suitable for the WiMAX Mesh topology should be proposed so that it can better support the mobility and application of different services.

  A Model for the Web Retrieval Model Based on Domain Ontology

  WANG Yan-Wen, TUN Gang

  2010, 32(2): 23-26. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 588 )   PDF (623KB) ( 428 )   　　

  In this paper, an ontologybased Internet search model is designed. This model is proposed in the background of the Internet's mass, changeable resources.On the one hand, its semantic inference solves the problems, such as the low recall, which appears in the search; on the other hand, it makes good use of ontology resources to do query excitation that provides more userfriendly search and advances the efficiency of the ontological inference. This model is validated in the case of the stock domain ontology, and the results prove that the model has more flexible and effective semantic search characteristics for the complexstructured and typerich everchanging resources.

  An Adaptive Clustering Algorithm Based on Orbit Prediction in Ad Hoc Networks

  LIN Yao-Hua-, HU Hua-Ping

  2010, 32(2): 27-30. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 717 )   PDF (668KB) ( 988 )   　　

  This paper studies deeply the technology of mobility prediction, analyzes the shortcomings of the existing clustering algorithms, and presents an adaptive clustering algorithm based on orbit prediction in the mobile MANET. According to a node’s moving history, its future moving track can be predicted using the algorithm. Besides, the stability of the connections between adjacent nodes can be estimated, and then the network topology based on the connection stability can be constructed by the algorithm.Finally, all the nodes in MANET are divided into several clusters by the graph coloring technique. The algorithm maintains the stability for the clustering framework and effectively prevents the interruption of communications, thus compensates for the flaws of the previous clustering algorithms.

  A New LoadBalancing Routing Algorithm for P2P Streaming

  HU Ying-Song, CHEN Liang-Ban, TU Qing-Sheng

  2010, 32(2): 31-34. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 694 )   PDF (621KB) ( 478 )   　　

  An important problem in P2P streaming is building and maintaining a highbandwidth routing structure. To accommodate the current situation that nodes can frequently join and leave the overlay session, this paper presenta a load balancing algorithm based on the link available bandwidth. This algorithm utilizes the known knowledge of the physical topological structure, chooses the routing path that has the least effect on the available bandwidth among multiple routing paths, so as to obtain a lightlyloaded overlay edge. The algorithm can dynamically maintain highbandwidth multicast trees, balance the load of nodes and traffic on the links in the overlay session. The results of the simulation demonstrate that the algorithm can mitigate routing congestion, and achieve a load balancing effect in the dynamic environment.

  A LinkStabilityBased Clustering Algorithm for  Wireless Sensor Networks in the Group Mobile Model

  XU Hua, CHU E-Qiang, GUO Bin, HU Gui-Bing

  2010, 32(2): 35-37. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 691 )   PDF (642KB) ( 371 )   　　

  The clustering technology is one of the key technologies in wireless sensor networks (WSNs). The existing clustering algorithm for WSNs in the group mobile model takes insufficient consideration of the application of its group characteristics. In this article, a linkstabilitybased clustering algorithm of WSNs, referred to as LSBC, is put forward. This algorithm allows the predication of a link’s residual lifetime by using the age of the link and the predication result is adopted as a parameter to measure the link stability. The nodes at the two ends of the link with higher stability are classified as one cluster from which the cluster head is selected. This algorithm takes advantage of the nodes’ mobility effectively and the simulation results show that this algorithm, compared with other algorithms, can achieve a better performance in the group mobile model.

  Application of the Supporting Vector Machine in P2P Traffic Identification

  PAN Shan-Rong, FU Meng, SHI Chang-Qiong

  2010, 32(2): 38-40. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 830 )   PDF (430KB) ( 520 )   　　

  This paper puts forward a method of P2P traffic identification based on SVM. Using the doublevalue classification and the essential characteristics of SVM, we classify the data packages of the Internet into the P2P class and the nonP2P class; using the method's ability to classify multiple category problems, we further denote the P2P stream as a specified protocol.The experimental results show that the method has a high identification rate,which shows the effectiveness of using the SVM technology in the identification of the P2P traffic.

  Research on the QoS Guarantee Strategy Realization of Big Campus Networks

  YANG Xiang, LI Ke-Jian, HE Dian-Hua, GUO Jia, HUANG Zhen-Hua

  2010, 32(2): 41-44. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 726 )   PDF (806KB) ( 1049 )   　　

  This paper studies the QoS arichecture of the triditional IP networks.Through analyzing the basic principle and working mode of the two models: IntServ and DiffServ,the paper points out their merits and demerits as well as their network application environment, respectively. This paper presents a new model which combines the extended RSVP protocol and the DiffServ model to provide the QoS guarantee strategy of big campus networks. From the emulation test results, we can see that the application of the new model can promote the QoS characteristics of big campus networks effectively.

  Research on the Design of a Distance Screen Scout System Based on Java

  HOU Jin-Biao, GUO Chang-You

  2010, 32(2): 45-48. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 552 )   PDF (833KB) ( 512 )   　　

  At present, distance screen scout is widely used in practice and daily life .In order to satisfy people’s need to it, a distance screen scout system which adopts the advanced Java technology is studied and designed. This system has allsided functions and is compact and practical. It can get screen images, windows or course lists of the target computer,and can record and distill the distance keyboard events, open or close any directory of the target computer, realize the resource share and upload/download files, and capture the audio or videl signals, and so on. The system has a very good generalization and application value.

  Hierarchical Secure Communications in Computer Networks

  QIAO Tong-Xu, LI Guo-Huan, ZHANG Wen-Zheng

  2010, 32(2): 49-52. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 498 )   PDF (434KB) ( 365 )   　　

  TCG（trusted computing group） develops a series of technical specifications［1］，but it does not deal with the problem of hierarchical key assignment. This problem has been solved in reference ［2］.But in ［2］,the users of the same level use the same key and any user of the higher level uses the key of the user(of the lower level) in secure communications,which is undesirable in some cases. In this paper，we give two methods，by which the users can securely communicate with the users of the same level or the users of the lower level using different keys,and can not securely communicate with the users of the higher level.

  UEGS:A New Key Management Scheme for Distributed Wireless Sensor Networks

  CENG Ying-Zhi, SU Jin-Shu

  2010, 32(2): 53-57. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 771 )   PDF (890KB) ( 525 )   　　

  Key management has long been a challenging problem in distributed wireless sensor networks(DWSN) due to their high security requirements and strict resource constraints. Many schemes have been developed to establish keys in DWSN, including the famous EG scheme. Based on this classic scheme, this paper proposes a new key establishment and rekeying scheme: UEGS. The analysis in this paper demonstrates its feasibility, efficiency and security for key establishment and maintenance in DWSN.

  A Signature Scheme with Message Recovery  Based on Elliptic Curves

  HAN Yuan-Beng

  2010, 32(2): 58-59. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 614 )   PDF (298KB) ( 345 )   　　

  In accordance with the concept of digital signature and the elliptic curve cryptosystem, this paper puts forward a signature scheme with message recovery based on elliptic curves. Analysis shows that the scheme realizes digital signature, and its security equals finding the solution to a discrete logarithm problem. Moreover, it bears the function of identifying the sender and the recipient of the message. Compared with the scheme proposed by paper ［4］, the presented scheme has no inverse operation and less calculation of the point multiplication on the elliptic curve, thus simplifying the computing complexity and optimizing the performance.

  Design and Implementation of a MultiLevel InterDomain Routing Security Monitoring System

  MA Dun, CA Kai-Yu, CAO Hua-Yang, LIU Xin

  2010, 32(2): 60-62. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 608 )   PDF (840KB) ( 431 )   　　

  As we all know, the interdomain routing systems, based on BGP (Boarder Gateway Protocol, BGP), has become the core routing facilities of the Internet. However, due to their lack of the BGP security mechanism, BGP is vulnerable to all kinds of human configuration errors or malicious attacks. We develop a multilevel interdomain routing security monitoring system, and it can achieve four levels of domain routing, including the Internet, national network, a particular ISP and a specific route. This paper describes the system's composition structure, software structure, design, implementation technology and the achieved results.

  Research on the Email Monitoring System in Wireless LANs

  HU Xiao-Ya, CAO Lian-Jie

  2010, 32(2): 63-66. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 606 )   PDF (1065KB) ( 1029 )   　　

  To solve the security problems of IEEE802.11 wireless LANS, a email monitoring system is designed based on Microsoft Visual C++ 6.0. The system consists of three parts:email message listening, email message disposing, and email message reverting. This paper mainly studies how to deal with the captured email messages.A triplex filtrating method is proposed to ensure the integrality and veracity of the email original information, which includes protocol filtrating, port and sequence number filtrating, and time control rule. The system is well validated by test results.

  A Novel Strategy Based on MemoryRelevance Feedback for Image Retrieval

  DIAO Jian-Min, WEN Chun-Yong, SHU Shen-Zhong, XU Hui-Yang

  2010, 32(2): 67-71. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 822 )   PDF (1092KB) ( 460 )   　　

  Relevance feedback has been proposed as an important technique to boost the retrieval performance in contentbased image retrieval (CBIR).However,since there exists a semantic gap between lowlevel features and high level semantic concepts in CBIR, typical relevance feedback techniques need to perform mutiple iterations of feedback for achieving satisfactory results. These procedures are timeconsuming and may make the users bored in the retrieval tasks. Therefore, this paper proposes a novel feedback strategy based on analysising the usage and storage process of the retrieval feedback historical information in contentbased image retrieval. The testing results of the prototype system show that the strategy we put forward significantly improves the efficiency of retrieval performance and performs better than typical relevance feedback strategies.

  Research on the Detection and Identification Technology of Early Oil Fire Images

  CHEN Dun, DU Yang, WANG Dong, LV Hang

  2010, 32(2): 72-74. doi: 10.3969/j.issn.1007130X.2010.

  Abstract ( 543 )   PDF (453KB) ( 409 )   　　

  An algorithm of early oil fire image detection and recognition is put forward. The flame color, brightness and movement characteristics are chosen as the criteria. The early oil fire images are further detected and recognized by the algorithm of the Discrete Fractal Brownian Incremental Random Field model based on an analysis of the flame model and the differential model. The results of the simulated tunnel experiments show that the algorithm can successfully detect and recognize oil fire.