Computer Science

New Research Advances in Technologies of Semantic Web Search

YE Yu-xin,OUYANG Dan-tong

Computer Science. 2010, 37 (1): 1-5.

Abstract

PDF(460KB) ( 710 )

RelatedCitation | Metrics

Technologies of semantic Web search are very novel and young. There are many theories and approaches contributed to semantic Web search, such as ontologies, information retrieval, natural language processing, and so on. Situadons of semantic Web and semantic Web search were introduced dividually. hhen, the general architecture of semantic Web search engine was concluded through analyzing research at home and abroad in detail. This architecture is composed of four models, including information extraction models, intelligent interface models, information query models,and user query models. The functions, technologies, and evaluation systems of models which arc different components of this general architecture were introduced in further. At last, the conclusion about our work was given. At the same time,the future challenge of semantic Web search technologies was discussed.

Review of Statistical Shape Prior-based Level Set Image Segmentation

DONG Jian-yuan,HAO Chong-yang

Computer Science. 2010, 37 (1): 6-9.

Abstract

PDF(354KB) ( 777 )

RelatedCitation | Metrics

Abstract Image segmentation problem often demands the incorporation of as much prior information as possible to help the segmentation algorithms extract the tissue of interest The model of image segmentation based on statistical shape prior level set was reviewed. The feature of mode is the energy function of the model composed by two terms. The first one is data term based on the image gradient or region gray intensity, the second one is the shape prior term which provides robustness against missing shape information due to cluttering,occlusion and gaps. How to construct the implicit shape model which aims to extract a compact representation for the structure of interest from a set of training examples, how to construct the evolve model to constrain an implicit surface to follow global shape consistence while prescrving its ability to capture local deformation were discussed intensively. The key problems such as shape registration and the correspondence problem were introduced. Finally the open issues and possible future research directions were pointed.

Survey on Routing Algorithms of Mobile Ad Hoc Networks

ZHANG Peng,CUI Yong

Computer Science. 2010, 37 (1): 10-22.

Abstract

PDF(1295KB) ( 999 )

RelatedCitation | Metrics

Mobile ad hoc networks are widely used and it is an important task to route efficiently. But traditional routing algorithms are not fit for mobile ad hoc networks due to their dynamic character and many routing algorithms of mobile ad hoc networks were proposed these years. Main classification standards for routing mechanisms were presented.Then basic routing protocols and routing algorithms for mobile ad hoc etwork were overviewed. After that, as the latest advance of routing algorithms, the prediction-based models, energy models, position-based models, QoS and security in routing algorithms were deeply researched. In prediction-based routing algorithms, using historical information, proper routes were rapidly chosen based on the prediction of nodes' motion. In energy-aware routing algorithms, the consumplion of energy was modeled based on the nodes' information. Then we introduced the special applications of the routing algorithms,including position-based routing algorithms,QoS routing algorithms and security routing algorithms. In position-based routing algorithms,forwarding decisions were made based on the geographical position of a packet's destination. hhe main idea of QoS routing algorithms was to take bandwidth, delay, energy and battery's life cycle into account while routing. In security routing algorithms, messages were encrypted based on secret keys, hash chains and electronic signatures. Last,more than thirty routing algorithms were compared and researched in the fields of computation overhead, communication overhead, storage overhead, optimal path, critical node and the types of the routing algorithms.The characters and applied areas of these routing algorithms were analyzed and the research trends in this field were proposed.

Present Research and Developing Trends on Industrial Control Network

HU Yi,YU Dong,LIU Ming-lie

Computer Science. 2010, 37 (1): 23-27.

Abstract

PDF(520KB) ( 2301 )

RelatedCitation | Metrics

Industrial control network is needed for industrial communication and advanced manufacture. An overview on the evolution process was given, and the real-time technology of industrial ethernet was summarized. I3y analyzing the real-time behavior of industrial ethernet, this paper divided it into soft real-time, hard real-time and isochronous realtime classes,and expatiated the realtime research methods of each class in detail. The architecture of real-time heterogeneous networks was described, and the researches of wireless industrial network were proposed. Finally, the prospects and the further research direction were pointed out.

Research on the Realization of Anti-collusion Fingerprinting

ZHOU Guo-rui,SUN Shi-xin,WANG Wen-jiang

Computer Science. 2010, 37 (1): 28-33.

Abstract

PDF(542KB) ( 494 )

RelatedCitation | Metrics

Considerable progress has been made in fingerprinting technology, particularly with regard to fingerprinting codes,embedding methods and colluder identification. But applying fingerprinting technology to engineering,it faces the problem on how to realize anti-collusion fingerprinting in real system. The concept of anti-collusion fingerprinting realination was introduced firstly, and then effects of three factors on it were studied. hhese factors arc anti-fingerprinting codes, content delivery system and user's multimedia consumption pattern. The tradeoff strategies between these factors and the problem were proposed. Some major anti-collusion fingerprinting realization algorithms were reviewed.hhcir performance was analyzed from following aspects: communication price, hardware and software costs which inelude computing and storage price, safety and real-time characteristics, etc. Finally, the future trend of the problem was discussed.

Research on System Software of Trusted Smart Space Based on SOA

MING Liang,KUANG Xiao-hui,HUANG Min-huan,JIN Qi

Computer Science. 2010, 37 (1): 34-38.

Abstract

PDF(485KB) ( 471 )

RelatedCitation | Metrics

A new architecture of trusted smart space based on SOA (Service Oriented Architecture) was proposed for the security requirement of pervasive computing and implicit interaction. This new architecture can provide scalable information services by using SOA,and support trusted and invisible interaction by adopting a mix-access control based on actor, policy, context, and turst/risk evaluation. Finally, a prototype application was built upon Web service platform,which validated the architecture successfullly.

k-DmeansWM:An Effective Distributed Clustering Algorithm Based on P2P

LI Liu,PANG Jiu-yang,GE Bin,XIAO Wei-dong,TANG Da-quan

Computer Science. 2010, 37 (1): 39-41.

Abstract

PDF(239KB) ( 496 )

RelatedCitation | Metrics

Centralized master node is established in normal distributed clustering algorithms in order to control clustering processes, which reduce the system's reliability, single point failure problem occurs easily. The effective distributed clustering algorithm k-Dmeans Without Master(k-DmeansWM) based on peer-to-peer concepts, which abandons the master node,uses peer nodes to achieve controlling of the clustering process entirely.Both theoretical analysis and experimental results show that k-I?meansWM not only ensures the accuracy and efficiency of cluster cases, but also greatly improves the reliability and scalability of the system.

Research on an Artificial Immune System-based Storage Anomaly Detection

HUANG Jian-thong,PEI Can-hao,XIE Chang-sheng,CHEN Yun-Bang,FANG Yun-f u

Computer Science. 2010, 37 (1): 42-46.

Abstract

PDF(445KB) ( 453 )

RelatedCitation | Metrics

Most authentication sub-systems can not guarantee the authenticity of the account, and an intruder using a stolen account may be regarded as a legitimate user. In order to filter out such illegal users, the storage system should be able to watch for the user access activities. In order to enhance the storage security, the paper proposed an immune anomaly detection scheme to identify the anomalous access behavior. When an access rectuest violates the access control rule,it is viewed as Non-self,so as to provide some storage early warning tips to the storage security subsystem. The proposed storage anomaly detection system (SADS) targets the anomaly detection at storage level and focuses on the read/write data requests, constructing two-layer detection together with the network intrusion detection system (KIDS). The simulation results show the proposed scheme can reach rather high detection rate and low false alarm rate,validating its feasibility. The overhead test exhibits that the computation time caused by SADS is acceptable, e. g below 11. 6% as to 3MB data.

Universally Composable Secure Roaming Authentication Protocol for Interworking Networks

LI Ya-hui,MA Jian-feng

Computer Science. 2010, 37 (1): 47-50.

Abstract

PDF(431KB) ( 623 )

RelatedCitation | Metrics

Security problems in interworking networks have attracted much attention of researchers recently. The typical interworking networks can be constructed with the core network WCDMA and the accessing networks WiMAX and WLAN. Considering security and efficiency issues,a roaming authentication protocol based on symmetric cryptographic algorithms was proposed for that network. It verified the identity of wireless access networks to prevent the redirection attack, and used the localized authentication mechanism to reduce the message transmission delay and improve the efficiency of rcauthentication process. The efficiency of proposed protocol was simulated in NS2. By analyzing the security of the proposed protocol with universally composable(UC)modcl, it proves that the proposed protocol satisfies the definition of UC security defined in the UC model.

Research on Dynamic Role-based Access Control Based on Trust Mechanism in Grid Environment

DENG Yong,ZHANG Lin,WANG Ru-chuan,ZHANG Mei

Computer Science. 2010, 37 (1): 51-54.

Abstract

PDF(421KB) ( 442 )

RelatedCitation | Metrics

In grid computing system, the trust relation between resource vendor and resource user is the premise of safe communication, for the traits on distribution and dynamic, the trust between each other in advance is impossible, however, it can be resolved by importing trust mechanism to RBAC in Grid community, the trust degree of each Grid entity can be counted based on the feedback from other entities,and then their roles are changed based on their trust degrees.Using this policy, the dynamic trait can be realized in Grid and the cheat behavior of Grid entity also can be avoided.

Research on Architecture of Grid Computing System

ZHANG Huan-jiong,ZHONG Yi-xin

Computer Science. 2010, 37 (1): 55-56.

Abstract

PDF(296KB) ( 513 )

RelatedCitation | Metrics

Grid computation is a branch of computer science researching, the main idea is how to integrate some different computation capabilities in different geographical positions,furtherrnore to unify them as a virtual super computer.We researched the architecture and performances of Grid Internet, especially, the characters of the new protocols and the sharing of resources were studied, the distributing computation and parallel computation, the co-operating performance and the intelligent performance were discussed too.

Predictive Control and Proportional Delay Guarantee in Web QoS

GAO Ang,MU De-jun,HU Yan-su,PAN Wen-ping

Computer Science. 2010, 37 (1): 57-59.

Abstract

PDF(238KB) ( 491 )

RelatedCitation | Metrics

Control theory is being used for QoS in Web server, but when workload varies significantly, the real-time performance of the feedback control needs to be improved. After analysing the heavy-tailed property of the number and size of embedded URL in the required pages,based on the ctueuing feature of the Web server and the HTTP 1. 1 persistent connection,a predictive approach was desired to achieve the relative delay guarantee by calculating the future delay of the different priorities and adjusting their ctuota of service threads. The experimental results demonstrate that the pro- posed approach achieves a better dynamic performance.

Frequency Assignment in Asynchronous FH Networks with Cosite Constraints

YUAN Xiao-gang,HUANG Guo-ce,NIU Hong-bo,GUO Xing-yang

Computer Science. 2010, 37 (1): 60-63.

Abstract

PDF(355KB) ( 485 )

RelatedCitation | Metrics

The cosite interference is aggravated when several frequency-hopping (FH) radios arc located on the same platform, and the small hop-set used in this condition results in low FH gain. The constraints in the in asynchronous FH networks with cosite restriction include constraints within subnets, constraints among subnets and no interference of the synchronization constraints,and cost function is based on the above analysis. The application of the simulated annealing (SA) algorithm was improved by the characteristic of the violations of the constraints. The method of frequency assignment problems (FAPs) in the asynchronous FH networks with cosite constraints was presented and the computing resups indicate this method can realize the FH communications in the full-band with given minimum gap. This approach gave a method to actualize the advantages of FH communications.

Composite P2P Storage System Based on Group Management

YANG Lei,HUANG Hao,LI Ren-fa,LI Ken-li

Computer Science. 2010, 37 (1): 64-67.

Abstract

PDF(373KB) ( 571 )

RelatedCitation | Metrics

By using the way of P2P, a P2P storage system was built. The nodes involved in the system were grouped by the predicted network distance, and then an overlay network which is maintained by super nodes was formed. By using the overlay network topological structure mechanism of maintenance, DHT data storage mechanism, data replication mechanism based on different frequency of data access and data restoration mechanism, the reliability of the system and the efficiency of data storage in system were promoted. The performance of the storage system was testified on the basis of the results of simulation experiment.

Security Routing Algorithm Based on Shortest Delay for Cognitive Radio Networks

XUE Nan,ZHOU Xian-wei,LIN Lin,ZHOU Jian

Computer Science. 2010, 37 (1): 68-71.

Abstract

PDF(337KB) ( 447 )

RelatedCitation | Metrics

The paper proposed a cross-layer routing selection algorithm, which considers the shortest delay and highest security as routing selection standard. The routing algorithm aims to solve the problem of both the dynamic spectrum and security problem in cognitive radio networks.The Algorithm reduces the data transfer delay by optimizing the number of forwarding nodes and selecting channel for transmitting and receiving, and ensures routing security by choosing a forwarding node of high level security. The results of theory analysis show that the algorithm is efficient and feasible.The computing complexity of the algorithm is O(N2)，where N is the number of nodes in the network topology graph.

Design and Implementation of a Standard Identity-based Authentication Protocol on Wireless Sensor Network

WANG Wei-hong,CUI Yi-ling,CHEN Tie-ming

Computer Science. 2010, 37 (1): 72-74.

Abstract

PDF(344KB) ( 450 )

RelatedCitation | Metrics

Identity cryptography has been widely used in Security Authentication. Bilinear pairing is an important tool for constructing the cryptography developed in recent years. Based on the research of the bilinear pairing, this paper proposed a standard identity-based authentication protocol. This protocol is secure against impersonation under passive attack and has high efficiency. This paper also implemented the authentication protocol in the platform of TinyOs. Finally,the simulation results show that the authentication protocol is effective in the wireless sensor network.

Research of Multi-level Security Policy Model Based on Information Flow

WANG Hui,JIA Zong-pu,SHEN Zi-hao,LU Bi-bo

Computer Science. 2010, 37 (1): 75-78.

Abstract

PDF(343KB) ( 680 )

RelatedCitation | Metrics

Insider threat is widely recognized as an utmost important issue for organization security management As the most important information asset (documents)，they are the chief target of insider misuse. The former coarse grained security policies that operate on "the principle of least privilege" or "separate of duty" are not enough to address documenu security about insider threat issue. We presented a novel multi-level security policy model and related algorithms,and defined the concept of document information flow and information flow graph. According to system context’s change,it will generate dynamic restriction conditions about information flow. And its aim is to prohibit these probable hiding channels of information flow.

Study on Intrusion Intention Recognition Based on the Probabilistic Inference

PENG Wu,YAO Shu-ping

Computer Science. 2010, 37 (1): 79-82.

Abstract

PDF(444KB) ( 575 )

RelatedCitation | Metrics

Intrusive behaviors of attackers always imply their objects and intentions, hereby a hierarchical model of intrusion intention recognition was proposed. In order to handle the uncertain message in the circumstance of computer networks, intrusion intention recognition algorithm based on probabilistic inference was presented. On this basis the future plans and goals of the attackers can be predicted in order to take actions in advance. Bayesian networks were constructed to cope with the concurrent intention recognition problems according to the relationship of network security events, goals and intentions. Feasibility and validity of this method were proved from the experiments.

Research on Handoff Strategy Based on Traffic QoS Guaranteed for Heterogeneous Hierarchical Wireless Networks

NING Guo-qin,ZHANG Jing,LIU Gan,ZHU Guang-xi

Computer Science. 2010, 37 (1): 83-86.

Abstract

PDF(327KB) ( 441 )

RelatedCitation | Metrics

A handoff strategy was proposed to guarantee the traffic QoS in heterogeneous hierarchical wireless networks. Based on the bidirectional call-overflow between different networks, guard channels were set for real-time traffic handoffs and buffer queues were set for non-real-time traffic handoffs. In order to further decrease the dropping probability of real-time traffic handoffs, preemption scheme was also used, which principle is that real-time traffic handoffs can preempt the channels used by the ongoing non-real-time traffic. Simulation results show that the proposed handoff control scheme can evidently decrease the dropping probability of each type of traffic, the channel utilization of overall heterogeneous system is slightly increased simultaneous.

Role Based Integrity Control Model for Operating System

LIU Zun,WANG Tao,LI Wei-hua

Computer Science. 2010, 37 (1): 87-90.

Abstract

PDF(416KB) ( 440 )

RelatedCitation | Metrics

Integrity is one of the security objectives of operating system. This paper analysed the integrity protection policies of security operating system, and elaborated user responsibility and process reliability arc important factors in determining the integrity level of the users,the processes and the files in operating system. Then we proposed an integrity control model with detailed description of its integrity axiom, model elements, system states, system configurations and state transition rules. The model employs roles to simplify user integrity management. The correctness of given model was proven. The application case and implementation method were given at the end of the paper.

Watermarking Protocol for Piracy Tracing

HU Yu-ping,ZHANG Jun

Computer Science. 2010, 37 (1): 91-94.

Abstract

PDF(352KB) ( 574 )

RelatedCitation | Metrics

A novel buyer-seller watermarking protocol was designed for piracy tracing. In the proposed protocol, a memoryless Watermark Certification Authority (WCA) was introduced, that can produce many watermarks for a buyer at a time and is not required to participate in each transaction of digital content between the buyer and seller. In order to guarantee the anonymity of the buyer, the WCA also can encrypt the digital certificate of buyer and hand over the encrypted message to seller via the buyer. In addition, the proposed protocol also can resolve other problems, such as the customer right problem, the binding attack problem, the digital conspiracy problem, the dispute problem. The analysis indicate that the proposed protocol is secure and practical.

Effective Trust-based Topology Evolution Mechanism for P2P Networks

HU Jian-li,WU Quan-yuan,ZHOU Bin

Computer Science. 2010, 37 (1): 95-98.

Abstract

PDF(412KB) ( 412 )

RelatedCitation | Metrics

Current unstructured peer-to-peer (P2P) systems lack fair topology structures, and take no consideration for malicious behaviors of peers. The main reason is that the topology is not sensitive to peer's trust, and cannot accommodate heterogeneity of peers over the network. hhus, a feedback credibility based global trust model was presented.Then, based on the trust model, an adaptive topology evolution mechanism for unstructured P2P networks was proposed. Through this mechanism, trusted peers can migrate to the centric position, while untrusted peers to the edge of the topology, guaranteeing fairness during topology evolution. On the other hand, the mechanism can effectively counter the malicious behaviors of peers, and also has the incentive functionality, which incents peers to provide more high-quality services in order to get more return on services. Analysis and simulations show that, compared with the current topologies, the resulting topology mechanism is more effective and robust in combating the selfish or malicious behaviors of peers.

Tag Mesh Networking to Resolve Shadow Area in the RFID Communication

GOU Hao-song,GOU Jing-song

Computer Science. 2010, 37 (1): 99-102.

Abstract

PDF(367KB) ( 404 )

RelatedCitation | Metrics

Wireless Sensor Network ( WSN) has been considered as a promising method for reliably monitoring both civil and military environments under hazardous and dangerous conditions. Research in WSNs performance improvement is very important burgeoning field in our daily life. This paper was formed after studying the problem‘Shadow area in the RFID Communication' in the port which the sensors are widely deployed in. The problem exists in the communicalion of Reader to Tag. It shows us the out of range communication area of Reader to Tag communication. Through the proposed solution,we can not only resolve shadow area in the RFID communication, but also can save the cost of the network infrastructure comparing with the common method which just adds the Reader.The simulation results and analysis show that our solution could achieve a perfect performance of WSN in terms of the shadow area in the RFID communication.

CAER:A Content-based Adaptive Event Routing Algorithm

ZHENG Li-ming,WANG Yi-jie,ZHENG Zhong,LI Xiao-yong,TAO Ke

Computer Science. 2010, 37 (1): 103-107.

Abstract

PDF(405KB) ( 467 )

RelatedCitation | Metrics

The content based publish/subscribe paradigm has received emerging attention due to providing efficient loosely communication channels for large-scale distributed systems. Event publishing nodes don't specify the destination addresses of events;in contrast,events arc progressively routed to the nodes which have subscribed them However,existing content based event routing algorithms can't efficiently adapt to the dynamics of subscriptions and incur higher network communication costs. A content based adaptive event routing algorithm (CAER) was proposed. The subscription was bound with the corresponding subscribers' addresses in the routing tables, and then the routing tables were efficiently constructed and maintained. Simulation results show that the accuracy of the event publishing is improved, and the network communication costs arc significantly reduced.

Localization Problem of Faulty Links Based on Simple Network Tomography

ZHAO Zuo,CAI Wan-dong

Computer Science. 2010, 37 (1): 108-110.

Abstract

PDF(341KB) ( 529 )

RelatedCitation | Metrics

Faulty links are a typical network fault symptom in wireless sensor network. It affected the operation of network and the quality of service seriously, therefore faulty links need to be detected and repaired. Localization technique of faulty links based on simple network tomography was introduced. This paper presented a binary disjunctive model for describing link faulty states; the localization problem of faulty links is referred to as the Maximum A-posteriori Probability problem. By making the Localization problem of faulty links mapped to weighting minimal set-cover problem, the algorithm based on heuristic strategy was proposed. The performance of inference algorithm was evaluated by simulation, and the simulation results indicated the feasibility and efficiency of the method.

Introduction for Heterogeneous Cooperative Communication System and DMT Analysis Based on Distributed Space-time Coding

ZHANG Peng,WANG Fu-rong

Computer Science. 2010, 37 (1): 111-113.

Abstract

PDF(337KB) ( 619 )

RelatedCitation | Metrics

Cooperative communication has been recently proposed as a way to provide dramatic gains in fading wireless environments. Most literatures about cooperative communication assume each node arc equipped with only one interface or several homogeneous interfaces. To date,mobile terminals with heterogeneous interfaces become more popular,which motivate us to study cooperative communication in heterogeneous cellular networks. We introduced a VAA_DSTC scheme, which is suitable to heterogeneous cooperative communication system, and demonstrated that the data rate difference between heterogeneous interfaces can bring extra spatial multiplexing gain,in addition to diversity gain.Moreover,we characterized the diversity-multiplexing tradeoff of VAA_DSTC scheme and compared it with other schemes.

MFALM:An Active Localization Method for Dynamic Underwater Wireless Sensor Networks

YANG Kui-wu,GUO Yuan-bo,WEI Da-wei,JIN Yong-gang

Computer Science. 2010, 37 (1): 114-117.

Abstract

PDF(330KB) ( 428 )

RelatedCitation | Metrics

We studied several fundamental key issues of localization in underwater sensor networks(UWSNs) , then proposed an multi-frequency active localization method ( MFALM) based on the time difference on arrival (TDOA) for UWSNs. The method, which is event driven, divides all the nodes into three groups: buoy nodes, relay nodes, and ordinary nodes. To guarantee the low-power and effectiveness of communication, the nodes communicate with each other by two types of frequencies ; low frequency and high frequency, the former is used in long distance communication, while the latter is used in short distance. In addition, package formats for communication between nodes were defined, which make the packages short and can reduce the costs of sensor nodes in localization. Analysis shows that our method, which has the characteristics of event driven, reduced packages and multi-frequency, has low communication cost and can extend the network lifetime.

Novel Security Protocol and Strand Space Analysis

PI Jian-yong,YANG Lei,LIU Xin-song,LI Ze-ping

Computer Science. 2010, 37 (1): 118-121.

Abstract

PDF(382KB) ( 410 )

RelatedCitation | Metrics

We proposed a novel PJY identity authentication and key agreement security protocol based on computational difficulty of discrete logarithm in finite field. The PJY security protocol can validate identity of the both parties with two session, and generate peer session key. We analyzed the correctness of the security protocol on strand space model. First we constructed infiltrate strand space, and proved the PJY security protocol has injective agreement and secrecy by authentication tests under arbitrary penetration strand, so proved the PJY security protocol is correct.

Distributed Energy Balancing Routing Algorithm in Wireless Sensor Networks

LIU Xiang-wen,XUE Feng,LI Yan,YU Hong-yi,HU Han-ying

Computer Science. 2010, 37 (1): 122-125.

Abstract

PDF(362KB) ( 415 )

RelatedCitation | Metrics

Aiming at the energy balancing issue in wireless sensor network(WSN)，based on a local energy balancing metric,predicting-based distributed energy balancing routing (PDEBR) algorithm was proposed. PDEBR gets to energy efficiency with geographic location information and power control. PDEBR was complemented with distributed style.The simulation results show PDEBR can prolong network lifetime efficiently.

Method of Network Security Situation Awareness Based on Artificial Immunity System

LIU Nian,LIU Sun-jun,LIU Yong,ZHAO Hui

Computer Science. 2010, 37 (1): 126-129.

Abstract

PDF(481KB) ( 487 )

RelatedCitation | Metrics

As a positive defense technology, Network Security Situational Awareness has become the orientation of research in the field of network security. Based on the analysis of the papers from domestic and foreign on technologies for network security situational awareness, a method of network security situational awareness based on the profound research of AIS was designed and built. I}he method uses network intrusion detection based on the theory of immunity as the base of situational awareness, to detect known and unknown intrusions with the help of biological technology. According to correspondence relations of density change of antibody in the artificial immune systems and pathogen invasion intensity, a novel network security situational evaluation model was also established. In the tendency prediction for network security situational,this paper used Urey Markov Model to make quantitative prediction. Experiment results show that this model is also helpful to resemble network security tragedy effectively, therefore, it is a better solution for network security initiatives defense.

Research on Dynamic Grid Workflow Model of ITS

YANG Hao-lan,LI Hua,LI Shi-chang,GAO Min,WU Zhong-fu

Computer Science. 2010, 37 (1): 130-132.

Abstract

PDF(330KB) ( 429 )

RelatedCitation | Metrics

Aiming at the problem that it is hard to scheduling vehicles in ITS, the characteristic and requirement of the workflow model for real-time vehicles scheduling in Urid environments which cross lots of domains and systems was presented. A meta-model of workflow was proposed for ITS vehicles scheduling. It utilizes Petri Net based flexible workflow model technique and top down layered decomposition approach to create the workflow model. And then a dynamic modifying strategy was proposed in order to suit the ideas and methods of ITS workflow modeling. At last, an algorithm was given to validate the model.

Implementation of Anycast Service in Ad Hoc Network Based on Mobile IPv6

WANG Xiao-nan,XU Jiang,YIN Xu-dong

Computer Science. 2010, 37 (1): 133-137.

Abstract

PDF(458KB) ( 461 )

RelatedCitation | Metrics

At the present time, the issue of how to perform Anycast service in Ad Hoc network has attracted more attendon. This paper proposed an Anycast communication model in Ad Hoc network which has been implemented in the simulation network. This model creates a new kind of IPv6 address autoonfiguration scheme which does not need detection of address duplication. In addition, in this model when an Anycast mobile member switches between different sulrnets it does not need to register its carcof address and to build a tunnel to transfer data. This paper deeply discussed and analyzed the model and the experimental data proved its validity and efficiency.

Application of Network Coding in Real-time Tactical Data Multicast

ZHANG Li-dong,QIN Guang-cheng,LI Zhen,YIN Hao

Computer Science. 2010, 37 (1): 138-141.

Abstract

PDF(370KB) ( 427 )

RelatedCitation | Metrics

Network coding technique is used to the multicast of tactical scene, which is abstracted in this paper. And based on analyzing the characteristic of this network with central node and many-to-all traffic, a transmission policy based on coding only upon the central node and corresponding greedy algorithm were proposed. The up-bound of theoretic gain for the transmission policy based on network coding was analyzed, and simulation results show that the gain of greedy algorithm is close to the optimum. Finally, the analyzing and simulation results provide important insights that the proposed policy based on network coding has better real-time performance than the classical policy in the multicast application.

Research on Deployment of Locating Unit in Ubiquitous Computing

LI Hai-cheng,ZHANG Rui,ZHOU Yan

Computer Science. 2010, 37 (1): 142-145.

Abstract

PDF(298KB) ( 434 )

RelatedCitation | Metrics

Vetting location information of an unknown mobile node is a key point of locating service in the ubiquitous computing environment. The errors due to locating an triangular area was studied in the 2-D space as a pervasive computing environment. And the theorem of a locating unit deployment(LUD) was presented. Based on the finding,a new physical property parameter of coverage, coverage base, was defined and introduced for the first time. And, the deployment issue between locating units was also studied, as a result, the theorem of coverage base about locating units was proposed in minimum error condition. hhe simulation tests indicate that the locating unit deployment theorem can meet the requirement of ubiquitous terminal's real-time localization and possesses the preferable localization precision.

Aspect-oriented Architecture Description Language Based on Temporal Logic

NI You-cong,YING Shi,ZHANG Lin-lin,WEN Jing,YE Peng

Computer Science. 2010, 37 (1): 146-152.

Abstract

PDF(641KB) ( 459 )

RelatedCitation | Metrics

There always exist some crosscutting behaviors and features in software architecture design based on traditional architectural description language, which tangle and scatter in different design elements of software architecture,leading to the difficulties in comprehension, evolution and reusability of software architectural design decisions. Aiming to these problems, under a unified temporal logic framework, an Aspect Oriented Architectural Description Language (AC2-ADL)was proposed based on the temporal logic language XYZ/E. The framework and syntax of AC2-ADL were presented and its semantics was explained using XYZ/E. Furthermore, how AC2-ADL can be used for specifying aspect oriented software architectures was illustrated through case study.

Memory Corruption Detection Based on Dynamic Taint Analysis and Access Control

WANG Lei,MAO Bin,XIE Li

Computer Science. 2010, 37 (1): 153-157.

Abstract

PDF(426KB) ( 499 )

RelatedCitation | Metrics

Memory corruption attacks account for most parts of malicious attacks toward software security. Recently dynamic taint analysis was proposed and was gaining momentum. This proposed technique attempts to defeat attacks by checking the taintedness and integrity of pointers when accessing memory. Unfortunately, there exists some class of attacks without tainting pointers, such as array bounds violation attacks using pointers. We proposed a novel approach to defeat this kind of undetected attacks using taint based tracking analysis. Our notion is based on the memory access control,dmt is, first,we will check the taintedness of the pointers when accessing memory like existing taint based approaches, second, we will check whether or not the memory area pointed by the pointer is in the legitimate range of the accessing pointer. Our implementation does not need source code and is based on Valgrind, hence works on commodity software. To demonstrate our idea,we performed a preliminary empirical experiments, the results arc quite promising;our system can effectively detect a wide range of attacks, and the average runtime overhead is close to Memcheck, a widely used memory error detector.

Rav-tree:An Efficient Index Structure for Reverse Approximate Nearest Neighbor Query

LI Bo-han,HAO Zhong-xiao

Computer Science. 2010, 37 (1): 158-162.

Abstract

PDF(418KB) ( 442 )

RelatedCitation | Metrics

Index structure is the precondition and foundation in the efficient data query. The reverse approximate nearest neighbor query is a new issue in the area of spatial query. This approach can avoid much metric distance computation in exact query, and acquire a better tradeoff between the efficiency and precision. The Rav-tree is different from the index structures based on the heuristic rules. It applies partial Voronoi cell approximation with estimated circles to filter the results of approximate nearest neighbor query. The final result set of RANK is reached through the algorithms of ANN query and Division ctuery with primary candidates. The experimental results indicate that the Rav-tree is an effective index structure and has better efficiency and recall for the query such as RANN query.

Copyright Protection System for Outsourced Database Based on Information Hiding

ZHU Qin,LU Zhi-Ming

Computer Science. 2010, 37 (1): 163-166.

Abstract

PDF(334KB) ( 597 )

RelatedCitation | Metrics

In outsourced database scheme, since the database server, which is provided by the third part, is not in the trust domain, the data files may be pirated, and the data contents may be tampered with. A copyright protection system for outsourced database using information hiding technique was built, which was based on algorithms of database watermarking, as well as their corresponding protocols, and was combined with technologies of PKI, digital signature and USB Key. Compared with traditional solutions which of data encryption or digital signature, the copyright protection system for outsourced database based on information-hiding possesses advantages of less redundant capacity either for storage or for communication, better imperceptibility, and validation information harder to be removed.

Semantic Solution to Conflict of Information Integration

QU Zhen-xin,TANG Sheng-qun

Computer Science. 2010, 37 (1): 167-169.

Abstract

PDF(258KB) ( 636 )

RelatedCitation | Metrics

When heterogeneous information systems were integrated, a great amount of conflictions appeared. After tidying up all kinds of cases,five kinds of conflicts were found out, which were analyzed in detail. Based on the work, a full solution applying semantic technology was proposed,which can eliminate conflicts at all levels.

Research on Call Semantic of Datalog Logic Program and its Application

ZHONG Yong,GUO Wei-gang,ZHONG Chang-le,LIU Feng-yu,LI Ning

Computer Science. 2010, 37 (1): 170-175.

Abstract

PDF(480KB) ( 665 )

RelatedCitation | Metrics

The paper firstly gave the call semantic and call predicates of datalog logic programs, and discussed the operational and fixpoint semantics of U-datalog programs with call predicates. Then the algorithms that evaluate U-Datalog programs with call predicates under definite stratified calls were presented. Finally, applications of the call semantic of datalog programs in digital rights management were explained and demonstrated.

Formally Analyzing Security Protocol Using Approximation of Tree Language

LIU Nan,ZHU Wen-ye,ZHU Yue-fei,CHEN Chen

Computer Science. 2010, 37 (1): 176-180.

Abstract

PDF(429KB) ( 457 )

RelatedCitation | Metrics

Formal method and automated tools arc both necessary and efficient for analyzing practical security protocols. Theorem proving theory has been studied intensively because it can verify infinite-state system. But there are still limitations when verifying large scale system automatically. A formal model for security protocols was proposed based on theorem proving and rewriting approximation theory. It utilizes term rewriting system and tree automata to model protocol and the knowledge of intruder. An automatic generated approximation algorithm was designed to calculate the fix-point tree automata. In this model, secrecy and authentication were discussed. By a case of study, NSPK was analyzed effectively.

News Video Automatic Retrieval Method Based on Multiple Keywords

ZHOU Sheng,HU Xiao-feng,LUO Pi

Computer Science. 2010, 37 (1): 181-183.

Abstract

PDF(327KB) ( 430 )

RelatedCitation | Metrics

Aiming at the problem of manual annotation in TBVR and the immatureness of CBVR and in-depth analysis on the requirements and features of video retrieval in virtual television news, a news video retrieval method based on multiple keywords was proposed. Annotation dictionary, tree annotation structure, automatic keywords extraction, similarity computation model and automatic retrieval algorithm were discussed in detail. An experiment was made on this method. The result shows that it has attained a very high recall rate with a comparative high accuracy rate and it can solve the problem of automatic video retrieval in virtual television news.

Nearest-neighbor Query Algorithm Based on Grid Partition of Space-filling Curve

XU Hong-bo,HAO Zhong-xiao

Computer Science. 2010, 37 (1): 184-188.

Abstract

PDF(367KB) ( 1842 )

RelatedCitation | Metrics

As the overlap between minimum bounding rectangles in the directory of R-tree is increasing very rapidly with growing number of the data, the performance of the nearest neighbor query algorithm based on R-tree deteriorates rapidly. To avoid the problem, the paper presented a nearest-neighbor query algorithm based on grid partition of space filling curve. Space-filling curve has the properties of dimension reduction and data clustering. Using space-filling curve,the algorithm divides 2D space into equal-size grids, and map points in grids into linear space. It only visits points in the grid which query point lies in and points in near grids of ctuery point. The paper implemented nearest neighbor query algorithms based on Hilbert curve, Z curve and Gray curve, and compared the difference of mapping algorithm and data clustering between three curves. Experimental results indicate that the algorithm is better than brutcforce algorithm and near-neighbor cauery algorithm based on R-tree.

Research on Role-based Access Control with Spatial Character

ZOU Zhi-wen,CHEN Chang-qian,JU Shi-guang

Computer Science. 2010, 37 (1): 189-191.

Abstract

PDF(317KB) ( 537 )

RelatedCitation | Metrics

In order to strengthen the capability of safety expression for RBAC with spatial character, the secure spatial DBMS-SECVISTA was taken as platform to research spatial character of the Spatial Rolcl3ased Access Control (SRBAC) model. First, the region coverage constraint of spatial object, duration constraint of spatial object, various spatial object separation of duty constraints and spatial object cardinality constraint of role activation were researched;after extending the traditional session, a non-blocked sessions set was confirmed at any spatial location, then the role hierarchy was discussed in spatial environment. So the theory of secure system.DBMS was optimized and afforded to build the stricter

Research on Normalization of XML Schema with XML Strong MVD

YIN Li-feng,HAO Zhong-xiao

Computer Science. 2010, 37 (1): 192-196.

Abstract

PDF(390KB) ( 392 )

RelatedCitation | Metrics

For solving data redundancy and abnormal manipulation for XML documents in schema design under incomplete information circumstances, the normalization of XML Schema existing XML strong multivalucd dependencies under incomplete information circumstances was researched. hhe concepts of XML Schema and incomplete XML document tree according with XML Schema were formalized. Based on the equivalence of the sub-trees' information and the consistent of the sulrtrecs' information, XML strong multivalued dependencies' definition was given. Weak keys and XML strong multivalued dependencies weak normal form were given. The reasons of data redundancies aroused by XML Strong multivalued dependencies in XML Schema via the instance were analyzed, the transforming rules and the arithmetic of normalization were presented. The production in this work can deal with data redundancies in XML documents coming forth a great deal incomplete information and achieve better XML Schema design under incomplete information cmcumstances.

Dual Description Logic for the Relations in Databases

MA Dong-yuan,SUI Yue-fei

Computer Science. 2010, 37 (1): 197-200.

Abstract

PDF(275KB) ( 438 )

RelatedCitation | Metrics

To represent the logical difference of tuples and attribute values, a dual description logic for databases was introduced, in which concepts are classified into two classes; tuple concepts and attribute-value concepts. The logical language, syntax and semantics of the dual description logic for databases were introduced; two transformations from the relations in databases into knowledge bases or models in DDLD were defined; and finally an extended dual description logic for databases was introduced, in which roles are classified into three classes; the roles between tuples, the roles between tuples and attribute values and the roles between attribute values.

Research on Modeling Action with Uncertain Results Based on Probability Interval

TIAN Zun-hua,ZHAO Long,JIA Yan

Computer Science. 2010, 37 (1): 201-204.

Abstract

PDF(287KB) ( 446 )

RelatedCitation | Metrics

To model the uncertain actions of virtual characters is helpful to improve the fidelity and credibility of its behavioral representation.I3ecause the existing behavioral models are lack of necessary uncertainty, the characters often behave in a repeated and monotonous way. With probability interval, a method to model uncertain results of characters'action was proposed. Based on interval algebra, the concepts of probability interval and its completion were defined.Completion methods of probability interval were given. centering on normal probability interval, some operations were defined, and additional and multiplicative compensation methods to complete normal probability interval were discussed.Compared with algebra interval, probability interval can describe effects of actions more precisely. As an example, the uncertain actions of virtual tanks were discussed and normal interval was applied to model the uncertain actions of it.The proposed methods were validated through modeling the uncertain actions such as shooting.

Selected Ensemble of Classifiers for Handling Concept-drifting Data Streams

GUAN Jing-hua,LIU Da-you

Computer Science. 2010, 37 (1): 205-207.

Abstract

PDF(338KB) ( 475 )

RelatedCitation | Metrics

In data streams concept is often not stable but change with time. We proposed a selective integration algorithm OSEN(Orientation based Selected ENsemble) for handling concept drift data streams. This algorithm selects a near optimal subset of base classifiers based on the output of each base classifier on validation dataset. Our experiments with synthetic data sets simulating abrupt (SEA) and gradual (Hyperplane) concept drifts demonstrate that selective integration of classifiers built over small time intervals or fixed-sized data blocks can be significantly better than majority voting and weighted voting, which are currently the most commonly used integration technictues for handling concept drift with ensembles. This paper also explained the working mechanism of OSEN from error-ambiguity decomposition. Based on experiments, OSEN improves the generalination ability through reducing the average generalization error of the base classifiers constituting the ensembles.

Word Sense Disambiguating Method Based on Concept Relativity

MIAO Jian-ming,ZHANG Quan

Computer Science. 2010, 37 (1): 208-210.

Abstract

PDF(269KB) ( 449 )

RelatedCitation | Metrics

As one of the most important and also the most difficult problem of Chinese information processing field,word sense disambiguation has not been entirely solved until now. On the basis of our research on HNC representing Chinese knowledge, this article proposed a word sense disambiguating method based on concept relativity. The method combines various features in concepts. The features include hierarchy, network and structure etc. The uniform representation formalizes the features. In this way, the problem of knowledge relationship expression among various concepts will be solved. 20 Chinese polysemous words were tested in our experiment. The result with average precision 87% shows that the method is effective.

Artificial Gene Regulatory Networks Construction Based on Complex Network

ZHANG Lu-wen,XIE Jiang,CHEN Jian-jiao,ZHANG Wu

Computer Science. 2010, 37 (1): 211-213.

Abstract

PDF(347KB) ( 835 )

RelatedCitation | Metrics

Along with the explosion of high-throughput genomic information produced from such as DNA microarrays,modeling the mechanism of gene regulations and constructing gene regulatory network become an urgent task. We proposed a novel method, GN-Simulator, to simulate gene regulatory networks based on complex network theory. The real gene networks were modeled explicitly in matrixes based on the scale-free topology. According to the robust biological mechanism,we constructed artificial gene regulatory networks related topology to the dynamic stability. We considered the features of real gene network structures into the artificial system The computing experiments illustrate that GN-Simulator can simulate largcscale gene networks efficiently with high confidence level. Moreover, GN-Simulator generates various synthetic networks for testifying different algorithms performance and provides reasonable estimation on them.

Parameter Optimization Based on Orthogonal Test Design in Speech Recognition System

HAN Zhi-yan,WANG Jian,WANG Xu

Computer Science. 2010, 37 (1): 214-216.

Abstract

PDF(321KB) ( 489 )

RelatedCitation | Metrics

The performance of speech recognition depends on the choice of speech parameters. A systematic and practical method of the parameters selection based on the variance orthogonal test design was proposed. Firstly,chose factors (speech parameters) and levels. And then according to the principle of mathematical statistics and orthogonality, picked out proper and representative points from a large number experimental points to construct orthogonal table. Finally, calculated and analysed the experimental results, and the optimal set of process parameter values was discovered. Moreover, the word error rate is reduced 5. 6% and response time is reduced 181. 37ms when compared with that of the traditional parameter selection method. The experiment results show that the application of orthogonal test design is effective in parameter optimisation. The conclusion provides guidance for the subsequent research.

Dynamic Multi-object Artificial Immune System Model Based on Mechanisms of Evolution and Immunity

TAO Yuan,WU Geng-feng,HU Min

Computer Science. 2010, 37 (1): 217-221.

Abstract

PDF(402KB) ( 441 )

RelatedCitation | Metrics

This paper proposed a new dynamic multi-object artificial immune system model based on simulating the principle of the biological immune system. The model consists of five elements, i. e. environment set, antibody set, antigen set, rule set and a Dynamic Multi-object Evolutionary Immune Algorithm(DMEIA). As a key clement of the model, DMEIA combines the feature of evolutionary algorithm which selects optimal non-dominated antibodies and makes them to join in evolution of next generation,and the characteristic of immune algorithm which has strong population diversity and adaptive searching ability to control and assorts with the operation of the model. Compared with the existed algorithms, DMEIA has better convergence, diversity, distribution of solution and stability of environment tracking, therefore the performance of the new model is proven to be available.

Multi-construction Ant Colony Optimization Algorithm for Permutation Flow Shop Scheduling

LIU Yan-feng,LIU San-yang

Computer Science. 2010, 37 (1): 222-224.

Abstract

PDF(241KB) ( 810 )

RelatedCitation | Metrics

A Multi Construction Ant Colony Optimization Algorithm for Permutation Flow Shop Scheduling was proposed. In this algorithm, solutions are constructed through two modes, which are based on Nawaz-Enscore-Ham heuristics and Rajendran heuristics respectively. Then the proportion of construction modes is adjusted adaptively according to quality of solution constructed. Simulation results and comparisons based on benchmarks demonstrate the effectiveness of the algorithm.

Biclustering Method for Detecting Scaling Patterns with Genetic Rough Algorithm

LI Gang,MIAO Duo-qian,WANG Rui zhi

Computer Science. 2010, 37 (1): 225-228.

Abstract

PDF(315KB) ( 426 )

RelatedCitation | Metrics

This paper addressed the problem of detecting scaling patterns in gene expression data. A mean ratio residue as a merit function for Scaling patterns was presented. Based on the mean ratio residue a biclustering method was proposed under genetic rough framework. In this method, initial population is generated by Heuristic Rough Biclustering Algorithm as well as random choice. And then it adjust the seeds with Genetic Rough Algorithm. We tested this method on yeast expression data. The experimental results show that the Genetic Rough based method well improves the performance of heuristic algorithm and biclusters found on the yeast data are biologically significant using online GO Term Finder.

Earley Algorithm Using Prediction Strategies

GU Bo,LI Ru,LIU Kai-ying

Computer Science. 2010, 37 (1): 229-232.

Abstract

PDF(347KB) ( 875 )

RelatedCitation | Metrics

There are two kinds of parsing algorithms in nature language processing: one based on statistics and the other based on grammar rules. Earley algorithm is based on grammar rules. It can parse any context free grammar (CFG) without changing the grammar. This paper used several predictive strategies in Earley algorithm. These strategies come from LL, SLR and LR algorithms. Experiments were made for these strategies and their combinations. Results indicate that these strategics can usually reduce number of items and make parsing faster.

Ant Colony Optimization Algorithm with Path Smoothing and Dynamic Pheromone Updating

GAN Rong-wei,GUO Qing-shun,CHANG Hui-you,YI Yang

Computer Science. 2010, 37 (1): 233-235.

Abstract

PDF(252KB) ( 573 )

RelatedCitation | Metrics

Ant colony optimization is a new heuristic algorithm which has been proven a successful technique for combinawrial optimization problems, but it still has some shortcomings such as stagnation behavior, needing much time and premature convergence. A new algorithm based on path smoothing and dynamic pheromone updating was proposed for overcoming those shortcomings. By path smoothing, in the curly convergence phase, ants will search towards the path with shorter distance; ants will more constructe pheromone in the later convergence phase. By dynamic pheromone updating, algorithm can avoid being trapped into local optimum. The experimental results show that the algorithm presented in this paper has more effective than classical ant colony algorithm.

High Efficiency Approaches for Computing Attributes Core and Reductions Based on SQL

FENG Lin,LI Tian-rui

Computer Science. 2010, 37 (1): 236-238.

Abstract

PDF(323KB) ( 421 )

RelatedCitation | Metrics

Calculating approaches for attributes core, attributes reductions and values core were developed in relation databases by combing rough set theory and SQL Experimental results illustrate these methods have the better performance in run-time efficiency comparing with the related approaches of the classical rough set theory in processing decilion tables.

Investigating the Role of Vaccine Operator in Artificial Immune System for Network Intrusion Detection

FANG Xian-jin,LI Long-shu,QIAN Hai

Computer Science. 2010, 37 (1): 239-242.

Abstract

PDF(438KB) ( 481 )

RelatedCitation | Metrics

On the basis of author's previous research works, this paper integrates Immune Algorithm (IA) consisting of vaccine operator,mutation operator and other operators with Negative Selection Algorithm (NSA),called IA-NSA algorithm, to perform evolution of detector population, whose purposes are to quicken the process of affinity maturation and improve the efficiency of network intrusion detection. The algorithms of vaccine operator and adaptive extracting vaccines were given in detail. A novel network intrusion detection model and its algorithm based on IA-NSA were presented to establish anomaly detection. The network intrusion detection experiments based on IA-NSA algorithm and Clonal Selection Algorithm (CSA) were designed respectively, the experimental results show that the IA-NSA algorithm consisting of vaccine operator quickens the affinity maturation process of detector population and stably increases the detection rate along with increasing evolutionary generation. But in CSA-NSA algorithm,the affinity maturation process of detector population takes more time, the detection rate has a little degradation and maintains invariant for a long time.

Issues Regarding ε in Formal Language and Automata Theory

CHEN Wen-yu,WANG Xiao-bin,CHENG Xiao-ou,SUN Shi-xin

Computer Science. 2010, 37 (1): 243-244.

Abstract

PDF(192KB) ( 524 )

RelatedCitation | Metrics

The paper discussed some issues regarding blank string a in the formal language and automata theory. After analysis of the influence of production c on grammar and language classification,the paper discussed the effect of starting symbol S and the starting state qo from the perspective of grammer and infinite state and proposed a simple method to increase language or decrease sentence ε. The paper also proposed a new method to transit ε-NFA to NFA after studying the essence of a state transition function of E-NFA. The method is:first transit ε-NFA to formal grammar and elimmate production ε and single production. After that, regular grammar was obtained. Then transited regular grammar to NFA. Examples were given to support the discussion.

Pronominal Anaphora Resolution within Chinese Text Based on Fuzzy Rough Sets Model

LI Fan,LIU Qi-he,LI Hong-wei

Computer Science. 2010, 37 (1): 245-250.

Abstract

PDF(526KB) ( 457 )

RelatedCitation | Metrics

Anaphora resolution is an important issue in natural language processing. This paper presented an approach based on Fuzzy Rough sets model combined with instance -based learning approach to resolve pronominal anaphora within Chinese text. The first phase of the presented approach is preprocessing. In this phase,after extracting noun phases and eliminating those whose number and gender features arc inconsistent with pronominal anaphora, the potential antecedents set was formed. hhen, the attri-butc values of every noun phase in this set were computed according to an attribute set which only involves shallow syntactic and semantic information. The second phase aimed to select representative examples from the potential antecedents set and reduce redundant attributes to improve the generalization capability of these examples. hhese tasks were done by using concepts of Fuzzy Rough sets model. The two phases above can be regarded as learning phase. In the last phase, those examples were used to estimate whether a new noun phase is the that antecedent of a pronominal anaphor. The presented approach was tested by People Daily corpus. The results show this approach is effective.

Text Classification Algorithm Based on Adaptive Chinese Word Segmentation and Proximal SVM

FENG Yong,LI Hua,ZHONG Jiang,YE Chun-xiao

Computer Science. 2010, 37 (1): 251-254.

Abstract

PDF(425KB) ( 652 )

RelatedCitation | Metrics

New words recognition and ambiguity resolving are key problems in Chinese word segmentation. The result of traditional dictionary-based matching algorithm largely depends on the representative of the dictionary so that it can not recognize new words effectively, especially in some professional domains. Chinese word segmentation method in this dissertation is based on 2-gram statistical model and can meet the rectuirements of application in accuracy and efficiency respectively. PSVM takes classification as a linear equality quadratic programming problem. This dissertation describes a text classification algorithm based on adaptive Chinese word segmentation and PSVM, which has faster training speed and smaller memory requirements advantages. Several data sets of experiments showed that the classification algorithm can automatically adapt to knowledge management of some professional domains and has better classfication performance under the condition of timcsensitive.

Hybrid Monotonic Inclusion Measure on Intuitionistic Fuzzy Sets

YAO Yan-qing,MI Ju-sheng

Computer Science. 2010, 37 (1): 255-257.

Abstract

PDF(282KB) ( 506 )

RelatedCitation | Metrics

In the theory of uncertainty reasoning, inclusion measure is an effective tool for measuring the degree to which one set is contained in another set, We introduced a new definition of inclusion measure; hybrid monotonic inclusion measure on intuitionistic fuzzy sets (IFH inclusion measure for short). Based on which, several kinds of IFH inclusion measures were constructed, and the rationalities of which were proved. Then cetain distributivity and T-transitivity of some special IFH inclusion measures were investigated. And the use of IFH inclusion measures in pattern matching of knowledge was discussed. So this paper provided us with operational methods for measuring the degree to which an intuitionistic fuzzy set is contained in another intuitionistic fuzzy set as well as enriching the theory of inclusion measure.

Heuristic Attribute Reduction Algorithm to Ordered Information Systems

WANG Feng,QIAN Yu-hua,LIANG Ji-Ye

Computer Science. 2010, 37 (1): 258-260.

Abstract

PDF(320KB) ( 499 )

RelatedCitation | Metrics

Attribute reduction is one of important problems in rough set theory, and attribute reductions in ordered information systems are also concerned in recent years. We first proposed a new information granulation based on dominance classes, which can be used to measure uncertainty of an attribute set in the context of ordered information systerns. Using this information granulation, a definition of attribute significance was then given in ordered information systams. At last, a heuristic reduction algorithm was put forward to ordered information systems, and the validity of this algorithm was exemplified by an illustrative example.

Research on Modeling of Exploratory Analysis Model for C2 Organization

LIU Zhen-ya,LUO Xu-hui,XIU Bao-xin,ZHANG Wei-ming

Computer Science. 2010, 37 (1): 261-264.

Abstract

PDF(353KB) ( 450 )

RelatedCitation | Metrics

Exploratory Analysis Model for C2 organization is the key for validating the methodology of organizational design based on the Exploratory Analysis. In order to modeling the Exploratory Analysis Model, this paper expounded the characteristics of the Exploratory Analysis Model for C2 organization, defined the requirement, and advanced the principle and methodology for modeling. The hierarchy mapping was also analyzed.

Edge Detection Based on GPU

ZHANG Nan,WANG Jian-li,WANG Ming-hao

Computer Science. 2010, 37 (1): 265-267.

Abstract

PDF(241KB) ( 527 )

RelatedCitation | Metrics

Edge detection is a highly parallel algorithm with great computation. It is difficult to increase the speed of the algorithm by CPU to satisfy the real time application. Aiming at the computcintensive character of image edge deteclion, this paper analyzed some methods of edge detection based on CPU, using the programmer friendly CUDA framework, and proposed a method based on GPU, to realize the image edge detection. The efficient architecture of GPU was introduced firstly. Then, two representative image edge detection algorithms, Roberts and Sobel, were implemented on GPU. At last, using the same market price level CPU and GPU as hardware platform, and using various resolution images as test data, compared the computational efficiency of GPU and CPU. Numerical experiments show that the speed of the algorithm can be improved by up to more than 17 times compared with CPU-based implementations,with the same processing results. It proves that the GPU is practical for some applications of image processing.

Adaptive Key-frame Extraction Based on Affinity Propagation Clustering

XU Wen-zhu,XU Li-hong

Computer Science. 2010, 37 (1): 268-270.

Abstract

PDF(272KB) ( 443 )

RelatedCitation | Metrics

Key-frame Extraction is important to content based video analysis. For extracting key frames efficiently from different video, this paper presented an efficient method for key-frame extraction in which affinity propagation clustering is applied to key-frame extraction. First get the shot comparability matrix by histogram differences, extract key frames adaptively through affinity propagation clustering. The algorithm which this paper proposed can get the key frames adaptively according to the content of video shot and fast The experiment indicates the algorithm is efficient in key-frame extraction and has enough robustness.

Simulation of Precipitation Phenomenon Based on Particle System Method

TAN Liang,LUO Yi-fan

Computer Science. 2010, 37 (1): 271-274.

Abstract

PDF(331KB) ( 513 )

RelatedCitation | Metrics

Precipitation phenomenon in liquid is one of general phenomenon in nature or in research. It has very important significance for reappearing precipitation course and addressing precipitation disciplinarians by simulation. Based on particle system method, the mathematics model for precipitation phenomenon in lictuid was brought out, and the implementation of precipitation phenomenon in liquid was particularly described,more over, the technologies for real-time and reality were discussed. It is demonstrated by experiment that the method is not only simply, easy, and satisfying for requirements of real-time and reality, but also has preferable commonality, it's fit for other occasion by little modify.

Image Asymmetry Computing Algorithm Using Affine Transformation

SONG An-ping,DING Guang-tai,ZHENG Han-yuan,ZHANG Wu

Computer Science. 2010, 37 (1): 275-278.

Abstract

PDF(352KB) ( 479 )

RelatedCitation | Metrics

It might be a sign of pathology when an organ of a person appears more asymmetric than others. Based on mathematically asymmetric theory, an improved asymmetric computing model of 3-D medical image was presented, and a numerical method and computing framework were studied. The method can be used to diagnose cholesteatomatous chronic otitis media by numerical experiments. Compared with the doctors' diagnosis, the ratio of the numerical diagnosis of patients' cholesteatoma confidence is up to 98 percent by testing of T distribution method. The asymmetry computing model has both measurable and objective properties so that it can be extended to other computer-assisted diagnosis related to medical asymmetry.

Shape Image Retrieval Based on Compensation Mechanism under Minimum Circumscribed Circle

SHU Xin,WU Xiao-jun,PAN Lei

Computer Science. 2010, 37 (1): 279-281.

Abstract

PDF(331KB) ( 420 )

RelatedCitation | Metrics

A novel shape image retrieval algorithms based on compensation mechanism under miminum circumscribed circle was proposed in this paper, which not only extract the features of the target area, but also compensate the background characteristics of the object under miminum circumscribed circle. Extracting the Hu invariant moments and Zernike moment invariants,commonly used in the description of the regional characteristics,as image features. Moment invariants have the unchanged characteristics of translation, scaling and rotating. Similarity is given by the Euclidean distance between two images' normalized moment vectors. hhe approch is implemented simply, and compensates the human eye's visual experience effectively. A large number of experiments shown that the approach can achieve better accuracy and recall rate than that based on an object area only.

Statistical Thresholding Method for Infrared Images

LI Zuo-yong,LIU Chuan-cai,CHENG Yong,ZHAO Cai-rong

Computer Science. 2010, 37 (1): 282-286.

Abstract

PDF(493KB) ( 570 )

RelatedCitation | Metrics

Classic statistical thresholding methods take class variance sum of some form as criterions for threshold selection. They don't take special characteristic of practical images into account and fail to get ideal results when segmenting a kind of image having similar statistical distributions in the object and background. In order to eliminate the above limitation of classic statistical approaches, a novel statistical criterion was defined by utilizing standard deviations of two thresholded classes, and the optimal threshold was determined by minimizing it. Experiments on a variety of infrared images and general real world images show that our method outperforms the existing classic thresholding methods in segmentation quality, especially for infrared images.

New Algorithm for Text Segmentation Based on Stroke Filter

SHI Zhen-gang,GAO Li-qun

Computer Science. 2010, 37 (1): 287-289.

Abstract

PDF(255KB) ( 837 )

RelatedCitation | Metrics

In order to solve segment text accurately and robustly from a complex background, this paper proposed a new algorithm for text segmentation in images by using a stroke filter. First, we described the stroke filter briefly based on local region analysis. Second, the determination of text color polarity and local region growing procedures were performed based on the response of the stroke filter. Finally, the feedback procedure by the recognition score from an optical character recognition (OCR) module was used to improve the performance of text segmentation. The proposed algorithm was compared with other algorithms. The experimental results demonstrate that the proposed algorithm obtaines satisfactory results.

Robust Road Detection Based on Vision Dynamic Modeling

SHI Lei,JIN Zhong,YANG Jing-yu,WANG Yu

Computer Science. 2010, 37 (1): 290-293.

Abstract

PDF(345KB) ( 491 )

RelatedCitation | Metrics

Road detection and identification are one of the essential problems for vision understanding in autonomous vehide systems. However, it becomes relatively harder with the surrounding conditions such as shadows, asymmetric illumination, and blurry edge information. To enhance the robustness and interference immunity, a Vision Dynamic Mode ling based road detection algorithm was proposed. This method structures camera observation systems on computer vision, and introduces deformablc line model for road geometry description. Based on continuity a dynamic model for the vehicle motion was constructed, which incorporates Particle Filtering into road tracking by parameters estimation and prediction. Besides,likelihood probability combined with road model was utilized to evaluate the fitness with sampled images. Experiments with road images from CMU lab and our system proved the effectiveness and applicability of this algorithm.

Simulation Algorithm for "Bamboo" Distributional Pattern by Improved Particle System

LUO Yan,WU Zhong-fu,GUO Xuan-chang,ZHOU Shang-bo

Computer Science. 2010, 37 (1): 294-298.

Abstract

PDF(443KB) ( 415 )

RelatedCitation | Metrics

Based on the basic idea of particle system that the individual behavior is controled by groups behavior, an improved simulation algorithm for "Bamboo" distributional pattern was presented. In this paper, the improved simulation algorithm makes the random distribution of particles to break the rules of rectangular shape. Then, the improved algorithm can simulate different "Bamboo" distributional pattern more really. This paper gave a new solution in plant simulation and made the particle system to be applied in the plant growth pattern.