Computer Science

P-reasoning and P-reasoning Discovery-identification of Information

SHI Kai-quan

Computer Science. 2011, 38 (7): 1-9.

Abstract

PDF(626KB) ( 396 )

RelatedCitation | Metrics

By embedding the dynamic characteristics into the finite Cantor set X and improving it, a novel mathematics structure and model was obtained,which is P-sets(Packet sets). P-sets is a set pair composed of internal P-set XI' (in- tonal packet set XF)and outer P-set XF (outer packet set XF) , or (XF , XF ) is P-sets. Based on P-sets, internal P-rca-soning(internal packet rcasoning),out p-rcasoning(outer packet reasoning)，Prcasoning(packet reasoning) and the dia-grammatic representation of reasoning model were given. The structure of P-reasoning is "if (ak，。穿.，)今(菇，，潇)，then((二)象1，(二)穿)今((二)奋，(二)象1)”.A few theorems were presented, which are internal P-reasoning and information de-leting theorem, outer P-reasoning and information supplementing theorem, P-reasoning and information deleting-supple-menting theorem, the relation theorem of P-reasoning and general reasoning and so on. A few concepts were given,which arc information unit circle, internal P-reasoning information circle and outer P-reasoning information circle. Thedynamic characteristics theorem of P-reasoning information circle was given. The internal discovery and outer discoverycriterion of reasoning information circlcunknown information were given. By using the results, the application of inter-nal packet reasoning in the unknown internal information discovery-identification was given. The attribute perturbationcharacteristic of P-reasoning,internal P-perturbation theorem of internal P-reasoning, outer P-perturbation theorem of outer P-reasoning and P-perturbation of P-reasoning were obtained using the dynamic characteristic of P-rcasoning.In the internal P-easoning,owing to the existence of internal P-perturbation, part information in the internal P-reasoning conclusion is lost. In the outer P-reasoning, owing to the existence of outer P-perturbation, part information is supple-mented into the outer P-reasoning conclusion. P-sets is a novel model and method with good application prospect in the intelligent information system.

Survey of Semantic Web Services Publication and Discovery Technology Based on P2P Network

SI Hua-your, NI Yu-1in, CHEN Zhong, YU Lian

Computer Science. 2011, 38 (7): 10-17.

Abstract

PDF(748KB) ( 324 )

RelatedCitation | Metrics

In service-oriented computing, service discovery is the key. Service discovery technology is the basis of Web services composition and real-time construction of application. In recent years it has obtained the academic and industry's attention,related technology after another. We firstly analyzed the current research results of Web Service Publication and Discovery technology and carried out a systematic classification. Next, focusing on the perspective of semantic fca-ture publication,we discussed current major research results of WSPD in P2P system in detail. Finally we concluded the general characteristics of current SWS publication and discovery based on P2P network, as well as the major research oblectives,research questions and challenges in the future research.

Keyword-based Search on Semantic Web Data: The State of the Art

LI Hui-ying,QU Yu-zhong

Computer Science. 2011, 38 (7): 18-23.

Abstract

PDF(659KB) ( 502 )

RelatedCitation | Metrics

The growth of the Semantic Web has seen a rapid increase in the amount of RDF data. Meanwhile, the de-mand for access to RDF data without detailed knowledge of RDF query languages is increasing. In this paper, some definitions in terms of keyword-based search on Semantic Web data were firstly presented. Then, a large number of popular existing solutions were surveyed according to hybrid and non-hybrid style(including K-A means and K-"A means). In addition,eight representative systems were introduced and compared with each other based on different facets,and their unique features were highlighted in details. Finally, some remaining challenges were discussed, and some future research directions were pointed out.

Survey of Research on Cloud Infrastructure Security

HUANg Ying，SHI Wen-chang

Computer Science. 2011, 38 (7): 24-30.

Abstract

PDF(759KB) ( 648 )

RelatedCitation | Metrics

Cloud computing is a worldwide hot topic nowadays. It will probably cause a new revolution to information technologies. But meanwhile it induces new security problems as well. Standing on the bottom layer of a cloud computing environment, this paper carried out research on security of cloud infrastructure. It observed the statcof-thcart of cloud infrastructure security, analyzed security problems existing in cloud infrastructure from a wide perspective. In combination with a technical framework of cloud infrastructure security, it discussed principle key technictues of cloud infrastructure security. The motivation of the paper is to lay a solid ground for solving security problems in cloud infrastructure as well as in the whole cloud computing environment

Research of Probabilistic Planning

LIU Ying，GU Wen-xiang

Computer Science. 2011, 38 (7): 31-34.

Abstract

PDF(471KB) ( 621 )

RelatedCitation | Metrics

The probabilistic planning is a hot spot of intelligent planning research and is paid attention to its own practical significance by more and more scholars. At present,many scholars have made the new expansion to the probabilistic planning. This paper emphatically introduced the development of probabilistic planning in recent years, as well as the probabilistic planncr(FF-Rcplan).Then the International Probabilistic Planning Competition was presented. It enables the general scholars to have a more comprehensive understanding to the probabilistic planning.

Advances in Automatic Image Annotation

BAO Hong， XU Guang-mei，FEND Song-he，XU De

Computer Science. 2011, 38 (7): 35-40.

Abstract

PDF(558KB) ( 602 )

RelatedCitation | Metrics

Automatic image annotation has emerged as a hot topic in the field of image semantic understanding due to its potential application on Web image search. To effectively access and retrieve images, a popular solution is to tag images with meaningful semantic keywords,which is considered as automatic image annotation. Various machine learning technictues have been employed extensively in the field of image analysis, and there is no exception for automatic image annotation. Existing image annotation algorithms can be roughly divided into three categories, i. c.，the classification based methods, the probabilistic modeling based methods, and the graph learning based methods, respectively. We surveyed nearly 50 key theoretical and empirical contributions in the current decade related to automatic image annotation, and discussed the spawning of related sub-fields in the process. I3y carefully analyzing what has been achieved so far,we also conjectured what the future may hold for automatic image annotation research.

Survey on Wireless Cognitive Radio Sensor Networks

WANG Yong-hua，YANG Jian,CHENG Liang-lun，WAN Pin，ZHANG Bo-wei

Computer Science. 2011, 38 (7): 41-45.

Abstract

PDF(504KB) ( 750 )

RelatedCitation | Metrics

Incorporating Cognitive Radio technology in Wireless Sensor Networks yields the Cognitive Radio Sensor Networks,which can reduce the interference of ISM band interference and improve data transfer rates. Cognitive Radio Sensor Networks have become a new hot research area. The concept of Cognitive Radio Sensor Networks was described. The research work on spectrum managemnet functionnalities such spectrum sensing, spectrum sharing, spectrum decision, spectrum mobility, and the network layer protocols, transport layer protocols, application layer protocols were introduced. The existing problems in the current researchwork and the new research issues were also discussed.

Constraint Particle Swarm Optimization Algorithm for Wireless Sensor Networks Localization

OUYANG Dan-tong，HE Jin-shcng ，BAI Hong-tao

Computer Science. 2011, 38 (7): 46-50.

Abstract

PDF(458KB) ( 403 )

RelatedCitation | Metrics

Node localization is a key technology of wireless sensor networks. Although radio-ranging is accuracy, using least squares algorithm for node localization may lead to big error. To increase the node localization accuracy of range-based wireless sensor network, in this paper, the node localization problem was transformed into a constrained optimization problem, and then the problem was solved by using particle swarm optimization algorithm. In the solution process,by setting the constraint fitness function and the distance fitness function, the search computation was reduced, the convcrgenee rate was speeded up,and a better solution was achieved faster. The result from simulation experiments show that compared with least squares algorithm, under the circumstances of different ranging error, different distance radius,different number of anchors and different number of nodes,applying this algorithm can achieve a solution with more accuracy. hhis shows that the algorithm has stronger anti error, better astringency and less investment in hardware, etc. In addition,it performs better in the sparse network node localization.

Security Extension on Strand Space Model for Ad-hoc Routing Protocols

GNU Xue-wen，NIU Wen-sheng，MA Jian-feng，SHENU Li-jie

Computer Science. 2011, 38 (7): 51-54.

Abstract

PDF(341KB) ( 355 )

RelatedCitation | Metrics

Abstract based on the characteristics of Ad-hoc mobile network and detail analysis of the consistency condition in strand space model, a concept of five routing segments model was brought up and intermediator credibility condition was changed into the arbitrary intermediator credibility condition, thus the strand space model was adapted to the for Ad-hoc routing protocols. Then an attack was brought to verify the correctness and superiority of segments model.

Attribute Encryption Based Hidden Credentials Extended Model

GE Wei-jin，CHENG Quan，HU Xiao-hui，Zhan Qian-qian

Computer Science. 2011, 38 (7): 55-57.

Abstract

PDF(338KB) ( 418 )

RelatedCitation | Metrics

The current identity-based encryption based hidden credential can not support 1-N communication, endures no identity fuzzy and lefts open to conspiracy crack, which brings great limitation to its availability. In this paper, an attributcbased encryption based hidden credentials extended model was presented which solves those dilemmas of basic hidden credential system, meanwhile, the extended model keeps the advantages in protecting all the certificates, resources and strategics safe in trust negotiation. Since in AI3E system, each participant has and only has one attributcset certificate, the extended model significantly improves the efficiency of decryption process. On the other hand, through the random control in certificate issue process, the possibility of conspiracy crack is also eliminated. In addition, we described their performance, security and applications in detail. Finally, the farther research directions were suggested.

Implementation of Pipeline Structure on FPGA for SHA-1

LI Lei， HAN Wen-hao

Computer Science. 2011, 38 (7): 58-60.

Abstract

PDF(239KB) ( 545 )

RelatedCitation | Metrics

Hash algorithm SHA-1 is used widely in cryptographic applications such as E-commerce and commercial encryption softwares. By making an overall analysis, implementation of pipeline structure was proposed. By shorting the critical path, using inside RAMs to realize the data translation instead of LE, we improved the frequency and the speed The performance on FPGAs of Altera is three times faster of commercial IP cores for SHA-1.

Service Differentiation Model for Performance Enhancement of 802. 15. 4 Sensor Networks

CAI Ya-ping ，BAI Guang-wei

Computer Science. 2011, 38 (7): 61-65.

Abstract

PDF(433KB) ( 397 )

RelatedCitation | Metrics

IEEE802. 15. 4 is one of the standards for a low rate, low power consumption wireless sensor networks. By assigning the same sets of contention access parameters for all data frames and nodes, the contention access period (CAP) of the slotted IEEE 802. 15. 4 medium access control(MAC) currently provides a priority-independent channel access functionality and no service differentiation. To solve this problem, this paper proposed a differentiated service mechanism to support high IEEE802. 15. 4 QoS rectuirements. The main idea of the mechanism is allowing different sets of access parameters and data frame lengths for different priority classes, we developed a Markov-chain-based analytical model of the CAP of the IEEE 802. 15. 4 MAC with service differentiation, under unsaturated traffic conditions. In parocular, given two priority classes, our analytical model was used to evaluate the performance of a simple and effective service differentiation strategy, in terms of throughput, average frame service time, energy consumption, as well as access probability for each class. Our analysis results show that this model for differentiated services performs well.

Definition and Equation Proof of Semantic Security in Information Security

HU Ruo,QIAN Xing-san

Computer Science. 2011, 38 (7): 66-69.

Abstract

PDF(299KB) ( 1420 )

RelatedCitation | Metrics

Our definition of semantic security was introduced,and we tried to integrate the frameworks of definitions of semantic security, non-distinguishability and non-expandability by analysing semantic security in comparison based framework. This facilitates the study of relations among these goals against different attack models, and we made a slightly modified definition of classifying based on comparison and then we studied our proof of equation of semantic security and classifying in the new framework. This way makes the proof of the equation of semantic security and non-dis-tinguishability easier and more understandable.

Network Security Situation Prediction Model Based on HHGA-RBF Neural Network

MEND Jin,MA Chi,HE Jia-lang,ZHANG Hong

Computer Science. 2011, 38 (7): 70-72.

Abstract

PDF(332KB) ( 453 )

RelatedCitation | Metrics

To address the prediction problem in the Network Security Situational Awareness, the RI3F neural network method was proposed to predict the situation value. In order to improve the prediction accuracy of RBF neural network,the HHUA was used for training to acquire the neural network parameters. Experimental results show the validity of this prediction method,and the accuracy of the proposed algorithm was verified by comparative experiment with the existing forecasting methods.

Improved Key Management Protocol for LEO Satellite Network

PAN Yan-hui, WANG Tao, GHAO Xin-jie,LI Hua

Computer Science. 2011, 38 (7): 73-75.

Abstract

PDF(248KB) ( 609 )

RelatedCitation | Metrics

Key management protocol is the base of secure communications between satellite network nodes. Character of key management protocol in the terms of satellite network environment was analyzed. A distributed key establishment scheme combined with centralized one was proposed for LEO satellite network. hhen, an improved key management protocol was given with new clustered rekeying mathod. At the end, it was analyzed and checked. The result indicates that it could help to tighten security with shorter key updating period.

Security Analysis of Resistance against Differential-linear Attack on BLAKE-32

MAO Ming,HE Qiang,ZENG Shao-kun,ZHANG Jun

Computer Science. 2011, 38 (7): 76-79.

Abstract

PDF(355KB) ( 391 )

RelatedCitation | Metrics

Hash function BLAKE is one of candidates for the second round SHA-3 competition,祠〕ch is based on modular addition, rotation and XOR that is called as ARX system commonly. It is a common belief that the mixture of the three operations gives a good primitive in designing cryptographic algorithm. By replacing modular addition with XOR,this paper researched lincarization of ARX system in BLAKE-32 algorithm, then, analyzed differential diffusibility of the algorithm after linearization and exploited some diffusion characteristics, furthermore, researched the probability of linear approximation of addition, and analyzed its validity on the algorithm. I}hc result shows that differential spreading does not satisfy the designers' declaration. On account of ARX in BLAKE, differential attack can be applied on BLAKE by local linearizing its core function.

LCESM, Location-aware Context Event Subscription Mechanism

LIU Han,YE Jian,ZHU Zhen-min,LIU Ren-ren,YU Chang

Computer Science. 2011, 38 (7): 80-84.

Abstract

PDF(511KB) ( 325 )

RelatedCitation | Metrics

Traditional location-aware publish/subscribe systems mainly limit at event publication and matching, and couldn't support the frectuent context events that happen in the same event source. A location-aware context event subscription mechanism that consists of location-aware context event subscription language(LACI)and dynamic binding approach was presented in the paper. Based on LACL, location event and event source event were defined and made the system capable of location awareness and event source awareness separately; dynamic binding consists of the mapping between the available context event source and the subscriptions and the dynamic binding relationship between agents and subscriptions. hhe former converts the matching between events and subscriptions to the matching between event sources and subscriptions, and reduces the matching and improves the system performance; agents' dynamic operations to subscriptions avoid redundant monitoring cost. The mechanism was implemented in context aware middleware ChK.Experiments show effectiveness of the mechanism.

EHiQ:A RFID Reader MAC Protocol Based on Enhanced HiQ

YANG Jian,WANG Yong-hua,CAI Qing-ling,ZHAN Yi-ju,WAN Pin

Computer Science. 2011, 38 (7): 85-87.

Abstract

PDF(347KB) ( 334 )

RelatedCitation | Metrics

Based on HiQ reader anti-collision algorithm, an enhanced algorithm called EHiQ was proposed. EHiQ not only imporves HiQ's system structure and cost function, but also imports the concept of simulated anneaning which is applied to balance exploitation and exploration of every phase of learning process and shorten learning time. Simulation shows that resource usage and frequency interference rate of EHiQ are slightly higher than those of HiQ. However,EHiQ's convergence time is just one third of HiQ's. Moreover, due to the improvement of system structure,EHiQ's blocking rate remains zero all the time.

Event-driven Based Distributive and Adaptive Network Management System

DING Zhen-guo，WANG Jing-y，ZHAO Yong-jie

Computer Science. 2011, 38 (7): 88-92.

Abstract

PDF(435KB) ( 359 )

RelatedCitation | Metrics

Referring to CORI3A event services scheme, the paper proposed an event driven based distributive and adaptive network management system named EDDM The system has very good flexibility in functions, realizing the buffering, filtering, and distributing capabilities of massive network events. Meanwhile distributed architecture has fine adaptability. The load and number of the management nodes can be dynamically ajusted by BP neural network algorithm. Then the management architecture can be changed according to the actual needs. Experiments show that EDDM can effectively reduce the latency of response and the bandwidth occupied by management data.

Key Exchange Solution of Mobile VPN Based on Improved IKE

CHEN Nan, YU Ding-guo, TAN Cheng-xiang

Computer Science. 2011, 38 (7): 93.

Abstract

PDF(380KB) ( 430 )

RelatedCitation | Metrics

One new solution of the key exchange introduced in this paper does improve the IKE protocol. I}his solution not only allows the remote mobile users security access to the intranet and get the intranet information on the premise of no reduction in security, but also extent the ways of IKE authentication, making it more efficient in negotiation and more controllable. Besides, the solution develops a corresponding mobile VPN access system which supports both dynamic distribution of intranet IP address and the extension of user identity authentication. Thus the access server can easily control and manage the intranet IP-based access.

Efficient Dynamic Algorithm for Computation of Shortest Path Tree

LIU Dai-bo,HOU Meng-shu,WU Ze-xu,QU Hong

Computer Science. 2011, 38 (7): 96-99.

Abstract

PDF(342KB) ( 1126 )

RelatedCitation | Metrics

The computation of shortest path tree in dynamic environment is a typical combinatorial optimization problem.Ball-and-String Model is an efficient algorithm which can dynamically update shortest path tree(SPT)，but exists redundant computation. This paper presented an a new dynamic SPT algorithm that optimizes the processing of edges in Ball-and-String Model. New algorithm raises the efficiency of dynamically updating SPT. Additionally, new algorithm implements deleting of node or adding of node in SPT, accordingly, can adjust for the topological variation of SPT. Experimental results show that new algorithm is more efficient than Ball-and-String Model.

Strategy of Closed-loop Rate Control and Traffic Shaping Based on RTCP

WANG Zheng-jun,WANG You-zhao

Computer Science. 2011, 38 (7): 100-102.

Abstract

PDF(277KB) ( 563 )

RelatedCitation | Metrics

The quality of real-time video in network monitoring system can be easily affected by network bandwidth.This paper introduced a Closed-Loop rate control strategy based on RTCP. This strategy analyzes the feedback of the receiver report(RR) in RTCP,adjusts the bit rate of encoding,adaptive network carrying capacity. This paper also proposed a stream traffic shaping policy to solve the problem of the data burst of H. 264 stream by inhibiting data burst,thus avoiding sudden transient network congestion and reducing packet loss. Experimental results show that the strategy effectively reduces the transmission stream packet loss rate and improves the quality of real-time video.

ChecKing Information Flow on Output Channel with Reachability Analysis of Pushdown System

SUN Cong,TANG Li-yong,CHEN Zhong

Computer Science. 2011, 38 (7): 103-107.

Abstract

PDF(436KB) ( 359 )

RelatedCitation | Metrics

We proposed an approach for analyzing information-flow security of imperative language with output channels. The program is abstracted with pushdown system, which is then self-composed in order to adapt noninterference as a safety property. The output operations in the two relevant runs are respectively modeled as storing and matching procedure by pushdown rules. Then the termination-insensitive noninterference is verified by a reachability analysis of illegal-flow state. A variation of this approach can deal with program containing divergent run. An upper-bound regression algorithm was proposed to find the maximum upper-bound in order to trigger coercive termination of divergent run. The experimental results show that the approach is more precise and efficient than existing work.

Slice Parallelization Research Based on Program Slicing Technique

SANG Chun-lei，ZHANG Zhao-qing

Computer Science. 2011, 38 (7): 108-112.

Abstract

PDF(440KB) ( 703 )

RelatedCitation | Metrics

Computer program consists of a lot of computation units, which depend on each other or not, and serve to final computation result. The independent computation can be parallelized to accelerate the whole program. Program slicing can extract independent computation from large program according to slicing criterion, which is one set of variable and one program position. Program slicing can help exploiting parallelism from serial program, which we call slicing parallelism. This paper extended OpenMP to model slicing parallelism, and we also developed one slicing analysis tool,which could recognize slicing parallelism and help programmer to parallclize serial program.

Research on Dynamic Job Configuration Framework for Mining of Software Repositories

SHI Dian-xi，YIN Gang，MI Hai-bo，YUAN Lin，WANG Huai-min

Computer Science. 2011, 38 (7): 113-116.

Abstract

PDF(446KB) ( 338 )

RelatedCitation | Metrics

Construction of datacenters for mining of software repositories(MSR) is a hot topic in current software enginecring area. Data processing jobs for software repositories arc highly diverse in execution-time and resourccconsuming, which bring many challenges for the job configuration in such environments. This paper proposed a job configuralion framework named TrustieSDC for mining of software repositories. TrustieSDC supports a new paradigm for remote deployment and execution of MSR jobs,and proposes a software-subversion-partition based job configuration algorithm to cut the response time of long jobs and increase the resource utilization. The experiments based on SVN repositories of Gnome projects shows that compared with paralleled Alithcia system, TrustieSDC gains remarkable improvement on both performance and resource utilization.

Exception Handling Mechanism for Semantic Process Based on Rules

ZHAO Kai，YING Shi， ZHANG Lin-lin， HU Luo-kai，JIA Xiang-yang，WANG Quan-yu

Computer Science. 2011, 38 (7): 117-120.

Abstract

PDF(467KB) ( 326 )

RelatedCitation | Metrics

Based on the Semantic Programming Language(SPI)，a mechanism was proposed for handling exception of semantic process. Firstly,focused on the external services invocation failure and processes internal logic failure during the execution of semantic process, the corresponding exception ontology for semantic process was given. On the basis of exception ontology,five exception handling operations were given to construct ECA rules for exception handling. Then, a prototype implementation based on ECA rules for handling exception of semantic process was given. Finally, a case study demonstrates the effectiveness and feasibility of the mechanism.

Design and Implementation of Software Reconfiguration Tool Based on AADL

LI Long，DONG Yun-wei，QIN Yang-sen，ZHANG Fan

Computer Science. 2011, 38 (7): 121-125.

Abstract

PDF(461KB) ( 402 )

RelatedCitation | Metrics

Modes represent alternative operational states of software. In multiple modes system, we can make different configuration such as resources and properties. Up to now, there isn't any tool which can abstract the modes of AADL architecture and establish blueprint in the program written in C language under the Vxworks system. hherefore, designing a software reconfiguration tool based on AADL architecture will provide great help to software architecture reconfiguration. To set up an expandable platform of software reconfiguration for AADI. architecture under Eclipse open source development environment, we designed SRM2 (Software Reconfiguration Middleware based on Mode) plug-in tool. SRM2 firstly scans AADL architecture in C program as to describe the static blueprint information of program architecture,then it generates dynamic blueprint according to code run-time information, which can be got by an code probe, under the VxWorks system, the tool can guide software reconfiguration.

Using RBAC-based Approach to Integrate Access Control Policies in Legacy Systems

LI Han,GUO He,WANG Yu-xin,LU Guo-ji,YANG Yuan-sheng

Computer Science. 2011, 38 (7): 126-129.

Abstract

PDF(467KB) ( 336 )

RelatedCitation | Metrics

Access control whose objective is to ensure the security of accessing to resources in software systems is an essential part for software systems. As access control policies in legacy systems seldom based on roles are represented in various forms, an RI3AC-based approach was proposed to integrate these access control policies. I}he approach maps permission of legacy systems to tasks of integrated system. Based on task trees and transformation rules of access control policy, various access control policies were reorganized in a unified form. Moreover, management rules were provided to achieve further authorization. A case study is demonstrated to depict the proposed approach is a feasible solution to integrate legacy access control policies and introduce RI3AC into legacy systems.

Adaptive Consensus Voting Algorithm

OUYANG Cheng-tian,WANG Xi,ZHENG Jian

Computer Science. 2011, 38 (7): 130-133.

Abstract

PDF(345KB) ( 574 )

RelatedCitation | Metrics

Fault tolerant technologies have been applied in many high reliability fields. N-version programming technology is one of those Fault tolerant technologies. In the software system, voting algorithm can mask the error outputs, redundant technology can prevent false results from propagating to the next sub-module in the software system, and enhance the reliability of the system. Many voting algorithms are widely used in fault tolerant technology, in which Consensus Voting Algorithm arc also widely applied, but the consensus voting strategy is particularly effective in small output spaces because it automatically adjusts the voting to the changes in the effective output space cardinality. But the consensus voting algorithm is more prone to cause identical-and-wrong(IAW) results in this case. I}herefore,to resolve this problem, this paper proposed an adaptive consensus voting algorithm. The history records information of modules are used to improve the power of consensus voting algorithm, the proposed algorithm reduces the probability of the incorrect results passing the voter and improves the system safety and reliability. Experiments demonstrate the power of our proposed algorithm.

Software Reliability Model for Component-based Dataflow Software

XU Qin-gui, LIU Gui-xiong

Computer Science. 2011, 38 (7): 134-138.

Abstract

PDF(425KB) ( 341 )

RelatedCitation | Metrics

Duc to the mechanism that component based dataflow software has its execution path determined by components activated by data fed into input interfaces,its software reliability can be influenced by distribution of input data,and is thus difficult to evaluate using traditional models such as statcbased or path-wise ones. A reliability model combining frectuency statistics and operational profile was put forward in this paper. Starting from analysis of dataflow software structure, programs were denoted as multi-layer structure by introduction of composite node. On the basis of data and control flow relations,activated components and their execution probability were estimated. The operational profile was transferred along the direction of data flow to components on the same or a lower layer. By means of depth-first rule, reliability of composite nodes on all levels was evaluated in opposite order until the actual reliability of the whole software was obtained. A case study shows that the model can effectively evaluate actual reliability of component based data flow software, reflecting ready state of effective data in input interfaces and data distribution attributes.

General Java Program Instrumentation Tool Based on Eclipse

ZHENG Xiao-mei

Computer Science. 2011, 38 (7): 139-143.

Abstract

PDF(591KB) ( 437 )

RelatedCitation | Metrics

Program instrumentation technique has been widely used in program analysis, testing and verification as an effective method for understanding the dynamic information of programs. However, for the absence of general instrumentation tools, different applications need to recreate the specific tool meeting their requirements, hence wasting time and energy. In addition, the debugging processes become more and more complicated because of the instrumented code inside the original program. To solve such problems, the paper presented a general instrumentation tool based on Eclipse for Java program. By defining the rules to match the execute points of program, a specific instrumentation tool could be customized to meet the different requirements of Java program analysis, testing and verification. By switching between visibility and nonvisibility of the instrumented code, users of our tool can't be disturbed by the planted fragments in comprehending and debugging java program. We believe that through utilizing our tool, the program instrumentation technique can be better applied in the Java program development.

Comonad Theory and its Applications in Functional Programming Language Haskell

SU Jin-dian，YU Shan-shan

Computer Science. 2011, 38 (7): 144-147.

Abstract

PDF(390KB) ( 489 )

RelatedCitation | Metrics

Monad theory in functional programming language Haskell has some disadvantages in describing the context dependent computations. As the categorical dual notion of monads, comonad theory can effectively improve Haskell's description ability of context-dependent computations. Firstly, we gave the categorical definitions and properties of Comonads, as well as their implementations in Haskell. Secondly, we discussed the CoKleisli triple and CoKleisli category, and used some examples to demonstrate how to apply them into the descriptions and reasoning of context dependent computations. Finally, we also discussed the distributive laws between Comonads and Monads, and showed its uses in merging the effectful computations and context dependent computations.

Outlier Mining of the High-dimension Datasets Based on Information Theory

ZHANG Jing，SUN Zhi-hui，SONG Yu-qing，NI Wei-wei，YAN Yan-hua

Computer Science. 2011, 38 (7): 148-151.

Abstract

PDF(432KB) ( 494 )

RelatedCitation | Metrics

Phenomena of "curse of dimensionality" deteriorate lots of existing outlier mining algorithms validity. Conconing thw problem, the outlier mining algorithm of high-dimension and large datasets based on information theory was proposed. This algorithm used the concept of information entropy and the mutual information in the information theory,carried on the feature selection after using estimated mutual information value objective basis entropy power sorting, and eliminated redundant attribute for dimensionality reduction. Outlier mining using information entropy as a measure standard to judge eliminated the drawbacks of distance and density metric. The experimental result in the real data sets indicates that the algorithm for outlicr mining in high-dimensional mass data is effective and feasible, its efficiency and accuracy arc significantly improved.

Continuous Probabilistic Skyline Queries Based on Road Network for Uncertain Moving Object

FU Shi-chang,DONG Yi-hong,CHEN Hua-hui,QIAN Jiang-bo

Computer Science. 2011, 38 (7): 152-156.

Abstract

PDF(412KB) ( 326 )

RelatedCitation | Metrics

Skyline queries are an important operator of LBS,which aim to find all data points that are not dominated by any others. Skyline inctuires for moving objects with uncertainty in road network were studied. After modeling road network and moving object, the dominant probability and skyline probability in road network envirorunent were defined.Then, two types of event that may affect p-Skyline and four pruning rules were devised. The dynamic incremental algorithm U-CPSQRN is supposed based on the above definition. By tracking and calculating these events, the operation of continuous updated p-Skyline can be achieved, which reduces search steps and system overhead. The experiments having positive results show effectiveness of the proposed algorithm.

Rotation Grid:A New Cluster Ensemble Method

CAO Qiao-ling,GUO Hua-ping, FAN Ming

Computer Science. 2011, 38 (7): 157-161.

Abstract

PDF(449KB) ( 360 )

RelatedCitation | Metrics

Although it is rapid and efficient to use the grid-based clustering approach to learn the partition of a data set,grid clustering is excessively dependent on the initialization of density threshold, and the margin of each cluster constructed by the approach presents zigzag manner, which prohibits the recognition of smooth boundary surface. Thus, this paper proposed a new grid-oriented cluster ensemble approach(RG) to solve this problem. Instead of constructing the partitions with diversity on a given data set by random sampling or initializing parameters of corresponding algorithm,RG randomly splits the features set into K subsets,uses feature transformation method on the subsets to learn K differrent rotation basis,and applies grid cluster algorithm to the new feature space formed by the K axis rotations to learn the partitions with diversities. Experimental results show that, compared with single grid clustering, RG not only partilions the data set with arbitrary shape or size efficiently, but also alleviates its dependence on the density threshold initialization and smoothes the rough boundary.

Link Prediction Based on Node Similarity

DONG Yu-xiao,KE Qing,WU Bin

Computer Science. 2011, 38 (7): 162-164.

Abstract

PDF(314KB) ( 1235 )

RelatedCitation | Metrics

Link prediction is an important issue in graph mining. It aimed at estimating the likelihood of the existence of links between nodes by the known network structure information. Currently, most link prediction algorithms based on node similarity consider only the individual characteristics of common neighbor nodes. We designed a new algorithm exploiting the interactions between common neighbors, namely Individual Attraction Index While maintaining low time complexity, this algorithm remarkably improved the accuracy of prediction. I}his paper proved well the best overall performance of this new algorithm by comparing three well-known node similarity algorithms on eight real networks with Individual Attraction Index.

New Data Preprocessing Method for Privacy-preserving Data Mining

LIU Liang,XIE Shu-ting,LI Shun-dong

Computer Science. 2011, 38 (7): 165-169.

Abstract

PDF(431KB) ( 335 )

RelatedCitation | Metrics

The Apriori algorithm is a milestone in the development of data mining. A number of other algorithms, which generate association rules by producing frectuent itemsets,are derived from it. This paper proposed a data preprocessing algorithm based on item closure, which is an absolutely new method for privacy-preserving data mining. According to the characteristics and the processes of Apriori-like algorithms, this method transforms different items in various ways.So the data mining applicant can only obtain the information of its own products correctly, but cannot obtain useful information regarding its potential competitors' products. Hence, through the application of this method, the cooperated data mining between the data provider and data miner will benefit both sides, and obtains a win-win result.

Hybrid Markov Prediction Model and Research of its Applications in Anti-money Laundering

LI Yu-hua,LI Dong-cai,BI Wei,LI Rui-xuan

Computer Science. 2011, 38 (7): 170-174.

Abstract

PDF(442KB) ( 770 )

RelatedCitation | Metrics

An important problem in anti-money laundering is to predict the possible transactions conducted by suspicious accounts. Markov model has a wide range of applications in economic predictions such as stock, commodity prices,market share and so on. But the prediction accuracy of the single markov model remains to be improved. A hybrid Markov model jointing with clustering,association rule and low order Markov model was proposed. In the process of constructing the model, the confidence-based pruning was conducted to reduce the time complexity. Finally, the model was used to predict the transactions among accounts in anti-money laundering. hhe experimental results show that this modcl has high prediction accuracy and is a good tradeoff between the prediction accuracy and the time complexity.

Social Tag Predication Based on Probabilistic Topic Model

YUAN Liu,ZHANG Long-bo

Computer Science. 2011, 38 (7): 175-180.

Abstract

PDF(498KB) ( 407 )

RelatedCitation | Metrics

Fagging information created by users is important to understand the Web resource semantics and to improve the intelligence of Web applications. Probabilistic topic model was exploited to deal with the incompleteness and inconsistence of tagging systems. A probabilistic topic model generating technique based on tag statistical characteristics was proposed. According to tag statistical characteristics of each resource, tag documents with different format can be created. By analyzing the performance generated by different tag documents, document format that is appropriate for a certwin dataset was confirmed. High relatedness between the vocabularies in the same topic was exploited to predicate the tag for resources with incomplete and inconsistence tags. Experiments on DeliciousT 140 and Wiki10+ show the effectiveness of the technique proposed.

Learning Bayesian Network from Small Scale Dataset and Application

LI Ya-fei,LU Qiang,SU Wei-feng,LIU Yi

Computer Science. 2011, 38 (7): 181-184.

Abstract

PDF(450KB) ( 367 )

RelatedCitation | Metrics

An efficient algorithm FCLBN for learning Bayesian network from small scale dataset was proposed. FCLBN uses the method of bootstrap to rcsample from the small scale dataset, and estimates the high confidence features of thesource small scale dataset from the Bayesian networks learned from the re-sampling small datasets. The high confidence features arc taken to guide the search of the best Baycsian network on the source dataset. After being evaluated on the standard benchmark dataset, FCLBN is applied to predict yeast protein localization. The result of the experiments indicafes that the FCLBN algorithm can learn relatively accurate network from small scale dataset.

Community Detection in Complex Networks Based on Vertex Similarities

JIANG Ya-wen,JIA Cai-yan,YU Jian

Computer Science. 2011, 38 (7): 185-189.

Abstract

PDF(433KB) ( 894 )

RelatedCitation | Metrics

One of statistical characteristics in complex networks is a community structure. Detecting communities in networks has aroused great interest among researches in recent years. Actually, community detection is very similar to the classical cluster analysis in machine learning field. Thus, the key point is how to define vertex similarities in complex networks. We first proposed an algorithm named SUN based on vertex similarities. Compared with UN, SUN is much better and faster than UN. Secondly, we used four classical clustering algorithms to detect community structure in networks based on some existing vertex similarity measures. I}he results on artificial networks and real social networks show that the similarity measures based on signal propagation and regular equivalence theory by using the whole topology structure of networks are better than the methods of Jaccard based on local vertex information. Therefore, if vertex similarities arc given well enough,proper clustering algorithms based on similarity matrices can be used to detect community structures fast and effectively in complex networks.

Measurement of PSO Diversity Based on L1 Norm

CHENG Shi,SHI Yu-hui

Computer Science. 2011, 38 (7): 190-193.

Abstract

PDF(418KB) ( 342 )

RelatedCitation | Metrics

A novel PSO population diversity based on norm was defined, which provides useful information of PSO search process. Population diversity based on and norms were analyzed as well as element wised and dimension-wised PSO diversity. Population diversities of PSO with different number of dimensions,different topology structure,and different population sizes were discussed and tested on benchmark functions.

Research of Hybrid Parallel Genetic Algorithm Based on Multi-core Cluster System

WANG Zhu-rong,JU Tao,MA Fan

Computer Science. 2011, 38 (7): 194-199.

Abstract

PDF(533KB) ( 654 )

RelatedCitation | Metrics

In response to the challenge of the traditional genetic algorithms facing the slow pace of the evolution and difficulties of unable to meet real-time requirements in handing large-scale combinatorial optimization problems, this paper proposed the coarscgrained-master-slave hybrid parallel genetic algorithm ( HPGA) model run on multi-core cluster system. Through integrating the features of the message-passing model and the shared-memory model, we used message-passing model-MPI among nodes to correspond coarse-grained PGA, used share-memory model-OpenMP within one node to correspond r工caster-slave PGA and thus cor工ibined the higher parallel cor工iputing ability of rnuhi core cluster system with inherent parallelism of PGA. Realized the HPGA based on two-laycr parallel both process and thread by using hybrid parallel programming models combing MPI and OpenMP on multi-core cluster system. Theoretical analysis and experimental results show that the HPGA model of this paper has high performance and overcomes traditional GA's defects. It put forward an effective and viable solution for using Parallel Genetic Algorithm based on simple multi-core PC cluster to deal with the large-scale combinatorial optimization problems.

OBDD-based Bucket Elimination Algorithm for Constraint Satisfaction Problem

XU Zhou-bo,GU Tian-long,CHANG Liang,LI Feng-ying

Computer Science. 2011, 38 (7): 200-202.

Abstract

PDF(316KB) ( 616 )

RelatedCitation | Metrics

Bucket elimination algorithm is a typical reasoning method for the constraint satisfaction problem(CSP). Aiming at the state explosion problem of bucket elimination algorithm, ordered binary decision diagram(OBDD) technique was combined with bucket-elimination algorithm, and a symbolic 013DI}based algorithm for CSP was proposed. By encoding each variable and each value in the domain as binary variables,CSP was encoded as a propositional satisfiability (SAID) problem,and then CSP was formulated symbolically by OBDD. Based on the ideas of bucket elimination algorithm and the symbolic OBDD representation of CSP, the CSP was solved implicitly by the ANl)operator and the EXIST operator of OBDD，so that the explicit enumeration of states in traditional algorithms was avoided. The simulation results show that the symbolic algorithm is more efficient than both the bucket elimination algorithm and the direct algorithm based on OBDD.

Action Theory Based on the Dynamic Description Logic DDL

CHANG Liang,CHEN Li-min

Computer Science. 2011, 38 (7): 203-208.

Abstract

PDF(552KB) ( 400 )

RelatedCitation | Metrics

There is a gap on expressive power and reasoning ability between the action theories which arc based on first or higher-order logics and the action theories which arc only propositional. As a kind of dynamic extensions of descriplion logics, the dynamic description logic DDL provides an approach for describing and reasoning about actions. An aclion theory based on DDI_ was presented and studied systematically. Firstly, based on a representation of static domain knowledge with description logics, the parameterized atomic action definitions and the parameterized complex action definitions were introduced for describing the knowledge of actions; both of these knowledge, and together with the knowledge on the state of the world,are unified as a DDL-based knowledge representation system. Secondly,many reasoning tasks on the knowledge represented in this system were formally defined; corresponding reasoning mechanisms were also provided. Finally,the application of this action theory for the modeling of intelligent agents was discussed. The action theory based on DDL offers not only considerable expressive power but also attractive reasoning services; it is suitable for the description and reasoning of actions in the environment of the semantic Web.

Cloud Reasoning Method and its Application in Prediction

CHEN Hao,LI Bing

Computer Science. 2011, 38 (7): 209-211.

Abstract

PDF(373KB) ( 343 )

RelatedCitation | Metrics

Uncertain reasoning is an important work in current artificial intelligence research field. The cloud model can establish transformation of uncertainty between the qualitative concept and the quantitative description. I}hc rule generafor constructed by cloud model can describe qualitative rules expressing by natural language effectively, and carry out uncertain reasoning. Finally,an uncertain reasoning approach based on cloud model was used for forecasting electronic products lifetime under practical working conditions, which demonstrated the validity and practicability of the cloud rcaconing method.

Model of Partner Selection for Dynamic Virtual Enterprise Under the Environment of Cloud Computing

ZHANG Yi-wen,NI Zhi-wei,SONG Jie,WANG Li

Computer Science. 2011, 38 (7): 212-215.

Abstract

PDF(340KB) ( 306 )

RelatedCitation | Metrics

The retrieval of massive enterprises on the Internet and the building of UDDI have become a serious obstacle in virtual enterprise construction in terms of medium-small sized enterprises. Via the Internet to provide virtual resource calculating method, cloud computing can provide fast resource deployment and information acquisition, enable mediumsmall sized enterprises to build enterprise alliance for dominance possible. A new model of cooperative partner selection for dynamic virtual enterprise based on cloud computing was proposed, which used parallel screening mechanism to reduce problem solving space, the problems of massive partner selection and building UDDI for medium small sized enterprises alliance were resolved. Finally, the rationality and feasibility of the model are proved with the simulated experimcnts.

Min-Max Multiway Cuts in Several Special Graphs

LI Shu-guang,XIN Xiao

Computer Science. 2011, 38 (7): 216-219.

Abstract

PDF(322KB) ( 632 )

RelatedCitation | Metrics

Given an undirected graph with positive edge weights and a subset of verticescalled terminals, the min-max multiway cut problem is to partition the vertices into clusters such that each cluster contains exactly one terminal and the maximum cost among the clusters is minimizcd,where the cost of a cluster is defined as the total edge weight at the boundary of the cluster. I}his problem is motivated by data plcacement in Peer-to-Peer networks and is a variant of the traditional multiway cut problem. It is strongly NP-hard even when the graph is a tree. For chains and rings, linear time exact algorithms were presented which minimize simultaneously both maximum cost and total cost of the clusters. For trees and graphs of bounded treewidth,a (2-1/2k2)-approximation algorithm was presented where“·the number of terminals.

Self-adaptive Step Glowworm Swarm Optimization Algorithm for Optimizing Multimodal Functions

HUANG Zhcng-xin,ZHOU Yong-quan

Computer Science. 2011, 38 (7): 220-224.

Abstract

PDF(430KB) ( 375 )

RelatedCitation | Metrics

Because the GSO algorithm has slow convergence and low precision defects when optimizing the multi modal function, a self-adaptive step glowworm swarm optimization(SASGSO ) algorithms was proposed in this paper. This algorithm can overcome slow convergence and low precision defects of the GSO algorithm simultaneously it can find all peaks of the multi-modal function. Experiments show that,the SASGSO algorithm has the advantages of simple operation,easy to understand,fast convergence rates and high precision.

Discrete Differential Evolution with Learning Mechanism

ZHOU Ya-lan,ZHU Yao-hui,ZHANG Jun

Computer Science. 2011, 38 (7): 225-227.

Abstract

PDF(327KB) ( 893 )

RelatedCitation | Metrics

How to apply differential evolution to discrete field is currently a hot research problem in this area. Estimation of distribution algorithms build a probability model which characterizes the distribution of the current promising solutions in the search space and generates new solutions according to the model. Using the global learning of estimation of distribution algorithms, a discrete differential evolution with learning mechanism was proposed for multidimensional knack problem. Simulation results show that the proposed algorithm has good performance.

Research on Properties of Approximations in Set-valued Information Systems when Objects Vary with Time

GU Xiao-guang,LI Tian-rui,CHEN Hong-mei,LIU Yong-wen

Computer Science. 2011, 38 (7): 228-230.

Abstract

PDF(230KB) ( 324 )

RelatedCitation | Metrics

The set valued information system is an extension of the general model of the information system. In real-life applications,the information system changes dynamically according to the variation of objects. I}his paper discussed the changing rules of approximations under the tolerance relation when objects vary with time and presents properties of the variation of approximations in set valued information systems and set valued decision information systems. Examples are given to illustrate the validity of the proposed properties.

Convergence Analysis of Two-membered Evolution Strategy

ZHANG Yu-shan,HAO Zhi-feng,HUANG Han

Computer Science. 2011, 38 (7): 231-234.

Abstract

PDF(305KB) ( 360 )

RelatedCitation | Metrics

The theoretical investigation to Evolutionary Algorithms, e. g. convergence analysis, runtime analysis, is currently a hot topic, and the related theoretical results arc few despite many experimental results. I}his paper established a homogeneous Markov process model associated with(1+1)ES .It is proved by means of the theory of Markov process with continuous state that the Markov process associated with(1+1) ES has exponential ergodicity in a class of continuous optimization problem,hence the(1+1)ES can finally converge to the optimal solution with probability 1 when solving such a problem. The proposed analytic approach provides a new thought to the theoretical research of evolutionary algorithms.

Multi-instance Clustering Based on EMD

LI Zhan,PENG Jin-ye,WHEN Chao

Computer Science. 2011, 38 (7): 235-239.

Abstract

PDF(431KB) ( 540 )

RelatedCitation | Metrics

In the setting of multi-instance learning, each sample is represented by a bag composed of multiple instances.Previous studies on clustering mainly deal with the single instance in traditional learning setting, so it can't be applied to multi instance problem directly. In this paper, based on earth mover's distance, a novel multiplcinstance clustering algothrim named ECMKIL was presented. Firstly we calculated the bag's instances' similarity, emerged the similarity ones, then regarded the two bags' instances as suppliers and consumers, calculated the goods and capacity. To deal with the supplier-consumer imbalance problem, we solved it by multiplying the goods. Finally, used k-medoids to cluster the multi-instance data. Experimental results on MUSK, Corel and SIVAL data set indicate that the ECMKIL method is effective.

Maze Problem's Solution Based on Timed Petri-net

YE Jian-hong,YE Shuang,SONG Wen,SUN Shi-xin

Computer Science. 2011, 38 (7): 240-242.

Abstract

PDF(330KB) ( 390 )

RelatedCitation | Metrics

M-TdPN, an improved search algorithm, was presented in this paper in view of the shortage of traditional solutions of maze problem. Redundancy points were eliminated to reduce the space complexity. The timed Petri net was applied to simulate the maze. Using the concurrency, every token has its own traces. The flags attached on tokens in the end place arc the feasible paths of maze problem. hhe searching efficiency is improved effectively. Results of experimental showed that the M-TdPN algorithm has better search results in case of more complicated maze or more impasse vertexes specially.

Dynamic Description Logic Based Approach to General Service Knowledge Modeling

ZHOU Ping,WANG Wei-min,LUO Wei-min,CHEN Qi-ming,ZHENG Yu-fei,CAO Cun-gen

Computer Science. 2011, 38 (7): 243-249.

Abstract

PDF(527KB) ( 323 )

RelatedCitation | Metrics

There is no uniform solution to knowledge management though various kinds of methods have been proposed. This paper introduced a sort of general structural modeling method of service industry drawing on the experience of Dublin Core Metadata Elements with a thought of hierarchy, for modeling knowledge of service industry. In order to describe the model clearly, Dynamic Description Logic(DDI)was used to model actions and state changes in the model,which can make knowledge description and management more rigorously.

Approach of NON-ISA Concept Relation in Ontology Concept Similarity Computation

WANG Xiao-man,YAN Jing-jing

Computer Science. 2011, 38 (7): 250-254.

Abstract

PDF(383KB) ( 365 )

RelatedCitation | Metrics

In the process of ontology concept similarity computation, there is few about the approaches of dealing with non-isa concept relation. Considering the existence of non-isa concept relation in ontology, summarize some traditional approaches of concept similarity computation, proposed a novel approach which adapts the non-isa relation. Compute the overlay-grade of ontology directed acycline concept graph information content based on tversky model, and combined the semantic distance approach, then sumed the weighted value. Experimental results show that our approach is effectively for isa concept relation and applicable for non-isa concept relation.

SWO: A Fast Search Algorithm Based on Small World Effect

HUANG Gang,LI Jin-hang,JIA Yan

Computer Science. 2011, 38 (7): 255-260.

Abstract

PDF(566KB) ( 710 )

RelatedCitation | Metrics

This paper designed a fast search algorithm called Small World Optimization(SWO) which was inspired by the hierarchical categorization tree model and multi categories method based on small world theory.The solution space can be divided into the hierarchical categorization tree model using mask rule in binary coding. Two bijective mapping solution space were adopted to establish multi categories method. SWO can find the optimal solution in the designed small world network by short and long distance neighbor relationship as pushing the mail to target. SWO was tested via a benchmark test functions in a simulation and the corresponding results show that two bijective mapping space can avoid the algorithm falling into early maturity and the long distance neighbor relationship can accelerate the convergence rate. Compared with the most popular optimization algorithm as genetic algorithm(GA)，Particle Swarm Optimization (PSO) and Difference Algorithm(DE)，the SWO algorithm is endowed with faster convergence ability to solve complex optimization problems.

Implement Method of Desktop Environment Supporting Multitask Based on Embedded System

YANG Zhi-yi,HUI Yong-li,LI Shi-ning,LI Zhi-gang

Computer Science. 2011, 38 (7): 261-264.

Abstract

PDF(337KB) ( 499 )

RelatedCitation | Metrics

No matter personal consumer electronic products,or automobiles entertainment systems, they all cry for the desktop environment supporting multitask. User-friendly multitask support brings easy and convenient experience. A strong embedded ecosystem should provide basic desktop functions and applications for daily needs as well as tools and documentation for developers to write stand-alone applications for the system. With Daemon thread and well-known Pipe mechanism, a desktop environment supporting multitasks was implemented, and the paper introduced a method to develop the SDK for programmers.

Application of L-M Optimized BP Algorithm in Short-term Power Load Forecast

DAI Xiao-hong,WANG Guang-li

Computer Science. 2011, 38 (7): 265-267.

Abstract

PDF(223KB) ( 358 )

RelatedCitation | Metrics

Analyzing the deficiency of the traditional BP algorithm, combining Levenbery-Marquardt optimized algorithm and a neural network forecasting method, this paper put forward a L-M optimized BP algorithm, which quickens the train, improves stability and avoids trapping into local minimum. For some area power supply load of Power Corporation in somewhere, a short term load forecast was simulated based on L-M optimized BP algorithm. Analyzing the simulation results, it shows the L-M optimized BP algorithm has better forecast precision and adaptive capacity.

Simulation of Treading on Grassland

QIU Hang,CHEN Lei-ting,CAI Hong-bin,Jim X.Chen

Computer Science. 2011, 38 (7): 268-272.

Abstract

PDF(543KB) ( 404 )

RelatedCitation | Metrics

Grass is one of the nature elements that is difficult to simulate in real-time due to its large number and wide covering range. In view of the deficiencies of existing methods, a method for simulation of treading on grass was proposed. Constructed largcscale static grassland based on hybrid representation. Computed the deformation of grass close to the viewer by using GPU-based real time collision detection algorithm and force vector diffusion mechanism. Simulated the deformation of tussock by adjust the slope of billboard. In order to avoid popping between different levels, a transition scheme was implemented. Experiments show that this method not only can realistically simulate the effect of treading on grass,but also can overcome the deficiencies of the traditional methods.

Motion-based 3D Reconstruction and Applications to Virtual Reality

LI Lin-yao,ZHANG Zhao-xiang,WANG Yun-hong,WANG Chao

Computer Science. 2011, 38 (7): 273-276.

Abstract

PDF(344KB) ( 488 )

RelatedCitation | Metrics

This paper presented a method to robustly and rapidly reconstruct general 3D scene from 2D images in camera, with the appropriate virtual application to show the effects of reconstruction. We presented novel camera calibralion and scene reconstruction using scale-invariant feature points. Scale invariant feature has impressive matching ability in camera's vidco,which greatly reduces the restriction of facility in camera calibration, broadens the area of video reconstruction. With some optimization, we enhanced the accuracy of reconstruction and speed. At last, we processed the visualized application, which shows the effects of our system. The result illustrates that our system is capable of producing accurate and fast scene structure from video in a very few seconds.

Stereo Matching Algorithm Based on Uncalibrated Color Images for 3D Reconstruction

HU Yan,GENG Guo-hua,ZHOU Ming-quan,WANG Xiao-feng`

Computer Science. 2011, 38 (7): 277-279.

Abstract

PDF(372KB) ( 348 )

RelatedCitation | Metrics

This paper proposed an stereo matching algorithm for 3D Reconstruction based on uncalibrated color images.First,this algorithm referred a limiting factor to eliminate the clustering phenomenon of Harris corners,and correlative matching method based on color was used for match. Then, removed the obvious outlier points according to the principle of slope consistency, and used RANSAC algorithm to estimate Fundamental matrix, at the same time, excluded outlicr points. Finally,found unmatching points by using Fundamental matrix The experimental results show that object structure can be well reconstructed by using this algorithm, and it is an effective stereo matching algorithm for 3D Reconstruction.

Multi-spectral Remote Sensing Image Registration Based on Feature Point

XU Li-yan,WANG Jing,QIU Jun,SUN Quan-sen,XIA De-shen

Computer Science. 2011, 38 (7): 280-282.

Abstract

PDF(363KB) ( 401 )

RelatedCitation | Metrics

A registration method for multi-spectral remote sensing image based on feature points was proposed. First,a two-degree regular mesh was formed on the image, and feature grids were chosen according to entropy and uniformity principle. Then feature points were detected by Forstner operate in feature grids. Due to the characteristic of multi spectral images, the course matching step was achieved by Cross-Correlation theory, and improved space distance constraint was used in fine matching step. Finally, registered image was obtained by affine transformation. RMSE was applied to evaluate the method. The experimental results clearly indicate that the approach we proposed is efficient with sub-pixel precision.

Bone Marrow Cell Segmentation Based on Color Feature Weighted Filter

HAN Yan-fang,YANG Na,MIAO Yan,XU Bo-qing

Computer Science. 2011, 38 (7): 283-286.

Abstract

PDF(352KB) ( 389 )

RelatedCitation | Metrics

Color feature plays an important part in bone marrow cell segmentation as well as classification. Firstly, a color feature extraction and analysis method was discussed based on the relationship among r, g, b values of the sampling data from known classes. Then,fcm segmentation and multi-level thresholding were used for cell segmentation considering the color features. With respect to the defect of loss edges and holes inside, a weighted range filter was proposed by introducing one coefficients called in-neighbor consistency. Experiments were done and results show that this approach is simple and effective.

Multi-sensor Image Registration Algorithm Based on SIFT Points and Canny Edge Features Matching

WANG Wan-tong,HAN Zhi-gang,LIU Peng-fei

Computer Science. 2011, 38 (7): 287-289.

Abstract

PDF(285KB) ( 573 )

RelatedCitation | Metrics

To resolve multi-sensor remote sensing image registration, an algorithm combining scale invariant feature transform(SIFT) point features and Canny edge feature matching was proposed. First,this algorithm uses SIFT to extract point feature and conduct image coarse registration. Then after getting the initial affine transformation parameters,this algorithm uses Canny algorithm to extract edge feature and then uses the cost function for edge matching. Last, after getting the effective matching feature points by filtering outliers, this algorithm conducts image accurate registration. I}his algorithm combines the advantages of SIFT and Canny, and solves the problem that the multi-sensor remote images are difficult for correct registration because of geometric differences and radiometric differences. Experimental results show that this algorithm has strong robustness, and achieves high registration accuracy.

Fast Calculation of CT Simulation Projection Based on Pixel Phantom

ZHANG Shun-li,ZHANG Ding-hua,CHENG Yun-yong,ZHANG Xiao-bo

Computer Science. 2011, 38 (7): 290-293.

Abstract

PDF(361KB) ( 472 )

RelatedCitation | Metrics

Improving the calculation speed of simulation projection has been the important research of CT(computed tomography) simulation technique. Aiming at the projection calculation of pixel phantom, a fast algorithm of generating simulation projection was proposed. Firstly,the original pixel index and its vertical distance intersected by the ray and the pixel phantom boundary were determined. Then, the other pixel index and intersection length intersected by the ray were calculated in turn using the distance, the projection data was obtained from cumulative sum simultaneously. Experimental result shows that the proposed algorithm can improve the calculation speed of projection greatly, we can get a speedup about 7 times compared with the Siddon algorithm. The reconstructed result also shows the proposed algorithm is correct.

Research on the Set Redundancy Compression of Document Image Based on Template Difference

YU Ping,YANG You,SHANG Jin

Computer Science. 2011, 38 (7): 294-297.

Abstract

PDF(386KB) ( 367 )

RelatedCitation | Metrics

Much redundant information between document pages exists in the document image information system. The research on the compression method based on the pagcpage statistical features is significant. Set Redundancy Compression(SRC) is such a technique. It reduced the total entropy of the whole image set through utilizing the image page's similarity. Compression based on template differential(CTD) is an improved SRC. The similar image set was constructed by the template. "hhe coding performance was improved by adding the template image to the Min-Max Differential (MMI))coding/decoding model. It was proved theoretically that CTD's coding performance is higher than MMD's. It was demonstrated by experiments that both the CTD and MMI)are benefit to increase the compression ratio of image set, furthermore, and ChD is better than MMD.

Research of the SIMD and Vector Math Library

XIE Qing-chun,ZHANG Yun-quan,WANG Ke,LI Yan,XU Ya-wu

Computer Science. 2011, 38 (7): 298-301.

Abstract

PDF(371KB) ( 1543 )

RelatedCitation | Metrics

Firstly,we introduced the single Instruction Multiple Data(SIMD) vectorization technology and the features separately, based on the processors of Intel AMD and IBM Cell. Secondly, some vectorization functions were tested in these three platforms, which were deve-loped by the three vendors separately. Our test results show that we achieve high performance with the technology of the vectorization, compared to the traditional methods of the scalar calculation.Especially, the speedup of the Cell SDK functions is 10 on average, which were achieved by the help of many processing elements and the special processor structure. Lastly,we also found that there are some differences between the vectorial functions,which are in different vector math libraries. We analyzed that there are some reasons caused the difference betwecn the math functions in different platforms, such as processor structure, the number of the processing elements, memery accessing and so on.

Computational Emergence and its Constrained Generating Procedure Model

ZHANG Hai-su,ZHANG Song-lin,CHEN Gui-sheng

Computer Science. 2011, 38 (7): 302-305.

Abstract

PDF(366KB) ( 475 )

RelatedCitation | Metrics

The complexity of computational models can also put up some emergence properties around several certain critical values. I}his paper surveyed some computational emergence behaviors, and built their Constrained Generating Procedure(CGP) model. We introduced the CGP mechanic participating times, participating depth and average participaling degree, and then analyzed the critical values of emergence properties on 3 typical Turing machine computation processes.