Computer Science

Survey of Data Transmission Reliability in Wireless Sensor Networks

ZHU Xiao-juan,LU Yang,QIU Shu-wei and GUAN Jun-ming

Computer Science. 2013, 40 (9): 1-7.

Abstract

PDF(815KB) ( 1145 )

References | RelatedCitation | Metrics

The reliability of data transmission affected by resource-constrained and environmental factors in wireless sensor networks is lower．However,it is an important indicator to evaluate network performance in some application fields such as national defense,industrial control and so on．Firstly,the significance of the reliability of wireless sensor networks was introduced．Then the recent research achievements were reviewed from two aspects:the reliability estimation and the technology of improving reliability of data transmission,and these achievements were classified and compared．Finally the future directions of reliable data transmission in wireless sensor networks were pointed out.

Survey on Real-time Traffic Classification

BAI Jun,XIA Jing-bo,WU Ji-xiang,REN Gao-ming and ZHAO Xiao-huan

Computer Science. 2013, 40 (9): 8-15.

Abstract

PDF(743KB) ( 4577 )

References | RelatedCitation | Metrics

Real-time traffic classification,identifying and classing online traffics according to their applications,is of fundamental importance to network managements,operations and correlative studies．The paper first analyzed the research statusquo of real-time traffic classificaiton from different levels.The definition and evaluation metric of real-time classif-caiton were given to address the situation that there is no measurable and accepted metrics.Next,the main techniques and research progresses of real-time traffic classifi-cation were summarized while the real-time property of eath techniques were analyzed from flow features and machine learning shcemes.Finally,we pointed out possible future research directions in real-time traffic classification according to the demand of future network.

Research on Software Defect Classification and Analysis

WANG Bin,WU Tai-wen and HU Pei-pei

Computer Science. 2013, 40 (9): 16-20.

Abstract

PDF(787KB) ( 1026 )

References | RelatedCitation | Metrics

Software defect is an inherent component of software products,and how to manage,decrease or prevent defects plays an important role in improving software quality and reducing the cost of projects．This paper introduced software defect from two aspects:defect classification and defect analysis,analyzed the advantages and shortcomings of various defect classifications,and then summarized the main research directions and methods of defect analysis．Finally,the strategy how to select suitable defect research methods was discussed．

New Model for Interconnection Networks:Multipartite Group-theoretic Model

SHI Hai-zhong

Computer Science. 2013, 40 (9): 21-24.

Abstract

PDF(317KB) ( 776 )

References | RelatedCitation | Metrics

An interconnection network is an important partite of a supercomputer．In 1989,S.B．Akers et al,proposed a group-theoretic model for interconnection networks and designed many interconnection networks,such as,star network,bubble sort network,etc．In particularly,star networks own many better performances than the popular n-cubes．However,they have a weakness:the size of all above networks is n!,that is,n! is very speedly increasing with adding of n．This results in that the scalability of supercomputers built by the interconnection networks is very difficult．That is,the scalability of the supercomputer isn’t good．We proposed a multipartite group-theoretic model based on the the group-theoretic model．By this model,we designed many interconnection networks,such as,(n,k)-multipartite star networks,(n,k)-bubble sort networks,etc．Furthermore,we showed star network is(n,1)-multipartite star network and(n,k)-multipartite star networks own following better performances:the size of the network increases with fixed increment,its diameter increases slowly,its dgree is fixed．Other(n,k)-multipartite networks designed here own also the performances.

Resources Aggregation Model towards Domain-specific Scientific Data Cloud

GE Jing-jun,HU Chang-jun,LIU Xin,LI Yang and LIU Zhen-yu

Computer Science. 2013, 40 (9): 25-29.

Abstract

PDF(1341KB) ( 568 )

References | RelatedCitation | Metrics

Data center as the core facilities of the scientific data resources,is becoming overly complex,expensive and inefficient.Along with the large scale application and the increase number of the users,the connectivity,stability and safety of the data center become a serious challenge.Researches on resource integration,automatic deployment and resources centralization management in data center,are concerned by the current academic and business fields.This paper constructed domain-specific scientific data cloud to connect multiple decentralized and self-organizing data center into a virtual data network.It provides environment and service support for scientific data access,data integration,data sharing and data management,in order to realize data resources gathered,the large-scale scientific data sharing and unified mana-gement.And on this basis,this paper studied the data center agent model,heterogeneous source sharing model and resources communication model emphatically.Study of these models plays a critical roles to improve the data center scala-bility and fault tolerance,to reduce the data center operation and maintenance cost,and to realize the transition from the traditional data center to cloud data center.Finally,this paper introduced the resources aggregation model towards domain-specific scientific data cloud into the development,deployment,operation and monitoring management of the oil and gas well scientific data sharing service platform.The practice demonstrats that the open scientific data cloud model technology is feasible,and has an important reference significance and application value to promote scientific data integration,sharing,and management research.

Trusted Compiler for Safe Subset of C Language

WANG Lei,SHI Gang,DONG Yuan,BAI Xiao-ying and WANG Sheng-yuan

Computer Science. 2013, 40 (9): 30-34.

Abstract

PDF(710KB) ( 665 )

References | RelatedCitation | Metrics

Using the safety standards of safety-critical areas as the basis,safety-related software coding and compiling requirements as the guideline,this paper introduced the following aspects of research and exploration:first analysed the formal verification technique for trusted compiler,especially emphasized the CompCert verified compiler which is widely concerned,then constructed test cases according to safe subset of C language for space armament software to test and assess the CompCert compiler with the new method．Based on the test results,the project added unsupported features of C language to CompCert and cuts off the features which do not meet the safe subset of C language requirements,built a verified compiler for the safe subset of C language at last．The experimental results show that the compiler we implemented complies with safe subset of C language,but it do not increase the execution time of C code.

Design and Implementation of Synchronization for Multithreading Based on Hardware Locks

LI Chun-jiang,TANG Tao and YANG Can-qun

Computer Science. 2013, 40 (9): 35-37.

Abstract

PDF(338KB) ( 1413 )

References | RelatedCitation | Metrics

Hardware locks provide the atomic operation mechanism of "load and add one" or "load and subtract one" by simple load instructions．Firstly,we introduced the implementation of hardware locks in general purpose multi-core multi-threaded FT processors,and compared with the software lock mechanism,then described the method for synchronization of multi-threading using hardware locks．In the GNU OpenMP runtime library,we designed and implemented the synchronization mechanism based on hardware locks for FT processors．Finally,we evaluated the performance of typical OpenMP programs using hardware locks versus using software locks and gave some valuable analysis.

Extension to OpenMP Task Scheduling Mechanism for DSWP Parallelization and its Implementation

LIU Xiao-xian,ZHAO Rong-cai and DING Rui

Computer Science. 2013, 40 (9): 38-43.

Abstract

PDF(507KB) ( 701 )

References | RelatedCitation | Metrics

While multicore processors increase throughput for multi-programmed and multithreaded codes,many important applications are single threaded and thus are not benefited．Automatic parallelization techniques play an important role in migrating singe threaded applications to multicore platform．Unfortunately,the prevalence of control flow,recursive data structures,and general pointer accesses in ordinary programs renders the existing techniques unsuitable．Ottoni et al．proposed an automatic parallelization algorithm called Decoupled Software Pipelining(DSWP)to exploit fine-grained pipeline parallelism at the instruction level．But it requires knowledge of micro-architectural properties and hardware support of a communication channel and two special instructions．The improved DSWP algorithm based on OpenMP increases the parallel granularity and does not rely on hardware support any more,but the existing OpenMP task scheduling mechanism cannot satisfy the need of DSWP．A new binding clause for the task construct in OpenMP was proposed to extend the task scheduling mechanism．It guarantees the correctness of the OpenMP DSWP parallelization．The new clause is implemented in the GCC runtime library libgomp,which provides support for the compilation of OpenMP DSWP programs．The experimental results show that loops failed to be parallelized by existing techniques can be parallelized by the improved automatic parallelization algorithm and gain significant performance improvement on dual-core CPU．The average performance speedup is up to 1.23．Compared with Intel and Open64compilers,the compiler with the improved algorithm can increase execution efficiency evidently and the average speedup of the OpenMP DSWP programs generated by it increases more than 22% and 26%.

Similarity-driven Fine-grained Parallel Task Reconfigurable Algorithm

HAO Shui-xia,ZENG Guo-sun,MA Xiao-xin and XU Jin-chao

Computer Science. 2013, 40 (9): 44-50.

Abstract

PDF(561KB) ( 703 )

References | RelatedCitation | Metrics

Heterogeneous computing is a novel trend of high-performance technology．Mismatch between parallel tasks and architecture leads to realize high performance hardly．Reconfiguration technique brings an opportunity for it．In order to realize their matching,this paper provided similarity-driven reconfigurable algorithm．Firstly the paper gave definitions of graph-similarity and graph-reconfiguration．Secondly through building a measurement between task graph and fixed architecture,the paper provided a graph-reconfiguration problem and a similarity-driven reconfigurable algorithm．The algorithm includes three processes:nodes-fused,node_edge_reconfiguration and node_edge_refined．Finally the paper gave a test on Greenwich benchmark with the typical architecture．The results show that the algorithm can realize the optimal match between the parallel tasks and the architectures on heterogeneous computing.

Model Based Dynamic Routing Algorithm in Delay/Disruption Tolerant Network

WANG Xing-wei,WEI Yong-tao,HUANG Min and WANG Jun-wei

Computer Science. 2013, 40 (9): 51-54.

Abstract

PDF(298KB) ( 659 )

References | RelatedCitation | Metrics

In this paper,the non-topological information routing algorithms in DTN was analyzed．Then,a model based dynamic routing algorithm(MDRA)for DTN was proposed． Aiming at the difficulty of getting node location in advance,a process was designed to dynamically update node visiting probability and encountering probability,which can help adaptively choose the relaying nodes．Simulation implementation and performance analysis of the proposed MDRA were done．Simulation results show that with the MDRA algorithm running in DTN,the delivery ratio of the transferred message is increased and the time delay is reduced,and at the same time the generation of numerous redundant copies of the transferred message is avoided.

Evaluation and Analysis of Load Balancing Mechanisms in Storage Systems

LUO Xiang-yu,WANG Yun and CHEN Xiao-mei

Computer Science. 2013, 40 (9): 55-60.

Abstract

PDF(548KB) ( 706 )

References | RelatedCitation | Metrics

For a large-scale storage system,load balancing is an important way to improve the utilization ratio of system resources and the energy efficiency．It could greatly reduce the provisioning and operational cost．However,most exis-ting studies are focused on the design or improvement of a specific method while mading no evaluation or comparison among different methods．In this paper,extensive simulations were conducted to evaluate the existing load balancing methods．The results reveal the main challenge and unsettled problem in the area．Besides,theoretical basis was founded for selecting the appropriate load balancing method in practice.

Task Scheduling Algorithm Based on Improved NSGA-Ⅱ for Wireless Ad-hoc Networks

YANG Hong-li and GUO Hua

Computer Science. 2013, 40 (9): 61-63.

Abstract

PDF(338KB) ( 542 )

References | RelatedCitation | Metrics

To solve the problem of the lower efficiency of task-performing caused by the mobility and failure-prone of Ad-hoc nodes,a multi-object optimization task scheduling algorithm(MOTA)was proposed for wireless Ad-hoc networks．This algorithm tries its best to make less make span,but meanwhile,it also pays much more attention to the fai-lure probability and the energy-consuming of nodes．MOTA avoids the task assigned to the failure-prone node,which effectively reduces the effect of failed nodes on task-performing.Simulation results show that the proposed algorithm can trade off these three objectives well．Compared with the traditional task scheduling algorithms,the simulation experiments obtain better results．

Memory Optimization Algorithm of Migration Based on Xen Virtual Machine

CHEN Ting-wei,ZHANG Pu and ZHANG Zhong-qing

Computer Science. 2013, 40 (9): 64-67.

Abstract

PDF(459KB) ( 643 )

References | RelatedCitation | Metrics

In order to migrate the virtual machine in the cloud computing environment,Xen compares the send-page bitmap with the skip-page bitmap when migrating,and determines whether memory pages are retransmitted or not．Aiming at the problem that multiple retransmissions increase network transmission overhead from the comparison,this paper proposed a memory optimization algorithm based on AR model．According to all recorded intervals in which a memory page is modified,the new algorithm predicts the next modification time of the memory page and retransmits the memory page when the next modification time is greater than a threshold．The results show that the optimization algorithm shortens the migration and network overhead of the virtual machine and ensures the network applications of other vir-tual machines on the same server.

Data Delivery Scheme of DTMSN Based on Comprehensive State of Node

WU Qian and YANG Ji-lin

Computer Science. 2013, 40 (9): 68-72.

Abstract

PDF(432KB) ( 616 )

References | RelatedCitation | Metrics

To better solve the problem of data collection in DTMSN(delay tolerant mobile sensor network,DTMSN),this paper proposed a new data delivery scheme-NCSD(Node Comprehensive State Data Delivery Scheme).NCSD calculates the node comprehensive state,according to fuzzy comprehensive evaluation when delivering messages.To optimize the management of information copy,NCSD adopts dynamic message queue and decides the principle of abandoning messages based on their survival time.Simulation results show that compared with the existing algorithm of DTMSN data transmission,NCSD has a higher delivery ratio in data transmission.

NetFlow Based on Parallel Single Multi-linked Lists

YE Di-qiu,CHENG Dong-nian and LI Yu-feng

Computer Science. 2013, 40 (9): 73-77.

Abstract

PDF(430KB) ( 644 )

References | RelatedCitation | Metrics

In the high speed network,the bottleneck of NetFlow is the memory access．The efficiency of traditional list item operation "read-process-write" mode is only 36%,and the important factors which restrict the efficiency are read and write conversion as well as read and write delay．To solve this problem,continuous reading and writing method was proposed. It makes the efficiency of traditional list item operation increase to 90% or more by merging multi-delay and multiplexing read and write conversion．We proposed the parallel single multi-linked list method to handle with the inactive timeout flows．Continuous reading and writing method is applied to deal with the inactivity timeout flow through multiple list alternating operation．Theoretical analysis and experimental results show that the parallel single multi-linked list method can be applied in the link OC-768(40Gbps)to manage tens of millions of list items,which is significantly superior to the management ability of auxiliary storage method and doubly linked list method in the link OC-192(10Gbps)to manage millions of list items.

Efficiency Filtering Scheme for False Data in Wireless Sensor Networks

ZHAO Jin-guo,ZHU Ling-zhi,LUO Qing-yun and LIANG Jun-bin

Computer Science. 2013, 40 (9): 78-83.

Abstract

PDF(519KB) ( 613 )

References | RelatedCitation | Metrics

The fundamental reason for low efficiency of false report filtering in wireless sensor networks is the lack of security assurance of the secrets while improving the degree of secret sharing.This paper presented an efficient false report filtering scheme.First we divided the network into clusters and constructed a cluster heads-tree,and then form relationships between the cluster heads were formed.By taking into account of the burden factor of nodes in the keys dissemination procedure,the scheme guarantees that the forwarding nodes closer to a cluster hold more authentication keys for the cluster than those nodes farther from it do,hence,the degree of secret sharing can be elevated.What’s more,several keys from the same oral cluster are hold by different nodes in a forwarding cluster so as to guarantee the securityof the keys.Analysis and simulation results show that the filtering procedure based on grouped cluster-keys outperforms existing schemes in terms of energy saves,storage overhead and filtering performances.

Resource Deployment Model Design in Cloud Computing under Bursty Workloads

CHEN Peng,MA Zi-tang,SUN Lei and SUN Dong-dong

Computer Science. 2013, 40 (9): 84-88.

Abstract

PDF(445KB) ( 633 )

References | RelatedCitation | Metrics

Aiming at the degrading system performance that bursty workloads bring in cloud computing,BWA(Bursty Workloads Allocation)model was proposed to resolve resource deployment problems．Firstly,BWA’s workload monitor model is responsible of monitoring the variation of tasks in cloud computing,judging the on-off in real time about bursty workloads．Next,BWA tries to avoid the appearance of partial hot dots by using new deployment strategy to speed up the system response．At last,the forecasting algorithm is used to improve response speed by deploying the computing nodes in advance．The results of simulation in CloudSim prove that using BWA model can obtain better system perfor-mance．

Energy-efficient Deployment Scheme Based on Queue Model and Differential Evolution Algorithm for Web Cluster

XIONG Zhi,LIU Wei-jun and CUI Zhang-wei

Computer Science. 2013, 40 (9): 89-92.

Abstract

PDF(320KB) ( 606 )

References | RelatedCitation | Metrics

Energy-saving of Web cluster is an urgent problem to be solved,hence an energy-efficient deployment scheme for Web cluster was proposed．The scheme adopts both CPU dynamic frequency scaling and server node dynamic swi-tching on/off mechanisms to conserve energy．It uses M/G/1PS queue model to simulate Web server,and transforms the energy-efficient deployment problem of Web cluster to a constrained programming problem．Aiming at the problem,a differential evolution algorithm was proposed to solve it．During the iteration processes,the greedy idea was used to correct the solutions which do not satisfy the constraints．The solutions got by the greedy algorithm were put into the initial population to fasten the convergence speed．Simulation tests show the feasibility and effectiveness of the proposed algorithm.

Prevention of Code Reuse Attacks through Return Address Protection

CHEN Lin-bo,JIANG Jian-hui and ZHANG Dan-qing

Computer Science. 2013, 40 (9): 93-98.

Abstract

PDF(890KB) ( 650 )

References | RelatedCitation | Metrics

Despite the numerous prevention and protection techniques that have been developed,the exploitation of memory corruption vulnerabilities still represents a serious threat to the security of software systems and networks．Because of the adoption of the write or execute only policy (W⊕X) and address space layout randomization (ASLR),modern operate systems have been strengthened against code injection attacks．However,attackers have responded by employing code reuse attacks,in which software vulnerability is exploited to weave control flow through existing code base.Solutions targeting different aspects of the attack itself have got some success,but none of them can be a silver bullet．Under this situation,a novel defense technique was presented in order to prevent code reuse attacks,especially return-oriented programming (ROP) attacks．This new defense technique,which was benefit from the protection of return address,could dynamically prevent the execution of gadgets ending with 0xc3．Without requiring access to side information such as source code or debugging information,this defense technique could prevent ROP attacks with low performance overhead.

Study on Collaborative Defence Model of DRDoS

ZHANG Ming-qing,CHUAI Ying-cai,TANG Jun and KONG Hong-shan

Computer Science. 2013, 40 (9): 99-102.

Abstract

PDF(599KB) ( 629 )

References | RelatedCitation | Metrics

According to the defects of existing defence method on response lag and incomprehensive filtering,one collabo-rative defence model of DRDoS was proposed,based on collaborative defence theory.A collaborative self-learning algorithm was designed,which made it possible to share defence knowledge with other agents and could filter attack flows from external network．Intrusion tracking technology was used,together with intrusion detection and intrusion filtering,attack source in the internal network would be located and blocked.Simulation results show that this model could timely detect and effectively eliminate attack flows from both Internal and external network.

Improved Gossip Algorithm Based on Mobile P2P Networks

ZHANG Guo-yin,LI Jun,WANG Xiang-hui and XU Guo-kun

Computer Science. 2013, 40 (9): 103-105.

Abstract

PDF(339KB) ( 1126 )

References | RelatedCitation | Metrics

With the wide use of intelligent mobile terminals,the research of mobile peer-to-peer networks came to be in-depth continuously．Although the classic Gossip algorithm can be applied to data dissemination in mobile peer-to-peer networks,it cannot adapt to the requirement of mobile networks,especially to the adaptability of churn．Therefore,in order to realize the effective transmission of topology control information and to maintain replica consistency of the node resource lists, an improved Gossip algorithm based on the specific topology was presented to update the resource lists of nodes which belong to the same k-clique,by dynamically adjusting data dissemination probability of each neighbor node．Simulation results show that data dissemination efficiency of the proposed algorithm is improved significantly than the classic Gossip algorithm．It achieves the efficiency of flooding data dissemination strategy and ensures the lower network load.

Message Policy Attribute Based Key Agreement Protocol

WANG Yong-tao,FENG Wei-duan,LIU Xiao-nan,SONG Jing and GUO Zhen-zhou

Computer Science. 2013, 40 (9): 106-110.

Abstract

PDF(431KB) ( 568 )

References | RelatedCitation | Metrics

We proposed a message policy attribute based key agreement protocol．Attribute based key agreement protocol has many characteristics,the main of which is fuzzy identification．This paper discussed the application of fuzzy identification in key agreement protocol,and constructed a two-party message policy attribute based key agreement protocol,where the message policy is And-gates on two valued attribute．In addition,the new protocol was proved secure in the standard model under the decisional bilinear Diffie-Hellman assumption(DBDH).

Research on Conflict-measurable Multiple Information Fusion Technology

CHEN Chao,CHEN Xing-yuan,WANG Yong-wei and WANG Yi-gong

Computer Science. 2013, 40 (9): 111-115.

Abstract

PDF(435KB) ( 783 )

References | RelatedCitation | Metrics

As traditional multiple source information fusion system based on D-S evidence theory can not measure conflict and solve other problems,a conflict-measurable multiple source information fusion system was presented．Firstly,we presented a method which takes subjectivity and objectivity into consideration to get BPA based on decision-making table．Secondly,we defined a conflict measure based on conflict value and consistent value followed in the conflict matrix and consistent matrix．Next,we determined the weight value of the evidence in the combination rules based on the avera-ge conflict value of the evidence with others．Finally,we made full use of information produced from conflict evidence,and presented a new combination rules based on conflict information．Simulation experiment shows that the conclusion obtained by the fusion method is not only in accordance with human logical reasoning,but also reduces the uncertainty of the conclusion greatly.

Predicate Temporal Logic Based Description and Detection of Malware Behavior

JIN Ran,FAN Rong-rong and GU Xiao-qi

Computer Science. 2013, 40 (9): 116-119.

Abstract

PDF(311KB) ( 593 )

References | RelatedCitation | Metrics

The behavior based security has become main-stream in the research of malware detection techniques．Although there have been some behavior based malware detection methods introduced in public papers,they are prone to suffer from mimicry attack or shadow attack．Towards these problems,a novel technique using predicate temporal logic to describe malware behavior was proposed in this paper．A variety of relations among system function calls,such as logic combination,precedence,parameter decencies and subject-object associations,can be depicted by one logic formula,therefore our method can describe malware behavior more subtlety and accurately．An algorithm of detecting malware behavior based on the logic was given and its feasibility was justified through real example test.

Research on Network Intrusion Knowledge Base Model Based on Ontology

WU Lin-jin,WU Dong-ying,LIU Sheng-li and LIU Long

Computer Science. 2013, 40 (9): 120-124.

Abstract

PDF(509KB) ( 676 )

References | RelatedCitation | Metrics

In the field of information security,network intrusion knowledge base plays an important role in effective analysis and defense of the illegal invasion,but network intrusion knowledge base construction is one of the difficulties of research．As a conceptual modeling tool to provide knowledge sharing for a specific area,ontology has been widely used in various fields．Because there is no complete network intrusion ontology,the paper aimed to study the ontology-based network intrusion knowledge base model and build network intrusion knowledge ontology．Firstly,on the basis of in-depth analysis of network intrusion technology,the paper gave formalized definition of various types of network actions and the multi-level and multi-dimensional network intrusion knowledge base classification system．Then combining with the principles of ontology modeling,the paper built the network intrusion knowledge ontology composed of the network intrusion knowledge domain ontology,task ontology,application ontology and atomic ontology,giving the logical relationship and organizational structure between them．Finally,through two network scenarios the paper verified the validityof the model in the application of network intrusion knowledge acquisition.

Design and Application of Information Security in EPLC

ZHOU Hao,HUANG Shuang,HUANG Xiong-feng,LI Ke and ZHOU Chun-jie

Computer Science. 2013, 40 (9): 125-129.

Abstract

PDF(406KB) ( 629 )

References | RelatedCitation | Metrics

With the deep integration of industrialization and informationization and the rapid development of system network interconnection,the issues of information security in EPLC attract more attention．This paper proposed a two-level control framework to ensure the information security．Detection based on attack signature database is employed to draw up the control schemes for the identifiable instructions at the inner level．And at the external level,redundancy techniques are used to conquer the unidentifiable instructions which are detected based on system models．Finally,a test EPLC platform on CortexA8-Linux was established to realize and verify the proposed approach,and the experimental results show that the EPLC not only could meet the real-time requirement,but also could detect and process those external instructions well．

Efficient Strong (n,t,n) Verifiable Secret Sharing Scheme

WU Chun-ying and LI Shun-dong

Computer Science. 2013, 40 (9): 130-132.

Abstract

PDF(225KB) ( 863 )

References | RelatedCitation | Metrics

In 2010,Harn and Lin proposed the conception of strong t-consistency and designed a strong (n,t,n) verifiable secret sharing scheme,but the scheme has a low efficiency．This paper proposed an efficient strong (n,t,n) verifiable secret sharing scheme．Based on the property of Vandermonde determinant,the scheme can resist and verify the fraud proposed in Harn scheme．Meanwhile,our scheme doesn’t need to select the kn sub-polynomials．The new scheme with strong t-consistency has the advantage of a low computational complexity.

Pixel Value Replacement Method to Encryption Image Based on Memrisitive Cellular Automata

WANG Xiao-dong,WANG Li-dan and DUAN Shu-kai

Computer Science. 2013, 40 (9): 133-135.

Abstract

PDF(893KB) ( 608 )

References | RelatedCitation | Metrics

Memrisitive cellular automata is a kind of new cellular automata,and its operation can produce transformation of cells state and make its links changed．We analyzed the varied change of cellular automata and the links between the cells,then we designed a new encryption algorithm:use a W×H cells automata arrays as Public-Key,the initial states and respond time as Private-Key,get the links between the cells as a encryption matrix．Meanwhile,we used the pixel value replacement method to encrypt a image．This algorithm has the advantage of confusion and diffusion properties,besides,it owes its advantage in fine security,large key space and diffusion．The algorithm analyzes turn out that this method is of high encryption efficiency and security,which could just meet the digital image demands on security．

VoIP Traffic Identification Based on UDP Statistical Fingerprinting Mixture Models

DING Yao-jun,CAI Wan-dong and YAO Ye

Computer Science. 2013, 40 (9): 136-140.

Abstract

PDF(410KB) ( 786 )

References | RelatedCitation | Metrics

Because it is difficult to identify encrypted VoIP traffic,we proposed a UDP statistical fingerprinting mixture models to enhance the accuracy and stability of VoIP traffic identification． We used the statistical features of UDP flow along with the anomaly score of traffic flow in which the statistical fingerprinting model is used to identify a traffic flow to train a Support Vector Machine(SVM)classification model,and used a hyperplane of high-dimensional space instead of a threshold point to classify the traffic．Because we use both the packet level features and flow level features in our mixture models,the impact of the deviation of traffic features which is caused by the instability of network will be decreased．The results of our experiment show that the precision of VoIP traffic is over 97% in our model,and our model is more stable compare with the statistical fingerprinting model and Support Vector Machine(SVM)．

Secure Palmprint Image Coding Algorithm for JPEG-LS

LI Heng-jian,WANG Lian-hai and ZHANG Jia-shu

Computer Science. 2013, 40 (9): 141-146.

Abstract

PDF(1385KB) ( 673 )

References | RelatedCitation | Metrics

Based on forward-feedback nonlinear dynamic filter(FFNDF),a secure JPEG-LS coder was proposed,and then the proposed algorithm was employed in palmprint image．According to the characteristics of JPEG-LS,FFNDF was employed to generate chaotic stream cipher and then encrypt the coded stream during in the regular mode and run mode．The security of proposed scheme was also analyzed and some other compression and encryption algorithms were also compared．The experimental results show that the proposed palmprint image encryption algorithm has the incomprehensible nature of the encryption with high efficiency and security,as well as low-complexity and no affect on image compression performance.

Multi-user Conjunctive Keyword Search Scheme over Ciphertext

WU Qi and WAN Chang-xuan

Computer Science. 2013, 40 (9): 147-151.

Abstract

PDF(781KB) ( 687 )

References | RelatedCitation | Metrics

Searchable encryption becomes increasingly significant with the prevalence of data outsourcing in cloud computing．To overcome the flaw in Yang et al.’s scheme that users have to provide all keywords,an improved one was proposed in this paper.The correctness,security and performance of the proposed scheme were analyzed in detail．It shows that the new scheme maintains the security properties of Yang et al.’s scheme but allows users to provide part of keywords．The new scheme performs well in terms of computational complexity,and can be applied to cloud computing where “thin client” is demanded．At last,a simple implementation of the new scheme was given.

Certificateless Anonymous Authentication Protocol for Vehicle Ad-hoc Network

XU Jun-feng,DING Xue-feng,SI Cheng-xiang and ZHANG Wei

Computer Science. 2013, 40 (9): 152-155.

Abstract

PDF(424KB) ( 638 )

References | RelatedCitation | Metrics

Vehicular Ad-hoc networks (VANET)have acquired more and more applications in dealing the aggravating traffic problems all over world．And their security is receiving increasing attention．Based on discrete logarithm problem,a certificateless secure vehicle-to-roadside anonymous authentication protocol was proposed．The proposed protocol combines the advantages of the protocol based on traditional public key cryptosystem (no key escrow)and that based on ID-PKC(implicit authentication)．Compared with the protocol at the same security level,the efficiency of the proposed approach is much more efficient．So it is more appropriate for VANET.

Attack Graph Construction Method Based on Intelligent State Transition and Permission Improvement

MA Yan-tu and WANG Lian-guo

Computer Science. 2013, 40 (9): 156-158.

Abstract

PDF(345KB) ( 632 )

References | RelatedCitation | Metrics

In view of the current attack graph in a wide range of network applications with time complexity and not accurate graphical description,this paper put forward a attack graph construction method based on intelligent state transition as well as permission improvement,which adopts the intelligent state transition way to build attack model,based on the model of planning the corresponding attack graph generation algorithm,and introduces the permissions in the attack graph generation algorithm improvement system.The method based on host connections between analysis,based on network attack state graph and network of different host original permissions,access permissions attack improvement path,can adjust the host network transformation condition in the permissions,then ensure network security,corresponding to large-scale complicated network attacks effectively attack graph,to ensure the overall network security and high-efficiency．By the final simulation experiment,this method can decrease consumption of system CPU time to complete the construction of network attack graph,has lower time complexity and high scalability,is a highly efficient attack graph method.

Real-time Optimizing Method Based on Independent Real-time Domain:Real-time Optimizing Technology Oriented to Multi-core Processors

FENG Hua,LU Kai and WANG Xiao-ping

Computer Science. 2013, 40 (9): 159-162.

Abstract

PDF(704KB) ( 611 )

References | RelatedCitation | Metrics

Multi-core processors with good performance per watt is an inevitable trend in real-time embedded systems．However,the multi-core feature of multi-core processors does no help to real-time performance improvement under the software structure currently existed,even,resource sharing between the multi-core processor cores make the impact of the program execution time factors complicated and the worst case execution time (WCET)of real-time tasks become more unpredictable and difficult to control．Based on domestic FT1000multi-core processor,real-time system building and real-time performance optimization were researched and “real-time optimization method based on independent real-time domain” was proposed.The processor was divided into “real-time domain” and “non real-time domain” through virtualization technology.Real-time tasks and non real-time tasks running on different cores take full advantage of the cores of multi-core processor,and the real-time tasks and non-real-time tasks are efficiently scheduled to run.

Efficient Modeling Method for Multidimensional OLAP Query Based on “C-Vine” Pair Copula

NI Zhi-wei,WANG Chao and GAO Ya-zhuo

Computer Science. 2013, 40 (9): 163-168.

Abstract

PDF(559KB) ( 739 )

References | RelatedCitation | Metrics

The rapid increasing dimensionality of database caused by the recent information explosion greatly impairs the accuracy and efficiency of On-Line Analytical Processing(OLAP)query models．In this paper,by first applying the statistical concept “C-Vine” Pair Copula in the study of OLAP query model,an effective solution to the “dimension curse” of higher dimensional OLAP query models was provided,and a specific parametric estimation method was proposed to extract the data synopsis from the original mass data for those higher dimensional OLAP models．Experimental results show that compared with existing methods,the proposed Pair Copula-based model can reduce storage space for data cubes while improving the relatively high query accuracy of OLAP models,and especially it provides a better query efficiency for higher dimensional data cubes compared with existing methods.

RDF Approximate Query Approach Based on Semantics

ZHOU Wen-jian and MA Rui-zhe

Computer Science. 2013, 40 (9): 169-173.

Abstract

PDF(707KB) ( 590 )

References | RelatedCitation | Metrics

To handle the problem of empty or few answers returned from RDF in response to a user query and the problem of synonyms in RDF queries,an approximate query approach by combining query relaxation and replacement with kinship words was proposed．RDF entailment to triple patterns is applied on the original query to relax it and then appropriate queries are chosen for replacement of kinship words so that more results can be obtained．The notion of semantic distance was introduced.The results that are semantically close to the original query can be determined．An approxiamate query algorithm based on semantics was hereby developed．The proposed approach was verified with experiments and it has good performances in its recall and precision.

Performance Modeling Approach for Resource Sensitive Distributed Systems

HUANG Xiang and CHEN Zhi-gang

Computer Science. 2013, 40 (9): 174-181.

Abstract

PDF(649KB) ( 782 )

References | RelatedCitation | Metrics

Performance prediction of software at early stage makes benefit to improve the quality of designs and reduce the performance problems．However,the huge costs of performance modeling approaches make it impossible to integrate the approach into the software development processes．In order to develop a cost effective modeling approach,we used the widely accepted UML model to support our work for distributed systems,and designed an intermediate model named Resource Scenario Model(RSM)for multi architecture model to multi performance model transformations．Therefore,designers can use their preferred tools to build and analyze software’s performance．To illustrate the effectiveness of our method,we given a case study,in which UML models can be easily transformed into performance models,such as LQN and LGSPN.

Personalized Query Expansion Based on Environment Information for Mobile Search

WANG Zhong-min,HUO Yi-wei and DENG Wan-yu

Computer Science. 2013, 40 (9): 182-184.

Abstract

PDF(624KB) ( 601 )

References | RelatedCitation | Metrics

Compared with traditional search,mobile search is sensitive to the users’ environment information such as location,temperature,speed and so on．In order to use the environment information to infer the user intents,an expansion approach based on environment information was proposed,which has been applied in the Clever Search Engine(CSE)——a mobile search system we developed．Using segmented queries and users’ environment information,the approach infers the user intents through the expert system and expands query in order to realize personalized query expansion．Experimental results show that the approach we proposed can significantly improve users’ search experiences and has better performance in precision than existing public search engine(e.g．Google).

Nested Model of Web Service Transactions Based on Alternative and Compensation

DANG De-peng and JIANG Xue

Computer Science. 2013, 40 (9): 185-189.

Abstract

PDF(721KB) ( 565 )

References | RelatedCitation | Metrics

As a new parallel computing model,Web Service,which has the advantages of platform independence,loose coupling and autonomy,has great potential and demand in service industry involved with complex business processes．Considering the limitations and characteristics of Web Service environment,traditional transaction model cannot be suited．This paper departed from the architecture of the Web service transaction systems,analyzed the characteristics of Web Service and Web Service transactional applications in an open Web environment,and finally proposed a nested mo-del of Web Service transaction combining alternative and compensation．We also illustrated the structure and processing of Web Service transaction combining with alternatives and compensation mechanism and presented a formal representation of this model．

Modeling Intelligent World Based on G-KRA Model Framework

WANG Nan,OUYANG Dan-tong and SUN Shan-wu

Computer Science. 2013, 40 (9): 190-193.

Abstract

PDF(877KB) ( 525 )

References | RelatedCitation | Metrics

Unified abstraction modeling framework and the formal representation help to realize automated reasoning．The General Knowledge Reformulation and Abstraction model(G-KRA model) was extended to characterize the mode-ling framework of the proposed intelligent world．Considering the diversity of the objects in the intelligent world,iterative primary perception process was defined to build the distinguishable objects and relations and the networked relationships between the three sub-worlds．The abstraction perception process of the intelligent world can be realized through constructing unified abstraction objects database or three different ones for the three sub-worlds．The extended G-KRA model considers the characters of the behaviors and relationships of various objects in the intelligent world．Each sub-world is constructed by the objects and relationships of same types so that the same unified reasoning mechanism can be built．The reasoning problems can be located in one(or more than one)sub-model(s)to reduce the reasoning spaces.

Rosenbrock Function Optimization Based on Improved Particle Swarm Optimization Algorithm

SHAO Peng and WU Zhi-jian

Computer Science. 2013, 40 (9): 194-197.

Abstract

PDF(618KB) ( 1092 )

References | RelatedCitation | Metrics

Rosenbrock function optimization belongs to unconstrained optimization problems,and its global minimum value is located in the bottom of a smooth and narrow valley of the parabolic shape．It is very difficult to find the global minimum value because of little information the function optimization algorithm provided．According to the characteristics of the Rosenbrock function,this paper specifically proposed an improved particle swarm optimization(PSO)algorithm(PSO-R)which introduces trigonometric factor with periodic oscillations of trigonometric functions so that each particle gets strong oscillation to expand the search space of each particle and guide particles to close nearly the optimal value and avoid converging prematurely and local optimum,and find the global minimum of the Rosenbrock function．A large number of experimental results show that the algorithm has good performance of function optimization,and provides a new idea for optimization problems similar with the Rosenbrock function for some problems of special fields.

Prediction Model for Airport-noise Time Series Based on GM-LSSVR

WEN Dong-qin,WANG Jian-dong and ZHANG Xia

Computer Science. 2013, 40 (9): 198-200.

Abstract

PDF(303KB) ( 599 )

References | RelatedCitation | Metrics

Along with the development of our country civil aviation,the airport noise is more and more serious.Aimed at the airport-noise time sequence prediction problem,this paper presented the prediction model based on GM-LSSVR,which is based on the ananlyses of noise’s characteristic．Namely,the model decomposes the noise time sequence into tendency and the residual,then establishes the GM(1,1)model of trend and the LSSVR model of residual,which are used for airport-noise prediction model.The application in the measured data of an airport shows that the accuracy of this model is better than other prediction models.

Simulation and Analysis of Radar Echo Based on Moving Scattering Center Model

XUE Ai-jun,WANG Xiao-dan,SONG Ya-fei and LEI Lei

Computer Science. 2013, 40 (9): 201-203.

Abstract

PDF(334KB) ( 776 )

References | RelatedCitation | Metrics

Because it is difficult to obtain Midcourse Target’s Wideband Radar Echo,the aspect angles and posture of warhead target modeling methods was further studied．Computational formula of scattering center locations were raised based on the moving scattering center model．Then,scattering intensities of every scattering center were calculated based on geometry diffraction theory．Lastly,after normalizing the high resolution range profiles(HRRP),relative ranges of scattering center were calculated and relative ranges variation array of scattering center was obtained．Experiment result shows that feature extraction of HRRP can get the information for Midcourse Target’s posture and provide important foundations to mid-course target recognition.

Particle Swarm Optimization Algorithm Combining Local Search and Quadratic Interpolation

QIAN Wei-yi and LIU Guang-lei

Computer Science. 2013, 40 (9): 204-207.

Abstract

PDF(300KB) ( 694 )

References | RelatedCitation | Metrics

To the problems of premature convergence frequently appeared in Particle Swarm Optimization(PSO)algorithm and its poor convergence accuracy,a particle swarm optimization algorithm combining local search and quadratic interpolation was proposed．Firstly,we randomly chose three positions from the N positions which are generated by standard Particle Swarm Optimization algorithm,and the new position was generated by using quadratic interpolation operator for each particle,and the previous best position of each particle and the global best position of swarm were updated.Then after some iteration steps,the Hooke-Jeeves local search technique optimized the global best position of the swarm found so far．Finally,simulation experiment on a set of 9benchmark functions was given,and the comparisons with other algorithms were provided．The numerical results show that the proposed algorithm has a fast convergence speed and good global search capability.

Research on Mongolian Spoken Term Detection Method Based on Segmentation Recognition

BAO Fei-long,GAO Guang-lai,YAN Xue-liang and WANG Wei-hua

Computer Science. 2013, 40 (9): 208-211.

Abstract

PDF(365KB) ( 749 )

References | RelatedCitation | Metrics

Mongolian is an agglutinative language．This special formation rule results in an amount of probably millions of words which are far beyond the coverage of the pronunciation dictionary of any current Mongolian Large Vocabulary Continuous Speech Recognition(LVCSR)System．On the other hand,even if the pronunciation dictionary is large enough to cover most of the Mongolian words,the recognition system still won’t perform well due to the sparseness of training data．To avoid the poor coverage problem of pronunciation Dictionary,we proposed a segmentation-based LVCSR approach and trained its acoustic model and language model．This approach is integrated into Mongolian Spoken Term Detection(STD)system and tested on Mongolian speech data．Experimental results show that our segmentation-based LVCSR approach can recognize most of the Mongolian words successfully and both the precision and recall of the Mongolian STD system are greatly improved by converting most of the out-of-vocabulary words into their in-vocabulary form.

Novel Multimodal Optimization Algorithm for Protein Structure Prediction

CHENG Zheng-hua,ZHANG Gui-jun,DENG Yong-yue and JIN Mei-mei

Computer Science. 2013, 40 (9): 212-215.

Abstract

PDF(441KB) ( 594 )

References | RelatedCitation | Metrics

Aiming at the multimodal demand of protein structure for the drug design,a multimodal optimization algorithm based on differential evolution was proposed．In order to reduce the computation complexity of the protein conformational space,energy minimization is applied to narrow the search space of feasible region.For balance local minima convergence and modal diversity of a multimodal optimization,under the framework of crowding differential evolution algorithm,under the premise of ensuring the convergence rate,the algorithm uses the principle of spatial locality and builds up procedures which randomly select a crossing strategy to increase the diversity of the population individual．Taking Met-enkephalin as benchmark,the new algorithm finds not only the global minimum energy conformation,but also many other distinct local minima.

Double-quantitative Boundary and its Algorithms Based on Variable Precision Upper Approximation and Grade Lower Approximation

ZHANG Xian-yong

Computer Science. 2013, 40 (9): 216-220.

Abstract

PDF(413KB) ( 608 )

References | RelatedCitation | Metrics

The double-quantification with precision and grade acts as a novel project in the approximate space．By the Cartesian-Product combination of quantitative information,this paper aimed to explore a double-quantitative boundary and its algorithms based on the variable precision upper approximation and grade lower approximation．First,a double-quantitative expansion-model was naturally constructed by the two approximations,and the double-quantitative expansion-boundary was correspondingly defined．Then,the double-quantitative semantics was analyzed for the boundary,and its precise description and mathematical properties were obtained,in order to calculate the boundary.The approximation-set algorithm and information-granule algorithm were proposed,analyzed and compared.The information-granule algorithm has more advantages on the space complexity．Finally,a medical example was provided to illustrate the boundary and its algorithms．The boundary expands the Pawlak-boundary,and makes the complete and fine double-quantitative descriptions for partial uncertainty,thus,it has a great significance for the uncertainty analyses and applications with respect to the double-quantification.

LM Based Training Algorithm for Quantum Neural Networks

ZHANG Yi-peng,CHENG Liang and HAO Huan

Computer Science. 2013, 40 (9): 221-224.

Abstract

PDF(335KB) ( 916 )

References | RelatedCitation | Metrics

Aiming at the question that the result trained by quantum neural networks is easy to fall into the local least value,the Levenberg-Marquardt algorithm was introduced into the original training algorithm to increase the training speed and improve the performance of the network．In addition,the conflict between different objective functions used for the training synaptic weights and quantum intervals can cause mean square error fluctuates．In order to solve this problem,the iteration order of the original training algorithm was adjusted．The experimental results show that,compared with the original training algorithm,the algorithm using LM can significantly reduce the number of iterations,significantly decrease the mean square error when the network convergences,and improve the classification results of quantum neural network.

Bat-inspired Algorithm:A Novel Approach for Global Optimization

LI Yu and MA Liang

Computer Science. 2013, 40 (9): 225-229.

Abstract

PDF(429KB) ( 966 )

References | RelatedCitation | Metrics

The study of bionics bridges the functions,biological structures and organizational principles found in nature with our modern technologies,and numerous mathematical and meta-heuristic algorithms have been developed along with the knowledge transferring process from the life forms to the human technologies．Recently,a new global optimization algorithm,called Bat-inspired Algorithm(BA),has been developed by Yang．The presented algorithm is inspired by fundamentals of echolocation of micro bats,and intends to combine the advantages of existing algorithms into the new bat algorithm．The first part of the paper was devoted to the detailed description of the existing algorithm．Subsequent sections concentrated on the performed experimental parameter studies and a comparison with efficient particle swarm optimizer based on existing benchmark functions．Finally the implication of the results and potential topics for further research was discussed．

FSVM Based on Pre-classification

SHEN Feng-shan

Computer Science. 2013, 40 (9): 230-233.

Abstract

PDF(337KB) ( 724 )

References | RelatedCitation | Metrics

A fuzzy support vector machine(FSVM)reduces the bad effect of outliers or noises on the classifier by using different training parameters for different training examples．This paper proposed an FSVM based on pre-classification,in which fuzzy weight for each training example is computed according to a classification hyperplane created for the training example．The method considers not only the effect of each individual training example,but also the sensitivity of the classifier to the outliers or noises．SVM with such fuzzy weights is able to make proper suppression for the training data．Experimental results on the standard training datasets from IDA and UCI repositories show the reasonableness and effectiveness of the proposed algorithm.

Study on Algorithm of Minimum Enclosing Ball Problem Based on Active Set Strategy

CONG Wei-jie and LIU Hong-wei

Computer Science. 2013, 40 (9): 234-236.

Abstract

PDF(331KB) ( 671 )

References | RelatedCitation | Metrics

Firstly,based on computing two furthest points from the current center at each iteration,a(1+ε)- approximation algorithm was proposed for solving the minimum enclosing ball problem of m points in n dimensions．The algorithm returns a core set of size O(1/ε)and achieves an O(mn/ε)time complexity for a given ε∈(0,1)．Secondly,an active set strategy was presented,which computes N furthest points from the current center at each iteration．By incorporating this strategy into the proposed algorithm,an algorithm based on active set strategy was obtained．Finally,the experiment results show that the algorithm based on active set strategy can quickly and effectively solve approximate minimum enclosing ball of the large-scale date sets with mn.

Algorithm Research on Multi-vehicle Ride Matching Problem Based on Passengers Classification and Transfers

MENG Chun-hua,WANG Hong-guo,SHAO Zeng-zhen,YU Hong-ling and DING Yan-hui

Computer Science. 2013, 40 (9): 237-242.

Abstract

PDF(560KB) ( 565 )

References | RelatedCitation | Metrics

Multi-vehicle ride matching problem(MRMP)is an important problem in the field of logistics and transportation.The existing MRMP algorithms mainly solve the basic MRMP.To improve the riding rate,an algorithm for the MRMP with transfers and passengers classifing was proposed.The algorithm with ant colony optimization algorithm as the core is divided into three steps:finding the sets of start-point and end-point；ant colony optimizes the routes,and two-way ant colony is proposed based on one-way ant colony,triming the vehicle routes．Simulation experiment shows that the algorithm can achieve riding rate of 80%.And the two-way ant colony shows stronger optimization ability than one-way ant colony.The results show clealy that the algorithm can find the matching routes effectively.

Multiple Dimension Data Modeling Method for Statistical Data Sheets Based on Semantic Analysis

SHEN Hui,LU Yu-feng,GUO Yi-dong and YANG Yong

Computer Science. 2013, 40 (9): 243-246.

Abstract

PDF(878KB) ( 683 )

References | RelatedCitation | Metrics

For implementing statistical analysis applications,it is the most fundamental task to design back-end database to support front-end data sheets processing,including data capturing,calculation and presentation．It always depends on technical designer’s personal experiences instead of rule-based step-by-step procedure and has been regarded as “art” instead of “scientific technology”．A risky result is that different technical designers might create different data models for one same set of statistical data sheets pre-defined by business end-users,which can cause big troubles both for system performance and on-going application maintenance．Through in-depth study on real case to summarize business rules,a formal method based on semantic analysis was proposed to support creating unified multiple dimension data model from pre-defined statistical data sheets and applied in a large SOE information system project as PoC(Proof-of-Concept).

Algorithm for Traffic Flow Based on Micro-variables

WEI Xiao-wei and JU Yong-feng

Computer Science. 2013, 40 (9): 247-249.

Abstract

PDF(254KB) ( 638 )

References | RelatedCitation | Metrics

With rapid development of urbanization in China,traffic condition has become even worsening,therefore,it is the primary task of the urban transport strategy to build up the modern transportation systems and traffic information management system earlier．It is desired aims of traffic flow research to find out the algorithm for traffic flow which could properly represent the fundamental laws of traffic flow so as to better guide the management and control of optimizing local traffic．Based on the theories of micro-variables and cellular automaton,the model of algorithm for traffic flow was put forward,after designing and numerically modeling the algorithm for traffic flow, the paper predicted the ideal traffic flow and trip distribution situation,deepened the understanding of the internal mechanism of traffic flow in various phenomenon and predicted the most actual situation of traffic flow and made fundamental work for traffic flow simulation and real-time traffic control as well.

Research on Permutation Flow-shop Scheduling Problem by Intelligent Water Drop Algorithm

ZHOU Ji-hua,YE Chun-ming and SHENG Xiao-hua

Computer Science. 2013, 40 (9): 250-253.

Abstract

PDF(326KB) ( 705 )

References | RelatedCitation | Metrics

Production scheduling problem is the one of the most basic,important and difficult theoretical research in a manufacturing system．This paper proposed a novel group intelligent optimization algorithm named intelligent water drop algorithm for permutation flow-shop scheduling problems(PFSP)．The intelligent water drop algorithm(IWD)is based on the processes that happen in the natural river systems and the actions and reactions that take place between water drops in the river and the changes that happen in the environment in which river is flowing．This paper analyzed the basic principle and mathematical model． Car1－Car6,Rec01and Rec13were tested by MATLAB．Compared to PSO,the results indicate that the intelligent water drop algorithm has better feasibility and validity for solving production scheduling problem.

Bias Sampling Data Stream Based on Sliding Window Density Clustering Algorithm Research

HU Zhi-dong,REN Yong-gong and YANG Xue

Computer Science. 2013, 40 (9): 254-256.

Abstract

PDF(311KB) ( 1100 )

References | RelatedCitation | Metrics

In management of the mobile object trajectory data stream in the field of mobile computing,the most commonly used technical means is sampling techniques,but the traditional uniform sampling is easy to lose some of the key changes in data,resulting in the phenomenon of loss of information．To solve this problem,we proposed a data stream based on the probability density clustering bias sampling algorithm．The algorithm in a sliding window model,makes full use of the distribution of characteristics of the the trajectory data stream itself,combines a bias sampling algorithm ideo-logy to overcome uniformly sampled data loss problems．Firstly the sliding window is divided into a strong cluster clustering techniques based on density data exists,weak clusters and excessive cluster,and then different sampling rates for different clusters biased sampling are given,thereby to obtain a final summary of the data stream．The experimental testing results of the set of actual data show that the algorithm ensures the sampling quality and has faster data processing capability.

Improvement of 2D LIC Algorithm for Vector Field Visualization

ZHAN Fang-fang,HU Wei and YUAN Guo-dong

Computer Science. 2013, 40 (9): 257-261.

Abstract

PDF(961KB) ( 724 )

References | RelatedCitation | Metrics

Line integral convolution(LIC)is a method for vector field visualization．This paper presented an improved LIC technique for 2D vector field．Firstly,focused on the problem that some vector field’s magnitude is too similar in regions of interest,we used nonlinear color mapping method to deal with it,and the final visualization result can highlight the vector field’s characteristics in regions of interest．Secondly,we extracted four parallel sub-modules from the original LIC serial computing tasks and implemented color enhanced LIC’s hardware acceleration relying on NVIDIA’s CUDA architecture．The results show that,with the increase of the input vector fields’ resolution,the speed of the accelera-ted algorithm increases as well．Therefore,the accelerated algorithm implemented in this paper can be well applied to interactive visualization of high-resolution 2D vector fields,and it has a good versatility,demanding no high hardware requirements．To conclude,the new method can improve visual quality with better performance.

Gait Recognition Based on Local Binary Pattern and Discriminant Common Vector

LIU Zhi-yong,FENG Guo-can and CHEN Wei-fu

Computer Science. 2013, 40 (9): 262-265.

Abstract

PDF(597KB) ( 628 )

References | RelatedCitation | Metrics

Recently,gait recognition for individual identification has been attracting increasing attention of biometrics researchers．It is well known that Gait Energy Image(GEI)is an efficient representation for gait,and Local Binary Pattern(LBP)can extract the local information efficiently．So,this paper used and Local Binary Pattern(LBP)to extract the local feature of gait energy image(GEI),and then it was used to identify．First of all,in order to extract local information better,the gait energy image(GEI) was segmented and the LBP features in each block were extracted and then each sub-block was fused in the feature layer to gain the whole gait energy image(GEI)’s features,at the same time,in order to explore the gait energy image(GEI)information better,this paper expanded the LBP model．Because the obtained LBP feature dimension is high,this paper used the Discriminant Common Vector(DCV)which has good dimensionality reduction and recognition ability to reduce the LBP features’ dimension．Finally,for simplicity consideration,we used the nearest neighbor classifier to classification．Experimental results on CASIA databases show that our algorithm is effective and obtains high recognition rates．

Semantic Image Retrieval Based on Sub-block Weight

XIA Li-min,ZHU Cheng,ZHANG Hai-yan and PENG Dong-liang

Computer Science. 2013, 40 (9): 266-269.

Abstract

PDF(587KB) ( 546 )

References | RelatedCitation | Metrics

The semantic gap between low-level visual feature and high-level semantic has become a primary problem．For improving the efficiency of semantic-based image retrieval system,this paper based on chunked weight and a visual vocabulary proposed a semantic image model which utilizes the correlation of low-level feature and high-level semantic．The model is used to interpret the image visual characteristic and detecte high-level semantic,which improves the efficiency of the semantic image retrieval．The experimental results show that this method improves the precision and recall of semantic image retrieval system.

Variational Fusion Strategy for Multiscale Analysis Domain

MA Ning,ZHOU Ze-ming and LUO Li-min

Computer Science. 2013, 40 (9): 270-274.

Abstract

PDF(1217KB) ( 561 )

References | RelatedCitation | Metrics

Conventional fusion strategies typically manipulate the fused coefficiented too directly,which may introduce artifacts and decrease the fused image quality．This paper presented a variational fusion strategy for multiscale analysis domain．An energy minimization problem was proposed to find the solution whose gradient is closest to that of the target gradient field,which is built from the gradient information of the source images’ decomposition coefficients．To keep the similarity with source images,the constraint condition was imposed into the highpass and lowpass coefficients separately．Since the highpass coefficients are sensitive to the noise,the total variation term is also introduced to assure the smoothness of the solution．The performance of the proposed fusion strategy was evaluated on multifocus images,infrared and visible images．Experimental results show that the strategy can inject more spatial detail information and meanwhile preserve more structural information of the original image than conventional fusion strategies.

Method of 3D Reconstruction Based on Given Axonometric Sketching

WU Xiao-liang and TIAN Huai-wen

Computer Science. 2013, 40 (9): 275-278.

Abstract

PDF(387KB) ( 976 )

References | RelatedCitation | Metrics

This paper presented a method of 3D reconstruction based on given axonometric sketching to realize 3D reconstruction of planar object conveniently．Firstly,according to the outline information of the patch organization of axonometric sketching,it establishes the relationship between all patches by interactive designated father-face．Secondly,it resolves the father-face of axonometric sketching by projection inverse transformation．Last,it rotates the child-faces around the edge linked to the father-face,and calculates them categorically when the area of the tridimensional reaches the maximum,generates the accurate target shape．The key technology and test drawings of the method realized by Matlab were given in this paper．The experiment results prove that the proposed method realizes 3D reconstruction of axono-metric sketching of general planar object with a certain availability and practicality.

Road Edge Detection Based on Heuristic Probabilistic Hough Transform

WANG Yan-qing,XIN Ke-jun,CHEN De-yun and WU Jian

Computer Science. 2013, 40 (9): 279-283.

Abstract

PDF(1087KB) ( 673 )

References | RelatedCitation | Metrics

Considering the difficulty of real-time,robustness for the edge detection in the unstructured road environment,a weighted Canny edges detection method and a heuristic probabilistic Hough transform method for weighted Canny edges image were proposed in this paper．According to the irregular characteristics of the unstructured road and the requirement of the real-time motion planning for intelligent vehicle,the double line segment models of the roadside were proposed．The experiment results in different road environments demonstrate the validity and real-time of the proposed method.

Key Technologies of Terrain Viewshed Analysis Based on DEM

ZHANG Bin,ZHANG Ze-jian,GUO Li and WANG Hao

Computer Science. 2013, 40 (9): 284-287.

Abstract

PDF(874KB) ( 1147 )

References | RelatedCitation | Metrics

Terrain viewshed analysis is an important content of spatial analysis．Based on DEM,the current algorithms have complexity in computation and bad applicability.Especially according to the multiresolution DEM data,they are immature this paper,from organization of terrain data and the necessity of defilade angles analysis,considering the influence of the data structure on the algorithm, discussed the main contents of the calculation principles of defilade angles and key technologies of terrain viewshed analysis in detail．DEM data was organized and managed with Quad-tree structure based on the Pyramid model in this paper．The matrix called “Advanced Matrix” was introduced so that the calculation efficiency was highly improved．Based on those above,the advanced algorithm based on DEM was used to do the analysis of terrain viewshed angles automatically．

Real-time Image Texture Substitution Algorithm

WAN Jin-liang,SONG Jin-bao,YE Long and LI Shu-hong

Computer Science. 2013, 40 (9): 288-292.

Abstract

PDF(1304KB) ( 635 )

References | RelatedCitation | Metrics

A real-time image texture substitution algorithm was proposed in this paper．The regional texture sample can be selected and be used to achieve medium level visual region texture substitution．The parallel image texture substitution algorithm with a hybrid MPI and OpenMP approach can be implemented on a dedicated cluster．The MPI paralle-lism,OpenMP parallelism and dynamic load balance under Master-Slave model were presented．Experimental results show parallel algorithm can ensure good performance of visual quality,greatly reduce the amount of data required by ima-ge reconstruction,effectively reduce the computing time and satisfy the time constraints of real-time application.

Study of Vehicle-surrounding Image Stitch Algorithm

LU Bin,QIN Rui,LI Qing and CHEN Da-peng

Computer Science. 2013, 40 (9): 293-295.

Abstract

PDF(493KB) ( 849 )

References | RelatedCitation | Metrics

The paper gave a way to stitch the Images captured by the cameras in four different positions．Mainly,we rebuilt the grant overlook-image through calculating the relationship between the four different visual fields and overlook field．Due to the more perfect image-stitching in overlap region,based on camera calibration theory,we focused on the overlap region,recalculating the external parameters,via inverse mapping,we built the relationships from four different visual fields to overlook field．Finally,we tested the algorithm,and the result gives a proof that the algorithm is effective.

Interactive Medical Image Segmentation Algorithm Optimized by Multi-thresholds

LAN Hong

Computer Science. 2013, 40 (9): 296-299.

Abstract

PDF(1146KB) ( 654 )

References | RelatedCitation | Metrics

Interactive image segmentation methods usually require users to provide much more initial seeds or more than once interactive when they are used for medical image segmentation with fuzzy boundaries．This paper presented an optimized interactive image segmentation algorithm with multi-thresholds technology．The proposed algorithm is based on GorwCut algorithm and improved in two aspects:one is automatically generating initial seeds templates by image gray histogram with multi-thresholds,and the other is raising iterative efficiency by improved cellular automaton iterative algorithm．Compared with GrowCut algorithm,the proposed algorithm simplifies the user interactive operations and improves the segmentation accuracy．Experimental results on clinical plaque and liver image segmentations demostrate the sound performances of the proposed algorithm.

Building Planar Recognition Based on GA-PSO Hough Transform

CHENG Hui and ZHANG Jian-pei

Computer Science. 2013, 40 (9): 300-301.

Abstract

PDF(245KB) ( 635 )

References | RelatedCitation | Metrics

A building planar recognition in digital images was described．Our approach uses GA-PSO Hough transform to improve the HT’s performance:A novel efficient limitation of the search space in combination with a hierarchical search strategy is implemented to reduce the computational effort．An empirical performance evaluation of the modified GA-PSO HT was presented by comparing it to build planar recognition techniques.

Design and Implementation of Explosive Devices Exclude Virtual Simulation Training System

LI Dai-wei and SUN Jian-shun

Computer Science. 2013, 40 (9): 302-306.

Abstract

PDF(961KB) ( 606 )

References | RelatedCitation | Metrics

To solve the problems of complication and reproduction of making training devices and huge cost of human and material resources in the explosive devices exclude training process,the paper designed an explosive devices exclude virtual simulation training system．The divisions of system hierarchical structure and the hardware architecture were designed based on the virtual reality technology,and the modular design of the system was based on object-oriented thin-king．The paper designed the explosive devices exclude tool module according to the design pattern of simple factory and template method．It solves the collision detection between the device nodes with Open Dynamic Engine (ODE),virtual hand grab objects with the Open Scene Graph (OSG)intersects testing and matrix transformations and the judgment of device excluding correctness key technology．And it implements the explosive devices exclude virtual simulation training system under the VC++ development environment．Practice proves that the system has high degree of simulation,reaches the effect of the explosive device exclude simulation training．And it is easy to operate with a strong scalability.

Segmented Low Rank Approximation Approach for Motion Capture Data Denoising

PENG Shu-juan,LIU Xin,CUI Zhen and ZHENG Guang

Computer Science. 2013, 40 (9): 307-311.

Abstract

PDF(484KB) ( 579 )

References | RelatedCitation | Metrics

The objective of motion capture data denoising aims to recover the frame sequences to better express the origi-nal data characteristics from the noise corrupted motion capture data．In general,the frame sequences within a short period of human motion capture data always reflect the same or similar motion semantic behavior．To this effect,this paper presented a Segmented Low Rank Approximation (SLRA)approach for motion capture data denoising．The proposed approach first divides the noise corrupted motion sequence into several continuous subintervals．Then,the inexact augmented lagrange multiplier method (IALM)is employed to decompose each subinterval batch matrix in terms of the low rank matrix approximation and sparse noise error estimation．Accordingly,the noise corrupted information within each frame subinterval can be removed．Finally,all the approximated low rank matrixes corresponding to the segmented subintervals are sequentially combined to represent the whole recovered sequence from the noise corrupted motion data．The simulation experimental results show that the proposed approach is able to well perform denoising of the human motion capture data with arbitrary topologies．The satisfactory performance demonstrates its universality and practicality.

Infrared Target Recognition Algorithm Based on Image Grayscale

GAO Jing,WU Kun,WU Yu-feng and CHEN Zhong-hua

Computer Science. 2013, 40 (9): 312-316.

Abstract

PDF(1227KB) ( 708 )

References | RelatedCitation | Metrics

A new infrared target recognition algorithm based on image grayscale was presented to solve the problems of complex background,the impact of similar goals on target identification and low recognition rate.Firstly,a multi-level morphological target detection operator was designed and combined with region growing method to detect the target region.Secondly,the introduction of sigmoid function was used to screen the candidate target region,narrowing the searching range,and the Haar wavelet method was used to consider the similarity relations surrounding the target region neighborhood of 8directions starting from the region texture in order to confirm the target extraction in infrared-red real-time target recognition.Finally,the detected results were precisely matched using the prior of target template knowledge．Experimental results show that the algorithm has a higher target detection recognition and get great improvement compared with template matching algorithm and otsu threshold detection algorithm．The new algorithm has a certain application value for infrared image target in complicated background.