Computer Science

LI Shao-hua,WANG Jinn-xin,FENG Qi-long,CHEN Jinn-cr

Computer Science. 2009, 36 (10): 1-4.

Abstract

PDF(0KB) ( 363 )

RelatedCitation | Metrics

Set cover and hitting set problems are two important W[2]-complete problems. Set cover problem is applied widely in the testing of VI_SI devices and crew scheduling. Hitting set problem has important applications in biological computation. In the parameterized computation and complexity theory, Set Cover and hitting set problems have been fo-cused on. This paper firstly introduced the categories and definitions of set cover and hitting set problems, then analyzed and summarized the computational complexity and the recent research results about these problems. hhe paper proved the computational complexity levels of (k,h)-Set Cover and (k,d)-Set Cover. Finally,the main points are concluded and some future research issues arc outlined.

Survey on Mobility Model

TONG Chao, NIU Jian-wei, LONG Xiang, GAO Xiao-peng

Computer Science. 2009, 36 (10): 5-10.

Abstract

PDF(573KB) ( 3014 )

RelatedCitation | Metrics

Abstract Mobility model decides how the nodes move and describe mobility manners of nodes. Hence it has been widely used in research on wireless network. This paper firstly introduced, classifies, and compares the current familiar mo-bility models. Secondly,it was discussed that current research focus on new mobility models,analysis of nodes mobility features, trace strategy, and evaluation of mobility model. Next, it was covered the existing problems of mobility models. Finally, this paper involved what calls for further study.

Dependencies Theory and its Application for Repairing Inconsistent Data

HU Yan-li , ZHANG Wei-ming, LUO Xu-hui ,XIAO Wei-dong , TANG Da-quan

Computer Science. 2009, 36 (10): 11-15.

Abstract

PDF(447KB) ( 937 )

RelatedCitation | Metrics

The theory of data dependencies was introduced and its application for repairing inconsistent data was discussed. Firstly, dependencies were analysized and the semantics for repairing inconsistent data were introduced. Secondly,the methods for repairing inconsistent data were analyzed. Finally, the future research topics related to repairing inconsistent data were also discussed.

Research and Survey on Algorithms of Blind Signal Separation Technology

ZHOU Zhi-yu , CHEN Hao

Computer Science. 2009, 36 (10): 16-20.

Abstract

PDF(513KB) ( 2742 )

RelatedCitation | Metrics

B1ind signal separation(BSS) is a useful technology which can recover the unknown source signals from the received signals. It has become a new research hot spot of neural network and signal processing. We first introduced the development of BSS and then introduced the BSS's math models which containe linear instant model,linear convolutive model and nonlinear model. We also analysed the algorithms' principal and characteristic of every model and introduced blind signal extraction technology which is closed to BSS. Finally, we pointed out the research direction and wide fore-ground of BSS.

Review on Digital Audio Authentication

LI Wei, WANG Zhu-rong, LI Xiao-qiang, LIU Ya-duo

Computer Science. 2009, 36 (10): 21-24.

Abstract

PDF(390KB) ( 756 )

RelatedCitation | Metrics

Modern audio processing techniques have made it pretty easy to make modifications like tampering, replace-ment, rearranging etc to audio content and time sectuence. It is more and more important to ascertain the integrity and authenticity of audio data. In toms of human auditory system, audio authentication aims to protect audio content rather than bit stream itself , this way, it should sustain some common signal manipulations that can maintain audio perceptual quality or semantic meaning while leaving authentication detector untriggered. This paper gave a vision on the back-ground, representative application scenarios and properties of audio content authentication, characterized hard authenti-cation and soft authentication, differentiated content preserving and malicious manipulations, and summarized most state-of-the-art audio content authentication algorithms published in the literature. Several technical characteristics in this research field were concluded and discussed, possible solutions were also pointed out for future research.

Summary for Control Structure of Multi-dimensional Intelligent Control Algorithm

HU Yang,GUI Wei-hua, CAI Zi-xing

Computer Science. 2009, 36 (10): 25-26.

Abstract

PDF(251KB) ( 731 )

RelatedCitation | Metrics

To simulate the characteristics of mankind intelligence, a multi-dimensional control architecture of intelligent algorithm was proposed. Integrating some algorithms with genetic algorithm, artificial neural network algorithm, immune algorithm and metabolic algorithm, the completed intelligent studying, reasoning and developing system was formed according to the structure theory and features of intelligent control. The function of traditional single algorithm can be perfectly improved by coordinating intelligent algorithms. Some mankind functions can be realized better by the integrated multi-dimensional intelligent control algorithm under the switching and integrating by a certain of algorithms.The control performance can be improved by computer programs controlled by the integrated artificial intclligence.

Research on Mobility Support in Wireless Sensor Networks

CHEN Chen, XIE Wei-guang, PEI Qing-qi, ZENG Xing-wen, FAN Ke-feng

Computer Science. 2009, 36 (10): 27-31.

Abstract

PDF(436KB) ( 618 )

RelatedCitation | Metrics

In traditional wireless sensor networks, users, sinks and sensor nodes are assumed static. However, moblility in a wireless sensor networks presents interesting challenges and requires special attentions. We first analyzed the challenges posed by mobility in wireless sensor networks. Then, existing proposals that support mobility in wireless sensor networks were summarized. And we pointed out the advantages and shortcomings of the existing proposals. Finally we concluded this paper with brief comments on promising future of wireless sensor networks.

Survey on Opportunistic Networks

HU Si-quan, WANG Hong-bing ,WANG Jun-feng

Computer Science. 2009, 36 (10): 32-37.

Abstract

PDF(548KB) ( 714 )

RelatedCitation | Metrics

Opportunistic networks is one of the newest hot research spots in wireless networks after mobile ad hoc net-works(MANET) and wireless sensor networks(WSN). As an exciting evolution of the legacy MANET paradigm,opportunistic networks have released the basic assumption of complete paths between data senders and receivers. With the support from devices movement and message forwarding, opportunistic networks enable users' communication in dis-connected environments, in which island of connected devices appears, disappears and reconfigures dynamically. In this paper, the concept, characteristics and typical applications of opportunistic networks were introduced. The hot spots ineluding routing strategy and mobility models were discussed and presented in detail. Other directions and foreground were also briefly presented.

Survey of Secure Clustering in Mobile Ad hoc Networks

WANG Heng-jun, WANG Ya-di HAN Ji-hong

Computer Science. 2009, 36 (10): 38-41.

Abstract

PDF(372KB) ( 867 )

RelatedCitation | Metrics

Mobile ad hoc networks composed of mobile wireless nodes,are particularly vulnerable due to their features of open medium and dynamic changing topology. Clustering is effective in reducing the spending on routing and control and improving the scalability in large mobile ad hoc networks. This paper surveyed the state of the art in secure clustering of mobile ad hoc networks.Firstly,the characters,proposed architectures and security threats of mobile ad hoc networks were analyzed. Secondly, secure clustering schemes were reviewed according to authentication model, evaluation of trust and identification of malicious nodes. We also made a comparison and discussion of their respective merits and faults. Finally, we also presented the challenges which arc still worth to further research in this area.

Dynamic Multi-path Routing Protocol for LEO/MEO Satellite Networks

TANG Jian, SHE Chun-dong, XU Zhi-ming

Computer Science. 2009, 36 (10): 42-45.

Abstract

PDF(357KB) ( 833 )

RelatedCitation | Metrics

According to the characteristic of LEO/MEO satellite constellation, a rapid and reconstructable routing protocol was proposed. At the same time, the model of the algorithm was simulated and features were analyzed. This protocol not only reduces the duration for end-to-end path establishment significantly, but also the multi path capability provides the flexibility for reliable transmission and load balance in satellite networks.

Improved Version of Concentric Anchor Beacon Localization Algorithm for Wireless Sensor Networks

JIANG Zhi-peng , GAO Sui-xiang

Computer Science. 2009, 36 (10): 46-48.

Abstract

PDF(305KB) ( 636 )

RelatedCitation | Metrics

Localization is one of the key technologies for wireless sensor networks. Concentric Anchor 13cacon proposed recently is a localization scheme with high precision. Each sensor node determines in which annular ring they are located within each anchor according to the broadcast information from the corresponding anchor. Each node uses the approximated center of intersection of the rings as its position estimate. This paper proposed an improvement for Concentric Anchor Beacon scheme. Based on the original algorithm, each sensor node can reduce the width of the annular ring they have determined through the broadcast information from the sensor nodes in the adjacent rings. So they can get a more accuracy position estimate than the original scheme. The improved scheme only needs to increase a little energy consumption than the original scheme. The simulation result shows that the improved scheme is more accuracy than the original scheme in an ideal environment or an interference environment.

Site Multihoming Path Failure Recovery Mechanism Based on Locator/Identifier Split

TU Rui, SU Jin-shu , CHEN Feng

Computer Science. 2009, 36 (10): 49-54.

Abstract

PDF(531KB) ( 642 )

RelatedCitation | Metrics

The multihoming is one of the effective ways to defeat path failure and increase the reliability of site network service. However, limited by the hCP/IP naming and addressing architecture, the mulithoming has not been well deployed. One of the most important reasons is the problem of semantic overloading of IP address. Current IP address hasdual semantic functions, which indicates both the network node' s routing locator and its endpoint identifier. We proposed a sit mulithoming path failure recovery mechanism(LISA-Recovery) based on the LISA(Locator Identifier Split Architecture) architecture. The results of simulation test prove that LISA-Recovery can effectively detect the path failure and performance reduction, and make a rapid path switch. It will greatly strengthen the reliability of application service. We also gave a theory analysis of the bandwidth overhead brought by LISA-Recovery. It shows that the bandwidth overhead is very small and our approach is feasible.

Fractal Statistic on the Self-similarity of Internet Router-level Topology

GHANG Jun , GHAO Hai, FU Da-yu, ZHANG Xin

Computer Science. 2009, 36 (10): 55-58.

Abstract

PDF(434KB) ( 623 )

RelatedCitation | Metrics

Because the statistical method with multi-angle and multi-measurement has many problems, a method to depict the overall Internet topology characteristics by using network fractal dimension was proposed in the paper. With the basement of the traditional fractal theory, combined with the self-similarity of Internet topology, the related concepts of the network topology dimension were given. I3y the mapping from Euclidean space to topology structure, the network topology dimension had been analyzed deeply and then the definitions of weighted network topology dimension and the computation method were given. By computing some main measurements in Internet topology such as power-law distribution and clustering, we analyzed the relationship between the network topology dimension and the traditional statistical method, described the advantage to depict network integral properties using network topology dimension.

User Cooperative Strategy and Performance Analysis Based on Limited Feedback

NING Yuan-hui, ZHU Guang-xi, SU Gang,TAN Li

Computer Science. 2009, 36 (10): 59-63.

Abstract

PDF(426KB) ( 573 )

RelatedCitation | Metrics

The cooperative schemes based on amplify-and-forward and decode-and-forward involve the user forward partner's data in the cooperative frame. However, it seems a waste of power and bandwidth when the based station can recover the user's data only based on the direction transmission. Thus, we proposed a new cooperative strategy, which is based on limited feedback from the base station and employed RCPC codes for its implementation. Based on our proposed scheme, the users adaptively change the data transmission in the cooperative frame. I}he simulation results reveal that our proposed scheme achieves impressive gains in contrast with non-cooperative system and also outperforms existing cooperative schemes while not only saving considerable power and bandwidth but also improving the throughput and spectral efficiency under various inter-user and uplink channel conditions.

Chosen Ciphertext Secure Identity-based Encryption in the Standard Model

LIU Zhen-hua, HU Yu-pu, ZHANG Xiang-song

Computer Science. 2009, 36 (10): 64-67.

Abstract

PDF(314KB) ( 835 )

RelatedCitation | Metrics

In Eurocrypt 2005,Waters' identity-based encryption scheme suffers from a drawback that the scheme only guarantees chosen plaintext security, which hampers its applications in higher security level environments. A chosen ciphertext secure identity-based encryption scheme was proposed to remedy this drawback. The proposed encryption scheme was regarded as the extended version of Waters' scheme with only one additional element in the ciphertext, and guaranteed chosen ciphertext security. The extended scheme's indistinguishability against adaptive chosen ciphertext attacks was proven in the standard model and rested on the hardness of the decisional bilinear Diffie-Hellman intractability assumption.

Quantum Secure Direct Communication Protocol Based on W State

YANU Xin-yuanl, MA Zhi, LU Xin

Computer Science. 2009, 36 (10): 68-71.

Abstract

PDF(397KB) ( 644 )

RelatedCitation | Metrics

A novel two-party quantum secure direct communication protocol was proposed. By using ordered four-particle W state as carrier of information, encoding the secret message with unitary transformation, we can directly transmit secret message with local Be11-basis measurement and classical communication. In ideal channel, the protocol is secure against incoherent attack. hhe advantage of this scheme is that using W state is more little than GHZ state in energy loss and no quantum bit carrying secret message is transmitted in quantum channel.

Delegation Authorization Model for Trust Management and its Application in Peer-to-Peer Security

ZHANG Zhi-yong, PEI Qing-qi, YANU Lin

Computer Science. 2009, 36 (10): 72-76.

Abstract

PDF(430KB) ( 567 )

RelatedCitation | Metrics

In trust management existing delegation authorization models were not involved with the definition of trust relations and the trustworthiness measurements among entities, such as roles and anonymous users, and a fine-grained formalized model and relevant delegation authorization security protocols were also absent, as could not effectively satisfy the requirements of trust management system applications. A trust management oriented formalized model that depicts trust relationships among entities,which is called DAM for TM,was presented and trustworthiness measurement metrics of entities could be dynamically adjusted through introducing the trust punishment function. Also, the trust computing-enabling trust delegation and role delegation security protocols,and an application case in P2P security were also addressed. The case shows that the proposed model and security protocols constructed peers' trust delegation reladons, and ensured the security of computing platforms and shared resources by the remote attestation on the terminal mtegmty.

BPML Decoding Algorithm of LT Codes

ZHU Hong-peng, LI Guang-xia, FENG Shao-dong

Computer Science. 2009, 36 (10): 77-81.

Abstract

PDF(380KB) ( 616 )

RelatedCitation | Metrics

For Belief Propagation(BP) decoding algorithm of LT codes,stopping set prohibits the improvement of decoding efficiency. This paper analyzed and simulated the size of stopping set. A Belief Propagation-Maximum Likelihood decoding algorithm(BPML)was proposed. BPML uses BP algorithm to decode firstly. When stopping set makes BP stop, Maximum Likelihood(ML) decoding algorithm is used to deal with the stopping set. It can overcome the negative influence of stopping set and improve the decoding efficiency of LT codes. The simulation showed that BPML combines the advantages of BP algorithm in low decoding complexity and ML algorithm in high decoding efficiency. hhe conclusion of research is practically valuable in improving efficiencies of data distribution applications in computer networks.

Power Optimization in IDMA with SNR and Differential Evolutions

TASSING Remil, ZHU Guang-xi, YANG Yong-li, TAN Li

Computer Science. 2009, 36 (10): 82-85.

Abstract

PDF(342KB) ( 549 )

RelatedCitation | Metrics

Interleave Division Multiple Access(IDMA) is a multiple access scheme in which users are separated by the only means of interleavers. However,the performance of IDMA depends on power optimization because equal power repartition significantly deteriorates the performance when the number of users increases. hherefore, we proposed the applacation of the genetic annealing Differential Evolution algorithm to SNR evolution for solving this optimum power allocanon problem. hhc results obtained arc similar to those with Linear Programming but there is no need to quantize orgroup power variables, the decay factor for controlling convergence speed is also not needed.

Analysis of the Design Principles of the QoS-capable Packet Network

LIU Wen-bo, GUO Yun-fei, LAN Ju-long, MA Hai-long

Computer Science. 2009, 36 (10): 86-88.

Abstract

PDF(362KB) ( 632 )

RelatedCitation | Metrics

By formularized analysis of the function of the ideal packet networks, the Internet's limitation in guaranteeing QoS was presented, and the reason why the Internet cannot provide guaranteed QoS to interactive real time communication the was analyzed. Guaranteeing QoS required to design a novelty packet network architecture. Hereby,from the view of network architecture,the necessity that QoS-capable packet networks should do an expansion to the Internet archi tecture was proved, and the expansion contents were put forward.

Multi-party Non-repudiation Protocol Using Key Chains

LI Lei, TAN Xin-lian, WANG Yu-min

Computer Science. 2009, 36 (10): 89-90.

Abstract

PDF(246KB) ( 600 )

RelatedCitation | Metrics

The previous multi-party non-repudiation protocols normally focus on one round message exchange. The multi-round message exchanges within the same participatores are not studied well. For the latter condition,we presented a multi-party non-repudiation protocol using key chains which can reduce the TTP's storage requirements. The protocol consists of a dispute resolution policy and four sub-protocols, i. e.，the initialization sulrprotocol, the exchange sulrprotocol, the abort sub-protocol and the recovery sulrprotocol. Technical discussions show that the protocol satisfies fairness, timeliness and confidentiality.

Multi-dimension QoS Evaluation Mechanism for Wireless Sensor Networks Based on Service-oriented Middleware

LIANG Jun-bin, CHEN Ning-jiang

Computer Science. 2009, 36 (10): 91-93.

Abstract

PDF(339KB) ( 538 )

RelatedCitation | Metrics

To meet the requirements of wireless sensor networks ( WSNs)，a QoS guarrantee mechanism based on serviccoricnted middleware was presented. Focusing on the problem of QoS evaluation, a QoS evaluation algorithm based on multi-dimension cloud was proposed. It evaluated quantitatively the tasks, which could make the network be scheduled properly by middlcware according to the results of the evaluation. Moreover, to conquer the weakness of cloud parameters that were set by expert's experiences, the algorithm also used benefit driven function to adj ust the parameters,which could improve the accuracy of evaluation. Experimental results show the algorithm can improve execution efficicncy of tasks and can extend the network lifetime.

Research on Modeling and Simulating the Threats in MANET Environment

LI Bing-xin, ZHU Li-na, YUAN Wei-Bong

Computer Science. 2009, 36 (10): 94-97.

Abstract

PDF(342KB) ( 572 )

RelatedCitation | Metrics

Comparing with the traditional network,the Mobile Ad Hoc NETwork(MANET) is more vulnerable to various proactive and passive attacks. Building an efficient and practical benchmark for researching anomaly behaviors through modeling the possible threats and ways of attacks in MANET environment is very beneficial for verifying and testing some proposed detection and provention related approaches. It also establishes a significative base for exploring a secure MANET environment. This paper presented some approaches of modeling and simulating some typical abnormal behaviors like Worm hole,l3lack hole and Routing Jams commonly occurred in MANET. Those approaches depict the abnormal behaviors exactly. The verification and tests in simulation scenarios were also conducted in experiments with better performance.

Analysis of Some Attacks against the Schnorr Signature Scheme

HU Guo-zheng, HONG Fan

Computer Science. 2009, 36 (10): 98-100.

Abstract

PDF(257KB) ( 806 )

RelatedCitation | Metrics

The Schnorr signature scheme is a digital signature scheme based on discrete logarithms. Recently some attacks against the Schnorr signature scheme were presented in the literature and they claimed that these new attacks had the greater success probability. However, these attacks were analyzed and the conclusion is that all these new attacks are essentially trivial exhaust search ones. Given certain system security parameters, the success tacks is negligible. Moreover, some mistakes in the probability analysis of these attacks were probability of theses at pointed out.

Backbone for Heterogeneous Ad hoc Networks and their Performance Analysis

GUO Pan-hong, YANG Yang,LI Xin-you

Computer Science. 2009, 36 (10): 101-103.

Abstract

PDF(314KB) ( 597 )

RelatedCitation | Metrics

Most of existing work are based on the concept of minimum connect dominating set(MCDS)，which is only target for minimizing the number of backbone nodes, but not take the real characteristics of nodes into account when constructing the backbone. In the way, some low performance nodes could be the bottleneck of the backbone. Selecting more capable nodes as candidates to construct a high performance backbone, a minimum connected dominating set with high node performance(MCDS-HNP) algorithm was proposed, and the approximation ratio of the proposed algorithm was also presented. The simulation results demonstrate that the MCDS-HNP algorithm achieves better backbone performance while maintaining approximately the same backbone size.

AS-level Internet Topology Degree and Connectivity Analysis

FU Da-yu, ZHAO Hai, GE Xin

Computer Science. 2009, 36 (10): 104-105.

Abstract

PDF(241KB) ( 639 )

RelatedCitation | Metrics

The Internet topology, especially the AS-level topology, is the hotspot of the research. We can comprehend well the inner connective mechanism of the network by researching the evolvement trend of the Internet topology. In this thesis, the research task is based on the massive data authorized by CAIDA(The Cooperative Association for Internet Data Analysis) Skitter project and the data's time span is from January 2004 to June 2008. This paper introduced the essential basic conceptions first, then carried out the evolvement analysis of the node average degree, the maximum node degree, the average degree of the top-degree node, the connectivity of rich-club and the clustering coefficient. It is discovered that the affect of the top-degree nodes and the connectivity between the top-degree nodes is descended by the time movement. However the network puts up still obvious rich-club character and clustering coefficient totally.

Study of the Hole of Strong Password Authentication Protocol

CHENG Ying, GAO Qing-de

Computer Science. 2009, 36 (10): 106-107.

Abstract

PDF(264KB) ( 593 )

RelatedCitation | Metrics

Studied a strong password authentication protocol based on hash function, by analyzing the security of SPAS protocol among this kind of protocols. Although the SPAS is designed meticulously and regarded as more secure, but we also find some secure vulnerability in it. Using single cryptography technology induces this protocol has secure leak. At last we can prove that if this kind of protocol uses single cryptography technology, it can't produce a secure protocol.

Pairing-based Traceable Neighborhood Anonymous Authentication Scheme for Wireless Ad hoc Networks

ZHOU Yao, PING Ping, XU Jia, LIU Feng-yu

Computer Science. 2009, 36 (10): 108-112.

Abstract

PDF(391KB) ( 629 )

RelatedCitation | Metrics

To improve the disadvantage in traditional anonymous authentication schemes for wireless ad hoc networks which are vulnerable to nodes intrusion and unable to trace malicious nodes, a pairing-based traceable neighborhood anonymous authentication scheme was proposed. Using an ID based public key system, the authenticating node chooses a random number from the private key space as its temper private key, and then multiplies this temper private key with his ID based public key and an all-known generator respectively to create its temper public key. Using such temper keys pair, the node can establish anonymous authentication with its neighbors in a secure and efficient way. By formally analyzing this scheme under random oracle model,it is shown that this scheme can efficiently resist the impersonation attacks as long as the BCDH problem is hard and the hostile nodes can be traced through their temper public keys by the network manager.

Position Information Automatically Binding Protocol of Network Nodes in the Industry Ethernet Control System

GHOU Chun-jie, YU Guang-can, QIN Yuan-qing, CHEN Kai

Computer Science. 2009, 36 (10): 113-116.

Abstract

PDF(371KB) ( 586 )

RelatedCitation | Metrics

Position based dynamic host configuration protocol(p_DHCP) was proposed based on the dynamic host configuration protocol(DHCP). The p-DHCP inherits the protocol for delivering host specific configuration parameters from a DHCP server to a host, and specifics mechanism for allocation of network addresses to hosts. If network nodes in the industry Ethernet control system are maintained following the given rules, p-DHCP can achieve position information automatical binding of network nodes.

Anycast Routing Protocol Based on Density and Distance

XU Xin, GU Yun-li, DU Jie ,QIAN Huan-yan

Computer Science. 2009, 36 (10): 117-119.

Abstract

PDF(329KB) ( 580 )

RelatedCitation | Metrics

Aiming at bad performance of robustness of shortest path anycast routing algorithm in highly dynamic networks as an example of wireless Ad hoc network, an anycast routing protocol based on density and distance was proposed. In the protocol, routing target selection was determined by both the distance factor and anycast members surroundings(i. e. density) of the target. Routing towards a dense anycast member population increased the probability that a packet eventually reaches any anycast member because packets could more easily be re-routed to neighboring anycast members when the targeted one became unrcachable,making this protocol's performance greater at robustness. In this protocol,parameter k adjusts the weight of distance and density hence influences respective selection priority of each anycast member. So, the protocol is characteristic of adaptable. Simulation was performed to evaluate this protocol. According to network status(link downtime and moving node speed)，by adjusting parameter k, routing robustness androuting efficiency are better balanced.

Distributed Formulation of a Low-latency Broadcast Tree in Multi-rate Wireless Mesh Networks

WANG Tai, YANG Zong-kai, DU Xu

Computer Science. 2009, 36 (10): 120-123.

Abstract

PDF(410KB) ( 592 )

RelatedCitation | Metrics

The multi-rate broadcast is a special problem in multi-rate wireless mesh networks. The broadcast tree formulation algorithms based on the minimal connected dominating set(MCDS)，which is commonly used, can not be directly applied in a multi-rate wireless mesh network. We presented a minimized latency broadcast formulation for multirate wireless mesh networks, and proposed a novel distributed formulation of a multi rate broadcast tree algorithm. It determined the proper broadcast rate only depending on the local topology information. Extensive results demonstrated that the proposed algorithm can reduce the network wide broadcast latency significantly, compared with the existing distributed algorithms.

Bandwidth Allocation of UWB Systems Based on Utility

MENG Wen-wu, ZHU Guang-xi, LIU Gan, ZHANG Liang

Computer Science. 2009, 36 (10): 124-126.

Abstract

PDF(241KB) ( 547 )

RelatedCitation | Metrics

In this paper, the optimal bandwidth scheduling problem for UWB networks was formulated as a utility maximization problem Concerning the system bandwidth allocation, utility function is an effective measure of QoS. Utility reflects the user's satisfaction level to the assigned resource. Due to the complex coordination among users and links of wireless UWB systems, we proposed the problem with a decentralized scheme, which adapted naturally to changing wireless network conditions. We allocated system bandwidth based on utility, which fulfilled high rate transmission of UWB systems.

Hierarchical and Ordered P2P Super-peer Topology Construction

FENG Jin-xiao, CHEN Gui-hai, XIE Jun-yuan

Computer Science. 2009, 36 (10): 127-131.

Abstract

PDF(528KB) ( 646 )

RelatedCitation | Metrics

Topology construction is one of the most essential problems in P2P network research. The current super-peer topology construction employs a fixed two-layer structure and an unordered approach based on the gossip-based paradigm, which not only restrains the system performance but also produces too many traffic loads and makes super-peer hotspot of the system. Meanwhile it brings about the higher cost and the security issue. The paper presented a hierarchical and ordered super-peer topology,called HOST,which established an adaptive hierarchy structure of super-peers according to the network scale and exploited an ordered algorithm to regulate peer joining and leaving. I}he simulation resups and analysis show that HOST can effectively control the generation and load balancing of super-peers,and remarkably reduce the topology construction and repair cost.

Using SPIN to Model and Analyze Security Protocol in Wireless Sensor Network

JING Chao, CHANG Liang, GU Tian-long

Computer Science. 2009, 36 (10): 132-136.

Abstract

PDF(412KB) ( 823 )

RelatedCitation | Metrics

Model checking has been successfully applied to design and analyze wired network security protocol. Comparcel with the wired network,wircless sensor network(WSN) also has the strict requirement on its security protocol.The vulnerabilities of WSN exist in communication environment and network node; those are great challenges for designing and analyzing a security protocol in WSN. This paper proposed an appropriate method to model and analyze security protocol in the WSN. Based on the method of wired network security protocol, with the thorough consideration of sensor network environmental factors and feature of sensor nodes,a general model was established and analyzed by our method. "hhe paper took WSN SPINS security protocol as an example to model and analyze via SPIN tool. "hhrough verifying the authentication and confidentiality of SPINS, we found flaws in the SPINS protocol. I}he work demonstrates the feasibility of using model checking to model and analyze security protocol in WSN, and expands the usage of model checking.

Research on Executable Test Sequence Generation of Mobile Node of Mobile IPv6

LI Hua , YE Xin-ming, LIU Jing, LIU Long

Computer Science. 2009, 36 (10): 137-140.

Abstract

PDF(313KB) ( 579 )

RelatedCitation | Metrics

Mobile IPv6 can offer convenient for host and make it connected to any link by a permanent IPv6 address.The conformance testing can guarantee the consistence between the specification and its implementation. In Mobile IPv6, there arc three roles, mobile node, home agent, correspondence node. In this paper, the control flow model of mobile node was analyzed and the part test sequences were generated. An EFSM(Extended Finite State Machine) was reconstructed from FSM by consideration of addition of data flow and the executable test sequences were generated at the same time. This method can reduce or avoid state exploding. Some test case obtained from an executable test sequence was performed to display the validity of the method. The conclusion and the research work in the future were introduccd.

Statistical Analysis of One-way Hash Function SHA-1 and its Algorithm Improvement

LIU Jian-dong, YU You-ming, JIANG Hui-na

Computer Science. 2009, 36 (10): 141-145.

Abstract

PDF(383KB) ( 677 )

RelatedCitation | Metrics

The degrees of completeness and avalanche effect and strict avalanche criterion for SHA-1 with increased number of steps were statistically analyzed. In order to improve the performance of collision resistance and nonlinear diffusion for SHA-1，the original algorithm was improved for its design defects and vulnerability indicated in the field of the current cryptology. The improved algorithm with mix function applied inverse message expansions sequence and inscrted Integer tent maps at the first round of mix function, to accelerate differential diffusion, to alter the original linked variables passing method,to correct the inner design architecture defects of the algorithm. The test and analysis results proved the reforming algorithm improved the degrees of nonlinear diffusion and enhanced the security of the algorithm.

Verification Mechanism for Web Service Composition Based on Extended Colored Petri Net

LI Jing-xia, YAN Chun-gang

Computer Science. 2009, 36 (10): 146-149.

Abstract

PDF(297KB) ( 593 )

RelatedCitation | Metrics

The development of Web service technology provides us a kind of platform-independent, self-described, localion-transparent software module. Utilizing Web service composition technology, business demand can be met ctuickly and flexibly. As Web service composition becomes more and more complexity, design of composition process became more and more error-prone. We put forward a model for Web service composition description based on extended colored Petri net. This model is independent of any process description languages, supports hierarchical process description and

Executive Model of Petri Net Applied in the PLC Control Program

MENG Qing-chun, LIU Yun-qing

Computer Science. 2009, 36 (10): 150-152.

Abstract

PDF(317KB) ( 714 )

RelatedCitation | Metrics

In existing industrial control system,PLC(Programmable Logic Controller) is often applied to realize sequential control, timing and etc. For the reason that control program often performs synchronous control and the occasions of input switches triggered are undermined, it is inevitably to cause the problem that control program cannot describe real executing process of control program fully. In view of it this paper put forward a method as follows: Firstly, we can use Petri net to construct net model representing the control logic during the process of compiling control program, this model represents executing logic of the PLC program. Secondly, we can dynamically run the net model mentioned above according to the actual execution route of the control program until the program stops executing. So we can find out some of the logical errors.

Extended Queueing System GIx/M/1/N for Evaluating the Performance of AQM Algorithms

WANG Hao,YAN Wei, HUANG Ming-he, GUO Bing

Computer Science. 2009, 36 (10): 153-159.

Abstract

PDF(556KB) ( 585 )

RelatedCitation | Metrics

In order to evaluate the performance of AQM(Active Qucuc Management) algorithms,an extended qucueing system was developed by embedding an AQM algorithm into the queueing system GIX/M/1/N. Using this model and the self-similar traffic of the Internet, a novel approach was proposed to analyze the performance of AQM algorithms with unresponsive traffic. Four classical AQM algorithms TD, RED, GRED and Adaptive RED were assessed by this approach. A series simulation was performed using NS2 to verify the correctness of this approach. The simulation resups arc consists with those obtained by this approach. This fact shows that the extended queucing system GIx/M/1/N can be used to evaluate the performance of AQM algorithms.

Research on a Dynamic Workflow Model Based on Geo-Ontology

ZOU Zhi-qiang,Hu Bin,LIU Lin-feng,WANG Ru-chuan

Computer Science. 2009, 36 (10): 160-163.

Abstract

PDF(440KB) ( 561 )

RelatedCitation | Metrics

The method of Web service composition proposed by open gcospatial consortium(OGC) can compose various Web services to form a sort of workflow(OGC Web Services 2 , OWS2) , but it is a static workflow with exact matching. The DWFMOR(Dynamic Work Flow Model based on Ontology and Rule) put forward by this paper can form a dynes mic workflow based on geo-ontology with the specified rules. This model was extended from OWS2. Firstly the domain spatial ontology-geo-ontology was created with the special spatial feature. Secondly the engine of workflow was designed for inference based on gco-ontology. Then this engine was implemented by using Stanford protcgc and HP Jena reasoner. At last , a DWFMOR prototype was developed on the Java platform. The experiment shows that DWFMOR can search-match Web ser-vices automatically and dynamically form a kind of workflow for spatial information Web services according to the business requirement.

Model Checking for Temporal Logic Programs

WANG Xiao-bing, DUAN Zhen-hua

Computer Science. 2009, 36 (10): 164-167.

Abstract

PDF(327KB) ( 945 )

RelatedCitation | Metrics

Formal verification of temporal logic programs plays an important role in improving program correctness. A framing projection temporal logic language, called Framed Tempura, is the research object as an executable subset of projection temporal logic. Propositional temporal logic was used to describe properties of a Framed Tempura program,thus the program p and properties ф were both formalized in projection temporal logic, then model checking needs to eva luate whether or not p→ф is valid which can be translated into evaluate whether or not p∧乛ф is unsatisfiablc, and this problem is solved by constructing the normal form graph of p∧乛ф. At last, an example of model checking a F ramed (hempura program was given.

Research on Software Fault Localization Based on Execution Trace

WANG Xin-ping, GU Qing, CHEN Xiang, GHANG Xin, CHEN Dao-xu

Computer Science. 2009, 36 (10): 168-171.

Abstract

PDF(440KB) ( 625 )

RelatedCitation | Metrics

Software reliability is directly relevant to the count of faults in software. Fault localization is the key to detect and eliminate the faults. Execution traccbased fault localization is of great significance because it can be integrated well with automatic software testing. Proposed the framework of execution traccbased fault localization FLOC, which can be divided into four components; organization of execution trace, selection of execution trace, computation of suspiciousness, and evaluation of the output. The typical current execution trace-based approaches were described and compared in FL0C. Finally some improvements were proposed according to FLOC. The purpose of this paper is to compare the advantages and disadvantages of those localization approaches in a unified framework,and provide some improvements on those approaches.

DTSArch: A Software Architecture Model Based on Decentralized Tuple Space

GHENG Xiang, QIN Zheng, XING Jian-kuan

Computer Science. 2009, 36 (10): 172-175.

Abstract

PDF(329KB) ( 646 )

RelatedCitation | Metrics

The collaborative systems lack descriptional support for dynamism, self-organization and collaboration. A new software architecture model named DTSArch was proposed based on decoupling of decentralized tuple space model. DTSArch provided formal semantics to describe component behavior and system configuration, dealing with the difficulty in applying decentralized tuple space model into system development A visual develop tool based on DTSArch was implemented. The practice shows that developers can quickly build system architecture on decentralized tuple space model,which improves the reusability of components and development approaches.

Research on Dynamic and Cyclic Optimization Technology between Process Management and Project Management

WANG Bo, ZHANG Li, HAN Zhao-gang

Computer Science. 2009, 36 (10): 176-178.

Abstract

PDF(363KB) ( 641 )

RelatedCitation | Metrics

By researching the internal relationship between project management theory and process management theory,an integrated framework on enterprise process management and project management was proposed. Then the dynamic and cyclic optimization model based on this framework was designed which involved both cycle in process management layer and cycle in project implementation layer. The process simulation optimization technology and the dynamic adjusting technology of project plan related in the framework were also discussed. I}he usage of cyclic optimization technology will strongly support the enterprise to shorten the manufacture process modified period, and improve the emergency ability to the manufacture environment.

Reliability and Performance Analysis of Web Service Composition Based on DTMC

CAO Ke-qiang, GU Qin, REN Ying-xin, CHEN Dao-xu

Computer Science. 2009, 36 (10): 179-182.

Abstract

PDF(409KB) ( 671 )

RelatedCitation | Metrics

Web service composition enables a new way to create new services by assembling independent service components. One of the important goals in service composition was analyzed aiming at reliability and performance attributes of service compositions. In this paper,a novel evaluating method based on DTMC(Discrete Time Markov Chain) was presented to solve this problem. By applying properties and formulas derived from the theory of DTMC, this method comprehensively estimated reliability and performance attributes of service compositions in view of different execution scenarios. It also assisted in identifying bottlenecks of given service compositions and suggesting strategics for improve-menu. Compared with previous researches, this method not only analyzed reliability and performance attributes more accurately,but also adapted better to complex structures and multiple execution scenarios which are distinct characteristics of service compositions. Experiments show that it has a good ability in terms of reliability and performance analysis of Web service composition.

Research of Normalization for XML Schema under Incomplete Information Circumstances

YIN Li-feng, HAO Zhong-xiao

Computer Science. 2009, 36 (10): 183-188.

Abstract

PDF(468KB) ( 655 )

RelatedCitation | Metrics

For solving data redundancies and abnormal manipulation for XML documents in schema design under incomplete information circumstances, the normalization theory of XML Schema under incomplete information circumstances was discussed. The concepts of XML Schema and incomplete XML document tree according with XML Schema were formalized. Based on the equivalence of the nodes, the consistency of the nodes, the equivalence of the nodes' informalion and the consistency of the nodes' information,XML strong functional dependency's definition was given,inference rules for XMLstrong functional dependency were presented. The arithmetic of path set strong closure and membership problem was proposed, its correctness was proved and its time complication was analyzed. The definition of XML normal form under incomplete information circumstances and the corresponding arithmetic of the normalization were formalized. The production in this work removes the redundancies of datum, eliminates update anomalies and achieves better design of XML Schema.

Semi-supervised Feature Selection Algorithm Based on Extension of Label

WANG Bo, JIA Yan, TIAN Li

Computer Science. 2009, 36 (10): 189-191.

Abstract

PDF(317KB) ( 619 )

RelatedCitation | Metrics

Feature selection is an important step during data mining and machine learning. With the lack of labeled instances, the problem of effective selection is worthy of consideration. This paper proposed a novel semi-supervised fealure selection algorithm based on the definition of inter-set and infra-set correlation,which starts from the original and small labeled samples and gains the final clusters by extension of labels. A complex evaluation was utilized as criterion to find optimal feature subset. Finally, the experimental results demonstrate the efficacy of the algorithm.

Extraction and Application of Discontinuous Phrase Templates in Statistical Machine Translation

SUN Yue heng, DUAN Nan HOU Yue-xian

Computer Science. 2009, 36 (10): 192-196.

Abstract

PDF(468KB) ( 841 )

RelatedCitation | Metrics

The discontinuous phrases are seldom taken into account in present phrase-based statistical machine translalion models,which leads to the distortion or omission of translation results. This paper took discontinuous preposition phrases for example and proposed a phrase template extraction algorithm. It first extracted phrase templates from chinese corpus based on some specified rules, and then got their english translations with a bilingual alignment corpus and a preposition and location-word translation table. The generated bilingual templates were then added into the translation table. Comparative experiments in standard test corpus indicate that when these discontinuous phrase templates arc applied in the translation system, the resulted translations are well comply with grammar specifications, and the translation quality is also improved.

Combination and Designation of Axioms in Proposional Encodings

JIANG Hong , LIU Da-you, LU Shuai , CAI Dun-bo , SHI Jing-jing

Computer Science. 2009, 36 (10): 202-208.

Abstract

PDF(606KB) ( 643 )

RelatedCitation | Metrics

In recent years,researches on planning as satisfiability have become a popular trend. We proposed partial relaxation and completed relaxation methods about mutex actions, appended the frame axioms finally. Faking SAhPLAN2006 as the base,we implemented these improved encoding methods respectively, tested them in the logistics track and the block world track which arc used in international planning competition, and analyzed the encoding scale and plan efficiency of different encoding methods, and then validated the improvements based on the graphplan encoding method in the overwhelming majority situation arc effective. Finally, we implemented the statcbased encoding method in SATPLAN2006 planner, tested on the above tracks and compared two extremely conditions, which eliminate actions and states respectively. The experimental results show that statcbased encoding method is more effective in logistics track than action-based encoding. The above improving strategi es about the SATPLAN2006 planner arc proved effective, and we should decide which combination of axioms in one encoding method is most suitable by consideing the characteristics of different tracks,rather than using certain encoding method absolutely.

BAM Models of Formal Contexts and Generating Concepts on the Models

QU Kai-she, TIAN Yong-sheng, ZHAI Yan-hui LIANG Ji-ye

Computer Science. 2009, 36 (10): 209-212.

Abstract

PDF(352KB) ( 717 )

RelatedCitation | Metrics

As a tool for data mining,formal concept analysis(FCA) has get rapidly developed and been widely used in machine learning, software engineering and information retrieving. Artificial neutral network(ANN) is a new field of artificial intelligence,whose evolvement is founded on simulating intelligent characteristics of human brain. In fact, the synthesis between FCA and ANN will be greatly advantageous to the further development of such domains as intelligent control, pattern recognition, knowledge processing etc. Based on the enactment toward a BAM neutral network, the paper presented the corresponding relation between formal contexts and Nk-BAM neural network models. Besides, we proved that the stable states of the Nk-BAM neutral network arc corresponding to the concept nodes of concept lattice built on the formal context,which establishes the theoretical foundation for the fusion of FCA and ANN. At the mean time, an algorithm for generating concepts on the models was brought forward and an illustrative example guarantees the availability of the algorithm.

Interval Scaling Algorithm and its Concept Lattice Construction from Extended Formal Context

LIU Yao-hua, ZHOU Wen, LIU Zong-tian

Computer Science. 2009, 36 (10): 213-216.

Abstract

PDF(317KB) ( 514 )

RelatedCitation | Metrics

The existing concept lattice model is unable to process data which contains not only fuzzy information but also scalar and Boolean information. The extended formal context includes many kinds of information such as scalar,fuzzy, l3oolcan, and interval. Therefore, how to build concept lattice from extended formal context, is a meaningful study. An interval scaling algorithm is proposed here to deal with the extended formal context and a corresponding concept lattice construction method is produced. In the end of this paper, the experimental results show that this algorithm is useful.

Optimized k-NN Text Categorization Approach

YAN Peng, ZHENG Xue-feng, ZHU Jian-yong, XIAO Yun-hong

Computer Science. 2009, 36 (10): 217-221.

Abstract

PDF(393KB) ( 581 )

RelatedCitation | Metrics

As one of the most classical TC approaches,k-NN is advantaged in tackling concept drift. However,to avoid curse of dimensionality,it has to employ FS(feature selection) method to reduce dimensionality of feature space and improve operation efficiency. But FS process will generally cause information losing and thus has some side-effects on the whole performance of approach. According to sparsity of text vectors, an optimized k-NN approach was presented in paper. This optimized approach greatly simplified euclidean distance model and reduced the operation cost without any information losing. So it can simultaneously achieve much higher both performance and efficiency than general k-NN approach. It then enhanced the advantage of k-NN in managing concept drift.

Statistical Language Model

ZHANG Yang-sen

Computer Science. 2009, 36 (10): 222-224.

Abstract

PDF(330KB) ( 705 )

RelatedCitation | Metrics

The training of statistical language model parameter is the key of language modeling. Chooseing how many training samples to meet the demand of the model parameter estimation error is one of concern problems of language modeling theory. We applied mathematical statistics theory to give the estimating method for training samples lower bound capability for Chinese model, the quantification estimation formula was suggested. By using this formula, the corpus sample capability needed to train model parameters can be calculated according to the demand of parameter estimation error.

Adaptive Functional Networks Loop Structures and Learning Algorithm

XIE Zhu-cheng , ZHOU Yong-quan

Computer Science. 2009, 36 (10): 225-229.

Abstract

PDF(374KB) ( 646 )

RelatedCitation | Metrics

The Banach contraction theorem not only plays a vital role in functional analysis, but also is an important theoretical basis for the algebraic ectuations of numerical analysis, the existence and uniqueness of ordinary differential equations and the integral ectuation of mathematical analysis. It is one of the most common methods in mathematical and engineering calculations. This paper presented adaptive functional networks loop structures which were designed based on the I3anach contraction theorem and the learning algorithm. These structures are used for the approximation of the fixed point of unknown functional relations(mappings) represented by training sets. Finally, the simulation results do monstrate that the structure presented in the paper has high precision and stable. The results obtained in this paper are very important for researching the methods of neural computation.

Research of Formal Concepts Rough Mining under Incomplete Electronic Patient Record Knowledge System

DING Wei-ping,GU Chun-hua, SHI Ghen-guo, CHEN Jian-ping,GUAN Zhi-jing

Computer Science. 2009, 36 (10): 230-233.

Abstract

PDF(342KB) ( 584 )

RelatedCitation | Metrics

Formal concepts analysison and rough sets theory are two different fast developing tools for data mining. The advantages of both the concept lattice in formal concepts representation and rough sets in knowledge reduction were cnough taken, and the algorithm of formal concepts rough mining(FCRM) under the incomplete electronic patient record knowledge system was put forward. The algorithm can carry on incomplete knowledge formal concept representation and rough positive approximate reduction with the decision rule lattice, and the corresponding consistent decision rule was extracted. Finally the expert system of the traditional Chinese medicine(TCM) patient record was designed. The experimental results show that algorithm of FCRM is better on the knowledge reduction and mining capability.

Constraint Projection-based Support Vector Machines Selective Ensemble Algorithms

WANG Lei

Computer Science. 2009, 36 (10): 234-236.

Abstract

PDF(338KB) ( 600 )

RelatedCitation | Metrics

This paper proposed two constraint project based selective ensemble algorithms of support vector machines.Firstly,projective matrices were determined upon randomly selected must link and cannot-link constraint sets, with which original training samples were transformed into different representation spaces to learn a group of base classifiers.Then, two selective ensemble techniques of genetic optimization and minimizing deviation errors were utilized to combine base classifiers. Experiments on UCI datasets show that both proposed algorithms improve the generalization performance of support vector machines significantly, which are much better than classical ensemble algorithms, such as Bagging,Boosting,fcaturc Bagging and LoBag.

De-noising Approach for Online Handwriting Character Recognition Based on Mathematical Morphology

SUN Yan, LIU Han-meng, RUI Jian-wu, WU Jian

Computer Science. 2009, 36 (10): 237-239.

Abstract

PDF(331KB) ( 714 )

RelatedCitation | Metrics

There arc lots of various noises when users are writing characters on the tablet. It's significant to eliminate these noises in order to make these characters be recognized accurately. According to the features of online handwriting characters, we analyzed all sorts of noises generated during writing on the tablet.By applying erosion, dilation, thinning operations of the mathematical morphology into preprocessing of online handwriting recognition,a dcnoising approach was proposed in this article. Using appropriate structure elements, we can eliminate large amounts of noises in the glyphs of characters. The experiment shows that this solution is valid and robust for online handwriting character recogration.

Algorithm of Simplifying the Image Emotion Semantic Rules

ZHAO Juan-juan, CHEN Jun-jie, LI Guo-qing

Computer Science. 2009, 36 (10): 240-243.

Abstract

PDF(362KB) ( 708 )

RelatedCitation | Metrics

This article discussed the significance of studying the semantic rules of image emotion. It gave the concept of rough set and described the learning algorithm about minimal-and-maximal rules. The author put forward the method of simplifying the semantic rules in image emotion by using this learning algorithm. Firstly, used the attribute reduction of rough set to simplify the train set. Then, got the rule set by using the decision tree algorithm. Finally, applied the lcarning algorithm about minimal-and-maximal rules to simplify the decision tree algorithm. The advantage of this method is while reducing the total quantity of the rules, it can narrow the scope of simplification, and ensure the accuracy of the semantic rules in image emotion and reduce the total number of rules.

Application Study of Hidden Markov Model Based on Genetic Algorithm in Noun Phrase Identification

LI Rong, ZHENG Jia-heng, GUO Mei-ying

Computer Science. 2009, 36 (10): 244-246.

Abstract

PDF(303KB) ( 698 )

RelatedCitation | Metrics

To increase further the accuracy of noun phrase(NP) identification and utilize features of the genetic algorithm(GA) and the hidden markov modcl(HMM)，a novel HMM identification method based on GA was proposed. The method was based on a high-performance POS(parts of speech) tagging. During the training phase, model parameters were gained by the genetic algorithm. And during the identifying phase, an improved Viterbi algorithm for dynamic programming was first presented to identify the same hierarchy noun phrase, then the combination method of hierarchical syntax parsing and Viterbi algorithm was brought forward to identify those recursive noun phrases. Experimental resups show that this combined algorithm has achieved a high precision and recall rate of 94. 78 0 0 and 94. 29 0 0，rcspcctively,fully inosculating the strength of genetic algorithms and hidden markov model. This proves that the combination method has much better identification effect than the unitary hidden markov model identification approach.

Improved Classification Algorithm of Domain Ontology

LI Gang, QIAN Xing-san, YE Chun-ming

Computer Science. 2009, 36 (10): 247-249.

Abstract

PDF(242KB) ( 555 )

RelatedCitation | Metrics

Ontology learning technology is still in the exploratory stage. The paper researched the techniques in the field of knowledge production, and proposed a classification algorithm of taking into account both the semantic similarity and the structure similarity of the concept,ontology. ontology learning As the algorithm using the "quantitative clas sification of value" as the classification standards, experiment shows that the algorithm is more effective than other al gorithms.

New Filled Function Applied to Global Optimization

YAN Jun-hua, HE Zhong-yue, WANG Yong-Jun

Computer Science. 2009, 36 (10): 250-252.

Abstract

PDF(201KB) ( 550 )

RelatedCitation | Metrics

The filled function method is an effective approach to find the global optima of multivariable and multimodal functions. Inspired by the previous filled functions, we proposed a new filled function with simple form and one parameter. The case of discontinuous point in computation needn't be thought. Moreover, it overcame the disadvantages that some previous filled functions were affected by exponential terms. Numerical results on benchmark functions show that the proposed function is effective.

Research and Implementation of Decision-making Mechanism Based on Identification of Multi-degree

LIU Min

Computer Science. 2009, 36 (10): 253-255.

Abstract

PDF(239KB) ( 548 )

RelatedCitation | Metrics

The genetic algorithm optimization back-propagation BP neural network decision-making mechanism was presented in order to overcome the shortcoming of the independent study model based on the digital technology supporled learning that can't identify a number of degrees. According to its feature, the genetic algorithm is adopted to produce sample groups and determine parameters of the neural network model. The relationship between input and output has been identified during adaptive learning and training of the neural network, so as to achieve the purpose of identificalion of multi-degree. It validates the proposed approach by experiments.

Prediction and Countermeasure of Chinese CPI Based on BP Neural Network

WANG Yu, LI Xu-dong, LI Zi-li

Computer Science. 2009, 36 (10): 256-257.

Abstract

PDF(270KB) ( 644 )

RelatedCitation | Metrics

Since 2007,CPI in our country has reached new high repeatedly. Using data published by State Statistical Burcau, and processing them, we applied BP neural network with momentum item to forecast separately CPI in 2008 and in 2009 will be respectively 104. 91 and 104. 88,CPI in the first quarter and second quarter of 2008 is respectively 106. 36 and 106. 53,CPI for food classification will be respectively 116. 52 and 116. 32,and also put forward some corresponding policy proposals.

Using MEASUR for Architecture Development of Information Systems

LUO Ai-min

Computer Science. 2009, 36 (10): 258-261.

Abstract

PDF(326KB) ( 595 )

RelatedCitation | Metrics

Information systems architecture is the guide of systems development and integration. The development of architecture has affected the quality of the system directly. Because the research mainly concerns design methods and tools for architecture products, the effective methods for analysis problem domain are lack. Based on DoD AF1.0, applying MEASUR to architecture development was studied. The research focused on applying problem articulation methods, semantic analysis method and norm analysis method to architecture design. Finally, the availability of the methods was verified through a case.

Image Quality Assessment Algorithm Based on High Frequency Band of Haar Wavelet Transform

JI Guo-li , NI Xiao-ming

Computer Science. 2009, 36 (10): 262-264.

Abstract

PDF(269KB) ( 688 )

RelatedCitation | Metrics

This paper proposed a image duality assessment algorithm based on wavelet transform. Here, both reference and test images were decomposed into four bands: LL, HL, LH and HH. The edge structure information similarity was obtained on other three bands, then we got a universal image quality assessment result(CSSIM) with weights based on the human property. The proposed metric performs better than MSSIM and MUSSIM by providing larger correlation coefficients and smaller errors after nonlinear regression fitting.

New Statistical Recognition Method for Road Signs

AN Ji-yao, WEN Gui-lin, LU Yuan-zhi, CHEN Zhong

Computer Science. 2009, 36 (10): 265-267.

Abstract

PDF(234KB) ( 551 )

RelatedCitation | Metrics

A new statistical recognition method of the road signs was proposed. In view of the basic characteristics of the road signs,and through a comprehensive analysis to turn road signs, we found that the image global features are more likely to be detected and not liable to be disturbed by noise and local distortion. An approach to the image feature selection and extraction based on the global feature was presented,in which the mean gray-scale of the image was taken as the major feature value. In what follows, the statistical recognition method was developed to accomplish these steps:the image was divided into several blocks; projective method with directional feature was adopted; the invariant of the feature marks was established and corresponding invariant values of nine turn road signs was computed. The experimental results show that the statistical approach has a good anti-interference ability by which the road sign in an image with moderate noise can be accurately and quickly recognized.

Fast Image Segmentation Based on NAMPD

WU Xue-li, CHEN Chuan-bo, XIA Hui

Computer Science. 2009, 36 (10): 268-273.

Abstract

PDF(533KB) ( 616 )

RelatedCitation | Metrics

With the concept of packing problem, the Non-symmetry and Anti packing pattern representation Model (NAM) uses a set of sulrpatterns to represent an original pattern. In this paper, we developed a new method for grey scale image representation based on NAM, called NAM-structured plane decomposition(NAMPD). In NAMPD, each sub-pattern is associated with a rectangular region in the image. The luminance function of pixels in this region was approximated by an oblique plane model. Image segmentation was a key method in image analysis. The traditional image segmentation algorithms was developed using the pixel representation. In this paper, we proposed a fast algorithm for segmentation of grey scale images based on NAMPD. Image processing using the NAMPD representation performed more quickly because it permited the execution of operations on image blocks instead of pixels. hhe experimental results presented in this paper show that the image segmentation method using NAMPD performs faster than the classical ones.

Natural Object Recognition Algorithm Based on SVM and Coloexture Combination Features

LEI Bao-quan, YANG Li-hua, CHENG Yong-mei, ZHAO Chun-hui, WU Yan-ru

Computer Science. 2009, 36 (10): 274-276.

Abstract

PDF(322KB) ( 795 )

RelatedCitation | Metrics

Affected by many factors, outdoor scenes vary greatly, so using a single feature(color or texture) to complete the outdoor scenes recognition can not achieve satisfactory results. A natural object recognition algorithm based on SVM and color/texture combination features was presented. Firstly, the color histogram based on the RGI3 color space was extracted as color feature. Then, the texture feature was extracted based on multi-channel Gabor filters. At last, the color/texture combination features were presented. An image database of training samples including sky, road, house, tree and grass was created,which is obtained from Pasadena Houses2000 database of California Institute of Technology. And the natural object recognition based on SVM using a single feature and color/texture combination features was completed respectively. Experimental results show that this algorithm has good recognition effect on the images in which each natural object varies greatly from each other.

Application Analysis of Color Spaces in Image Colorization

TENG Sheng-hua, SHEN Yi-ping

Computer Science. 2009, 36 (10): 277-279.

Abstract

PDF(253KB) ( 972 )

RelatedCitation | Metrics

Local color similarity is the basis of image colorization, where choosing the suitable color space is a key problem. According to the standard whether luminance and chrominance could be separated from each other, common color spaces were divided into two categories, i. e. color-component space and color-attribute space. Smoothness of color distribution differs in different color spaces. The applicability of different color spaces in colorization was evaluated. Theoretic and experimental results show that color-attribute space is more appropriate to use in colorization. Furthermore some conclusions on color space selection are drawn by analyzing the essence of colorization.

Iris Recognition Based on Wavelet Transform and 2D-PCA

DONG Qin-ke, WANG Xiang-hai

Computer Science. 2009, 36 (10): 280-283.

Abstract

PDF(330KB) ( 652 )

RelatedCitation | Metrics

Recently, as tow effective methods for image analysis, 2D-PCA and wavelet transform get extensive attenlions. A fast iris recognition arithmetic was proposed in this paper based on 2D-PCA and wavelet. Firstly,we pre-dealt with the image; then applied 2D-PCA on the two third level middlcfrequcnce subband of wavclet coefficient, then got the feature vector by combination and symbol uantify;finally, we applied multi-templet matching between unknown class sample and feature database, in the same time, in order to increase the robustness against rotation of the original image, we applied a anti-rotation method on the unknown class image; then got the ecognition result by K-Neighbor and threshold. The experiment result validates the efficiency of the arithmetic proposed.

Research on RD Optimized Intra Refreshment Algorithm for Region of Interest

RUAN Ruo-lin, HU Rui-min, LI Zhong-ming

Computer Science. 2009, 36 (10): 284-288.

Abstract

PDF(425KB) ( 613 )

RelatedCitation | Metrics

The importance of video sectuence varies with the regions, and human visual sensitivity too varies with the importance of region of video sequence, so the subjective duality requirement of region of interesting is not the same as non-region of interesting. Although there arc very many researches on region of interesting, it is few research which joints the overall Rl)optimal theory, the rate control, infra macroblock refreshment and the region of interesting. So the paper proposed the RD optimized infra refreshment algorithm based on region of interesting, and it was used the overall end to end distortion theory to analyse the region of interesting and non-region of interesting, and gave more infra refreshment chances to the marcoblock in region of interesting, as the bit rate was constrained, so the bit rate of the region of and non-region of interesting must be reallocated. At last, the algorithm was implemented in reference software JMl2. 2 of J VT. The simulation results showed that, comparing with the traditional RD optimized infra refreshment algorithm, the proposed infra refreshment algorithm based on region of interesting could obtain a better subjective quality of the reconstructed video in the region of interesting at the decoder.

Curve Fitting of B-spline Based on Particle Swarm Optimization

ZHU Qing-sheng. ZENG Ling-qiu, QU Hong-chun, LIU Ji

Computer Science. 2009, 36 (10): 289-291.

Abstract

PDF(311KB) ( 1341 )

RelatedCitation | Metrics

Curve fitting plays very important role in preprocess of object recognizing. A particle swarm optimization (PSO) based multi-object optimization algorithm was proposed in this paper to implement the smoothness fitting quickly for image with complicated noise around the target-area. The external repository and strategy of diversity were employed to prevent the PSO from converging too quickly. Moreover, the search policy of split and-merge made the selection of knots parameter more flexibly in l}spline bases computation while getting the discrete control points set of the target area. I}herefore, curve fitting can be achieved by the multi-resolution interpolation. As shown in experiments, this algorithm can get the approximation curve quickly, eliminate the noise from the target area, and satisfy the requirement of image based 3-D reconstruction as well.

Image Zoom Algorithm Based on the Spring Model

KANG Mu, LI Yong-liang

Computer Science. 2009, 36 (10): 292-295.

Abstract

PDF(346KB) ( 758 )

RelatedCitation | Metrics

To avoid the phenomenon of jaggy and blur edges during zoom image,after analyzing the nearest neighbor interpolation model and the surface fitting modcl,an image zoom algorithm was proposed based on the spring model. The interpolation formula was presented and the output results concerning on image zoom were simulated based on different algorithms. The result of experiment shows that the proposed method is able to maintain clear borders of source image and the algorithm is efficient in computation for making image zoom.

Coding Technology of the ShuangMaSanBi Chinese Input Method

YAN Yu, HUA Ze-xi

Computer Science. 2009, 36 (10): 296-298.

Abstract

PDF(210KB) ( 602 )

RelatedCitation | Metrics

To resolve the problem, that is the balance between easy to learn and fast to input and the switch between input methods, a new coding technology of the Chinese character was proposed. I}hat is the code of the ShuangMaSanBi input method. I}he design idea is as follows. Sound and shape of the Chinese character have been combined together. The sound adopts the pronunciation of its first letters. The shape adopts five universal strokes of Chinese characters(HengShuPieDianZhe). The coding r}rate is reduced in the coding design and its key mapping. The flip frequency of the selected Chinese character is reduced by the frequency index Finally, the user can input easily and quickly.