Current Issue

• 2013, No. 6　Published:25 June 2013 Last issue    Next issue
  Columns in this issue: 论文

• 论文

  Remotely sensed image directory system
  based on the segmentation of partwhole ontology       

  CUI Jifeng1,2,DU Hongyue2,ZAHNG Yong1,XING Chunxiao1,XUE Benxin2

  2013, 35(6): 1-6. doi:

  Abstract ( 111 )   PDF (887KB) ( 230 )   　　

  We built a file catalog system of data application based on the partwhole ontology of spatial relation to manage the massive remotely sensed image files effectively. Firstly we analyzed the property of images’ metadata and calculated the weight for application. Secondly, we built the concept level relation, and calculated the similitude degree of the image property to build the catalog tree. Thirdly, we stored the files into the corresponding directories. Finally we designed and realized the directory experiment system. Experimental results show that our method is effective for image data’s high efficient integrative management.

  Modeling continuous data protection           

  WU Jiangjiang1,CHENG Yong2,ZHU Yaokai3,WANG Zhiying2,REN Jiangchun2

  2013, 35(6): 7-14. doi:

  Abstract ( 97 )   PDF (880KB) ( 188 )   　　

  Data protection is a vital issue in information age when the data size is rapidly expanded. As a useful data protecting scheme, continuous data protection mechanism can address data outage raised by software and hardware failure, user errors and so on. In order to guide the continuous data protection mechanism design, this paper proposed a universal model for continuous data protection schemes. And then we analyze and proposed the formal description of data replication frequency, recovery time and storage overhead. At last, we evaluated the effectiveness of our model by actual trace in online transaction processing systems.        

  A novel cache replacement algorithm for iSCSI storage    

  ZHAO Jianming1,YAO Nianmin2,HAN Yong2,CAI Shaobin2

  2013, 35(6): 15-23. doi:

  Abstract ( 132 )   PDF (1297KB) ( 234 )   　　

  Due to the bandwidth restriction and the long delay of IP network, iSCSI initiator and iSCSI target have to wait for a longer time until they acknowledge the status or the control command sent by the other party. The communication of iSCSI protocol takes a large part of the total cost. Therefore, the cache data's loading cost of iSCSI storage is different from the traditional direct attached storage. The existing researches focused on achieving high cache hit ratios, but rarely paid attention to reducing the loading cost of miss data for improving the storage system performance. In the paper, a cache replacement algorithm called CFL-LRU (LRU Combined with Frequency and data Length) was proposed for iSCSI storage. The algorithm not only considers the time and the frequency but also takes into account the logical block address continuity of pages, thereby compromising the hit rate and the miss penalty. Our tracedriven simulation results show that, for different kinds of workloads and cache size, the CFL-LRU outperforms the others.

  Research on framework for
  distributed software runtime monitoring          

  LIU Donghong,ZOU Peng

  2013, 35(6): 24-29. doi:

  Abstract ( 101 )   PDF (773KB) ( 203 )   　　

  Software monitoring is an important means to check whether a largescale distributed system complies with its intended behavior, which is the foundation of further online adjustment and evolution activities. This paper proposes a framework named Mega that supports the development and running of monitoringenabled distributed software systems. On the one hand, a dedicated language is introduced into the framework to describe where and what should be monitored at runtime, which separates the monitoring logic and the functional logic and enables the former to be developed efficiently and maintained independently. On the other hand, an eventbased model for monitoring data processing is introduced, which is scalable and extensible in largescale distributed systems. Based on the description of the feature of the Mega framework, this paper presents a set of monitoring design patterns as well, which can support the development of largescale distributed systems with monitoring capability.

  An optimization method for XML
  query data service in cloud environment               

  WEI Yongshan1,ZHANG Feng1,CHEN Xin1,HAN Yanbo2,SUN Zhonglin1

  2013, 35(6): 30-36. doi:

  Abstract ( 99 )   PDF (723KB) ( 174 )   　　

  Providing XML data service in could environment attracts more and more attentions. By introducing existed XQuery engine (such as saxon) into cloud environment, a XML data query service framework was implemented. To improve the efficiency of XML data query in cloud environment, an optimization method based of XPath projection is applied in the procedure of splitting XML data, only leaving the data related to XQuery statements. Experiments show that the optimization method effectively reduces both the data size and the query processing time of XML data service.

  Research of P2P service discovery algorithm based on QoS

  GAO Xiaoyan

  2013, 35(6): 42-46. doi:

  Abstract ( 114 )   PDF (463KB) ( 203 )   　　

  The existing P2P network service discovery algorithms are lack of service quality assurance consideration, and cannot fully utilize the specific service quality attributes of some nodes in the networks. In order to improve the efficiency of P2P networks service discovery, this paper proposed a P2P service discovery algorithm based on QoS by combining QoS attributes and P2P characteristics. we first defined description of P2P service based on QoS and introduces semantic information into the service description in order to improve the accuracy of service matching. On this basis, we built up the P2P service discovery model with QoS guarantees and provided a genetic algorithm for P2P service discovery. At last, the feasibility and effectiveness of the service discovery algorithm was performed through the experiments.

  Exploring SVMbased intrusion
  detection through information entropy theory           

  ZHU Wenjie1,2,WANG Qiang2,ZAI Xianjun1

  2013, 35(6): 47-51. doi:

  Abstract ( 132 )   PDF (996KB) ( 231 )   　　

  In traditional SVM based intrusion detection approaches, both core function construction and feature selection use prior knowdege. Due to this, they are not only inefficient but also inaccurate. It is observed that integrating information entropy theory into SVMbased intrusion detection can enhance both the precision and the speed. Concludely speaking, SVMbased entropy intrusion detection algorithms are made up of two aspects: on one hand, setting sample confidence vector as core function's constructor of SVM algorithm can guarantee the mapping relationship between training sample and optimization classification plane. Also, the intrusion detection’s maximum interval can be acquired. On the other hand, simplifying feature subset with samples's entropy as metric standard can not only shrink the computing scale but also improve the speed. Experiments prove that the SVM based entropy intrusion detection algoritm outperfomrs other tradional algorithms.

  Spectrum sensing algorithm based on
  frequency domain Lempel-Ziv complexity analysis          

  LUO Mengyu1，CHEN Zhigang1，KUANG Zhufang1，2

  2013, 35(6): 52-56. doi:

  Abstract ( 118 )   PDF (701KB) ( 195 )   　　

  Research on related tasks scheduling in mobile grid           

  JU Hongjun1,3,DU Lijuan2

  2013, 35(6): 57-64. doi:

  Abstract ( 100 )   PDF (724KB) ( 197 )   　　

  Taking mobile grid as background, the problem of scheduling related tasks on dynamic resource environment was studied. During scheduling, not only dependencies between tasks, but also resources’ dynamic behavior such as joining, leaving and performance changes are considered. Dynamic scheduling policy was proposed, which includes task subset scheduling and rescheduling. Dynamic partition of subset mainly focuses on dependencies between tasks and helps reduce the number of rescheduling. Subset scheduling objective and constraints were described and particle swarm scheduling algorithm was put forward, which integrates the idea of simulated annealing. Rescheduling further improves the adaptability of scheduling policy to the dynamic behavior of resources. Rescheduling trigger conditions was explained. The complete process of related tasks scheduling in mobile grid was given, as well as complexity analysis and experimental analysis of the proposed algorithm. Experiment results demonstrate the effectiveness of scheduling policy.

  Identitybased authenticated key agreement protocol
  featuring strong security and no bilinear paring operation      

  LI Fang1,2，CHEN Ming3

  2013, 35(6): 65-71. doi:

  Abstract ( 96 )   PDF (758KB) ( 181 )   　　

  To reduce the computation cost, a novel Identity-based Authenticated Key Agreement Protocol (ID-AKA) was proposed. Based on the elliptic curve discrete logarithm problem, the new protocol minimizes message exchange times with only one exchange by using an implicit authentication, and removes computationintensive bilinear paring operation. Based on the ID-AKA security model proposed by Chen et al. and the eCK model proposed by LaMacchia et al., by adding EphemeralSecure query, a stronger model for ID-AKA protocols was proposed. Meanwhile, the paper pointed out that the security model proposed by Shu Jian was ineffective to analyze the key-compromise impersonation in ID-AKA protocols, because the model constraints are too strong. Under the new model, the new protocol is provably secure. Comparative analysis shows that the new protocol with not only strong security but also lower computational overhead is suitable for establishing a secure end-to-end connection in mobile communication environment.

  Research on secure payment in online banking          

  ZHANG Lina1,YAN Wenbin2

  2013, 35(6): 72-77. doi:

  Abstract ( 106 )   PDF (687KB) ( 216 )   　　

  With the increase of the number of users and the total amount of transactions of online banking, it has become the preferred target of some attackers. The security issue has become one of the main considerations of users. To some extent, this has hindered the development of online banking. The related security has become one of the focused issues of the most banks and researchers. Based on analyzing the basic processes of online payment and the mode of attack, we discussed its security and proposed related prevention techniques. The processes of tampering with the signature data were analyzed, then corresponding system model and design strategies were discussed in detail.

  Security analysis of two traitor tracing schemes   

  WANG Qinglong1,2,XU Li1

  2013, 35(6): 78-81. doi:

  Abstract ( 120 )   PDF (420KB) ( 216 )   　　

  ZHANG Xuejun et al presented a multioriented traitor tracing scheme with fully collusion resistance. They asserted that it was computationally infeasible for any number of users to get a different key by collusion. Here, this scheme was cryptographically analyzed. We found that it does not meet the collusion resistance. Also, we proposed a concrete attack in which the three or more traitors can build more than one valid decrypt key unidentified by black box tracing method. MA Hua et al also proposed a traitor tracing scheme that can trace all traitors undoubtedly. However, we presented a simple attack in which traitors can easily avoid tracing by changing their keys.

  Radar range profile’s recognition
  based on an improved F-SVM algorithm        

  FANG Ning1,2,TAN Fei2

  2013, 35(6): 82-87. doi:

  Abstract ( 111 )   PDF (644KB) ( 194 )   　　

    Fuzzy support vector machines (F-SVM) algorithm is effective for pattern classification on unclassifiable sample sets condition. For the sake of enhancing the algorithm’s applicability, it was improved in this paper from two aspects. One hand, nonequidistant margin hyperplane (NM) in high dimension feature space is introduced to improve on study precision; On the other hand, effectual training sample sets in high dimension feature space are filtrated, via algorithm introduced by this paper, to reduce study time. This paper gave the theoretical derivation of improved FSVM and the filter algorithm of effectual training sample sets. The improved methods were applied to Radar Range Profile’s Recognition of two planes. Experimental results show that these methods can obtain very excellent recognition effect and reduce the algorithm study time.

  Research and development trend of path planning
  based on artificial potential field method            

  ZHANG Dianfu，LIU Fu

  2013, 35(6): 88-95. doi:

  Abstract ( 379 )   PDF (3477KB) ( 742 )   　　

  The artificial potential field method is used in mobile robot path planning extensively because of its simpleness, high efficiency and smooth path. In this paper, the principle of path planning based on artificial potential field model was introduced and its main shortage is also analyzed detailedly. Then, the methods of resolution to the main problems are discussed, which contain GNRON, loiter and jitter phenomenon and poor adaptive ability in dynamic environment. Finally, technology development trends of robot path planning based on artificial potential field are given.

  Study and application of incremental learning
  Bayesian network guided by node order confidence             

  JIA Songhao，YANG Cai，ZHANG Haiyu

  2013, 35(6): 96-100. doi:

  Abstract ( 95 )   PDF (536KB) ( 226 )   　　

  By introducing the node order of confidence in the procedure of incremental learning, so the NOCLBN algorithm is proposed. For the learning procedure of Bayesian network under largescale data set, the algorithm enhances the accuracy of the study of each batch of data, thus improving the quality of the final network model. Experimental results show that the NOCLBN algorithm can obtain high quality for the learning results of Bayesian network under largescale data set.

  Integrated feature location in objectoriented program       

  SHU Shaoxian1，XU Xiaoping2，RONG Ni3,GUO Min1

  2013, 35(6): 101-106. doi:

  Abstract ( 94 )   PDF (783KB) ( 167 )   　　

  Locating codes for software features efficiently and accurately is very important for software maintenance. The paper proposed an integrated approach to feature location in objectoriented (IAFLO) system, which uses software reconnaissance to build trace relations from features to classes, and regards these classes as seeds to construct concern graph for objectoriented program. Furthermore, under the slice criteria of attributes and statements in concern graph, the objectoriented slicing method was adopted to compute codes that directly affect features. The method has a high degree of automation, and can improve the efficiency of program understanding in software maintenance.

  Method for software trustworthiness
  evaluation supporting dynamic and multiple attributes       

  ZHONG Xiaomin,ZHAO Xuefeng

  2013, 35(6): 107-112. doi:

  Abstract ( 95 )   PDF (578KB) ( 161 )   　　

  In order to support dynamic and multiattribute characteristics in software trustworthiness evaluation, a method was proposed with the grey number and the Minkowski distance function and fuzzy clustering algorithm. Firstly, trustworthy indicators are represented by gray numbers, and the ideal trustworthy indicators are calculated from the positive and negative ideal points. Secondly, the trustworthiness for each evaluation period is computed through the weighted grey number Minkowski distance function. Finally, the fuzzy clustering algorithm is adopted to aggregate the evaluations from each period. Experimental results and analysis show that the proposed method is effective and high accuracy, and does not cause more computational burden.

  Saliency detection based on global structure similarity metrics       

  LU Kai,XIONG Zhenhai,LI Chongfei,LI gen

  2013, 35(6): 113-117. doi:

  Abstract ( 128 )   PDF (1057KB) ( 229 )   　　

  Saliency detection is an important step image processing with computer, which includes the target recognition, image segmentation, and it is widely used in many fields such as network graphics and fingerprint recognition. In this paper, we proposed a global approach of structural similarity measure to significant target detection: using structural similarity to the human visual system’s highlevel abstraction, and taking into account the human visual characteristics and psychological feelings. We used the theory of structural similarity to measure the significant level of the target object, and get the saliency map after weight averaging the results. At last, the method of threshold was used to extract the salient target. Compared with the classic Itti algorithm, our approach can not only overcome the mosaic and take into account the global spatial information on the impact of structural similarity, but also select the parameters according to computing power and precision.

  Aircraft target recognition method based on
  discriminant locality preserving projections algorithm      

  ZHANG Shanwen,ZHANG Yunlong,SHANG Yijun

  2013, 35(6): 118-122. doi:

  Abstract ( 103 )   PDF (586KB) ( 201 )   　　

  Dimensionality reduction and feature extracting are important preprocessing steps in pattern recognition. Since the images of an aircraft target are much different from each other in the various conditions of the geographic transformations (such as rotation, translation, affine transform, etc) and the influence of different observed angle, locality and illumination in the real scene, many classical dimensional reduction and feature extracting methods are not effective to recognize the aircraft target by aircraft images. The discriminant locality preserving projection (DLPP) algorithm is an effective supervised dimensional reduction algorithm. The DLPP algorithm outperforms the LPP method, because it constructs the subspace by maximizing the betweenclass distance and minimizing the withinclass distance. In this paper, based on DLPP, an aircraft target recognition method was proposed. The recognition results show that the proposed method is very effective and feasible.

  Technique of keyframe extraction and
  motion editing based on wavelet transform            

  LIANG Jiahong,LI Ke,LI Meng,HU Yinglong

  2013, 35(6): 123-128. doi:

  Abstract ( 87 )   PDF (831KB) ( 208 )   　　

  For the problems of large motion capture data and the difficulty of editing key frames, the paper proposed a technique to introduce wavelet transform into key frames extraction and motion editing. The proposed technique deals with the motion capture data as signal processing, decomposes the signal into high frequency signals and low frequency signals, and disposes the coefficients by semisoft threshold function, and extracts key frames through setting different threshold. After extracting the key frames, the key frames animation is reconstructed by the interpolation technique in order to generate new human motion sequence. In motion editing, the high frequency signals are adjusted so as to enhance the characteristics of motion, further improving the technique of motion editing. The experiment demonstrates this technique is of good application, the extracted key frames have recapitulation, and after the motion editing, the reality of motion can be ensured to some extent.