Current Issue

• 2011, No. 10　Published:25 October 2011 Last issue    Next issue
  Columns in this issue: 论文

• 论文

  An Overview of the Techniques for Reliable and Secure EndtoEnd Communications

  HOU Jie,LIU Yaping,GONG Zhenghu,HUANG Jian

  2011, 33(10): 1-8. doi:

  Abstract ( 507 )   PDF (621KB) ( 545 )   　　

  With the rapid development of the Internet, endtoend communications are facing  new challenges on connectivity, mobility and security, mainly due to the widely deployed middleboxes on the edge of networks and the popularity of some new applications. Based on this, researchers bring forth many related solutions. This paper gives a review of the related works, and analyzes some challenging problems of the research, and then makes an analysis, and compares some representative solutions. Finally, the development trends on endtoend communications are discussed.

  Research of the Criteria of Available Bandwidth Estimation in Wireless Ad Hoc Networks

  SONG An,ZHAO Haitao,WANG Shan,WEI Jibo

  2011, 33(10): 9-14. doi:

  Abstract ( 381 )   PDF (633KB) ( 325 )   　　

  The criteria of available bandwidth estimation in wireless ad hoc networks are studied and the standard in judging the feasibility of bandwidth requirements is proposed where the overall QoS provision should be taken into consideration. An analytical model of IEEE 802.11based wireless networks is built and the QoS metrics such as delay, packet loss ratio and throughput are derived, and based on which the detailed criteria of available bandwidth estimation are presented. Simulations verify the accuracy of the analytical model and the validity of the criteria of available bandwidth estimation.

  A Routing Protocol with Network Life Time and Other Network Performance Balancing Based on Ant  Colony Optimization for Mobile Ad Hoc Networks

  REN Jingan,TU Yaqing,ZHANG Min,JIANG Yinhua,XIE Hongtao

  2011, 33(10): 15-24. doi:

  Abstract ( 472 )   PDF (1533KB) ( 382 )   　　

  This paper puts forward a routing protocol for mobile Ad Hoc networks called AntBased EnergyAware Routing Protocol (ABEAR), which is based on ant colony optimization (ACO). ABEAR starts the route setup procedure reactively by sending out artificial ants to find paths to the destination node. In the routing computation for data packets, ABEAR considers not only the global information but also the local information of every node, including the pheromone values, the linkquality and congestion metric, and the remaining energy of the next hop. Incorporating these information in the routing computation makes the neighbors with less remaining energy and links with high congestion be less selected. Nevertheless, based on the crosslayer methods, ABEAR turns off the idle network interfaces safely to conserve energy while guaranteeing the basic connectivity of the ad hoc network, and avoids network partitioning. In this way, ABEAR can balance life time and other network performance metrics, including packet delivery ratio and average endtoend delay. The simulation results on the NS2 platform show that ABEAR outperforms AODV (Ad hoc On Demand Distance Vector Routing) greatly in terms of life time, packet delivery ratio and average endtoend delay.

  Pairing Computation on the Jacobi Intersections

  TANG Chunming1,XU Maozhi1, 2,QI Yanfeng1

  2011, 33(10): 25-29. doi:

  Abstract ( 515 )   PDF (512KB) ( 378 )   　　

  So far, pairing computation is implemented on elliptic curves in the plane, such as the Weierstrass, Edwards and Jacobi quartic curves. This paper discusses pairing computation on elliptic curves in the threedimensional space for the first time. As elliptic curves in the threedimensional space for cryptography, intersections of quadric surfaces have important relations with the Edwards curves and the Jacobi quartic curves, which gives a deep comprehension of the Edwards curves and the Jacobi quartic curves. For simplicity, we just consider pairing computation on the Jacobi intersections. However, our results can be generalized to other intersections of the quadric surfaces. We first analyze the geometric properties of the Jacobi intersections and construct efficiently computable endomorphisms for the Jacobi intersections. Finally, we give pairing computation and optimization for the Jacobi intersections.

  The Autocorrelation Values of Several Cyclotomic Sequences Over GF(3)

  LEI Mingliang,YUE Qin

  2011, 33(10): 30-33. doi:

  Abstract ( 397 )   PDF (383KB) ( 409 )   　　

  Cyclotomic sequences with a few correlation values have wide applications in communication systems and cryptography. Let p≡1 mod 3. We calculate the autocorrelation values of a class of unbalanced cyclotomic and periodic p sequences of order three over GF(3). Moreover, we support a condition of prime p such that it has 3level autocorrelation values.

  Relationship Between Algebraic Immunity and Propagation Characteristics of the Boolean Functions

  ZHOU Yu,CAO Yunfei,ZHANG Wenzheng,ZHU Shixiong

  2011, 33(10): 34-38. doi:

  Abstract ( 390 )   PDF (451KB) ( 391 )   　　

  Using the relationship between the sum of square and algebraic immunity by GAC, the divisibility properties concerning the autocorrelation coefficient of the Boolean functions with propagation criterion is derived by Walshspectrum and coefficient, and the inequality among variables, algebraic immunity, propagation criteria and algebraic degree is deduced by this divisibility, and by the computer search methods a compactly expression among these four parameters is given from 4 variables to 30 variables. Finally, the relationships between the propagation criterion and the liner structure and normality are discussed.

  A Multimodel Differential Fault Analysis on PRESENT

  TANG Ming1,2,SHEN Fei2,DENG Hui2,YIN Peng2,QIU Zhenlong2,MA Xiao2,ZHANG

  2011, 33(10): 39-44. doi:

  Abstract ( 416 )   PDF (462KB) ( 398 )   　　

  PRESENT is an ultralightweight block cipher which is suitable for lightweight hardware such as the RFID tags and sensor networks. In this paper, the strength of PRESENT against the differential fault analysis is explored. We present four kinds of fault models of differential fault analysis on PRESENT. Comparing these methods, we come up with the best method to analyse PRESENT using differential fault analysis. Up to now, our method is proved to be more efficient than the existing differential fault analysis on DFA in the published papers. The best result is, by introducing a 8 bit random error between the 28th round permutation and the 29th round permutation, we can recover a 64 bit postwhitening key on an average of 17 fault samples.

  Design and Implementation of a XSS Vulnerability Whitebox Testing Framework

  FENG Kai,LIN Bogang

  2011, 33(10): 45-50. doi:

  Abstract ( 401 )   PDF (514KB) ( 389 )   　　

  With the popularity of the Web2.0 applications, crosssite scripting vulnerability which can cause a serious threat to the safety of users exists in a wide range of websites, due to the lack of proper verification mechanisms. Currently, the mixture of the serverside and clientside codes makes accurate and effective detection of crosssite scripting vulnerability more difficult. In our study, based on a static data flow analysis combined with a string constraint solving technique, we design a whitebox testing framework for detecting crosssite scripting vulnerability. We implement our framework in a prototype tool called XSSExplore, and the experimental results show that the system can better generate a more effective attack vector and detect crosssite scripting attacks compared with similar products.

  The Nonlinearity of Complementary Symmetric Boolean Functions

  CHEN Yindong1,LU Peizhong2

  2011, 33(10): 51-56. doi:

  Abstract ( 421 )   PDF (555KB) ( 322 )   　　

  Complementary symmetric Boolean functions are a special class of symmetric Boolean functions. A high proportion of symmetric Boolean functions with optimum algebraic immunity are complementary symmetric Boolean functions. Especially for the case of 2〖WTBX〗m〖WTBZ〗 variables, it reaches a high proportion of 2/3. By the relationship between the nonlinearity and the Walsh spectrum of the Boolean functions, and that between the Walsh spectrum of the Boolean functions and the Krawtchouk polynomial, the nonlinearity of complementary symmetric Boolean functions is determined. As a result, the nonlinearity of all complementary symmetric Boolean functions with n variables is2n－1－ 1/2(nn/2)。

  All Fiber Discrete Modulation ContinuousVariable Quantum Key Distribution

  WANG Xuyang,BAI Zengliang,LI Yongmin,PENG Kunchi

  2011, 33(10): 57-59. doi:

  Abstract ( 435 )   PDF (529KB) ( 422 )   　　

  In this paper, we give an all fiber discrete modulation continuousvariable quantum key distribution system. We introduce a quaternary discrete modulation protocol, and show the schematic of our experimental setup. The principle of the pulsed homodyne detector is analyzed and the test result about it is given. Finally we present the structure of pulses, the rules of encode, and the procedure to acquire a raw key.

  An Intrusion Detection System Based on Neural Network Ensembles

  XU Min，SHEN Xiaohong，GU Qi

  2011, 33(10): 60-63. doi:

  Abstract ( 393 )   PDF (628KB) ( 441 )   　　

  With the problems of the low detection rate and the insufficient sensitivity to the new intrusion，the current intrusion detection systems affect the functions of the entre system. Based on the very deep research, this paper proposes a new neural network ensembles method for intrusion detection. The method is used to train the individual networks on the basis of data reduction. Neural network techniques are used to combine the different classification results. Theory and experiment show that the model is effective.

  Full Custom CORDIC Arithmetic Unit Design

  BI Zhuo,DAI Yijun

  2011, 33(10): 64-69. doi:

  Abstract ( 393 )   PDF (984KB) ( 394 )   　　

  Floatingpoint trigonometric computing is a fundamental operation in navigation systems, 3D image processing, radar signal preprocessing and so on. This paper presents a floating point trigonometric computing circuit using a CORDIC algorithm and a full custom layout method, and the output data are compatible with the IEEE754 signal precision floatingpoint standard. It describes the CORDIC algorithm principle and chooses a pipeline structure based on the principle of priority on performance. A static circuit structure and a full custom layout are given in the SIMC 0.13μm 1P8M CMOS process. The silicon area of data path is 605 284μm2 and the critical path delay is 3.013ns in the SS (SlowSlow) corners.

  Simulation and Analysis of 10Gbps Signal Transmission on the SerialLink Backplane

  ZHENG Hao,JIN Lifeng

  2011, 33(10): 70-75. doi:

  Abstract ( 398 )   PDF (1210KB) ( 465 )   　　

  In this article, we firstly introduce the obstacles in the design of a 10Gbps serial backplane and how to estimate the performance of a channel accurately with specification. And then, we build a model and simulate its performance by the S parameter and eye diagram based on a typical backplane channel. In order to analyze the significant impact of crosstalk, we especially build a model of two sidebyside channels and analyze its NEXT and FEXT from the S parameter, and compare the eye openings of the signal transmitted with different crosstalks. The simulation results indicate that the impact of crosstalks with two different analysis methods can keep consistent and may be tolerated without errors. Finally, the performance of the backplane channel which has been verified by a test board can meet the specification of 10Gbps signal transmission.

  A Soft Error Tolerance Method for the I/O System of MultiCore Systems

  GUO Yufeng,GUO Songxin,GONG Rui

  2011, 33(10): 76-79. doi:

  Abstract ( 376 )   PDF (509KB) ( 357 )   　　

  With the development of integrated circuits, the transistor dimensions are reducing and the density is increasing, and ICs depend on work environment servers, so the possibility of soft error occuring becomes more often, which has important effect on the reliability of chips. Multicore microprocessors have become the mainstream over the past few yesrs, the redundant resources in the multicore microprocessors provide a redundant resource to enhance the reliability. A soft scrub method for multicore processors is proposed to tolerate soft errors. The experimental results demonstrate that this technique can greatly increase the reliability of the I/O systems.

  Pshare:A TwoLevel Adaptive Branch Predictor Algorithm and Its Implementation

  ZHANG Qibin,LI Qiang

  2011, 33(10): 80-84. doi:

  Abstract ( 583 )   PDF (472KB) ( 925 )   　　

  A branch prediction algorithm and predictor called Pshare is proposed, which has the benefit of PAs and the Gshare predictor,and modifies their flaws due to using a global history pattern by using a separate branch history shift register(BHSR).Moreover it also can reduce the area and delay overhead than PAs by accessing a small PHT table using the hash of address and history patterns instead of accessing two dimensions PHT table using the address and history patterns.The Pshare predictor can obtain the branch prediction accuracy of PAs ,but the implementation overhead is near to Gshare.This predictor is applied to high performance processors which have a  superscalar deep pipeline and focuses more on higher core performance.It reduces the pipeline’s vertical waste and improves the core performance and efficiency by using the Pshare predictor.

  Research and Development of the GeneralPurpose Computation on GPUs

  LIN Yisong,TANG Yuhua,TANG Tao

  2011, 33(10): 85-92. doi:

  Abstract ( 356 )   PDF (1082KB) ( 331 )   　　

  With the development of the semiconductor technology, the number of transistors integrated on a chip keeps increasing. Consequently, the computation and memory capacity of graphics processing units improve rapidly. So far, the floatingpoint computing capacity of GPUs has greatly exceeded that of CPUs, and the potential of GPUs in the nongraphic computing field, especially in high performance computing, has attracted more and more researchers’ attention. This paper gives an introduction to the principles of the general purpose computation on GPUs and the latest research results about architecture and the programming model of GPGPU from both the research community and industry.

  Design and Implementation of a Compiler for the SelfAdaptation Strategy Description Language

  DONG Menggao1，MAO Xinjun1，YANG Hua2，QI Zhichang1

  2011, 33(10): 93-98. doi:

  Abstract ( 379 )   PDF (1071KB) ( 368 )   　　

  The software entity in complex adaptive systems should implement the enterprise and adaptation functions. To implement the adaptation functions, the software entity senses the environment continually and adjusts its structure and behavior according to the environmental changes. However, the adaptation logic and enterprise logic of selfadaptive systems are often tangled together in existing approaches, which makes it difficult and complicated to develop and maintain selfadaptive systems. In this paper, we abstract the autonomous running entity in selfadaptive systems as the selfadaptive agent, and believe it is necessary to separate the selfadaptation logic and enterprise logic of selfadaptive systems. A SelfAdaptation strategy Description Language SADL is therefore presented to express how agents adapt to the changes. To compile the adaptation strategy into the executable program, we design and implement the SADL compiler. Furthermore, for illustrating the feasibility and effectiveness of our proposed approach, a case study is presented to describe how to define the selfadaptation strategy, and show the compilation results.

  A Software Runtime Verification Method Based on the 3Valued Semantics

  SUI Ping1,ZHAO Changzhi1,DONG Wei1,LI Bingpeng2

  2011, 33(10): 99-104. doi:

  Abstract ( 373 )   PDF (722KB) ( 296 )   　　

  Runtime verification complements the two traditional approaches for ensuring that a system is correct, namely model checking and testing. Unlike these approaches which try to ensure that all possible executions of the system are correct, runtime verification concentrates on the correctness of the current execution of the system. This paper presents a runtime verification method based on the 3valued semantics. One hand, this method provides a complete solution from code instrumentation and system information extraction to monitor generation and verifying requirement specification against runtime tracing. On the other hand, the monitor based on the 3valued semantics has the ability to find the smallest good (bad) prefix, so the monitor can find the violation as early as possible. Meanwhile, we have developed the prototype tool and have applied it in an example.

  Application of the Faade Pattern to the Data Persistence Layer

  ZHANG Li1,ZHANG Weixi2

  2011, 33(10): 105-110. doi:

  Abstract ( 360 )   PDF (769KB) ( 390 )   　　

  Data persistence is an important problem of development in the process of designing enterprise application systems. Based on the analysis and comparison of the application of the Faade pattern, the generic DAO pattern, the Hibernate framework and the Spring framework, a universal solution is proposed for the data persistence layer based on the Faade pattern, which makes the loose coupling possible. And the data persistence layer is highly stable, flexible and expansible. Finally, the technology of the data persistence layer is applied to the information system of a telecom customer manager.

  A Survey on Theoretical Combination Techniques of SMT Solvers

  LI Jing，LIU Wanwei

  2011, 33(10): 111-119. doi:

  Abstract ( 585 )   PDF (1238KB) ( 501 )   　　

  Satisfiability modulo theories solver is a decision procedure for the satisfiabilily of formulas of the firstorder logics. It is the verification engine for many formalization methods. Theory solvers solve problems with different theoretical backgrounds; however, the problems in the realworld applications often are supported by multiple theories. This paper mainly introduces the methods of theory combination, summarizes the status quo of the SMT solvers. Besides，it analyzes several major techniques on the theoretical combination for SMT solvers. Testing with the industrial cases, we finally evaluate methods of combination theory with experimental results, and compare the  performances of several popular SMT solvers which  support theoretical combination techniques.