Loading...
  • 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

Current Issue

    • 论文
      An Integrated Group Key Management Scheme with Hierarchical Access Control
      GU Xiaozhuo
      2012, 34(7): 1-5. doi:
      Abstract ( 287 )   PDF (401KB) ( 364 )     

      The group key management which is efficient to perform the key updating is the foundation to guarantee the security and privacy of Mobile Ad Hoc Networks(MANET). Inspired by the hierarchical access control in the centralized environment, this paper proposes an integrated group key agreement scheme (IGK) to provide hierarchical access control in MANETs when members have accesses to multiple resources. Performance analyses on computation and communication overhead demonstrate that the integrated group key agreement is more efficient than the single group key agreement scheme.

      A Distributed Key Management Scheme for the Group Signature Based on Authentication in VANETs
      SUN Yipin,HU Qiaolin,SU Jinshu
      2012, 34(7): 6-11. doi:
      Abstract ( 265 )   PDF (469KB) ( 341 )     

      Groupsignature based authentication is a promising approach for addressing the security and privacy issues in vehicular ad hoc networks(VANETs). However, it is prone to causing huge revocation overhead in VANETs with millions of nodes and serious security risks. To solve this problem, we develop a distributed key management scheme(DKM)where the whole domain of VANET is divided into several subregions, and any vehicle has to update its group secret key periodically from the regional group manager who manages the region where the vehicle stays. In this way, a revoked membership is just notified in a subregion but not the whole domain. Therefore, the average size of the revocation list in each subregion decreases. Moreover, the proposed key updating process which guarantees a vehicle can obtain an updated group secret key from a regional authority without leaking the value of the group secret key to the regional authority. Performance analysis demonstrates that DKM can reduce the revocation cost significantly while keeping the authentication overhead the same as the the original group signature algorithm.

      An EnergyEfficient Multipath Routing Scheme for Wireless Sensor Networks
      FAN Zhiping1,2,JIN Zhengzhe1,XIE Dongqing2
      2012, 34(7): 12-17. doi:
      Abstract ( 237 )   PDF (575KB) ( 339 )     

      The energy supply of nodes will be limited strictly in wireless sensor networks (WSNs). Considering the characteristics, a new scheme called energyefficient multipath routing (EMR) is presented. Based on AODV,this scheme analyses path hopcount, residual energy of nodes and energy status of sensor networks. The packets can be transmitted along the multiple paths according to the minimum hop or their Key Energy Ratio,reduces the network energy consumption, and avoids heavy traffic on some critical nodes. The simulation results show that compared with AODV the scheme effectively improves the packet delivery ratio, and the endtoend delay,delays the emergence of the death node, and prolongs the network lifetime.

      Research on the Handoff Mechanisms for Wireless Bus Networks
      KUANG Luobei,XU Ming,YU Wei,CHEN Yingwen
      2012, 34(7): 18-23. doi:
      Abstract ( 232 )   PDF (613KB) ( 307 )     

      Providing high quality Internet services in public buses allows the bus passengers to entertain themselves and even to work remotely during their travel time, thus significantly improving their quality of life. This paper focuses on the handoff mechanisms for bus networks. Firstly, a data transmission rate based handoff trigger is proposed based on an improved bus network structure. The trigger can be fired for a handoff accurately with low cost and high communication stability. Further, an adaptive handoff mechanism called  MHandoff is studied for the scenario in which there exist multiple APs (Access Points) in a bus station. The mechanism aims to guarantee load balancing in all APs to execute AP selection and also can achieve high average bus throughput for each associated bus. Through simulation experiments, the above mechanisms can improve system performance more greatly and support bus communication better than the traditional methods.

      A Modified Approach for Byte Frequency based Payload Anomaly Intrusion Detection
      WENG Guangan1,YU Shengsheng2,ZHOU Jingli2
      2012, 34(7): 24-28. doi:
      Abstract ( 216 )   PDF (813KB) ( 281 )     

      The content characteristics of datasets have strong effect on the detection accuracy of network anomaly intrusion detection systems. The influences impacted on byte frequency distribution based models by the differences between content characteristics of the training packets are analyzed, revealing that those differences would lead the models calculating the average frequency of grouped packets to a higher false alarm rate. Based on this, a modified model named single packet frequency distribution is proposed, which uses the frequency distribution data of the unitary packet to form normal profiles instead of using their average values, and controlls the size of that normal set by clustering techniques. Experiments are  carried out respectively on the simulation dataset and the DARPA99 real network dataset. The results indicate that the great difference between packet contents in deed makes the average byte frequency value based models generating more false alarms, whereas the single packet frequency distribution model is not affected by that, and it gets higher detection accuracy, generating an equal detection rate with the lower false alarm rate. The average value based model even becomes invalid at the worst case. The single packet frequency distribution model can be considered having good adaptability to those network services with rich dynamic contents.

      FISDR: A New Fault Injection Performance Evaluation System for Wireless Sensor Networks
      HUANG Xu1,2,CHEN Dongyan1,LI Hui1,SHAO Zhuyu1,YU Leilei1
      2012, 34(7): 29-34. doi:
      Abstract ( 248 )   PDF (1257KB) ( 274 )     

      In wireless sensor network (WSNs), the reliability and fault tolerance are the important metrics to evaluate the stability of WSNs. In actual use of WSNs, there often happen many faults and interferences, so we can use the Fault Injection (FI) technology to inject these faults and interferences into WSNs artificially. Through the observation of network reaction we can evaluate the reliability and fault tolerance of the network, so as to make the changes of the network mechanism in order to improve reliability and stability. In this paper we put forward a new performance evaluation system for WSN reliability evaluation with FI, and its inject method based on software, and connect the WSN nodes through a special port in a onetoone manner, so we can inject the fault command to the WSN nodes. First, the FISDR can effectively inject the various practical faults and interferences which WSNs likely encounter, and we can observe the network reaction; Second, the FISDR can receive including WSN nodes and other various equipment through special ports and store them; Finally, the FISDR is equipped with the PC software to monitor the network topology structure, record the transmission rate and analyze the enormou amount of information stored  in devices, so that we can test WSNs and evaluate the reliability. We test this system in a fivestorey office and we use dozens of WSN nodes and FISDR nodes separately. Our experiments mainly include such contents. We use FISDR to inject the mass number of faults to WSNs and record the network reaction, so we can verify the effect of FISDR fault injection, so as to test and analyze the FISDR performance. The experimental results show that FISDR can inject various kinds of faults to WSNs effectively and evaluate the WSNs reliability effectively, and it has high an application value in WSN testing and reliability evaluation.

      The Optimization of the WRR Algorithm in MultiClass RealTime Data Scheduling
      XIONG Liyan,ZHANG Shenghui
      2012, 34(7): 35-38. doi:
      Abstract ( 263 )   PDF (697KB) ( 331 )     

      With the development of converged networks, the quality of service (including the available bandwidth, end to end delay, jitter and packet loss rate) for some realtime data streaming applications (voice flow, video flow, etc.) become more and more important. The traditional WRR algorithm can only meet the fairness of the realtime queue, but it can not assure that the multiclass realtime data is low latency and low jitter. The BSTLRR algorithm is based on the WRR scheduling algorithm. The BSTLRR scheduling algorithm not only in the low delay and low jitter multiclass realtime data stream frame is superior to the WRR scheduling algorithm, and to some extent, the priority queue ensures the fairness of scheduling.

      Research on Alert Records Perception of Network Situation Awareness
      WANG Juan 1,PENG Jing2,WANG Can3
      2012, 34(7): 39-45. doi:
      Abstract ( 235 )   PDF (1634KB) ( 354 )     

      The alert perception of network situation awareness is different from that of the traditional intrusion detection area in particle size, scale, target, and so on. It pays more attention to human understanding. Based on the existing similarity based alert analysis method, a "similarity based macro network alert awareness algorithm" is proposed. It gives a new definition of attribute similarity, and uses an "optimal sequence method" to improve attribute weight setting. Finally, a threshold selection scheme is proposed  based on human instantaneous understanding. The experimental results show that this method can help network managers get the whole awareness of network situation including the time, range, and type of network abnormality.

      Transactional Memory:A Concurrent Control Mechanism with FaultTolerant Characteristics
      SONG Wei
      2012, 34(7): 46-53. doi:
      Abstract ( 254 )   PDF (525KB) ( 302 )     

      With the development of the multicore processor, developing threadlevel parallelism becomes a necessary means to improve the execution performance of the application programs. It makes transactional memory, which is a promising mechanism to support threadlevel parallelism, attract more and more attention. In this paper, firstly, we classify the transactional memory systems according to the conflict detection mechanism and the dataversioning mechanism. After that, we  summarize the main implementations of the transactional memory. Finally, we reexamine the transactional memory from the perspective of faulttolerance. We believe that the transactional memory has good characteristics of faulttolerance, and it can combine with some faulttolerant techniques naturally to achieve efficient fault isolation, fault detection and fault recovery.

      Runtime Partitioning Technique of Hybrid Distributed Shared Memory Space in Multicore Processors
      CHEN Xiaowen1,CHEN Shuming1,LU Zhonghai2,Axel Jantsch2
      2012, 34(7): 54-59. doi:
      Abstract ( 287 )   PDF (811KB) ( 372 )     

      In multicore processors, Distributed Shared Memory (DSM) offers ease of programming by maintaining a global virtual memory space as well as imports the inherent overhead of translating virtual memory addresses into physical memory addresses, resulting in negative performance. We observe that, in parallel applications, different data have different properties (private or shared). Even for the same datum, its property may be changeable in different phases of the program execution. This paper firstly introduces a hybrid DSM, aiming at supporting fast and physical memory accesses for private data and maintaining a global and single virtual memory space for shared data. A runtime partitioning technique is proposed to change the hybrid DSM organization during the program execution. It ensures fast physical memory addressing on private data and conventional virtual memory addressing on shared data, improving the performance of the entire system by reducing virtualtophysical address translation overhead as much as possible. The experimental results show that the hybrid DSM with runtime partitioning demonstrates performance advantage over the conventional DSM counterpart. The percentage of performance improvement depends on network size, problem size, way of data partitioning, etc. In our experiments, the maximal improvement is 13.14%, and the minimal improvement 6.98%.

      A BlockMatching Algorithm Based Accelerator for SAD Computation
      GU Huitao,CHEN Shuming
      2012, 34(7): 60-64. doi:
      Abstract ( 285 )   PDF (649KB) ( 245 )     

      Blockmatching based motion estimation is one of the most important techniques in image and video applications. The sum of absolute difference (SAD) is the major computation in motion estimation and requires huge computation complexity and transmission bandwidth. This paper proposes a reconfigurable SAD accelerator, in which a 16×1 processing elements (PE) array and an adder tree structure are used to improve the execution speed of SAD computation. The pipeline partition of PE array and adder tree is performed carefully in order to increase the work frequency. In order to reduce the performance loss caused by data transfer delay, a DMA event mechanism is employed to transmit data when the SAD accelerator is working. The experimental results show that, the proposed architecture needs 4102 cycles for searching a 16×16 search window. With a 0.13μm CMOS standard cell technology, the proposed accelerator requires only 16.8 k gates and 3.5 KB of memory at the 750MHz operation frequency.

      Design and Implementation of an Asynchronous Embedded Processor:TengYueⅡ
      SU Bo,SHI Wei,WANG Zhiying,REN Hongguang,WANG Yourui
      2012, 34(7): 65-70. doi:
      Abstract ( 244 )   PDF (918KB) ( 325 )     

      In embedded systems, the power consumption of processors is restricted to a low limit. Asynchronous circuit techniques can be effective methods to design low power processors. In order to meet the performance and power requirements of the multimedia applications in embedded systems, a low power asynchronous microprocessor named “TengYueⅡ”  is designed and implemented. A 32bit onchip bus, two transport triggered architecture cores, two memory controllers and several communication interfaces form the processor. One processor core is based on asynchronous circuits and the other one is its synchronous counterpart. TengYueⅡ processor is implemented in the UMC 0.18μm CMOS technology and the supply voltage is 1.8V. The area of the chip is 4.89mm×4.89mm. The experimental results show that both cores can run correctly at the frequency of 200MHz and the asynchronous core’s power consumption is lower than 50% of the synchronous core.

      An Automatic Vectorization and Data Permutation Oriented Intermediate Representation
      CHEN Xiang,SHEN Li
      2012, 34(7): 71-77. doi:
      Abstract ( 246 )   PDF (967KB) ( 263 )     

      Compilers or programmers have to use a permutation instruction to reorganize the element of vectors to get the correct operands for the SIMD instructions in the design of a SIMD program recently. But these redundant permutation instructions result in performance loss. This paper proposes a new intermediate representation, which contains enough address messages for the operands of scalar and vector. It makes the generation of data permutations be postponed and the redundant permutation instruction be reduced. We utilize this new intermediate representation to implement an optimization algorithm on data permutation, which can effectively reduce the performance loss from permutation instruction. The test result to a group of typical multimedia program shows that the algorithm can achieve 7% performance acceleration on average.

      Optimization of Sparse Diagonal MatrixVector Multiplication Based on the CUDA Program Model
      QIN Jin,GONG Chunye,HU Qingfeng,LIU Jie
      2012, 34(7): 78-83. doi:
      Abstract ( 315 )   PDF (829KB) ( 357 )     

      Sparse matrixvector multiplication is often an important computational kernel in many scientific applications. This paper faces the ndiagonal sparse matrix, uses the CUDA program model and describes a new compress format of sparse matrix based on the DIA compress format (CDIA), and gives each thread finegrained task distribution. In order to fulfill the characteristics of the align access of memory in CUDA, we transpose the compress matrix and design a  finegrained algorithm and program and do  some optimization to the program. In the data experiment, our best implementation achieves up to 39.6Gflop/s in singleprecision and 19.6Gflop/s in doubleprecision, and enhances the performance by about 7.6% and 17.4% that of Nathan Bell’s and Michael Garland’s respectively.

      A Quick Method for  Inversing a Circulant Matrix on GPU
      ZHENG Zuoyong,ZHANG Ruixia
      2012, 34(7): 84-88. doi:
      Abstract ( 518 )   PDF (652KB) ( 498 )     

      Circulant matrix is a special case of the Toeplitz matrix, which is widely used in many domains of specialization, especially in image and digtial signal processing. Calculating the inverse of this category of matrices consists of the following three steps: (1) transform the first row vector to frequency space by using DFT; (2) calculate the inverse of each amplitude in the spectrum; (3) apply IDFT to the adjusted spectrum to acquire the first row of the inverse matrix, and finally reconstruct it. In this algorithm, the computational process involving each matrix element is entirely identical, and independent of the process of any other element, therefore, it adequately adapts to the modern GPU. This paper implements such a fast algorithm on the GPU, which is transferred to rendering a square. The experimental results prove that the GPU version is ten times faster than the CPU one.

      A Formalized Representation of the OntoUML Model Based on Description Logic
      QI Yudong,YANG Bin,LI Ying,XIE Xiaofang
      2012, 34(7): 89-92. doi:
      Abstract ( 265 )   PDF (488KB) ( 296 )     

      The formal ontological theory is one of the main methods applied to UML formalization and improvement. On the basis of Ontology UML has been extended to OntoUML,and it provides a richer, better version able to express the semantics of realworld modeling primitives, but its expression is not used in information system design and development. Based on description logic, the OntoUML primitives and relationship between them are  formally represented with SHIQ, and a case study is given. The method of expression is concise, has clear semantics, and is not only able to express the logical model clearly, but also to ensure that the model represents things of domain correctly. This study promotes OntoUML to be more widely used, and to some extent, it provides a theoretical and application support to improving ontologybased information systems’ conceptual modeling approach.

      Research on Behavioral Semantics for the Subset of the AADL Process
      MIAO Decheng1,2,XI Jianqing2,SU Jindian2
      2012, 34(7): 93-98. doi:
      Abstract ( 279 )   PDF (651KB) ( 264 )     

      AADL is a semiformal modeling language based on components, which models software and hardware for largescale complex software systems uniformly by structured description, and describes effectively system functional behaviors, nonfunctional attributes and architecture dynamic evolution in runtime, but some questions of AADL are required to further stduy and improve. This paper analyzes firstly the status quo of research for the AADL formal semantics, defines the formal language for the subset of the AADL process, makes a communication model for the subset of the AADL process, defines the concept of event formally, and clarifies the important role of event during the system status transformation, studies the behavioral semantics for the subset of the AADL process, and finally illustrates our advantages by contrasting with other literatures. This paper provides beneficial consultation for the development of AADL and its formal semantics and further improves the technology of architecture modeling and analysing for largescale complex software systems.

      Research on Flight Trajectory Recreation and ThreeDimensional Flight Playback
      ZHAO Xiangling
      2012, 34(7): 99-103. doi:
      Abstract ( 344 )   PDF (517KB) ( 580 )     

      For the analysis after flight, flight trajectories are  gotten using civil flight airborne data by integral approximation and are corrected by latitude and longitude transformation. Geometric attitude is used as flight trajectories attitude for the first time in flight playback. For dynamic flight playback in threedimensional scenes, a flight playback program is  designed, especially displaying the change of flight trajectories and gestures. The program demonstration indicates that this method can make the flight playback precisely and fluently, which is easytoobserve and easytoanalyze, and can be used to flight inspection and research after flight.

      An Improved Sperm Image Sequence Segmentation Method Based on Kalman Filtering
      YU Dong,HUANG Wenming,WEN Peizhi,NING Ruhua,HUANG Jinfang
      2012, 34(7): 104-108. doi:
      Abstract ( 281 )   PDF (571KB) ( 376 )     

      Image segmentation in the analysis of sperm's motility plays an important role. Firstly,the collected sperm image sequence are carried out a preprocessing,such as graying and denoising,then the first piece of pretreated sperm image’s binarization using the Otsu algorithm and the threshold range of binarization of the subsequent images can be determined by the Kalman filter, which improves the Otsu algorithm to calculate the images’ appropriate threshold and binarization under the predicated threshold range,and reduces time. In order to get highquality segmentation images for the analysis of sperm’s motility,we apply morphology to eliminate sperm tail and part of the phenomenon of sperm adhesion,and by comparing with the area and shape factors to remove the tiny noisy particles and impurities in the sperm image.

      A ZeroWatermarking Algorithm Based on Image Affine Invariant Feature Points
      NIE Xuelian,DAI Qing
      2012, 34(7): 109-113. doi:
      Abstract ( 263 )   PDF (633KB) ( 321 )     

      This paper proposes an image zerowatermarking algorithm based on lifting scheme wavelet and affine invariant feature points.The image is firstly decomposed by threelevel lifting wavelet transformation,and then the affine invariant feature points and affine covariant feature regions are  extracted from the low frequency subband by utilizing the HarrisAffine point detector,with which the zerowatermarking information is constructed.It is demonstrated by the experiments that the proposed algorithm is not only robust against common image processing operations and simple geometric attacks,but also with strong resistance to complex geometric attacks such as shearing,aspect ratio change,row or column removal,local distortion,etc.