Computer Engineering & Science

Recent progress in high-performance
parallel computing of the aerospace area

GONG Chunye1,2,3，BAO Weimin1,2，TANG Guojian2，WANG Ling1，SUN Xuegong1，LIU Jie

2014, 36(09): 1629-1636. doi:

Abstract ( 348 )

PDF (622KB) ( 266 ) 　　

The numerical simulation of the large scale science and engineering problem in the aerospace area relies on the support of HighPerformance Parallel Computing (HPPC) and is the engine of the development of HPPC. The paper gives a survey of recent progress of HPPC in the aerospace area and a brief introduction of the HPPC environment. The relative research areas of HPPC of aerospace, including aerodynamic, aerothermal, chemical nonequilibrium, structure, heat protection, Monte Carlo method and turbulence et al., are classified and described in detail. The dilemma of high parallel efficiency in scientific computing and low use in engineering computing, and the diversity of parallel applications and the lack of scientific parallel methods are concluded. Finally, the prospects of the future HPPC development in aerospace area are presented.

Research of digital predistortion in nonlinear power amplifier
with memory based on parallel evolutionary computation

LIU Zhao,HU Li

2014, 36(09): 1637-1643. doi:

Abstract ( 117 )

PDF (938KB) ( 187 ) 　　

Adaptive digital predistortion is the most promising technique to overcome the nonlinearity of High Power Amplifier (HPA). In order to improve the efficiency and effectiveness of the predistortion, the evolutionary computation techniques of the parallel computing platform are introduced, the method of training neural network in advance based on the PSO algorithm is proposed, and the basic process of the algorithm is given. Based on the above, a threelayer forward neural network predistorter with two inputs and two outputs is proposed for HPA with memory. The predistorter is realized using indirect learning architecture associated with the Backpropagation algorithm .This technique allows us to correct for general nonlinearities and memory effects simultaneously. Simulation results show that the new approach is more efficient than the conventional BP algorithm, without training in advance based on PSO.

A tasks scheduling algorithm for grid
computing based on fuzzy multi-objective

QIAO Fu

2014, 36(09): 1644-1649. doi:

Abstract ( 123 )

PDF (714KB) ( 192 ) 　　

Users, resource administrators, group managers are entities of heterogeneous grid computing in the grid computing platform. They require grid management, use and maintenance, security, reliability, and so on. These objectives are not often quantitative. Aiming at the tasks scheduling problem for grid computing based on multiobjective, a tasks scheduling model for grid computing based on fuzzy multiobjective is proposed. Fuzzy equation is used to process multi-objective, and the theorem of the fuzzy equation is proposed and proved. Utilizing the merit of the differential evolutionary algorithm that the objective function can be not continuous and differentiable, a fuzzy differential evolutionary algorithm is proposed to process tasks scheduling for grid computing based on fuzzy multiobjective. In the experiments, the proposed algorithm is compared with the existing algorithms. Although the proposed algorithm is not better on the execution time, it outperforms the existing algorithms in terms of reliability, security and drooped task numbers.

Research and implementation of the
fast burning technology of smart card COS

HUANG Yiping1,NONG Liping2,SU Jiande1

2014, 36(09): 1650-1655. doi:

Abstract ( 114 )

PDF (735KB) ( 175 ) 　　

Through studying and analyzing the existing reading and writing technology of USB and serial readerwriter of smart card COS, a master hardware platform based on ARM processor is proposed. This method is a kind of fast burning copy method, i.e. fast burning the other smart cards’ COS through special interface of smart cards. Furthermore, the method features multiple advantages such as simple hardware design, fast burning, high stability and high safety, and it is easily extended to be the burning platform for a batch of smart card COS. At present, this technology has been extensively applied by the paste smart card manufacturers and increased the COS burning efficiency considerably, proving that it has a good application prospect.

Channel allocation and link scheduling
in variable-width wireless networks

CHEN Xiaohui,XU Jing,LIU Wei,CHENG Wenqing

2014, 36(09): 1656-1661. doi:

Abstract ( 103 )

PDF (722KB) ( 139 ) 　　

Taking account of the impacts of channel-width on the transmission range and interinterference of links, the existing cumulative interference model is improved and the channel allocation and link scheduling problem in variable channel-width wireless networks is analyzed. A twolayer algorithm is proposed to decouple the channel allocation and link scheduling in the problem and a heuristic approach considering the links-traffic satisfactions is devised to obtain the links priorities and construct the channel allocation matrices for concurrent transmissions. Simulation results show that the proposed algorithm can get convergence in reasonable time and the heuristic approach can efficiently construct the channel allocation matrices.

A token-ring-like real-time response algorithm
of Modbus/TCP message based on μC/OS-II

SUN Zhonghao,ZHOU Xingshe,ZHANG Kailong

2014, 36(09): 1662-1667. doi:

Abstract ( 116 )

PDF (696KB) ( 134 ) 　　

Modbus/TCP industrial Ethernet standard cannot guarantee the realtime capability by itself. A token ring scheduling algorithm is designed, which uses priority to achieve the realtime response to highpriority Modbus/TCP messages. Based on that, a weighted token ring scheduling algorithm and a weightvariable token ring scheduling algorithm are developed to meet the customizable and changeable time delay demands. Finally, the response delay on a μC/OSII+ARM7 platform is measured to be 1.15±0.15ms, which is superior to most current researches that aim at improving the realtime capability of Modbus/TCP, and what’s more, it reaches the level of POWERLINK working in open mode.

A method of measuring the overall resource
efficiency of the MIMO-OFDM system

LIU Jinzhu1，WANG Meng2

2014, 36(09): 1668-1671. doi:

Abstract ( 121 )

PDF (455KB) ( 146 ) 　　

The concepts of the normalized bandwidth and the normalized signal-to-noise ratio are defined, which are used to construct the plane model. Based on the model, a method of measuring the overall resource efficiency of the MIMOOFDM system is proposed. This method obtains the overall resource efficiency of the MIMO-OFDM system under different detection algorithms and different modulation. The main conclusion is that the MIMO-OFDM system with the maximum likelihood detection algorithm has the relatively high overall resource efficiency, and using Mary quadrature amplitude modulation (MQAM) has higher overall resource efficiency than employing Mary phase shift keying (MPSK) while its overall resources efficiency improves with the increase of M.

Spectrum allocation algorithm in cognitive radio
based on immune-clonal-selection and difference

LIU Sheng1,CHEN Zhigang1,KUANG Zhufang2

2014, 36(09): 1672-1677. doi:

Abstract ( 102 )

PDF (728KB) ( 152 ) 　　

Spectrum allocation is an important issue in cognitive radio, but the spectrum difference is rarely discussed in traditional spectrum allocation algorithms. A new spectrum allocation algorithm is proposed, which is based on Immune-Clonal-Selection (ICS) and takes account of the spectrum difference. A reliability matrix is introduced to model the time difference of spectrum. The algorithm shows a rise in total network bandwidth by using the Difference-Constraint Satisfaction Operator when two cognitive users have interference using the same spectrum. The DifferenceConstraint Satisfaction Operator can allocate spectrum in higher credibility to cognitive users. The algorithm also shows an improvement in fairness by using the Fairness-Constraint Satisfaction Operator to deal with the interference. The Fairness-Constraint Satisfaction Operator can allocate spectrum to cognitive users with more intensive conflicts. Simulation results show that, compared with the traditional ICS algorithm and color-sensitive graph coloring (CSGC) algorithm and Genetic algorithm (GA), DICS can achieve higher total network bandwidth, higher credibility and better fairness.

Method of NetMagic hardware development

CAO Chengzhou,MAO Jianbiao,SUN Zhigang,YIN Jiabin,LIN Qi,GONG Xiaolin

2014, 36(09): 1678-1683. doi:

Abstract ( 142 )

PDF (700KB) ( 224 ) 　　

NetMagic has been widely applied as an experimental platform to support network research. In order to further reduce the complexity of the hardware development of the NetMagic platform, the principal of separation of decision and forwarding is adopted to expand the interface of the Rule. The development methodology based on UM (userdefined logic) and its development interface specification not only simplify the design of UM, but also support the optimization of NetMagic resources efficiently. NetMagicbased NProbe network probe is an example to introduce the basic process of UM development.

Measurement and analysis of end-to-end #br# latency in high bandwidth multicast network

LI Shanxi, MA Qiang, CHEN Wenbo

2014, 36(09): 1684-1689. doi:

Abstract ( 125 )

PDF (720KB) ( 175 ) 　　

With the rapid development of IP multicast technology, the increased user demand for multimedia video stream leads to the performance degradation of the IP multicast network. To solve the IP multicast network latency issues in performance measurement, in the highspeed campus network, the accurate measurement of endtoend latency in multicast network is achieved according to the multicast performance testing and measurement methods defined in the RFC standard document. And the principle of clock synchronization is used to do clock frequency difference analysis on the measurement results. The contribution of the paper is that the linear programming algorithm is applied into the measurement model's analytic procedure. Through measurement experiments in the highspeed campus network, the results indicate that the measurement results amended by the algorithm can have an accurate endtoend network latency. The study lays the groundwork for popularizing the multicast streaming media applications in the campus network, and is more conducive to find network bottlenecks that lay the foundation for the build of future networks.

HMM risk assessment model on the exploitation levels

CHEN Yongyan,DAI Wei

2014, 36(09): 1690-1696. doi:

Abstract ( 83 )

PDF (710KB) ( 154 ) 　　

The security dependency exists among the nodes in the network vulnerability exploitation graph.Three types of dependencies are calculated to obtain dependencies among nodes.The transition matrix of the dependencies among nodes is constructed.At last, the Hidden Markov Model (HMM) model is used and the risks in different exploitation levels are given.The experiments show that the method can be more accurate to express the actual vulnerability security risks.

Research of workflow access control
mechanism based on object status

WU Ling,WANG Hao

2014, 36(09): 1697-1704. doi:

Abstract ( 106 )

PDF (874KB) ( 145 ) 　　

To overcome the insufficiency of object level in access control mechanism, the access control mechanism of object level is integrated in the taskrolebased access control.This mechanism takes into account the effects of object status changes for user access permissions. Granularity of access control is accurate to express individual object status.When designing the access control mechanism,the access rules under different conditions separates from the access management structure. By this way, the accesscontrol mechanism can be used in different workflow processes.Objectoriented design patterns are used in the system design. Finally,the mechanism is applied onto the online authoring and publishing system of a multimedia digital museum to assure that the digital content is correct and appropriate.

Reflections on community e-government
network security and data sharing

CHEN Daqian,REN Guangwei

2014, 36(09): 1705-1710. doi:

Abstract ( 120 )

PDF (965KB) ( 155 ) 　　

A brief description of the VPN and Web Services technologies and their characteristics is provided.Using VPN technology to solve community e-government network security issues on the public network and using Web Services technology to solve issues of data sharing among heterogeneous systems are analyzed.Then,a constructional idea of community e-government private network based on VPN and Web Services technology is proposed.A VPN design case of a community e-government private network in Guiyang city is given,and the advantages and disadvantages of VPN technology are analyzed.Finally,a community e-government application integration platform example of a community e-government private network in Guiyang city is given,and the advantages and disadvantages of Web Services are analyzed.

An secure and efficient dynamic password
authentication protocol based on smart card and fingerprint

HUANG Chaoyang,WU Zhengying,LUO Shaolan

2014, 36(09): 1711-1715. doi:

Abstract ( 108 )

PDF (391KB) ( 173 ) 　　

The shortages existing in Khan et al’s scheme are analyzed carefully and its possible attack methods and steps are expounded.Then,an improved remote authentication scheme is introduced,which is based on biologic character and smart card.The new scheme ueses simple and effective hash and exclusiveor operations and can fill the many loopholes in the original scheme.Therefore,the proposal is an efficient and feasible remote authentication protocol.

Multi-strategy particle swarm optimization algorithm

CAO Ju,CHEN Gang,LI Yanjiao

2014, 36(09): 1716-1721. doi:

Abstract ( 119 )

PDF (624KB) ( 197 ) 　　

Like many other intelligent optimization algorithms,Particle Swarm Optimization (PSO) algorithm often suffers from premature convergence,especially in multipeak problems.In addition,the convergence accuracy of its local search is unsatisfactory.To overcome these problems,a novel improved PSO,named Multistrategy Particle Swarm Optimization (MPSO) algorithm,is proposed.In the evolution of particle swarms,each particle chooses its own contemporary optimal search strategy from multiple alternative strategies according to the changes of optimal position it finds.Among these strategies,the steepest descent strategy,the corrective decline strategy and the random mobile strategy are able to be chosen by the optimal particle,while the aggregation strategy and the diffusion strategy are available for nonoptimal particles.In the end,the performance of MPSO is tested with four typical test functions and numerical results indicate that the proposed MPSO algorithm has a stronger and more stable global search ability than the standard PSO algorithm.

Critical chain project management
based on the improved ant colony algorithm

LUO Yuebin,ZHAO Yingjun,WANG Hua

2014, 36(09): 1722-1727. doi:

Abstract ( 124 )

PDF (556KB) ( 164 ) 　　

The progress is important in project management.As one of the commonly used schedule management methods in current project management,the critical chain method is a multiconstrained optimization problem in its nature.Chaos Ant AlgorithmGenetic Algorithm (CAAGA) is used to solve the optimal scheduling problem based on Critical Chain Project Management (CCPM). By making use of the characteristics of ergodicity and randomness of chaotic motion, the traditional Ant Colony Algorithm (ACA) is improved.It can get rid of the ACA’s faults that more time is required to seek the solution because of lacking pheromone at the beginning and it is easily trapped into the local solution.It extends the search area,improves the evolution speed of ant colony,and maintains the robust and convergence the ant colony algorithm.It has higher computing accuracy and speed.An example shows that the CAAGA method is better than the AGA method in solving this issue.

A hybrid algorithm of bee colony and invasive weed
optimization and its application in irrigation schedule optimization

GONG Yaxing1，WANG Lianguo1,2

2014, 36(09): 1728-1735. doi:

Abstract ( 88 )

PDF (710KB) ( 150 ) 　　

Firstly, aiming at the premature convergence problem of Invasive Weed Optimization (IWO) algorithm,a hybrid algorithm of Bee Colony and Invasive Weed Optimization (BCIWO) is proposed by introducing the optimization mechanism of Artificial Bee Colony (ABC) algorithm. Every individual in the weed colony is mutated by employed bees’ search behavior and the global best individual is mutated by onlookers’ search method in this improved algorithm. The better result of mutation is used to replace the original individual.Those improve the convergence speed and the accuracy of IWO.Then the optimal performance of BCIWO is verified by test functions.Finally,the new algorithm is applied into the irrigation schedule optimization problem,thus providing a promising way to allocate irrigation water.

Fault diagnosis based on intuitionistic fuzzy Petri nets

SUN Xiaoling,WANG Ning

2014, 36(09): 1736-1741. doi:

Abstract ( 94 )

PDF (540KB) ( 184 ) 　　

Fault diagnosis is often affected by many uncertain and fuzzy factors. In view of the uncertain fault diagnosis, an intuitionistic fuzzy Petri net model is constructed by considering that intuitionistic fuzzy sets can better express uncertainty information and Petri networks has strong parallel processing and graphics processing ability. Because nonmembership parameters are involved when Intuitionistic fuzzy reasoning is transformed into matrix operations, the reasoning result can offer more information. According to the actual fuzzy inference problem in the process of fault diagnosis, a new fuzzy reasoning algorithm with weights, threshold value parameters is given. Through obtaining and processing the uncertain and fuzzy knowledge in the process of fault diagnosis, the algorithm converts the fault diagnosis to intuitionistic fuzzy reasoning by using intuitionistic fuzzy Petri nets. The actual gas turbine fault diagnosis model shows that the intuitionistic fuzzy reasoning algorithm is effective.

Survey on complex electromagnetic
environment modeling and visualization

GAO Ying1,ZHANG Zheng2,WANG Fenghua1,GUO Shuxia3

2014, 36(09): 1742-1749. doi:

Abstract ( 228 )

PDF (1067KB) ( 378 ) 　　

The research of complex electromagnetic environment modeling and visualization is widespread in the field of civil communication and electronic war. Firstly, the issues and classification are introduced by focusing on the modeling of complex electromagnetic environment and the theoretical methods of visualization. Secondly, the current situation home and aboard of complex electromagnetic environment modeling and visualization are analyzed. The radar's electromagnetic environment modeling of complex electromagnetic environment is studied and the complexity of the model, the prediction accuracy is summarized and compared. For the massive amount of data characteristics of the data field of the electromagnet, the basic idea of the electromagnetic volume data visualization and the scope of application are summarized from the aspects of graphics hardware acceleration and software algorithms. Finally, the development trend of complex electromagnetic environment modeling and visualization is predicted.

Fuzzy C-means clustering based microscopic
image segmentation with illumination robustness

WEI Weiyi，LIN Xianghong，GAO Zhiling，YANG Xiaodong

2014, 36(09): 1750-1753. doi:

Abstract ( 130 )

PDF (565KB) ( 173 ) 　　

It influences the microscopic image segmentation results that traditional fuzzy Cmeans (FCM) ignores the spatial intensity variations caused by uneven illumination, so a novel FCM based microscopic image segmentation algorithm with illumination robustness is proposed, which models the uneven illumination as a linear combination of a set of orthogonal basis functions. Meanwhile, the energy function in FCM is modified by means of the proposed model to support image fuzzy segmentation. The algorithm can alleviate the impact of uneven illumination on segmentation effect and estimate the uneven illumination fields simultaneously. The experimental results show that the algorithm is high efficient.

Current Issue

Author center

Review center

Online journal