ISSN 
2010 Vol. 39, No. 5
2010, 39(5): 644-650.
doi: 10.3969/j.issn.1001-0548.2010.05.001
Abstract:
Based on Barabási's paper in the special issue "Complex Systems and Networks" published byScience on 2009, some problems, e.g., the BA model, hierarchical networks, network dynamics, degree correlation,and network Markov chains, are discussed. Under an Markov chain framework, we provide a simple iterativemethod instead of simulation to compute network degree distribution for a class of growing networks, and thensome general criteria for judging whether a set of network generating mechanisms can ensure the existence of thesteady-stage degree distribution and whether the steady-stage degree distribution is scale-free. These results showextreme importance of strengthening basic theory and applied research for futher development of complexnetworks.
Based on Barabási's paper in the special issue "Complex Systems and Networks" published byScience on 2009, some problems, e.g., the BA model, hierarchical networks, network dynamics, degree correlation,and network Markov chains, are discussed. Under an Markov chain framework, we provide a simple iterativemethod instead of simulation to compute network degree distribution for a class of growing networks, and thensome general criteria for judging whether a set of network generating mechanisms can ensure the existence of thesteady-stage degree distribution and whether the steady-stage degree distribution is scale-free. These results showextreme importance of strengthening basic theory and applied research for futher development of complexnetworks.
2010, 39(5): 651-661.
doi: 10.3969/j.issn.1001-0548.2010.05.002
Abstract:
Link prediction aims at estimating the likelihood of the existence of links between nodes. Thepredicting of existent yet unknown links is similar to the data mining process, while the predicting of future linksrelates to the network evolution. The traditional methods are based on Markov Chains and machine learning whichusually involve the node attributes information. Although these methods can give good prediction, the highcomputational complexity limits their applications in large-scale systems. The approaches based on maximumlikelihood approximation also suffer this shortcoming. Another group of methods is based on the node similaritythat is defined solely based on the network structure. Extensive experiments on many real networks show that thesimilarity-based methods can give good prediction while with lower computational complexity comparing with theabove two kinds of methods. This article introduces and compares many representative link prediction methods,and outlines some important open problems, which may be valuable for related research domains.
Link prediction aims at estimating the likelihood of the existence of links between nodes. Thepredicting of existent yet unknown links is similar to the data mining process, while the predicting of future linksrelates to the network evolution. The traditional methods are based on Markov Chains and machine learning whichusually involve the node attributes information. Although these methods can give good prediction, the highcomputational complexity limits their applications in large-scale systems. The approaches based on maximumlikelihood approximation also suffer this shortcoming. Another group of methods is based on the node similaritythat is defined solely based on the network structure. Extensive experiments on many real networks show that thesimilarity-based methods can give good prediction while with lower computational complexity comparing with theabove two kinds of methods. This article introduces and compares many representative link prediction methods,and outlines some important open problems, which may be valuable for related research domains.
2010, 39(5): 662-665.
doi: 10.3969/j.issn.1001-0548.2010.05.003
Abstract:
Aiming at the multipath diversity reception in non-spectrum-spreading system, a RAKEcombining receiver is presented in this paper. The estimate of the transmitted data is obtained by the ⅡR (infiniteimpulse response) equalizer with decision feedback (DF) structure, and each multipath signal can be resumed bythis initial solution. Then the multipath elements in the received signal are obtained, and the components arecombined together using maximal ratio combining (MRC) before the final decision, so RAKE combining receiveris accomplished. The cumulated round off noise phenomenon is eliminated by DF-ⅡR, however, error propagationexists in the decision directed filter. While the signal-to-noise ratio (SNR) can be increased evidently after MRC,the error propagation problem is depressed dramatically in the combining receiver. Simulations show that theproposed algorithm can achieve the performance advantages effectively.
Aiming at the multipath diversity reception in non-spectrum-spreading system, a RAKEcombining receiver is presented in this paper. The estimate of the transmitted data is obtained by the ⅡR (infiniteimpulse response) equalizer with decision feedback (DF) structure, and each multipath signal can be resumed bythis initial solution. Then the multipath elements in the received signal are obtained, and the components arecombined together using maximal ratio combining (MRC) before the final decision, so RAKE combining receiveris accomplished. The cumulated round off noise phenomenon is eliminated by DF-ⅡR, however, error propagationexists in the decision directed filter. While the signal-to-noise ratio (SNR) can be increased evidently after MRC,the error propagation problem is depressed dramatically in the combining receiver. Simulations show that theproposed algorithm can achieve the performance advantages effectively.
Analysis of Joint Polarization-Space-Time ProcessingMethods with Applications in Phased Array Radars
2010, 39(5): 666-669,675.
doi: 10.3969/j.issn.1001-0548.2010.05.004
Abstract:
Most of previous conclusions on the target detection ability increase of phased array radars byutilizing the joint polarization-space-time processing methods were obtained under the environment of Gaussianclutters. The real clutter environments are much more complex than that showed by Gaussian clutter models. Bymeans of the theoretical analysis and the simulation experiments, the target detection ability of phased array radarwith the polarization-space-time generalized likelihood radio (PST-GLR) algorithm and joint polarizationzspace-time localize GLR (PSTL-GLR) algorithm are analyzed under the environment of non-Gaussian clutters.The results show that the PST-GLR algorithm is robust against non-Gaussian clutters, but it requires largenumbers of echo data to estimate the covariance matrix of clutters and noises in the process of target detection.Although PSTL-GLR algorithm requires less echo data in the process of target detection, its target detection abilitydegraded seriously so that it is invalidation under the environment of non-Gaussian clutters.
Most of previous conclusions on the target detection ability increase of phased array radars byutilizing the joint polarization-space-time processing methods were obtained under the environment of Gaussianclutters. The real clutter environments are much more complex than that showed by Gaussian clutter models. Bymeans of the theoretical analysis and the simulation experiments, the target detection ability of phased array radarwith the polarization-space-time generalized likelihood radio (PST-GLR) algorithm and joint polarizationzspace-time localize GLR (PSTL-GLR) algorithm are analyzed under the environment of non-Gaussian clutters.The results show that the PST-GLR algorithm is robust against non-Gaussian clutters, but it requires largenumbers of echo data to estimate the covariance matrix of clutters and noises in the process of target detection.Although PSTL-GLR algorithm requires less echo data in the process of target detection, its target detection abilitydegraded seriously so that it is invalidation under the environment of non-Gaussian clutters.
2010, 39(5): 670-675.
doi: 10.3969/j.issn.1001-0548.2010.05.005
Abstract:
Multi-input multi-output (MIMO) systems can achieve more capacity than single input singleoutput systems. For MIMO systems, maximum likelihood detection is the optimum detection but its exponentialcomplexity limits its application. In this paper, a complex lattice reduction algorithm based on householdertransform is proposed and combined with MIMO suboptimal detection algorithm. Moreover the complexity of thiscomplex lattice reduction algorithm is lower than that of real lattice reduction algorithm. Simulation results showthat MIMO suboptimal detection algorithm based on this complex lattice reduction can approach the optimumperformance of maximum likelihood detection by optimizing channel matrix to get better decision domain.
Multi-input multi-output (MIMO) systems can achieve more capacity than single input singleoutput systems. For MIMO systems, maximum likelihood detection is the optimum detection but its exponentialcomplexity limits its application. In this paper, a complex lattice reduction algorithm based on householdertransform is proposed and combined with MIMO suboptimal detection algorithm. Moreover the complexity of thiscomplex lattice reduction algorithm is lower than that of real lattice reduction algorithm. Simulation results showthat MIMO suboptimal detection algorithm based on this complex lattice reduction can approach the optimumperformance of maximum likelihood detection by optimizing channel matrix to get better decision domain.
2010, 39(5): 676-679,705.
doi: 10.3969/j.issn.1001-0548.2010.05.006
Abstract:
To improve performance of single frequency networks employing distributed transmit antennas,some transmit signals are introduced in this paper. Pair-wise error probability is derived and one transmit signal isproposed to obtain the achievable diversity order. Then, two full-diversity (maximum diversity) transmit signaldesign methods are proposed. To optimize the first proposed transmit signal, one interleaved-partition subcarriergrouping scheme, establishing a tradeoff between receivers' location and performance, is introduced. Simulationresults are presented in a representative scenario to demonstrate the performance gains offered by the proposedtransmit signals.
To improve performance of single frequency networks employing distributed transmit antennas,some transmit signals are introduced in this paper. Pair-wise error probability is derived and one transmit signal isproposed to obtain the achievable diversity order. Then, two full-diversity (maximum diversity) transmit signaldesign methods are proposed. To optimize the first proposed transmit signal, one interleaved-partition subcarriergrouping scheme, establishing a tradeoff between receivers' location and performance, is introduced. Simulationresults are presented in a representative scenario to demonstrate the performance gains offered by the proposedtransmit signals.
2010, 39(5): 680-683.
doi: 10.3969/j.issn.1001-0548.2010.05.007
Abstract:
This paper proposes a high efficient finite-difference time-domain (FDTD) parallel algorithmscheme with no exchanging data process in shared memory systems for the electrically large electromagneticproblems, which takes the advantage of shared memories of the general-purpose multicore processors architecture.FDTD parallel algorithm is realized by directly reading the field component data on the parallel boundary and themultithreaded programming technique is adapted to develop the parallel computation code of FDTD. Through aseries of the numerical tests, a way of the splitting region is proposed to implement the maximum parallelcomputational efficiency. It is verified that the proposed scheme is an idea parallel FDTD scheme in sharedmemory systems.
This paper proposes a high efficient finite-difference time-domain (FDTD) parallel algorithmscheme with no exchanging data process in shared memory systems for the electrically large electromagneticproblems, which takes the advantage of shared memories of the general-purpose multicore processors architecture.FDTD parallel algorithm is realized by directly reading the field component data on the parallel boundary and themultithreaded programming technique is adapted to develop the parallel computation code of FDTD. Through aseries of the numerical tests, a way of the splitting region is proposed to implement the maximum parallelcomputational efficiency. It is verified that the proposed scheme is an idea parallel FDTD scheme in sharedmemory systems.
2010, 39(5): 684-687.
doi: 10.3969/j.issn.1001-0548.2010.05.008
Abstract:
A precoding scheme for solving the co-channel interference (CCI) problem of multiusermulti-input multi-output(MIMO) downlink is proposed. This scheme can implement the joint designs of transmitterand receiver and avoid iteration operation under the minimum mean square error (MMSE) criterion by regardingthe receiver as a matched filter. The CCI in the proposed scheme is eliminated fully by introducing the zero-force(ZF) method into the MMSE criterion. To simplify the joint design further, the vector optimal design problem istransformed into scalar optimal design problem with the singular value decomposition(SVD) of the equivalentchannel matrix. Simulation results show that proposed precoding scheme can achieve better average bit error rate(BER) and sum capacity performance than other non-iterative design scheme.
A precoding scheme for solving the co-channel interference (CCI) problem of multiusermulti-input multi-output(MIMO) downlink is proposed. This scheme can implement the joint designs of transmitterand receiver and avoid iteration operation under the minimum mean square error (MMSE) criterion by regardingthe receiver as a matched filter. The CCI in the proposed scheme is eliminated fully by introducing the zero-force(ZF) method into the MMSE criterion. To simplify the joint design further, the vector optimal design problem istransformed into scalar optimal design problem with the singular value decomposition(SVD) of the equivalentchannel matrix. Simulation results show that proposed precoding scheme can achieve better average bit error rate(BER) and sum capacity performance than other non-iterative design scheme.
2010, 39(5): 688-691.
doi: 10.3969/j.issn.1001-0548.2010.05.009
Abstract:
The orthogonal performance of between signals is crucial for implementing multiple inputmultiple output (MIMO) radar system. Orthogonal discrete frequency coding waveform (DFCW) can be utilized toMIMO radar. This paper firstly analyses the ambiguity of DFCW and investigates the autocorrelation sidelobe peak(ASP) and cross-correlation peak (CP) level. The genetic algorithm is also proposed to optimize DFCW with lowercross-correlation peak. Then the designed DFCW is utilized to MIMO radar and its matched filter performance isanalyzed. Simulation results validate the investigation of cross-ambiguity and the analysis.
The orthogonal performance of between signals is crucial for implementing multiple inputmultiple output (MIMO) radar system. Orthogonal discrete frequency coding waveform (DFCW) can be utilized toMIMO radar. This paper firstly analyses the ambiguity of DFCW and investigates the autocorrelation sidelobe peak(ASP) and cross-correlation peak (CP) level. The genetic algorithm is also proposed to optimize DFCW with lowercross-correlation peak. Then the designed DFCW is utilized to MIMO radar and its matched filter performance isanalyzed. Simulation results validate the investigation of cross-ambiguity and the analysis.
2010, 39(5): 692-695,705.
doi: 10.3969/j.issn.1001-0548.2010.05.010
Abstract:
Because of the importance of the discrete cosine transform (DCT) in the field of image processing,various algorithms and architectures for 2-D DCT processor design have been proposed. In this paper, a novel fastcomputing mechanism for 2-D 8×8 DCT and quantization for JPEG or MPEG codec is presented. By an effectivejudging mechanism for an 8×8 image block, the algorithm will adjust DCT computing time depending on thedistribution of an image block. This algorithm costs a few adders, shifters and comparators, but it reducessignificantly the number of DCT computing times which dominates the computing performance of imageprocessing. The simulation result shows that the proposed algorithm could save more DCT calculation thanconventional DCT and integer DCT, and when quantization parameter is large, such as 32, the performance ofproposed algorithm is better than that of small one.
Because of the importance of the discrete cosine transform (DCT) in the field of image processing,various algorithms and architectures for 2-D DCT processor design have been proposed. In this paper, a novel fastcomputing mechanism for 2-D 8×8 DCT and quantization for JPEG or MPEG codec is presented. By an effectivejudging mechanism for an 8×8 image block, the algorithm will adjust DCT computing time depending on thedistribution of an image block. This algorithm costs a few adders, shifters and comparators, but it reducessignificantly the number of DCT computing times which dominates the computing performance of imageprocessing. The simulation result shows that the proposed algorithm could save more DCT calculation thanconventional DCT and integer DCT, and when quantization parameter is large, such as 32, the performance ofproposed algorithm is better than that of small one.
2010, 39(5): 696-700.
doi: 10.3969/j.issn.1001-0548.2010.05.011
Abstract:
In this paper, the impact of Gaussian estimation error on the performance of vertical bell labslayered space-time (V-BLAST) architecture is investigated by using maximum likelihood (ML) receiver inindependently and identically distributed (i.i.d.) Rayleigh fading channel. Based on the statistical property ofsquared Euclidean distance, the pairwise error probability (PEP) for V-BLAST system over i.i.d. Rayleigh fadingchannel with channel estimate error is derived. The approximate expression of the PEP for high signal-to-noiseration (SNR) region is used to analyze the loss of diversity order and penalty of SNR. It shows that the diversityorder achieved by ML receiver is not affected by channel estimation error and the loss of SNR due to imperfectchannel estimation is 3 dB. Simulation results are given to corroborate the theoretical analysis.
In this paper, the impact of Gaussian estimation error on the performance of vertical bell labslayered space-time (V-BLAST) architecture is investigated by using maximum likelihood (ML) receiver inindependently and identically distributed (i.i.d.) Rayleigh fading channel. Based on the statistical property ofsquared Euclidean distance, the pairwise error probability (PEP) for V-BLAST system over i.i.d. Rayleigh fadingchannel with channel estimate error is derived. The approximate expression of the PEP for high signal-to-noiseration (SNR) region is used to analyze the loss of diversity order and penalty of SNR. It shows that the diversityorder achieved by ML receiver is not affected by channel estimation error and the loss of SNR due to imperfectchannel estimation is 3 dB. Simulation results are given to corroborate the theoretical analysis.
2010, 39(5): 701-705.
doi: 10.3969/j.issn.1001-0548.2010.05.012
Abstract:
In the sinusoidal pulse width modulation (SPWM) inverters, the output voltage contains lots ofhigh frequency harmonics. These harmonics may bring about serious electromagnetic interface problem and reducethe reliability of the system. According to output harmonic analysis of the conventional SPWM inverter, this paperproposes a harmonic suppression method adopted random frequency modulation in the SPWM inverter. Thismethod can make the harmonic spectrum of inverter output voltage be continuously distributes without affectingthe fundamental frequency component, and thus the harmonic energy of inverter output can distribute more equably.By adopted this method, the peak value of output voltage harmonic spectrum decreased and electromagneticinterface reduced. Based on the theoretical analysis and simulation, a 1 kW experimental prototype is designed.Comparative test shows that the peak value of harmonic in the SPWM inverter adopted our proposed method isreduced about 10dB than the conventional one.
In the sinusoidal pulse width modulation (SPWM) inverters, the output voltage contains lots ofhigh frequency harmonics. These harmonics may bring about serious electromagnetic interface problem and reducethe reliability of the system. According to output harmonic analysis of the conventional SPWM inverter, this paperproposes a harmonic suppression method adopted random frequency modulation in the SPWM inverter. Thismethod can make the harmonic spectrum of inverter output voltage be continuously distributes without affectingthe fundamental frequency component, and thus the harmonic energy of inverter output can distribute more equably.By adopted this method, the peak value of output voltage harmonic spectrum decreased and electromagneticinterface reduced. Based on the theoretical analysis and simulation, a 1 kW experimental prototype is designed.Comparative test shows that the peak value of harmonic in the SPWM inverter adopted our proposed method isreduced about 10dB than the conventional one.
2010, 39(5): 706-710,724.
doi: 10.3969/j.issn.1001-0548.2010.05.013
Abstract:
In this paper, an innovative air-borne forward-looking SAR system using antenna array is studied,and three-dimensional imaging principle is analyzed. A three-dimensional imaging process is proposed forair-borne forward-looking SAR based on the image geometry and the characteristic of echo signal, and theresolutions in three dimensions are derived. Raw data of forward-looking SAR in X-band is simulated andthree-dimensional image is achieved. The imaging result is analyzed, and simulation results show that theforward-looking SAR could achieve three-dimensional imaging, and also confirm the validity of the proposedalgorithm.
In this paper, an innovative air-borne forward-looking SAR system using antenna array is studied,and three-dimensional imaging principle is analyzed. A three-dimensional imaging process is proposed forair-borne forward-looking SAR based on the image geometry and the characteristic of echo signal, and theresolutions in three dimensions are derived. Raw data of forward-looking SAR in X-band is simulated andthree-dimensional image is achieved. The imaging result is analyzed, and simulation results show that theforward-looking SAR could achieve three-dimensional imaging, and also confirm the validity of the proposedalgorithm.
2010, 39(5): 711-715,730.
doi: 10.3969/j.issn.1001-0548.2010.05.014
Abstract:
Efficient compression for hyperspectral imagery has been the research focus for the developmentof remote sensing technique. The small targets information protection during the compression process without anypreknowledge should be necessarily considered. This paper presents a new lossy compression method forhyperspectral imagery based on fast independent component analysis (FastICA). Virtual dimensionality isintroduced to determine the number of target endmembers. The mixing matrix of FastICA is initialized by targetendmembers. Minimum noise fraction is employed for dimensionality reduction of original data volumes, andFastICA is performed on the selected principal components to generate independent components. Then, constantfalse alarm rate detection is performed on each IC, which is followed by morphologic filtering. Karhunen-Loevetransform is used to decorrelate the spectral redundancy, general scaling-based method is selected to upshift thewavelet coefficients of interested targets. Finally, each principle component is allocated optimal rate andcompressed by SPIHT algorithm. Experimental results on AVIRIS data show that the proposed method not onlyprovides high compression performance, but also preserves targets interested effectively.
Efficient compression for hyperspectral imagery has been the research focus for the developmentof remote sensing technique. The small targets information protection during the compression process without anypreknowledge should be necessarily considered. This paper presents a new lossy compression method forhyperspectral imagery based on fast independent component analysis (FastICA). Virtual dimensionality isintroduced to determine the number of target endmembers. The mixing matrix of FastICA is initialized by targetendmembers. Minimum noise fraction is employed for dimensionality reduction of original data volumes, andFastICA is performed on the selected principal components to generate independent components. Then, constantfalse alarm rate detection is performed on each IC, which is followed by morphologic filtering. Karhunen-Loevetransform is used to decorrelate the spectral redundancy, general scaling-based method is selected to upshift thewavelet coefficients of interested targets. Finally, each principle component is allocated optimal rate andcompressed by SPIHT algorithm. Experimental results on AVIRIS data show that the proposed method not onlyprovides high compression performance, but also preserves targets interested effectively.
2010, 39(5): 716-719,730.
doi: 10.3969/j.issn.1001-0548.2010.05.015
Abstract:
The goal of this paper is to characterize the nonlinear dynamic behavior of RF power amplifiersusing the general memory polynomial (GMP) model whose coefficients are extracted by the multichannel inverseQR decomposition recursive least-squares (IQRD-RLS) algorithms. The IQRD-RLS approach solves directly forthe time-recursive least squares weight vector, while avoiding the highly serial backsubstitution step required inpreviously derived QRD approaches. Furthermore, the approach employs Givens rotation operations to recursivelyupdate the model coefficients and is suitable to be implemented using systolic array to achieve fast convergenceand good numerical properties. The model is validated by means of ADS simulation data. The result indicates thatmodel is capable of satisfactorily describing the nonlinear dynamic behavior of wideband RF power amplifiers.
The goal of this paper is to characterize the nonlinear dynamic behavior of RF power amplifiersusing the general memory polynomial (GMP) model whose coefficients are extracted by the multichannel inverseQR decomposition recursive least-squares (IQRD-RLS) algorithms. The IQRD-RLS approach solves directly forthe time-recursive least squares weight vector, while avoiding the highly serial backsubstitution step required inpreviously derived QRD approaches. Furthermore, the approach employs Givens rotation operations to recursivelyupdate the model coefficients and is suitable to be implemented using systolic array to achieve fast convergenceand good numerical properties. The model is validated by means of ADS simulation data. The result indicates thatmodel is capable of satisfactorily describing the nonlinear dynamic behavior of wideband RF power amplifiers.
2010, 39(5): 720-724.
doi: 10.3969/j.issn.1001-0548.2010.05.016
Abstract:
The top-down electromagnetic compatibility (EMC) design flow and its applications inmulti-vehicle communication system, which is based on EDA software, are introduced. For the EMC problemssuch as antenna interaction, enclosure leakage and radiation of printed circuit board (PCB) in a four-vehiclecommunication system, the design flow from system level and device level to PCB level is present with simulationprocedure. The results show that this design flow can obviously improve the electromagnetic immunity of thesystem and make the design first pass success.
The top-down electromagnetic compatibility (EMC) design flow and its applications inmulti-vehicle communication system, which is based on EDA software, are introduced. For the EMC problemssuch as antenna interaction, enclosure leakage and radiation of printed circuit board (PCB) in a four-vehiclecommunication system, the design flow from system level and device level to PCB level is present with simulationprocedure. The results show that this design flow can obviously improve the electromagnetic immunity of thesystem and make the design first pass success.
2010, 39(5): 731-736.
doi: 10.3969/j.issn.1001-0548.2010.05.018
Abstract:
A method about 3D visualization of radar detection range in complicated environment is presented. Based on the formula of radar, this method considers the influence of terrain and electronic jamming. First, a radar detection range model in the free space is built by using uniform sampling method, and then according to the terrain elevation around the radar, an algorithm of 3D visualization of radar detection range is proposed to represent the 3D radar detection range affected by terrain. Based on jamming formula, the radar detection range considering the influence of multi-electronic jamming is implemented. Besides, by using view-dependent simplification of model the speed of rendering is improved. The result of experiment shows that this method not only is efficient but also can visualize radar detection range.
A method about 3D visualization of radar detection range in complicated environment is presented. Based on the formula of radar, this method considers the influence of terrain and electronic jamming. First, a radar detection range model in the free space is built by using uniform sampling method, and then according to the terrain elevation around the radar, an algorithm of 3D visualization of radar detection range is proposed to represent the 3D radar detection range affected by terrain. Based on jamming formula, the radar detection range considering the influence of multi-electronic jamming is implemented. Besides, by using view-dependent simplification of model the speed of rendering is improved. The result of experiment shows that this method not only is efficient but also can visualize radar detection range.
2010, 39(5): 737-741.
doi: 10.3969/j.issn.1001-0548.2010.05.019
Abstract:
The time-frequency transformation for detecting high-frequency Continuous Wave(CW) signalwith heavy background noise has a noticeable decline in performance. A new method for detecting weak CW signalby use of the nonlinear bistable stochastic resonance model is proposed. Firstly, the received CW signal ismodulated to low frequency, then filtered by stochastic resonance model and processed by Wigner-VilleDistribution (WVD) transformation and Hough transformation in turns, thus the detection of CW signal withinnoise can be easily implemented by a certain peak value in (ρ,θ) parameter space. Simulation studies show thismethod performs very well in the detection of CW signal with low SNR. The new method can be applied todevelop the automatic receiver of high-frequency CW telegraph signal.
The time-frequency transformation for detecting high-frequency Continuous Wave(CW) signalwith heavy background noise has a noticeable decline in performance. A new method for detecting weak CW signalby use of the nonlinear bistable stochastic resonance model is proposed. Firstly, the received CW signal ismodulated to low frequency, then filtered by stochastic resonance model and processed by Wigner-VilleDistribution (WVD) transformation and Hough transformation in turns, thus the detection of CW signal withinnoise can be easily implemented by a certain peak value in (ρ,θ) parameter space. Simulation studies show thismethod performs very well in the detection of CW signal with low SNR. The new method can be applied todevelop the automatic receiver of high-frequency CW telegraph signal.
2010, 39(5): 742-746,773.
doi: 10.3969/j.issn.1001-0548.2010.05.020
Abstract:
A mathematical model of precise harmonic drives transmission system including the effects offriction and transmission flexibility is developed and identified experimentally. In our model, the characteristics ofthe harmonic transmission are expressed as a function, the frictions in both motor and load side are analysed,nonlinear friction is described as a function of position and velocity, the average friction is described by a secondorder polynomial, and the periodical part of friction is simulated by Fourier series. The consistency of simulationresults and experimental outcomes confirm the effectiveness of the proposed model.
A mathematical model of precise harmonic drives transmission system including the effects offriction and transmission flexibility is developed and identified experimentally. In our model, the characteristics ofthe harmonic transmission are expressed as a function, the frictions in both motor and load side are analysed,nonlinear friction is described as a function of position and velocity, the average friction is described by a secondorder polynomial, and the periodical part of friction is simulated by Fourier series. The consistency of simulationresults and experimental outcomes confirm the effectiveness of the proposed model.
2010, 39(5): 747-751.
doi: 10.3969/j.issn.1001-0548.2010.05.021
Abstract:
On the basis of the analysis of the present service processes and data analysis method ofautomobile service station, the service capability elements of service station are researched and defined. Theservice capability model is build and calculated base on the structural equation models and the maximum likelihoodmethod by using after-sale service data of one automobile 4S shop. The quantitative relationships model among theservice capability elements is built and the quantitative impacts of these elements on customer satisfaction arereflected. According to the above research, the relative suggestion is put forward to improve the service quality andcustomer satisfaction purposefully, and the suggestion has been accepted by some 4S shop showing better results.
On the basis of the analysis of the present service processes and data analysis method ofautomobile service station, the service capability elements of service station are researched and defined. Theservice capability model is build and calculated base on the structural equation models and the maximum likelihoodmethod by using after-sale service data of one automobile 4S shop. The quantitative relationships model among theservice capability elements is built and the quantitative impacts of these elements on customer satisfaction arereflected. According to the above research, the relative suggestion is put forward to improve the service quality andcustomer satisfaction purposefully, and the suggestion has been accepted by some 4S shop showing better results.
2010, 39(5): 752-756.
doi: 10.3969/j.issn.1001-0548.2010.05.022
Abstract:
1. Sichuan Electric Power Test and Research Institute Chengdu 610072;2. School of Computer Science and Engineering, University of Electronic Science and Technology of China Chengdu 610054;3. Physical Education College, Zhengzhou University Zhengzhou 450044)
1. Sichuan Electric Power Test and Research Institute Chengdu 610072;2. School of Computer Science and Engineering, University of Electronic Science and Technology of China Chengdu 610054;3. Physical Education College, Zhengzhou University Zhengzhou 450044)
2010, 39(5): 757-761,773.
doi: 10.3969/j.issn.1001-0548.2010.05.023
Abstract:
Most frequent item sets mining is the focus and the difficulty of text association rules mining, andit directly determines the performance of text association rules mining algorithms. Firstly, several most frequentitem sets mining algorithms are analyzd and summarized. And then, traditional inverted list is improved. Based onthe improved list and set theory, a new TOP-N most frequent itemset mining algorithm combined minimum supportthreshold dynamic adjustment strategy is presented. In addition, several propositions and deductions for improvingthe performance of the performance of the provided algorithm are offered. Experimental results show that theprovided algorithm is better than Napriori and IntvMatrix.
Most frequent item sets mining is the focus and the difficulty of text association rules mining, andit directly determines the performance of text association rules mining algorithms. Firstly, several most frequentitem sets mining algorithms are analyzd and summarized. And then, traditional inverted list is improved. Based onthe improved list and set theory, a new TOP-N most frequent itemset mining algorithm combined minimum supportthreshold dynamic adjustment strategy is presented. In addition, several propositions and deductions for improvingthe performance of the performance of the provided algorithm are offered. Experimental results show that theprovided algorithm is better than Napriori and IntvMatrix.
2010, 39(5): 762-768.
doi: 10.3969/j.issn.1001-0548.2010.05.024
Abstract:
Aimed at the numerical computation of singular value decomposition (SVD) of large scale matrix, the characteristic of single direction shrink quadrature right-triangle (QR) algorithm is summarized systematically, and it is revealed by an example that this QR algorithm may not converge when it is used to process the SVD of large scale matrices created by some small amplitude signals. The variation characteristic of Givens matrices in QR iteration process is studied theoretically and it is found out that the essential reason of non-convergence of this QR algorithm lies in the attenuation of elements in diagonal belt of first row of large scale matrix. This attenuation will finally make the first right Givens matrix become an identity matrix, and then all the back Givens matrices will become identity matrices too, so the invalidation of QR algorithm is caused. On this basis, the influence of attenuation of elements of first row on convergence speed of QR algorithm is further studied. All the theoretical analysis is verified by the real data, and therefore the convergence characteristic of this QR algorithm is ascertained essentially.
Aimed at the numerical computation of singular value decomposition (SVD) of large scale matrix, the characteristic of single direction shrink quadrature right-triangle (QR) algorithm is summarized systematically, and it is revealed by an example that this QR algorithm may not converge when it is used to process the SVD of large scale matrices created by some small amplitude signals. The variation characteristic of Givens matrices in QR iteration process is studied theoretically and it is found out that the essential reason of non-convergence of this QR algorithm lies in the attenuation of elements in diagonal belt of first row of large scale matrix. This attenuation will finally make the first right Givens matrix become an identity matrix, and then all the back Givens matrices will become identity matrices too, so the invalidation of QR algorithm is caused. On this basis, the influence of attenuation of elements of first row on convergence speed of QR algorithm is further studied. All the theoretical analysis is verified by the real data, and therefore the convergence characteristic of this QR algorithm is ascertained essentially.
2010, 39(5): 769-773.
doi: 10.3969/j.issn.1001-0548.2010.05.025
Abstract:
A new LSB steganography method is proposed. According to the parity of carrier's image pixel tobe embedded and the content of embedded information, the embedded way of the pixel adding 1 or 2 is chosen tomake the hypothesis of SPA (sample pair analysis) attack keep balance before or after the embedding, causing SPAattack disable. When the proposed algorithm is used to make the embedding rate reaching 100%, the estimatedvalue of the SPA attack is still smaller than judge threshold. Experimental results validate that the new method notonly can resist the SPA attack, but also can resist the test such as RS. The embedding mode of new method isconvenient to achieve without additional disposal after embedding.
A new LSB steganography method is proposed. According to the parity of carrier's image pixel tobe embedded and the content of embedded information, the embedded way of the pixel adding 1 or 2 is chosen tomake the hypothesis of SPA (sample pair analysis) attack keep balance before or after the embedding, causing SPAattack disable. When the proposed algorithm is used to make the embedding rate reaching 100%, the estimatedvalue of the SPA attack is still smaller than judge threshold. Experimental results validate that the new method notonly can resist the SPA attack, but also can resist the test such as RS. The embedding mode of new method isconvenient to achieve without additional disposal after embedding.
2010, 39(5): 774-777,792.
doi: 10.3969/j.issn.1001-0548.2010.05.026
Abstract:
The phenomenon of coherence resonance (CR) in the coupled FitzHugh-Nagumo (FHN) neuronmodels driven by the colored noise is investigated. Based on the method of forth steps stochastic Runge-Kuttamethods, the interval-spike-interval (ISI) of the neuron firing and the coefficient of variation (CV) have beenobtained. It is shown that the CV is a non-monotonic function of the correlation of colored noise, the intensity ofcolored noise, and the coupling numbers. Furthermore, by choosing the appropriate the correlation of colored noise,the intensity of colored noise, and the coupling numbers, the minimum of the CV can be attained, the phenomenonof coherence resonance can occur. Meanwhile, the effects of the correlation of colored noise, the intensity ofcolored noise, and the coupling numbers on coherence resonance are discussed.
The phenomenon of coherence resonance (CR) in the coupled FitzHugh-Nagumo (FHN) neuronmodels driven by the colored noise is investigated. Based on the method of forth steps stochastic Runge-Kuttamethods, the interval-spike-interval (ISI) of the neuron firing and the coefficient of variation (CV) have beenobtained. It is shown that the CV is a non-monotonic function of the correlation of colored noise, the intensity ofcolored noise, and the coupling numbers. Furthermore, by choosing the appropriate the correlation of colored noise,the intensity of colored noise, and the coupling numbers, the minimum of the CV can be attained, the phenomenonof coherence resonance can occur. Meanwhile, the effects of the correlation of colored noise, the intensity ofcolored noise, and the coupling numbers on coherence resonance are discussed.
2010, 39(5): 778-782,799.
doi: 10.3969/j.issn.1001-0548.2010.05.027
Abstract:
A vulnerability quantitative evaluation model based on grey evaluation method and analytichierarchy process (AHP) is proposed. Evaluating indexes are created according to influencing factors ofvulnerability severity. AHP is utilized to establish a hierarchical model and get the weight of vulnerability factors.Grey evaluation method with a central point triangular whiten weight function is used to compute and assess eachlevel index of vulnerabilities. Then the quantitative result of certain vulnerability is achieved. The experimentalresults show that this model can evaluate the vulnerabilities severity of computer systems effectively andaccurately.
A vulnerability quantitative evaluation model based on grey evaluation method and analytichierarchy process (AHP) is proposed. Evaluating indexes are created according to influencing factors ofvulnerability severity. AHP is utilized to establish a hierarchical model and get the weight of vulnerability factors.Grey evaluation method with a central point triangular whiten weight function is used to compute and assess eachlevel index of vulnerabilities. Then the quantitative result of certain vulnerability is achieved. The experimentalresults show that this model can evaluate the vulnerabilities severity of computer systems effectively andaccurately.
2010, 39(5): 783-787.
doi: 10.3969/j.issn.1001-0548.2010.05.028
Abstract:
This paper proposes a rough set model of logical difference operation of grade and precision, and defines the concepts of rough set regions. In the new model, basic structure of rough set regions is obtained by transformation formulas between variable precision approximations and graded approximations. Regular algorithm and structural algorithm are proposed and analyzed to calculate rough set regions, and application prospect of the new model in decision table is explored and obtained. The new model has extended graded rough set model and classical rough set model.
This paper proposes a rough set model of logical difference operation of grade and precision, and defines the concepts of rough set regions. In the new model, basic structure of rough set regions is obtained by transformation formulas between variable precision approximations and graded approximations. Regular algorithm and structural algorithm are proposed and analyzed to calculate rough set regions, and application prospect of the new model in decision table is explored and obtained. The new model has extended graded rough set model and classical rough set model.
2010, 39(5): 788-792.
doi: 10.3969/j.issn.1001-0548.2010.05.029
Abstract:
A built-in self-test (BIST) scheme for analog to digital converters is proposed based on a linearramp signal and efficient output analysis. The proposed analysis method is an alternative to histogram-basedanalysis techniques to provide test time improvements, especially when the resources are scarce. In addition to themeasurement of DNL and INL, non-monotonic behavior can also be detected with the proposed technique. Twoimplementation options are presented based on how much on-chip resources are available. Experimental resultsdemonstrate the effectiveness of the proposed approach.
A built-in self-test (BIST) scheme for analog to digital converters is proposed based on a linearramp signal and efficient output analysis. The proposed analysis method is an alternative to histogram-basedanalysis techniques to provide test time improvements, especially when the resources are scarce. In addition to themeasurement of DNL and INL, non-monotonic behavior can also be detected with the proposed technique. Twoimplementation options are presented based on how much on-chip resources are available. Experimental resultsdemonstrate the effectiveness of the proposed approach.
2010, 39(5): 793-799.
doi: 10.3969/j.issn.1001-0548.2010.05.030
Abstract:
Aiming to the high-reliability and high-safety requirements of steam turbine protection system, anemergent trip system (ETS) based on the triple modular redundant (TMR) tolerant technology is presented in detailin this paper. The signal sampler, system bus, main controller, and output module are designed in TMR structure. Amethod of self-monitoring of intra-channel and cross-monitoring of inter-channel with synchronization strategy ispresented to realize fault processing and isolation. Through setting voting process in input data, output data, andH-pattern hardware structure, and adopting preset adaptable voting mechanism, the two failures-operation and threefailures-safety of the ETS system are realized effectively. Taking into account of maintenance, reliability and safetyreguirment of the system are analyzed based on the Markov model. Simulation results show that the system has ahigh reliability and high safety. Field commissioning and longtime electrified trials prove that the system satisfiesthe requirement of ETS safety protection system.
Aiming to the high-reliability and high-safety requirements of steam turbine protection system, anemergent trip system (ETS) based on the triple modular redundant (TMR) tolerant technology is presented in detailin this paper. The signal sampler, system bus, main controller, and output module are designed in TMR structure. Amethod of self-monitoring of intra-channel and cross-monitoring of inter-channel with synchronization strategy ispresented to realize fault processing and isolation. Through setting voting process in input data, output data, andH-pattern hardware structure, and adopting preset adaptable voting mechanism, the two failures-operation and threefailures-safety of the ETS system are realized effectively. Taking into account of maintenance, reliability and safetyreguirment of the system are analyzed based on the Markov model. Simulation results show that the system has ahigh reliability and high safety. Field commissioning and longtime electrified trials prove that the system satisfiesthe requirement of ETS safety protection system.
