摘要:Desktop virtualization usually runs mixed workload,and is more sensitive to the interactive performance.Current virtual machine scheduler cannot meet the two demands.This paper presents a full-time preemption CPU scheduler.It uses grey-box technology to inspect information inside virtual machine to support virtual machine scheduling,and considers the characteristics of remote desktop workload for optimization.The evaluation results show that when 5 Windows XP virtual machines running mixed workload concurrently,the display latency of slides presentation is reduced by at least 60% with our optimization.
摘要:Solving an equation system is a very important step in algebraic attack.For a cryptosystem,after being transformed to equations,we often need to employ guess-and-determine algorithm to estimate computational complexity of this attack.In this paper,we introduce a model to estimate average time in solving subsystems more accurately,and propose some criteria on selecting specific guessed variables to speed up the solving efficiency,which based on static weight and dynamic weight etc.For comupting Gröbner bases,we use serveral varible order which are AB,S,S-rev etc.Meanwhile,we introduce the concept of conflicting equations,and show the importance for correct analysis and narrow guessing space.In the end,we estimate the time of attacking Bivium.Experiments showed that,in the worst cases,guessing 60 varibles in the Evy3 position and with DM-rev varible order will have the optimal result,that is about 2 exp(39.16) seconds.
摘要:Group law of rational points E(Zn) on elliptic curves E over Zn and application of this group law are studied in this paper.We present explicit addition formulae to make E(Zn) an abelian group.This solves the problem that some points addition can not be defined using other addition laws in the literature.For the convenient cryptographic application of such elliptic curves,we give the group structure of E(Zn) under our addition law.We propose a remedy for the original elliptic curves Paillier schemes which have security defectiveness.
关键词:elliptic curve over ring;addition law;group structure;Paillier schemes
摘要:This paper proposed to decrease the structure of probabilistic neural network based on Gaussian potential of data field.Core idea is following:Introduce data field to estimate probabilistic density of training set of each class and select their maximum to construct the network;iteratively train the initial network by appending the maximum density sample unrecognized of each class to pattern layer and modify the weight of samples until satisfying desired accuracy.Incremental computing density ensures faster iteration and higher possible convergence.And introduce resampling technique to boost the generalization accuracy.Experiments show that the proposed algorithms have concise explanation,moderate fitness and effective calculation.
摘要:In order to deal with the classification problem for high-dimensional and small-sized data,a kind of support vector machine based on random subspace and orthogonal locality preserving projection was proposed.The random subspace method was used to select a feature subset from the original feature space randomly for several times.Based on the selected feature subset,several base support vector machine (SVM) classifiers were generated.The orthogonal locality preserving projection method was adopted to carry out feature extraction on the samples of each base classifiers,which can,effectively,realize dimensionality reduction.We applied the processed samples to train each base classifiers.The results of the base SVM classifiers were integrated to obtain the final classification result,using a bayesian sum rule.Results on two publicly available face databases show the feasibility and validity of our proposed method.
摘要:Posture modeling is critical for action description and recognition,a posture modeling and action recognition method is proposed in this paper.Spatial Temporal Interesting Points (STIPs) are extracted from learning samples,in fact,one posture consists of a set of STIPs;a unsupervised clustering method is adopted to classify salient postures from these posture samples,then a GMM model is established for each clustering result;transitional probability among salient postures are calculated,and a Visible state Markov Model(VMM) is learnt to describe various actions.Bi-gram method is put forward for action recognition,Extensive experiments are conducted and the results prove its robustness and validity.
摘要:This paper uses complex network theory to analyze the shape characteristics.Based on inner distance of shape,a shape is modeled into a small-world complex network,through analyze degree and cluster coefficient characteristics of nodes in the dynamic evolution of complex network,multi-scale histograms on nodes distribution is proposed for shape descriptor;describe contour points of shape with multi-scale degree、cluster coefficient characteristics and shape context features based on inner distance,this paper combines the improved optimal subsequence bijection to achieve shape matching.Experiment analyzes the capacity of complex network characteristics for shape descriptor,experimental results show that the proposed algorithm can achieve shape matching with non-rigid transformations and the high precision in shape retrieval.
摘要:To solve the slow convergence rate of the existing active set methods applied into optimization formulation of support vector machine,an active set algorithm based on the secondary descent method and the speculative assignment method is proposed.At each iteration of the algorithm,a projection operator is used to restrict the iterative vector onto the inequality constraints of optimization formulation,and then an adjustable step size is used to ensure the functional value of optimization formulation make further descent compared to the traditional active set method.As functional value ensure rapid and strictly descent at the end of each iteration,the global optimum solution can be obtained with rapid convergence rate.Experimental results show that iterations time and training time of the proposed method have been decreased obviously.
摘要:The change of base program structure in aspect-oriented software will lead to the accidental join point miss.The reason is that the definition of the join point is tightly dependent on the base program's structure.Such a close coupling hinders the evolution of the aspect-oriented software heavily.This paper proposes an approach to aspect-oriented software evolution based on metadata and reflection.A conceptual level is added between the join point definition and the base program's structure in order to achieve decoupling.The conceptual model is used to describe the logical information,physical information,as well as the relationship between the two kinds of information.The structure of base program is annotated by logical information based on Java annotation metadata.Simultaneously,join points are also defined by the logical information.The base program is transformed based on the defined metadata and meta objects are generated automatically.On the basis of reflection,the join points can be matched correctly after the evolution of base program.Transformation algorithms for method and constructor are described in detail.This approach solves the accidental join point miss problem effectively,and can be helpful to construct adaptable aspect-oriented software.
关键词:aspect-oriented software;accidental join point miss;software evolution;metadata;reflection
摘要:In view of the inaccurate segmentation in the traditional two-dimensional (2-D) Otsu's thresholding method mainly owing to the supposition that the sum of probabilities of main-diagonal districts in 2-D histogram is approximately one,afast and precise 2-D Otsu's image thresholding method is presented in this paper.A 2-D histogram was created with the select neighborhood and Otsu's method was used on the 2-D histogram in order to obtain better segmentation performance.The probabilities and the mean gray levels in the objects and the background of 2-D histogram main-diagonal districts were calculated preciselyto get a more accurate threshold.The 2-D histogram projection was analyzed to get its features,three theorems were proved,and a novel recursive approach was inferred with the features and the theorems to reduce the computational complexity.Experimental results show that the proposed method achieves more accurate segmentation results and more robust anti-noise than the current 2-D Otsu's thresholding methods,and that its running time is much less,almost the same as that of the current Otsu's recursive algorithm based on 2-D histogram oblique segmentation.
摘要:A new method is proposed in this paper for feature extraction based on geometric flow of images and the second generation of Bandelet transformation,where Bandelet coefficients and their statistical values were extracted as the feature of human images.Afterwards the full body and body parts classifier were trained on AdaBoost algorithm.At last,likelihoods of each body parts were computed combined with Bayesian decision-based approach to perform human detection.The results of human detection experiments indicate our proposed feature extraction method's better capability in describing human characteristics while effectively improving the performance of classifier.Combined with body parts detection,our proposed human detection method well enhanced the robustness of human detection task in both static and moving images.
关键词:human detection;feature extraction;Bandelet;Adaboost;body parts
LIU Wei, YU Hong-fei, YANG Heng, DUAN Bo-bo, YUAN Huai, ZHAO Hong
Vol. 39, Issue 8, Pages: 1793-1799(2011)
摘要:Obstacle detection is one of the key problems in driver assistance and robot navigation,etc.A monocular generalized obstacle detection method based on ground projection displacement vectors of feature points is proposed.The motion parameters of camera are estimated by feature points under the ground plane assumption and the rotation of camera is compensated by these parameters.The relationship between displacement vectors of camera and ground projection displacement vectors of both ground points and obstacle points are deduced by IPM (Inverse Perspective Mapping) respectively.An interval statistical method is proposed and the displacement vectors of camera are robustly estimated by this method.The generalized obstacle is detected by analyzing the relationship between ground projection displacement vectors of feature point pairs in image sequence and displacement vectors of camera.The experimental results under various scenes illustrate that the method can detect obstacle in arbitrary type and shape.Comparing with traditional obstacle detection method based on motion compensation,this method is more robust and accurate.
SUI Xiao-hong, GAO Yong, ZHAO Jie-xi, YANG Yuan, REN Qiu-shi
Vol. 39, Issue 8, Pages: 1800-1804(2011)
摘要:A novel pixel circuit of high dynamic range CMOS image sensor (CIS) was presented for the applications of visual prosthesis,which used the conditional overflow capacitors and multiple integration technologies to achieve high dynamic range,high sensitivity and high signal to noise ratio (SNR) requirements.The designed circuit was simulated using Chartered Semiconductor 0.35 μm CMOS process model parameters,the simulating results showed that the dynamic range was increased by about 62dB than the traditional 4 Transistor Active Pixel Sensor (4T APS) circuit,and the SNR and sensitivity were as high as the 4T APS in weak signal conditions.
摘要:Different perceptual organization (PO) algorithms focus on different global cues,so people have to firstly judge the object's global cue from the image,and then apply the corresponding PO algorithm.This paper proposed a PO algorithm with global cue selection based on prior model to automatically select the global cue with biggest probability.The prior model firstly obtained the prior knowledge and the sample distribution respectively from the statistical quantity of specific field and the generalized Laplacian distribution,then expressed the importance of each kind of global cues as the normalized information which was calculated by the posterior probability and the cue probability.To accelerate the computation,the grouping seed ranking method was incorporated.Finally,application of the algorithm in coal-mine field is analyzed.The results show this algorithm is efficient.
摘要:Kd-tree,due to its adaptive division,low memory consumption and fast traversal,has become the most widely used accelerating structure of the virtual scenes.A fast and optimal kd-tree construction method for virtual scenes is proposed in this paper.By analyzing the SAH function of the scene,simulated annealing is used in the optimal split plane search procedure so as to speed up this procedure.In this way,the kd-tree construction is finally accelerated.Experimental results show that the construction of kd-tree is efficiently accelerated with no loss of kd-tree quality.Furthermore,a multi-core parallel computing extension is implemented in this paper,and with the use of multi-core CPU's parallel processing capabilities,the kd-tree construction is further accelerated.
摘要:The open environment in cloud computing is much more complex and unpredictable,so how to identify untrustworthy cloud end-user by analyzing abnormal user behaviors is an important topic in cloud security.This paper proposes a model for behavior analysis based on incomplete information multi-stage dynamic games,in which current action and historical action,false negative and false positive in network detection methods are considered to improve the accuracy and comprehensiveness of the dynamic judgment of end-use trustworthiness.The experimental results show that it can discriminate potential untrustworthy cloud end-user,and decrease intrusion effectively while perfect Bayesian equilibrium is reached,laying the foundation for active safety mechanism.
关键词:cloud computing;user behavior analysis;dynamic game theory;belief updating
SUN Da-wei, CHANG Gui-ran, LI Feng-yun, WANG Chuan, WANG Xing-wei
Vol. 39, Issue 8, Pages: 1824-1831(2011)
摘要:Focusing on the problem of high efficiency and effectiveness resource scheduling in cloud computing environments,the model of cloud resource scheduling is systematically analyzed in theory at first,the application preferences and the user utility of multi-dimensional QoS is quantified,and the objective function of multi-dimensional QoS is given at last.Combining with the immune clonal algorithm of rapid multi-objective optimization,a heuristic cloud resource scheduling algorithm with application preference is proposed.The non-dominated antibodies are proportionally immune cloned according to their preference priority,which are defined by their cloud application preferences.It is beneficial to enhance the forecasting accuracy of the immune gene manipulation,and to increase the speed of finding the optimal solution based on the application preference.Experimental results conclusively demonstrate the efficiency and effectiveness of the improve system availability,load balancing deviation and valid time brought by the proposed algorithm in cloud computing environments.
摘要:Ant colony algorithm is one of the solutions to task mapping,which optimizes the communication energy and the distribution of link load.Because the ant colony algorithm is very sensitive to the initialization of the parameters,we use genetic algorithm to set the parameter.To avoid getting the local optimal solution,we use chaos module to optimizing the parameter.The algorithm we improved gives a solution,which is 11% lower than the traditional algorithm on power,1% better than the traditional algorithm on load balance and 4% better when optimizing both of them.
摘要:Space information networks,which have become a popular research focus,are a new type of Ad Hoc networks constituted by information systems of land,sea,air and space.This paper proposed a topology reconfiguration scheme combing protection and recovery strategies according to the features and application demands of space information networks.Detection of key nodes in topology and topology failures both trigger reconfiguration process.In reconfiguration process,selected redundant nodes move under the guidance of virtual forces to reconfigure topology.Redundant nodes jointly optimize topology next with targets of minimizing communication cost of designated repairing areas.Simulation results show the proposed topology reconfiguration scheme can improve network performances as well as network invulnerability.
关键词:space information networks;topology control;node position control;virtual force
摘要:Ontology evolution is a research hotspot in ontology engineering.Nowadays,most of the researches are related with evolution strategies and the technology used in evolution,but there are no clear conclusions for the problem how to scale the ability of ontologies for evolution.This paper solves the problem from a new point of view,that it uses ontology cohesion to determine the degree of ontology evolution.We propose a new Change Transmissibility Model that contains the information of ontology evolution;we define an encoding method called ETMC,which is used to represent the model;we analyze the factors that affect the ontology cohesion and propose an approach to compute the ontology cohesion based Change Transmissibility Mode and ETMC encoding method;we carry out our experiments on two ontologies which describe the same domain knowledge,and the experimental results show that our approach is feasible on the problem of determining the ability of ontology evolution.
摘要:To address the issue of task co-schedule with deadline constraint in grids,a metric called probability of deadline guarantee is proposed to quantitatively evaluate the deadline-guarantee of a co-scheduling scheme.Based on this metric,an extensible co-scheduling framework is implemented,which combines multiple policies to obtain the optimal deadline guarantee.Extensive experiments are conducted to examine the performance of the proposed model.The results show that the probability of deadline guarantee metric can accurately quantify the degree of deadline guarantee provided by a specific co-scheduling scheme,which in turn provides an enhanced QoS for those applications with deadline constraint.
关键词:grid computing;deadline constraint;Quality of Service;stochastic service theory
WANG Chuan-chen, ZHANG Run-tong, LI Dan-dan, CAO Feng
Vol. 39, Issue 8, Pages: 1858-1863(2011)
摘要:The delay and loss of data packets,which caused by the links congestion,will lower down the users' QoS.While the present link state routing algorithms don't possess congestion avoidance mechanism.This paper proposes an ant-based multiple-path routing algorithm with congestion avoidance,in order to solve the above problems.It improves the ant colony algorithm,and adopts double-way ant-exploration method to speed up the exploration of optimal routes.The new searching routings guidance better satisfies the demands of QoS.Simulation contrast with OPNET shows that the algorithm is valid and effective at controlling packet loss ratio,delay and residual bandwidth.
摘要:The concept of strong regular implication operator and the n-valued S-MTL propositional logic system are introduced.Based on probability measure the truth degree of formula is defined and its integral expression is given and the inference rules w.r.t the truth degrees is proved.Moreover,similarity degrees among formulas are proposed and a pseudo-metric is defined therefrom on the set of formulas.The continuity of logical operators w.r.t the pseudo-distance is proved,and hence a possible framework suitable for developing approximate reasoning theory in n-valued S-MTL propositional logic is established.
ZHANG Jian, QI Hao-chen, XU Dong-liang, HU Zhi-wen
Vol. 39, Issue 8, Pages: 1869-1872(2011)
摘要:We used unit vectors of the parametric surface cells as an important parameter to achieve the surface evolution algorithm.Instead of the normal vectors of facet cells,Gauss integration is adopted to calculate the normal vectors of the surface cells,so as to determine the evolvement directions of the cells.The simulation results of this algorithm were 3-D visible of variable perspective.If a physical model of given fabrication is provided,it can be applied to other processes with surface evolution direction dependence.Some isotropic etching simulations were performed as examples,and the results indicate the accurate simulated surface.
摘要:A novel broadband circularly polarized stacked patch antenna is proposed for Chinese UHF radio frequency identification (RFID) reader/writer applications.The antenna is composed of two corner truncated patches and an S-shaped horizontally meandered strip (HMS),all of which are printed on the upper side of the FR4 substrates.One end of the S-shaped HMS is connected to the main radiating patch by a probe,while the other end is connected to an SMA connector.Using the proposed S-shaped HMS feed technique,a good impedance matching and symmetrical broadside radiation patterns can be achieved.Simulation results are compared with the measurements,and a good agreement is obtained.The measurements show that the antenna has an impedance bandwidth (VSWR<1.5) of about 26.6% (722-944MHz),a 3dB axial ratio (AR) bandwidth of about 10.8% (838-934MHz),and a gain level of about 8.5dBi or larger within the 3dB AR bandwidth.Therefore,the proposed antenna can be a good candidate for UHF RFID readers/writers operating between 840 and 928MHz.
摘要:The rigorous couple wave analysis(RCWA) is one of the most common methods for the study of grating diffraction.For the incident beam normal to the surface of the grating,the positive and negative N diffraction orders are same due to the symmetry,which lead to the great simplified in by decreasing matrix size from 2N+1 to N+1.In this paper,the simplified method is extended to the case that the incident plane parallel to the grating line.In this case,the problem of matrix size 2(2N+1) (due to the coupling of different polarization states) can be reduced to the problem of matrix size 2(N+1).Furthermore,some more simple approach for the small incident angle is discussed,in which the problem of matrix size 2(N+1) can be reduced to (N+1).The typical simulation shows that the error is acceptable in quite a large range,which is important in practical use.
GAO Ke-cun, YANG Jin-hui, DAI Yu-xing, HONG Tian-yu
Vol. 39, Issue 8, Pages: 1883-1888(2011)
摘要:According to the characteristic of three-phase inverter based on αβ rotating frame,this paper presented a control scheme of inverter output waveform control based on neural network internal model theory.The control scheme took a neural network estimate model as the internal model of three-phase inverter to estimate the actual output waveform.The simulation and experiment verify the scheme can work well.Highly accurate steady output,fast response and less distorted waveform is gained.
关键词:three-phase inverter;inverter waveform control;internal model theory;neural network
摘要:The channel mismatches degrade the reconstruction performance of hybrid filter banks (HFB) analog-to-digital converters (ADC).This paper presents a comprehensive model that takes gain,phase,and bandwidth mismatches into account.Based on this model,a method that calculates the value of the channel mismatches by solving overdetermined equations is developed.An improved architecture is introduced based on the model and the perfect reconstruction expression is derived.Finally,a two-channel 160MHz 12-bit HFB ADC system is implemented.The experimental results show that the proposed calibration technique produces about 23 dB of spurious-free dynamic range (SFDR) enhancement.
摘要:Reports a method of design 2nth-order Voltage-controlled Voltage Source(VCVS) low-pass filter,based on analyzing the error isogram by MathCAD,the paper put forward the optimal design plan of 2nth-order VCVS unit-gain Butterworth f0 Hz low-pass filter:firstly,decompose 2nth-order filter into n 2nd-order filters.For the ith one,if monomial coefficient of the polynomial expression is αi ,select two capacitances C1i and C2i,making C2i/C1i 0.25 αi2,and then calculate value of the two resistances by the same expression 1/παif0C1i.Obviously,the design plan has the advantages of convenience,low-cost.And with the aid of Monte Carlo analysis,parameter sweep analysis and real circuit experiment the design plan has also been verified its advantage of stability.In addition,analysises also shows different C,R have different influence on the filter features.
关键词:optimal design;low-pass filter;2nth-order;stability;MathCAD;Monte Carlo
摘要:Random noise perturbs objective functions in many practical problems,and genetic algorithms (GAs) have been widely proposed as an effective optimization tool for dealing with noisy objective functions.However,there are few theoretical studies for the convergence and the convergence speed of genetic algorithms in noisy environments (GA-NE).In this study,Objective functions are assumed to be perturbed by additive random noise.We construct a Markov chain that models elitist-worst genetic algorithms in noisy environments (EWGA-NE).Then the convergence of EWGA-NE is deduced based on the absorbing state Markov chain.Next,the convergence rate of EWGA-NE was studied.The upper and lower bounds for the number of iterations that EWGA-NE selects a globally optimal solution were derived.
关键词:genetic algorithm;noisy environment;absorbing state Markov chain;convergence;convergence rate
摘要:One simple and efficient method for the removal of high-density salt and pepper noise is proposed in this paper.In the proposed method,we calculate the mean value of the non salt and pepper pixels in the window centered at the current salt and pepper pixel.Then,we update the salt and pepper pixel with the mean value at the different phases.Although the principle in the proposed method is quite simple,the experimental results show that the proposed method behaves excellently for all noise ratios,from 10% to 90%.The proposed method can preserve the details of the original image well.Furthermore,the run time for the proposed method is quite short,which makes it possible for the real applications.
关键词:impulse noise;median filter;details-preserving;image process
MAO Jia-fa, NIU Xin-xin, YANG Yi-xian, SHI Shu-jian
Vol. 39, Issue 8, Pages: 1907-1912(2011)
摘要:In the domain of steganalysis,the theory of DCT coefficients statistical distribution model has been addressed in many previous papers.Based on the statistical characters of image's data,this paper introduces the symmetric alpha-stable model of the cover image's DCT coefficient systematically.Because of the adequate flexibility of SαS,its shape of distribution will be changed as the characteristic index α changes.According to SαS model flexibility,a new novel steganalysis method based on cover image quantitative describing is proposed in this paper.Compared with the traditional model's steganalysis method,the performance of the new model's is much better.
摘要:Evolutionary programming adopting single mutation operator can not give attention to exploration and exploitation both.So two kinds of mixed mutation strategy including Cauchy+chaotic mutation operator and Cauchy+Gaussian mutation operator are given.A novel adaptive cultural algorithm adopting above mixed mutation strategy is proposed by adopting dual structure in cultural algorithm.Implicit knowledge extracted from the evolution process is used to control occasion and proportion of each mutation operator in mixed mutation strategy.Simulation results indicate that the algorithm can effectively improve the speed of convergence and have better convergence.
摘要:A broadband (the first band is 877~960 MHz,and the second band is 1680~3189 MHz) planar phone antenna with weak ground surface current and stable performances is proposed,which has two resonant paths by eroding a slot on a monopole antenna.The radiation patterns at 2600 MHz of the antenna are still similar to the ones of monopole,and the size of ground plane is compact.Moreover,the antenna has a simple construction with a compact-size of 39mm in width and 13mm in length,and it can be easily fabricated.Simulated and measured results of the design are in a good agreement,so the antenna can be equipped in the system of mini mobile phone.
ZHANG Xi-hui, LIU Lian-fu, CAI Jing-ye, YANG Yuan-wang, CHEN Xing-hua
Vol. 39, Issue 8, Pages: 1923-1926(2011)
摘要:A PHPT-based algorithm is proposed for parameter estimation of high-order polynomial phase signal (PPS).The basic idea is to translate the phase parameters estimation into the estimation of the corresponding Taylor coefficients of the signal,and Taylor coefficients are estimated by using PHPT.As a linear algorithm,the proposed method has a good performance for the estimation of the lowest-order phase coefficient,and the computational complex is about three FFT operations,which is lower than the exist ones when the phase order is more than three.Computational simulations are presented to illustrate the effectiveness of the proposed algorithm.
JIANG Yun-hao, MA Wei-ming, ZHAO Zhi-hua, TANG Jian, LI Wen-lu, XIAO Huan
Vol. 39, Issue 8, Pages: 1927-1931(2011)
摘要:The time domain performance of analog adaptive interference cancellation system with delay mismatching is analyzed.Through theory calculation,the stability criterion of the system with delay is obtained.In the stable region,the convergence speed is redeced with the increase of the delay.Through solving and analyzing of the simplification model,the relation of stable state weights and delay is gained.The variation rule of residual error and interference cancellation ratio with delay is calculated.The simulations and experiments are agreement with the theory analyses.
摘要:Information about user preferences plays a key role in automated decision making.As a intuitive graphical tool for representing statements over multiple attribute qualitative decision,CP-nets has attracted many researchers to study it.Its higher complexity of dominance query algorithm is a difficult problem to solve.We study how to reduce dominance query complexity.A general-purpose framework for solving constraint satisfaction problem is introduced,namely,SCSP (c-Semiring Constraint Satisfaction Problem),and point out that conditional preference table of CP-nets just is a type of dynamic constraint.An algorithm transforming conditional preference table of CP-nets to constraint of SCSP is given,which can judge a solution is better or worse.Its polynomial time complexity property is given,so based on c-Semiring,dominance query for acyclic CP-nets is solved effectively.
摘要:In ADELIN (ADaptive rELIable traNsport protocol),a redundancy forwarding node is used to improve the reliability of data transmission.The downstream node has already received most of the relayed data packets of the redundant node.IPool-ADELIN protocol based on IPool (Improved Pool) node is proposed.By overhearing the data packet transmissions,IPool node not only does local link maintenance when the BER (Bit Error Rate) of a link is high,but also relays the data packets,which are not implicitly acknowledged by data packtes.The results of mathematic analysis and simulations show that,compared with ADELIN,IPool-ADELIN has higher data delivery ratio and lower energy consumptions.
摘要:To better protect personal privacy against background knowledge attack and homogeneity attack,single sensitive value and multi sensitive values (α,k)-anonymity models were defined respectively.For achieving this purpose,two clustering algorithms were designed.At the same times,we made correctness and complexity analysis for the algorithms.Since the data sets contain continuous attributes and classification attributes,a detailed mapping and processing method was given,that make the distance between data points can calculate easily,and avoid completely the case that confusion data points distance and information loss.Experiment results and detailed theory analysis demonstrate that our methods are effective on both information loss and execution time comparing with existing methods.
摘要:Geometry configuration of a 30.5GHz TE01-TE11 mode converter for Gyroklystron is obtained to realize optimal mode conversion by a code written based on the mode coupling theory,and a 30.5GHz TE11-HE11 mode converter is designed by High Frequency Structure Simulator (HFSS).Results show that complex TE11-HE11 mode converter consisting of the two converters has a transforming efficiency of over 97% within bandwidth of 1% at frequency 30.5GHz.Test indicates that the complex mode converter acts very well in conversion of TE01 to HE11 mode.
摘要:Using different test stimuli have an affect on the testability of the circuit under test (CUT).According to the Fourier series expression theory,which is described as the periodic function can be decomposed into a direct current component and a series of sine functions,a new optimization method for test stimulus generation of analog circuit is proposed.In this method,the random stimulus function is used as the optimal object,the optimal target is to obtain the maximal distance of different classes of faulty samples in the kernel space,and the constraint conditions are designed on the base of the relationship of amplitude,frequency and phase between the input and the output of the CUT,so a linear optimal model is constructed with one target and multiple constraints.This method has general adaptability.The optimal test stimuli succeed in improving the diagnosis results.
关键词:analog circuit;test generation;fault diagnosis;Fourier series
摘要:Cancer gene expression data is a typical data with high dimension and small sample,identifying it directly will encounter the curse of dimensionality,so needs dimensions reduction.This paper proposes a kind of classification approach based on Spectral Regression (SR)analysis and Kernel space K-Nearest Neighbor(KKNN) classifier for gene expression data,it gets the projection matrix through Spectral Regression Analysis witch can extract effectively discriminative characteristics of low dimensions,and reduces the dimensionality of gene expression data by projection matrix,then identifies the low-dimensional data reduced with the Kernel Space K-Nearest Neighbor Classifier.As the experiments operated on the cancer datasets Prostate-Tumor and 4-Tumors demonstrate the effectiveness of the proposed algorithm;simultaneously,compared with the K-Nearest Neighbor(KNN) classification approach,The Kernel space K-Nearest Neighbor has a better classification result.
关键词:gene expression data classification;kernel space k-nearest neighbor;spectral repression analysis;dimensions reduction