摘要:The performance of Capon beamformer degrades sharply in the presence of array steering vector errors.To solve this problem,a robust beamforming algorithm,which combines generalized sidelobe canceller(GSC) and robust least-square,is proposed.The Capon beamformer is converted to a least-square problem based on the principle of generalized sidelobe canceller,and then converted to a second-order cone program (SOCP) problem under the constraint on the norm of the error matrix.The well-established interior point method is adopted in order to obtain the optimal solution.Deduction proves that the proposed algorithm belongs to the diagonal loading sort.Simulation results show the efficiency of the proposed algorithm in the case of steering vector mismatch and snapshot deficiency.
摘要:The notion of Choquet integral truth degrees of propositions in Lukasiewicz propositional logic is introduced,by means of the Choquet integral of McNaughton functions with respect to uncertainty measures on the set of all valuations.When the involved uncertainty measures satisfy finite additivity property,the notion of Choquet integral truth degrees can induce in a natural way a pseudo-metric,with which the set of all propositions becomes a pseudo metric space and thus several graded reasoning methods can be established.The notion of Choquet integral truth degrees will reduce to the existing notion of Borel probability truth degrees in probabilistically quantitative logic when the uncertainty measures are Borel probability measures.This paper is a continuation of probabilistically quantitative logic and provides a possible framework for reasoning about non-linear uncertainty of propositions.
关键词:Lukasiewicz propositional logic;probabilistically quantitative logic;Choquet integral truth degree
摘要:The feature selection for software birthmark has a direct bearing on software recognition rate.We apply constrained clustering to analyze software features.The within-and between-class distances of features are measured based on mutual information.Information gain and penalty functions are constructed using homogeneous and heterogeneous software features respectively.Then the software birthmark features with high class distinction and minimum redundancy are selected.It is shown the algorithm provide an effective approach for software birthmark feature selection and optimization by analysis and comparison.
摘要:A method for micro-motion parameter estimation of radar target based on high-order moment function is presented,which is suitable for the narrowband radar.Firstly,the radar echo is mixed with the reference signal,and then its direct current is eliminated.The high-order moment function of the signal with different time delays is calculated firstly,then the Fourier transform results of the high-order moment's imaginary parts are calculated and the peak value of their product results is detected to obtain the rotation frequency of target's micro-motional parts.The radius and the initial phase are estimated by successive cancellation.A computer simulation is given to illustrate the effectiveness of the proposed method.
关键词:narrowband radar;micro-motion;parameter estimation;high-order moment
摘要:Considering the high complexity of HTE(Hierarchical Topology Estimation)and its performance degradation under the condition of large correlation estimation variance,a method based on agglomerative hierarchical lustering is proposed.The method employs bottom-up agglomerative hierarchical clustering,which only uses the data related to the node pair with the largest correlation,so it has lower computation complexity than HTE.A modified finite mixture model is established,increasing the amount of effective data,which improves the accuracy of parameter estimation.The simulation demonstrates that the proposed method infers the topology more rapidly,with higher accuracy when the correlation estimation variance is large.
摘要:By minimizing the energy functional,we can obtain the variational image decomposition which decomposes image into different characteristic components,and can be used for image restoration and reconstruction.An idea of multiscale image restoration and reconstruction in the framework of variation is proposed.Based on this idea,firstly,a single-parameter (BV,G,E) trituple decomposition model is proposed,and the relationship between the parameter and the scale of each component is studied theoretically.And then,by replacing the parameter with a binary sequence,we achieve a multiscale (BV,G,E) decomposition which can decompose an image into a sequence of image structure,texture and noise.The convergence of this multiscale decomposition is proved,and an efficient numerical method based on the duality theory and alternate iteration algorithm is introduced to solve it.At last,the proposed multiscale (BV,G,E) decomposition is applied for image restoration and reconstruction.Numerical results support the theoretical results and show that the proposed model is efficient for multiscale image restoration and reconstruction,and is superior to some other decomposition models.
摘要:Fast analysis of electromagnetic scattering problems of three-dimensional(3D) objects over a wide incident angle is always a difficult problem in computational electromagnetics.Previous study shows that introducing compressed sensing(CS)technique into method of moments(MoM)can reconstruct currents over all incident angles from only several measurements,thereby it can improve the computation efficiently.For the purpose of acquiring total number of measurements and choice of sparse transform when using compressed sensing,a priori technique based on physical optics(PO)is proposed in this paper.By this technique,sparsity of the projection of current coefficients over the wide angle and the number of times of measurement can be pre-estimated,thereby apriori knowledges can be provided for the fast algorithm of solving scattering problems over a wide angle by CS(especially the choice of sparse bases),thus the fast algorithm can be applied into engineering practice in a real sense.
关键词:physical optics(PO);prior knowledge;compressed sensing(CS);method of moments(MoM)
摘要:When the compound-Gaussian clutters are cluttered into groups of data with one and the same value of the texture,the homogeneous clutter-clustered model is generalized,and a generalized clutter clustered model is established.Moreover,a generalized likelihood ratio test detector based on the generalized clutter-clustered (GCC-GLRT) is proposed,and the formula relating false alarm probability to detection threshold is deduced.The GCC-GLRT solves the problem that the CC-GLRT is not CFAR in inhomogeneous clutter-clustered background.The simulation results show that the GCC-GLRT can overcome the collapsing loss of the NSDD-GLRT detecting spare scatterers target and improve the detection performance of the detector.
关键词:compound-Gaussian clutter;spherically invariant random vector;clutter-clustered;generalized likelihood ratio test
摘要:Multifunction radar(MFR)is a large scale dynamical system which exhibits complex signal forms.The problems in deciphering the signals and inferring the MFR's policy of operation really challenge the electronic intelligence(ELINT)society.By drawing lessons from the ideas of gene engineering,the research framework and study methods for MFR signal analysis are proposed.From the viewpoint of system,an analogy is made between the MFR and the organism's cell,which suggests their high level parallelism in system architecture and policy of operation.The key techniques in gene engineering,especially reverse engineering,which can be applied for MFR signal analysis,are summarized,and pre-well-prepared work demonstrates that the proposed method is reasonable.
摘要:The principles and characteristics of fully polarized microwave scatterometer was analyzed and a simulation model for fully polarized microwave scatterometer system was established.Wind retrieval quality of fully polarized and co-polarized system were compared under parameters of the SeaWinds scatterometer.The results show that fully polarized scatterometer has better performance than conventional co-polarized system in the nadir region and outlying region in swath.And it can improve wind inversion accuracy under high wind-speed conditions.Finally,the effects of polarization isolation on the measurement performance of fully polarized scatterometer system were analyzed.
摘要:The test application time becomes too long due to the heterogeneous-core interconnection in Network-on-Chips.The concurrent test technology is the optimal scheme to solve the problem.A novel test project for NoC is proposed in this article,and efficient concurrent test can be achieved in core-based NoC.Firstly,the folding partition method is used to change the network partition,and then it can improve the parallelism of test transmission.Secondly,the test sets are merged through the sequential shift matching algorithm to make them simple and isomorphic.Finally,the test sets are injected into the NoC by multicasting.The experimental results show that the test application time is reduced by 17.6%-40.47% compared with those in the reference[13] and [14],and our method is implemented easily.
关键词:network-on-chips;concurrent test;isomorphic;folding partition;test application time
摘要:There exist integration loss and computational cost drastic increasing in long-term integration of weak DSSS (Direct Sequence Spread Spectrum)signal in large dynamic Doppler frequency scenarios.A motion compensation-double block zero padding algorithm is proposed to this problem,which uses frequency spectrum cyclic shift for Doppler frequency searching,envelope shifting and Keystone transform for code Doppler frequency compensation based on the double block zero padding algorithm.Theoretical analysis and simulation results show that,the proposed algorithm can eliminate the influence of large dynamic Doppler frequency effectively,and has high computational efficiency,the computational cost is 16% of the similar algorithm in 20ms coherent integration,and the longer the time,the higher the efficiency.
摘要:A gaze tracking system based on single camera and four IR light sources is proposed in this paper.With pupil's center and four reflection centers on corneal generated from the IR light sources,a novel pupil-corneal reflection vector is constructed,which is invariant to head movements.In addition,the region of interest is applied to improve the speed of eye location;template matching and convex hull are used to locate pupil's center;point matching is used to identify the reflection centers.These methods guarantee the real-time performance and accuracy of the gaze tracking.The experiment results demonstrate the improved accuracy and the real-time performance of the system with less constraint on head movements.
摘要:In order to identify video copies on the Internet,a video fingerprinting algorithm based on binary tree and stochastic neighbor embedding is proposed in this paper.In this scheme,representative frames are selected based on binary tree and normal cut,and the Discrete Cosine Transformation (DCT) coefficients of luminance of representative frames are taken as the high dimensional features of the video.The features are mapped into three-dimensional space using stochastic neighbor embedding.A matching key is generated based on the mean and variance of distance vector between adjacent points in this three-dimensional space,and the video fingerprint is generated by binarizing.Moreover,during video fingerprint matching,the matching key is used in the first-stage matching to reduce the search range,and then a further matching is carried out in the candidate fingerprint to identify video copies.
摘要:The track correlation algorithm should be able to adjust its control parameter to adapt the complicated and changing track data with unknown system error and other unknown features.The concept of the local track correlation uncertainty was proposed in this paper,two types of local track correlation uncertainty was defined and an adaptive track correlation algorithm was proposed based on traditional sequential track correlation algorithm,which included a new anti-system-error method of calculating correlation probability,a method to adaptively adjust the length of sequence according to the first local track correlation uncertainty,and a method to adaptively adjust correlation threshold according to the second local track correlation uncertainty.The simulation results show that with the new algorithm,the system error has little effect on the correlation probability under certain conditions,the sequence length the range of correlation threshold can be adaptively adjusted according to the change of average distance of targets and random error of sensors,and the correct correlation rate can be effectively improved in complicated conditions.
摘要:Due to the complexity and amplitude-phase inconsistencies in traditional multi-channel interferometer and array,two phase difference observation models for direction finding based on the rotated long baseline interferometer(RLBI)is analyzed.The unambiguous direction finding conditions for the proposed models are derived.A multiple hypothesis pseudolinear iteration least square(MHPILS)algorithm is proposed.The Cramer-Rao low bound(CRLB)and mean square error(MSE)is also analyzed.The simulation results show that this method achieves moderate computation amount and can attain the CRLB.
摘要:A fast computational method of the multi-scale autoconvolution(MSA)transform is proposed in this paper.In order to reduce the times of MSA transform,the method deduces the smallest benchmark transform size according to the fast Fourier transform theory,and replaces the different transform sizes of the same scale transform within the minimum range of MSA transform scale.Then,for reducing the computational complexity,this method reduces the MSA transform by using the MSA transform symmetry outside the range of MSA transform scale.Several experiments on the aspects of time efficiency and accuracy of eigenvalue using typical sample data are given.The results demonstrate that computation speed of the fast proposed computational method is three times faster than that of the original method while maintaining eigenvalue accuracy.
摘要:Target micro-motion conveys vital information which is favorable for understanding synthetic aperture radar (SAR) images.However,image defocusing also ensues.Therefore,techniques on micro-motion effect analysis,micro-motion target detection and imaging are gaining more and more attentions in the remote sensing field.These techniques are generally named SAR/micro-motion target indication (SAR/MMTI).In this paper,topics on SAR/MMTI are introduced including its concept,key difficulties,existing experiments,processing approaches,and its application promise.Some directions of future work on SAR micro-motion are predicted at last.
摘要:Based on the researches of rate distortion optimization (RDO) and rate control (RC) techniques for H.264,this paper divides the development of RDO and RC into three phases based on RC intention,namely rate distortion modeling,considering the effect of quality fluctuation on subjective quality,subjective distortion guided RC;analyzes the characteristics of representative RC techniques of each phase,and focuses on the key and difficult issues of RC needed to be further researched.Then summarizes the RC schemes for transmission oriented error resilient video coding,scalable video coding (SVC) and multiple description coding (MDC) as well as stereo and multi-view video coding,and analyzes further research trends.Finally,how to develop the RC schemes for high efficiency video coding (HEVC) is discussed.
摘要:Heterogeneous data permeate in nearly every application and research field.Modeling and correlation analysis about heterogeneous data is the bottleneck of the conventional concept lattice as well as its various type of extended model.This paper presents a generalized concept lattice model by altering the definition of Galois connection function f(A) and g(B) given by Ganter and Wille,and discusses the ordering on heterogeneous attributes.After a survey of event-oriented knowledge representation,an event model formalized by generalized concept lattice is presented as an application,which is capable of dealing with different types of attributes and mining useful knowledge.This example gives a clue of generalized concept lattice model application in other fields.
摘要:The advantages of differential evolution(DE) are its simple structure,easiness of implement,fast convergence and robustness.However,DE often suffers from premature convergence and stagnation problems.A framework of the recurring two-stage DE is proposed to balance global exploration and local exploitation.The proposed framework is based on repeated and alternated execution of two different stages,namely,the local exploitation and global exploration stages.The parent individuals for the mutation operation at each stage are selected from neighbors or strangers of the target vector,respectively,based on the Mahalanobis distance matrix.The simulation results on the CEC2005 real-parameter optimization benchmark functions show that the proposed framework can make DE more efficient.
摘要:A new algorithm to jointly estimate the DOD-DOA and Doppler parameters of bistatic multiple-input multiple-output(MIMO)radar is proposed based on the fractional power spectrum density.For moving target,the echo often contains time-varied Doppler frequency.A new signal array model of bistatic MIMO radar system is constructed in this paper.Doppler parameters are estimated by searching the peak of fractional power spectrum density(FPSD).Two subarrays are constructed based on FPSD.Both direction of departures(DODs)and direction of arrivals(DOAs)of the multiple targets are estimated by employing the FPSD-MUSIC algorithm and FPSD-ESPRIT algorithm.Furthermore,the Cramér-Rao bound(CRB)for target parameters estimation is derived.Simulation results are presented to verify the effectiveness of the proposed method.
摘要:With the multi-scale binary sampling methods on the basis of Gaussian scaling function,the model of multi-scale quantum harmonic oscillator algorithm for high-dimensional function global optimization problems is proposed.High-dimensional function optimization process is divided into two steps,scale convergence and quantum harmonic oscillator convergence.This algorithm model is based on the same convergence process probability interpretations between function optimization problem and quantum harmonic oscillator from high-energy state to ground state.This algorithm,which has explicit physical model,can realize high-dimensional function optimization without coding or complex initial conditions.Experiments and analysis are done for 15 typical two-dimensional optimized test functions and 6 typical high-dimensional optimized test functions.The results show that multi-scale quantum harmonic oscillator algorithm gets precise global optimum for high-dimensional function quickly,and with "reduced frequency"approach,the search speed of function with "high frequency" component improves significantly.
摘要:Otsu's thresholding method is widely applied in image segmentation,but it is not suitable to segment the images that have large difference in intra-class variance of object and background,this paper presents a regularization Otsu's segmentation method based on the information entropy of posterior probability of object and background obtained by thresholding image.From the point of view that image segmentation is essentially pixel clustering problem,the classical Otsu's thresholding method is firstly interpreted as a kind of weighting hard C-means clustering algorithm.Secondly,considering that the clustering segmentation has typical ill-posedness,the object and background from raw image are obtained by segmentation and their posterior probability information entropy is taken as the constraint item,the regularized modification of Otsu's thresholding criteria function is realized,and regularization Otsu's thresholding method is obtained.In the end,the reasonability of the proposed thresholding method is explained by mathematical analysis,and the selection method of its regularization parameter is put forward.Experimental results show that the proposed regularization Otsu's thresholding method is effective,and the traditional Ostu's thresholding method can be viewed as a special case of the proposed method.
摘要:The adaptivity and real-time performance of feature extraction method are crucial in brain-computer interface.Based on Hilbert-Huang transform (HHT) and common spatial subspace decomposition (CSSD) algorithm,a novel feature extraction method,denoted as HCSSD,was proposed.Firstly,the motor imagery electroencephalography (EEG)/ electrocorticography (ECoG) was preprocessed,and a relative distance criterion was defined to select the optimal combination of channels.Secondly,Hilbert instantaneous energy spectrum and marginal energy spectrum of EEG/ECoG were calculated to extract time feature and frequency feature respectively.Then CSSD was applied to extract spatial feature.Furthermore,serial feature fusion strategy was adopted to obtain time-frequency-spatial feature.Finally,learning vector quantization neural network was designed to classify the EEG/ECoG data.The average recognition accuracy was 92% for the left small finger and tongue motor imagery ECoG tasks.Experiment results show that HCSSD can enhance the adaptivity and real-time performance of feature extraction,with the recognition accuracy improved.This method provides a new idea for the application of portable BCI system in rehabilitation field.
摘要:The geodesic active contour model(GAC)can't identify the object of the images with complex background such as noise and intensity inhomogeneities successfully.For this reason,this paper proposes GAC model driven by the local entropy.First of all,the local information entropy of image is abstracted to describe the local intensity variation.Then,the signed pressure force function based on the local entropy are structured,which guides the contour curve close to the boundary of the object and achieves the segmentation of the object.In order to reduce the computational complexity and improve the robustness of the proposed model to different level sets,the proposed method is implemented by the binary level set method.The experimental results show that this method can overcome the influence of complex background to the segmentation results,and realize fast and accurate segmentation.
关键词:geodesic active contour model;local entropy;signed pressure force function;binary level set method
摘要:The diving squinted SAR (Synthetic Aperture Radar,SAR) has the feature of non-uniform resolution in different direction.The resolution description is seriously restricted in the traditional SAR,which is not applied to the analysis of resolution in the diving squinted SAR.In order to illustrate a complete description of the resolution characteristics,a method of resolution analysis of squinted SAR based on distinguishable ellipse is presented.The expression of distinguishable ellipse on the slant and ground plane is derived respectively by GAF (Generalized Ambiguity Function,GAF).The simulation results show that this method is adaptive to the conditions of different geometry and system parameters,which can provide a theoretical basis for the design of system parameters and geometry configuration in squinted SAR.
摘要:This paper discusses the design of phase coded waveforms with ultra-low autocorrelation sidelobes in desired intervals.These waveforms can be used as interference suppression and signal synchronization waveforms for radar,sonar and communication systems.An algorithm based on power spectral density (PSD) approximation is proposed.According to the relationship between aperiodic autocorrelation sequence and the waveform PSD,the autocorrelation design object is translated into a PSD design object.The method transforms and projects the design waveform in the time and the frequency domain iteratively.Thus the PSD of the design waveform approximates the ideal PSD iteratively in order to generate the desired waveform.Compared with the general nonlinear programming methods,this method is derivative-free,thus requires less computer memory,and the iteration can be implemented by FFT efficiently.
关键词:waveform design;phase code;autocorrelation sidelobe;power spectral density approximation
摘要:Focusing on the problem of high counter memory overhead and frequent overflow of counter mode encryption,this paper proposes an efficient scheme of protecting data confidentiality and integrity.Based on the locality character of data accessing,the scheme set different counter lengths for memory area of different accessing frequency and the counter lengths can be dynamic adjusted.Analysis and simulation results indicate the scheme can decrease memory overhead and the number of overflow.
摘要:A radio frequency (RF) link selection method based on Bayesian compressive sensing (BCS) is presented to efficiently avoid selecting redundant links with random measurement for compressed RF tomography.Firstly,link entropy and redundancy are defined,and then the minimum differential entropy link decision model is established,which indicates the relationship between the RF link information and redundancy.Secondly,the presented model is combined with BCS based adaptive projections measurement framework,to realize link selection and target estimation.Localization experiments illustrate efficiency and feasibility of the proposed method,which can efficiently avoid the selection of redundant links and improve the sensing efficiency compared with the random selection method.
关键词:radio tomographic imaging;link selection;Bayesian compressive sensing;redundant link
摘要:The rain attenuation time series are simulated by the EMB model based on rain measured data and parameters at Haikou,Qingdao and Beijing.The power spectra estimated with simulative series are in good agreement with the measured result in references.The probability of fade exceeding a given fade duration value are calculated based on simulative attenuation time series,and compared with ITU-R predictive and measured values at 12.5GHz,in Haikou,Qingdao and Beijing.At Haikon,the statistics agree with predicted and measured results.At Beijing and Qingdao,as fade duration is less than about 100s,the simulative statistics are in agreement with measured results.As fade duration is more than about 100s,there are the difference among them.At 30/20GHz,for the smaller fade threshold and duration,the statistics based on simulation approach to the results predicted by ITU-R.As the larger fade threshold and duration,the statistical results by simulation are different to the prediction statistics by ITU-R.The results show that the simulation method is available for the rain region of abundant rainfall.However,at insufficient rainfall region,the method or ITU-R prediction may be modified as longer fade duration.It is significant groundwork for rain fade mitigation techniques development in higher frequency wireless systems.
关键词:microwave;rain;attenuation time series;fade statistics
摘要:A cooperative immune network algorithm used for virus detection is proposed,in which detectors are optimized through cooperation and incentive between different kinds of immune cells.The non-self set is introduced,and mature detectors are selected and cloned according to the fitness between detectors and non-self set to enhance the incentive for antibodies.The mutation step is updated by the generation number of evolution to change the mutation way of mature detectors adaptively.Furthermore,the network suppression strategy based on concentration partition is proposed according to the incentive between antibodies and antigens in the entire immune network.Experimental results show that the proposed algorithm can increase the diversity of detectors,and improve the virus detection ability of the entire immune network effectively.
摘要:Flag problem is a key problem in search based software test data generation.This study proposed a testability transformation method based on the dominant relationship of target statements.The basic idea is that:for a target statement involving flag variables,if another target statement (or target statement set) dominates the original one,then the original target statement is substituted with the new one to generate test data.Experimental results showed that the proposed method can effectively solve the flag problem,therefore improve the efficiency of generating test data.
关键词:test data generation;flag variable;testability transformation;genetic algorithm