摘要:Narrowband interference (NBI) and wideband interference (WBI) are common jamming signals against synthetic aperture radar (SAR),which can degrade the imaging quality severely.According to the time-frequency characteristic of NBI and WBI,an interference suppression algorithm based on Wigner distribution (WD) and sliding window masking technique is proposed.First,a fast signal synthesis algorithm from Wigner distribution (WD) is developed using the inherent relationship between original signal and synthesized signal.Then,the smoothed pseudo Wigner distribution (SPWD) is utilized as a time-frequency mask to suppress the cross-terms.By combination of signal synthesis algorithm and time-frequency masking technique,the interference can be extracted and synthesized.Finally,the reconstructed interference is removed from the original echoes to realize the interference suppression.The proposed algorithm can not only suppress the NBI and WBI,but also preserve the useful information as much as possible.The effectiveness of the proposed algorithm is proved by processing the simulation data and measured data.
摘要:Parameter-perturbed orthogonal matching pursuit is an effective technique for estimating the time-delays of off-grid targets.However,the technique consumes large computational loading because it only searches one target at each iteration.This paper develops a kind of low complexity methods,which is called parameter-perturbed band-excluded greedy reconstruction algorithms,to estimate the off-grid targets.The proposed technique combines the band-excluded technique and the greedy reconstruction methods to detect the nearest discrete grids of several off-grid targets and exploits the parameter-perturbed technique to estimate the time-delay bias between the off-grid targets and the nearest discrete grids.Taking the quadrature compressive sampling radar as an example,this paper studies the estimation performance of the proposed technique through the backtracking adaptive orthogonal matching pursuit method.Simulation results show that in comparison with other related methods,the proposed technique reduces the computational time more than one time without affecting estimation accuracy.
摘要:Quasi-optical mode convertor is an important component for a high-power gyrotron.The transverse area of collector in gyrotron can be increased and the wave beam can be separated from the electron beam with a high conversion efficiency quasi optical mode convertor.To improve the conversion efficiency of the whole gyrotron,a quasi optical mode convertor with small size must be used to high-efficiently convert the cavity mode to fundamental Gaussian beam.This paper presents a quasi-optical mode convertor for TE6,2 94GHz gyrotron.The converter consists of a dimpled-wall Denisov launcher,a quasi-elliptical mirror,two parabolic mirrors and a plane mirror.The coupled-mode theory is used to analyze the operation of the pre-bunching waveguide launcher.The code of the vector diffraction theory is used to calculate the radiate fields of the mirrors.The efficiency of more than 97.2% is achieved for converting the rotating TE6,2 mode at 94GHz into a fundamental Gaussian beam.Simulations with feko 6.0 are done to compare with the calculation result.The structures is fabricated and placed in the 94GHz gyrotron.Experimental measurements show that close agreement with theoretical predictions is obtained.
摘要:Space-time adaptive processing (STAP) is widely used for clutter mitigation in airborne radar.However,STAP shows significantly performance degradation for lacking sufficient independent identically distributed (IID) training samples in heterogeneous environment.To solve this problem,we propose a STAP approach based on jointly sensing of multiple measurements.The method sets the radar work with orthogonal and identical waveforms alternately,and achieves the clutter information by current and previous environment echoes.Then the clutter information and platform parameters are used,and a clutter covariance matrix is obtained incorporating system parameters.Finally the space-time processor can be built based on the combination of the estimated clutter covariance matrix and the sample covariance matrix.The simulation results show that the new approach can achieve better clutter mitigation performance under the circumstance of inaccurate environmental knowledge.
摘要:It always tends to assume that the noise contained in signal spread over high frequency domain in the traditional wavelet threshold de-noising techniques.However,it doesn't hold for different noise categories,and threshold de-noising methods in most literatures rarely mention the noise influence spread over low frequency domain.Thus,a new framework for noise reduction base on full frequency domain using wavelet decomposition and noise-type detection are proposed.In this framework,the noise type is firstly to be detected by analyzing autocorrelation coefficient for different noise,and then noise reduction is performed both in low and high frequency domain.The experimental results show that:(1) when signal-to-noise ratio is low,our method not only always achieves better de-nosing performance,but needs fewer decomposition layers than the traditional methods;(2) when the signal-to-noise ratio is high,our method can obtain the same performance as the traditional methods,but our method needs less decomposition layers.
关键词:wavelet decomposition;adaptive de-noising;de-noising in whole bands
摘要:Aiming at solving the problem of multipath time-delay estimation for narrow-band signal in underwater acoustic channel,this paper studied the cepstrum method and proposed a time-delay estimation algorithm based on logarithm domain homomorphic filtering.With the thoughts of spectrum and homomorphic filtering,the received signal is turned into logarithm domain and then subtracted by the stored signal.After the residual signal and noise is filtered,the signal is turned back into the time domain to get the multipath time-delay estimation.Compared with the conventional method filtering/correlation and cestrum methods,this algorithm has advantages in time-delay estimation precision and noise suppression capability.The validity of the proposed method is verified with simulation results and processed lake trial data.
摘要:Instead of the traditional state transition matrix eigenvalue estimation methods,the convergence rate estimation of real coded artificial immune algorithm(RCAIA) is studied based on the stochastic processes theory.The method begins with analyzing the necessary condition for probability-based strong convergence of artificial immune algorithm and takes it as the sufficient condition of a class of RCAIA,and the probability-based strong convergence exponential rate estimation method of RCAIA is proposed.The final convergence of the best antibody is taken as convergence judgment,which can overcome the conservative defect of traditional estimation methods.The method can be used to analyze the convergence and convergence rate of a class of artificial immune algorithms.The research can be used to optimize the convergence rate in the practical application of artificial immune algorithms.
摘要:Addressing the problem of channel order estimation,a blind algorithm for channel order estimation is proposed based on the equalization cost function,which utilizes the architecture of single-input multiple-output (SIMO) finite impulse response (FIR) channels and the statistical characteristic of input/output signals.First,the theoretical asymptotic value of normalized least squares equalization (NLSE) cost function is calculated under the ideal condition,and the link between its corner and channel order is pointed out.Then,its approximate value is analyzed under the practical condition.Finally,a corner optimization factor is introduced,and a channel order estimation algorithm is proposed by examining the NLSE cost function corner.Theoretical analysis and simulation results show that the algorithm is more robust than the other existing techniques in the case of low signal-to-noise ratio (SNR) and small leading/trailing channel coefficients,which leads to smaller equalization error.
关键词:channel order estimation;equalization cost function;single-input multiple-output (SIMO);blind channel identification;blind equalization
摘要:This paper focuses on sensor nodes rapid deployment strategy for event monitoring scenario in mobile sensor networks.Firstly,we deduced conclusion that sense error of Voronoi subdivision is minimum based on nodes error model.Then the node effective cover weight is defined by priority function of event area.We proved that coverage effectiveness is maximization when all of nodes coverage weight are identical.A node deployed algorithm SDOA(Sparse Deployment Optimization Algorithm) is presented.It combined coverage weight and virtual force.It guaranteed network connectivity during algorithm running.Finally,we use simulations to show that our deployment algorithm is accurate and effective and with more coverage effectiveness.
摘要:Optimal sensor placement is foundation and guarantee for design of (Prognostics and Health Management,PHM)system for avionics.The fault-sensor dependency matrix is improved which considers the failure probability of the sensors firstly.Based on this,the constraint optimization model is established and the improved discrete PSO algorithm is used to solve the problem.The algorithm designs the fitness function by the characteristics of optimal sensor placement,and the inertia weight is adjusted adaptively based on the swarm's premature degree which can avoid algorithm limits to local extremum and accelerate the convergence speed.The simulation examples demonstrate that the proposed method is effective,and the optimization results meet all the testability index requirements of system,and it can provide effective direction to the optimal sensor placement of PHM system for avionics.
关键词:optimal sensor placement;testability index;failure probability of the sensor;discrete PSO algorithm
摘要:Due to the frequency reuse,the outage performance of the interference-limited two-hop multiple access-channel (TH-MAC) has become one of the problems to be solved.With interference-limited destination,the outage probability of the amplify-and-forward (AF) TH-MAC is derived in closed form.Firstly,the outage probability is divided into two parts:one is the probability that an outage happens at the relay;the other is the probability that outage does not happen at the relay,but happens at the destination.Then,based on rational approximation for the second part of the outage probability,a closed form outage probability is derived for the channel,and it provides a theoretical framework for the practical system design.Numerical results show that the approximation fits well with the Monte Carlo results,which validates the accuracy of the approximation.
摘要:Recently,most local invariant features for image matching problems center around the invariance of rigid deformation.However,there is a widespread and growing need of the image matching for the non-rigid deformations.The complexity of non-rigid deformation makes the corresponding points' local support area difficult to maintain the structure similarity.So how to build the more pertinent local invariant feature descriptor is the key to solving non-rigid deformation image matching problems.We,therefore,propose an oriented non-rigid deformation local invariant feature descriptor and its structuring method to better specialize the direction of the local feature and divide local support area.We first estimate the elliptical neighborhood of each feature point and indicate the main direction according to its second moment matrix,and then normalized the local support area to eliminate differences in the anisotropic deformation.Finally,we embed the local area into 3D space and construct a descriptor in terms of a heat kernel signature,which is invariant to deformation.Comparative experiments show that,under the non-rigid deformation and photometric changes,our algorithm maintains good rotation invariance and a higher matching accuracy compared with the SIFT(Scale Invariant Feature Transform) and GIH(Geodesic Intensity Histogram) algorithm.
关键词:local invariant features;non-rigid deformation;second moment;heat kernel signal
摘要:Cuckoo search (CS) algorithm is a meta-heuristic optimization algorithm based on Lévy flights.We propose an improved cuckoo search algorithm to enhance the accuracy,avoid the local optima and accelerate the convergence speed.The proposed algorithm has three main characteristics.Firstly,pattern search method enhances the exploitation ability of the basic CS algorithm.The proposed algorithm combines the random exploration of the CS algorithm and the exploitation capacity of pattern search method.Secondly,the optimal solution is obtained by self-adaptive competition mechanism.Hence,the proposed algorithm has a trade-off between searching speed and the diversity of solution.Finally,we realize the cooperation of the optimal solution to share,and strengthen the advantage of experience learning in the use of the optimal solution set search mechanism.The experimental results conducted on 52 benchmark functions show that the proposed algorithm is promising in terms of accuracy,success rate and robustness.And it is also suitable for multimodal and high-dimensional numerical optimization problems.Therefore,in terms of the global search ability and solution accuracy,our algorithm performs better than other modified CS algorithms,such as ICS (Improved Cuckoo Search algorithm),CSPSO (Cuckoo Search algorithm and Particle Swarm Optimization),OLCS (Orthogonal Learning Cuckoo Search algorithm),etc.
摘要:Aiming at the deficiency of the functional differentiation of LUT-based embryonics array,a novel functional differentiation method was proposed.Firstly,based on the functional description,the target circuit was converted to a new form with electronic cell as basic node through front synthesis,logic optimization,logical mapping and packing.Then,the circuit was mapped into embryonics array,and the function and connection of each electronic cell were determined through physical mapping process.At last,the genome was generated and the expressed gene of each electronic cell was confirmed with the genome generated process,and the functional differentiation of embryonics array was completed.The new functional differentiation method does not require the functional evaluation of each generation circuit,and the computational cost is reduced,the calculation speed is improved.The presented method provides a design method for the self-repair circuit application based on LUT-based embryonics array.The functional differentiation is described with an example,and the differentiation speed is verified through various circuits' functional differentiation.
摘要:The existing group aggregation algorithms based on AHP(Analytic Hierarchy Process) judgment matrices pay more attentions on the determination of the weights of experts,and neglect the relationships among the values of judgment matrices.Therefore,a new aggregation algorithm based on the collective operator of spanning trees is proposed in this paper.The algorithm obtains complete consistency collective matrix using collective operator of spanning trees and the relationship between a judgment matrix and a spanning tree of a simple undirected graph.In the algorithm,a complete consistency collective matrix and a group incidence matrix,which represent the individual consistency and group incidence of the values of judgment matrices,are used to compute group collective result under the restraints of consistency and similarity.Finally,a numerical example shows that the method is feasible and effective.
关键词:group aggregation;analytical hierarchy process;judgment matrix;spanning tree
摘要:Based on an innovative two-layer PCB(Printed Circuit Board) medium,a monopole antenna is proposed for inter-chip wireless interconnection communications.The antenna is inserted into the PCB medium as an extended chip pin.The lower plate of the PCB is built up with a mushroom-type metamaterial to construct an electromagnetic wave absorbing layer.The interconnect communication model is analyzed and its scattering parameters are extracted.One antenna performs as a transmitter with a frequency 20GHz pseudo random binary signal with the high and low voltages at 1V and -1V,respectively.And others work as receivers whose received signals are represented through eye diagrams.The result shows that the mushroom-type metamaterial absorbs transmitted electromagnetic waves in the PCB medium,which reduces reflected waves and improves the system return loss for the two antenna system.In comparison with a metal supported PCB system,the eye-diagram derived for the metamaterial grounded PCB system can be improved.It is apparent that the proposed PCB system provides a good wireless inter-chip communication channel with the application of the absorbing feature of the metamaterials.
关键词:inter-chip wireless connection;metamaterial;chip's pin antenna;radio frequency communication;eye diagram
摘要:MacWilliams identity is an useful tool in studying weight distributions of linear codes and their duals.Weight distribution is also an important topic of coding theory.This paper defines the m-ply Lee weight enumerators for linear codes of length n over Z4+u Z4.We give a type of Mac-Williams identity for linear codes of length n over Z4+u Z4 on Lee weight.We prove that this identity is the MacWilliams identity on Lee weight for linear codes over GR(4,m)+uGR(4,m)having generator matrix over Z4+u Z4.Furthermore,by means of Krawtchouk polynomials the equivalent form of the type of MacWilliams identity for linear codes of length n over Z4+u Z4 is obtained.
摘要:An interest group based trust model (IGTrust) for ad hoc networks is presented in this paper.Using the income and costs,IGTrust masks the difference of a transaction's trust from different evaluating peers.Trust relationship in IGTrust which employ the relative income of a peer (or system) as its trust information is divided into three kinds and each solution is also put forward in view of transactions,discount factor and time.Subsequent experimental results show that IGTrust is robust on trust security and also low costs and failure rate when resource search happens.
摘要:In the mobile sink based sensor networks,the energy consumption issue becomes the key point because of energy constraint on sensor nodes.We established a maximizing minimum probability of energy consumption model,and gave a strong mathematical proof.We also proposed a maximizing minimum probability of energy consumption (MMPEC) data collection algorithm based on this model.MMPEC optimize the length of path distributed between child nodes and the aggregation nodes,and achieves low energy consumption of the whole network.The simulation results show that MMPEC is far superior to other hierarchical data collection methods based on a mobile sink in energy consumption.
摘要:The current computer aided diagnosis system classifies benign or malignant lung nodules mainly according to the low-level features of lung CT images.However,clinicians use the high-level semantic features of lung CT images.To overcome the inconsistency between the low-level features and high-level semantic features,a new approach of benign or malignant lung nodules classification based on semantic attributes is proposed.Firstly,lung nodule images are extracted using the threshold probability-map method.Secondly,on the one hand,some features including shape,gray,texture,size and position are extracted from lung nodule images to constitute the low-level feature set;on the other hand,according to the experts' annotation of lung nodules,the attributes are extracted to constitute the high-level attribute set.Thirdly,attribute prediction models are built to map the low-level features to the high-level attributes.Finally,the benign or malignant classification of lung nodules is performed using the predicted attributes.Experimental results on the LIDC dataset show that the proposed classification method possesses high classification accuracy and AUC value.
摘要:Coding aware routing reduces transmissions number and improves network throughput by exploiting network coding opportunities of a route,and has become a hot topic of routing algorithm research in recent years.The network coding condition failure and node energy neglecting are two problems of current coding aware routing.Therefore,current coding aware routing cannot be applied directly in wireless sensor network.A cross layer coding aware energy efficient routing (CAER) is presented for wireless sensor network.The revised network coding condition is proposed and proved to avoid network coding condition failure.Based on the idea of cross layer,the network coding aware mechanism explores potential network coding opportunities through combing topology control mechanism and coverage control mechanism.Besides,a cross layer coding aware routing metric (CCRM) is presented,which jointly considers coding opportunity and energy of nodes.Simulation results demonstrate that,compared with current network coding aware routing,CAER improves the accuracy of coding aware mechanism,increases coding opportunities number by 5%~15% and prolongs network lifetime by 8%~12%.
摘要:The development of Internet of vehicles provides a new traffic information collection technique for the study of intelligent transportation.In this article,we propose an N-order hidden Markov model to approach the vehicle travel time prediction problem,utilizing the Markov nature to model the internship of road network.We also promote an N-order neighbor road network to address the heterogeneity of road.A non-trivia update algorithm is applied to handle the real time data approaching issue.We also prove the temporality of the N-order hidden Markov model in travel time prediction.Experimental results on authentic data indicate the effectiveness and accuracy of this approach.
关键词:travel time prediction;hidden markov model;cluster
摘要:In microblog network,emotion propagates among users and infects user behavior.These emotional behaviors have an important impact on the microblog network or even the real world,so finding emotional influencers in microblog network is very significant for society management or marketing strategies.In this paper,a heterogeneous microblog network that contains two types of nodes (user,microblog) and three types of relations (forwarding,following,posting) is constructed.Utilizing the similarity of microblogs' emotion and users' emotional behavior,the heterogeneous network is transformed to be a homogeneous network that contains only users.Random walk model is employed to find emotional influencers in this network.Our contribution is summarized as follows:we verify emotional homophily in the microblog network constructed by our dataset,which confirms the existence of emotional influence in this network;we propose a novel model (EmotionRank) to find emotional influencers;experimental results effectively illustrate the utility and superiority of EmotionRank.
摘要:Line matching is a difficult problem due to reasons such as incomplete lines,inaccurate locations of endpoints,and so on.To deal with these challenges,we proposed an affine invariant line descriptor.An initial candidate match is dispersed to a set of correspondences. As a result,the problem of inconsistent support region size is resolved because we need only construct descriptors of correspondence points instead of lines. In order to make the descriptor affine invariant,the dominant orientation and the scale of the descriptor are created according to the direction and the length of the line,and gradients of the discrete points set in the local neighborhood are calculated. To speed up line matching,epipolar constraint is used before constructing line descriptors,and the number of potential matches is limited. Then,line matching is preceded accurately by the nearest neighbor distance ratio approach. The experimental results show that the proposed descriptor has accurate line matching under the changes of affine,illumination,viewpoint,and partial occlusion.
摘要:In view of the convergence of narrowband active noise control (NANC) algorithm,a novel adaptive filtered-X weighted accumulated least-mean-square algorithm termed variable forgetting-factor and variable step-size (VFVS-FXWALMS) is proposed for NANC systems.On the basis of FXWALMS algorithm,a scheme using variable forgetting-factor and variable step-size is proposed to obtain better step-size and forgetting-factor values,which can not only better solve the contradiction among convergence,tracking capabilities and steady-state error of the algorithm,but also enhance the anti-interference capability of the system.Thus,the proposed algorithm improves considerably the overall performance of the system.Computer simulations are conducted to demonstrate the superior performance of the proposed algorithm in both stationary and non-stationary scenarios.
关键词:FXLMS;FXRLS;narrowband active noise control (NANC);variable step-size;convergence speed
摘要:The Debye model with multiple relaxation times is most suitable for highly accurate description of the dispersion characteristics of many media,such as biological tissues,soil,and water,among the developed empirical models.In order to reconstruct the electromagnetic properties of these dispersive media simultaneously,a modified time-domain inverse scattering technique is presented.Firstly,in the proposed technology,the nonlinearity and ill-posedness of the inverse problem is circumvented by an iterative method and the Tikhonov's regularization,respectively.Secondly,a set of closed gradients (Fréchet derivatives) of its cost functional with respect to objective parameters are derived for the aforementioned inverse problem.Then,the finite-difference time-domain (FDTD) method and any conjugate gradient (CG) method is applied,at its each iteration,to solve the resulting forward and backward sub-problem,respectively.Lastly,the feasibility and robustness for the inversion technique are preliminarily confirmed by two one-dimensional (1-D) numerical examples where a model of the additive white Gaussian noise (AWGN) is considered.
摘要:The linear canonical transform(LCT)is a generalization of the fractional Fourier transform(FRFT)and has more flexibility than FRFT sinice it contains three free parameters.As a special case of the short-time Fourier transform(STFT),the Gabor transform(GT)is an important time-frequency analysis tool for signal processing.According to the relationship between STFT and LCT,as well as the invariance of Gaussian function in LCT domain,we present a new form of GT and obtain that the LCT is equivalent to an affine transformation for the new GT when parameters meet certain conditions.Furthermore,we investigate the application of this relationship for filter design in LCT domain.Finally,the simulation results verify the correctness and effectiveness of the proposed theory and technique.
摘要:Most of the existing sentiment analysis was performed in the sentiment classification for subjective texts;the emotional evolution analysis for complex interactive texts was not yet thoroughly targeted by the research community.This paper is concerned on short-text Chinese online reviews collected from Tianya forum.Firstly,an efficient affective computing framework is proposed to capture the underlying emotions of online reviews.Secondly,a game theory based emotional evolution prediction algorithm combining the affective computing is proposed,in which the mixed Nash Equilibrium strategies are calculated as the future emotional behavior of interactive users.Experimental results on the large-scaled review dataset are provided to demonstrate the effectiveness and accurateness of our approaches.
摘要:The existing classic anchor-free localization (AFL) algorithms have relatively large localization error and tend to get local optimal value.Therefore,a two-phase improved anchor-free localization (I-AFL) algorithm with phase advantages is proposed.In the first phase of the I-AFL algorithm,a precise coordinate system is established through hop-count quantization,which uses the measured distance information to break the length of node communication radius into multiple hops.The second phase of the I-AFL algorithm uses an adaptive mass-spring optimization algorithm to optimize the node position and prevent the position from falling into local optimum value during the iterative optimization process.Both the theoretical analysis and simulation results show that compared with the traditional algorithm,the proposed algorithm is more precise,robust and converges faster.
关键词:Wireless sensor networks;anchor-free localization algorithm;mass-spring model
摘要:This paper presents a method for DC analysis of IR-drop in power/ground (P/G) networks.The proposed method explores three-dimensional structure of multi-layer P/G grids.By inputting the coordinates of each metal layer and the process rules of vias,the proposed method analyzes the impact of different multi-layers P/G grids on IR-drop.Experimental results demonstrate that the method has high consistency with SPICE simulation results.The relative error is less than 0.4% and the time complexity has the linear relation to the number of vias.Moreover,the topology of inter-layers has an important relation to the value and distributions of IR-drop.