最新刊期
卷 48 ， 期 10 ， 2020
- 摘要：The accurate evaluation of higher order nearly singular integrals is one of the key technologies in accurate simulation by electromagnetic surface integral equations (SIE).However,the present methods mainly focus on low order nearly singular integrals for the planar element modeling,rather than on 3-order nearly singular integrals in higher order geometry modeling.Based on the former research of the Double Arctan-Transformation (DAT),the Improved Double Arctan-Transformation (IDAT) is proposed to improve the stability and accuracy of the nearly singular integrals with singular kernel RR/R5, R/R4 and 1/R3 for higher order geometry modeling.Specifically,the exponential transformation is utilized to stabilize the integrals when the field points are extremely close to the source surface.Furthermore,the shape-function transformation is adopted to stabilize the integrals when the projection point approaches to the border of source surface.The proposed IDAT is also effective for the lower orders of the nearly singular integral kernels.With theoretical analysis and typical testing cases,the accuracy and stability performance of IDAT is fairly evaluated.关键词：surface integral equations;higher order geometry modeling;singular integral;double-Arctan Transformation53|143|0更新时间：2023-02-24
- 摘要：In this paper,we propose an improved Wasserstein generative adversarial network (WGAN),de-aliasing Wasserstein generative adversarial network with Gradient Penalty (DAWGAN-GP),for magnetic resonance imaging (MRI) reconstruction.This method uses WGAN to replace the traditional GAN,and combined the gradient penalty to improve the training speed and to solve the slow convergence problem of WGAN.In addition,for better preservation of the fine structures in the reconstructed images,we incorporate perceptual loss,pixel loss and frequency loss into the loss function for training the network.Compared with other state-of-the-art deep learning methods for MR images reconstruction,DAWGAN-GP method outperforms all other methods and can provide superior reconstruction with improved peak signal to noise ratio (PSNR) and better structural similarity index measure (SSIM).关键词：magnetic resonance imaging (MRI);image reconstruction;WGAN-GP;perceptual loss23|64|1更新时间：2023-02-24
- 摘要：In recent years,deep learning based methods have shown preferable results for the task of inpainting corrupted images.However,the existing standard convolutional neural network approaches often cause problems with excessive color discrepancy,image texture loss and distortion.A deep network based image inpainting model is proposed in this paper,consisting of two generative adversarial network modules.One of the modules is used to inpaint the missing area of the image,where the generator is constituted with partial convolutions.The other module is the image optimization network,which is applied to solve the problem of local chromatic aberration after image restoration,and in which the generator is originated from the depth residual network.These two modules cooperated to improve the visual effect and image quality of inpainted images.Using MOS,SSIM and PSRN as the evaluation criteria,the experimental results of qualitative and quantitative comparisons with other state-of-the-art methods have shown that the proposed model performed better.关键词：partial convolution;generative adversarial networks;residual network;image inpainting29|88|6更新时间：2023-02-24
- 摘要：The bottleneck problem of single image de-raining is how to preserve the detailed structure information of image background with a lack of time series data between frames when removing the multi-density rain fringe.The new single image de-raining algorithm based on the coder and decoder structure is proposed in this paper.Firstly,the positional relationship information of various pixels between the points is obtained by non-local operation which can obtain the global information of the image representation.Secondly,the spatial attention is applied to recalibrate the global information in the spatial dimension position,and the channel features are nonlinearly modeled in the channel dimension to aggregate similar characteristics and useful information.Finally,the original size of rainy image is obtained by utilizing the de-convolution and long distance residual connection.The experimental results and analysis show that the proposed algorithm can obtain significant de-raindrop effect.The proposed algorithm can also resolve the difficulties of removing raindrops with various densities while maintaining the details of the image and edge information.关键词：attention mechanism;non-local;useful information;de-convolution;edge information21|86|10更新时间：2023-02-24
- 摘要：RoIs extraction is a challenging task due to scene complexity,less image texture and unstable image quality.Paying more attention to local pedestrian details and neighborhood relations of pixels,threshold segmentation is prone to give rise to mis-segmentation such as adhesion,omission,breakage and uncontrollable RoIs amount.Imitating human’s eyes,we focus on an image saliency region,size and location and propose a new RoIs extraction method using probability map.Design convex-concave curves to map an image pixel gray for enhancing the image contrast;get a saliency map based on image signature manner;fuse intensity and saliency probability images and then extract the image foreground;design an algorithm to search the probability map region limited by a road horizon and generate RoIs.Experimental results show that our method can improve RoIs locating accuracy,control RoIs amount and reduce non-pedestrian RoIs when comparing with threshold segmentation method.Our recall increases no less than 9% when same RoIs amount are extracted.关键词：RoIs extraction;saliency probability map;intensity probability map;vehicle-mounted thermal pedestrian detection14|50|0更新时间：2023-02-24
- 摘要：This paper aims to solve several underlying problems arising from the application of current single-cell power domain non-orthogonal multiple access (NOMA) structure.Specifically,to solve the problems lying in the current single-cell structure such as high decoding complexity,poor user fairness,and bad quality of service of cell edge users,this paper integrates NOMA technique with the orthogonal resource allocation strategy to present a parallel multiple-pairing operation unit,where the frequency band orthogonally allocated to each central user and the cell edge user share the total bandwidth by pairing with each central user to perform NOMA and a full-duplex cooperative communication strategy is further introduced to the central users for they have the priori-knowledge of the cell edge user.Simulation results verify the validity of the proposed scheme.关键词：non-orthogonal multiple access (NOMA);achievable rate;outage probability;fairness;complexity13|38|0更新时间：2023-02-24
- 摘要：To tackle the difficulty of imbalanced traffic load when forwarding data packets through multiple hops in LEO satellite network,we propose a collaborative game theory routing strategy with incomplete information.Each node determines the current optimal routing path of the data packets through cooperating with neighbors for locally optimized coalition gain.This strategy distributes and balances the traffic workload among neighbor nodes effectively.Simulations show that compared with satellite routing based on the DSP shortest path or the TLR intelligent routing,the proposed algorithm reduces the average transmission delay by 18.5%,and improves the load balancing of nodes by 65.6%.关键词：LEO satellite network;coalition game theory;optimal revenue;routing algorithm9|44|1更新时间：2023-02-24
- 摘要：Cross-domain recommendation can study the knowledge of the auxiliary domains to enrich the knowledge in the target domain,so as to improve the recommendation precision in the target domain.This paper proposes an aggregation and transfer collaborative filtering algorithm for cross-domain recommendation (ATCF).In order to represent the sharing knowledge in different domains,the knowledge in the auxiliary domain and the target domain are fully aggregated,through two levels of matrix concatenation.Moreover,the personalized knowledge of the target domain is represented by knowledge transferring from auxiliary domain.By fusion the sharing and the personalized knowledge,we can obtain the final rating.Two different cross-domain datasets are used for the experiments.Our efforts show that the ATCF algorithm has better recommendation performance.关键词：cross-domain recommendation;matrix factorization;transfer learning;knowledge aggregation13|57|4更新时间：2023-02-24
- 摘要：Secure multiparty computation (SMC) has become research focus in the international cryptographic community in recent years.Privately determining equality of data is a basic problem in SMC.It is widely applied to fingerprint matching,keyword searching and so on.Studying privately determining equality of data has important theoretical and practical significance.Protocol I which is based on the Paillier cryptosystem can privately determine equality of two rational numbers.Protocol Ⅱ which is based on elliptic curve cryptosytem can privately determine equality of multiple rational numbers.Finally this paper gives the protocol for privately determining equality of rational numbers in the malicious model.关键词：cryptography;secure multiparty computation;data equality;rational number;fingerprint matching13|28|2更新时间：2023-02-24
- 摘要：Sensitivity is the ability of microwave radiometer to detect the minimum temperature change and is an important index of microwave radiometer.In this paper,a method of calculating the sensitivity of each channel by means of the output of the correlator is proposed for the digital correlation full polarized microwave radiometer.The effects of quantization noise and crosstalk on sensitivity of analog channel and digital channel are analyzed.关键词：fully polarized microwave radiometer;digital correlation;sensitivity10|44|1更新时间：2023-02-24
- 摘要：Dense heterogeneous networks (DenseHetNets) is one of the most promising technologies to enhance the 5G network capacity.However,with a large number of small cells,the problem of co-channel interference and energy efficiency of small cell base stations (SeNBs) becomes more and more severe.In order to improve the power efficiency and alleviate the co-channel interference,small cells need to be switched on or off accordingly.This paper proposes a SeNB switching on/off algorithm based on the interference contribution rates of the small cells that involve less signaling information in the procedure of switching on/off operation and require less computational complexity under the dense HetNet environment.Simulation results show that the algorithm proposed in this paper can effectively reduce the co-channel interference of small cells and improve the total network rate,at the same time,maintain a lower loss rate of load in small cells.关键词：HetNets;small cell on/off;energy efficiency;interference contribution ratio44|8|0更新时间：2023-02-24
- 摘要：Multipath signals can be used to realize localization since they are abundant and contain geometry information of indoor environments.Based on this,this paper proposes a multipath-assisted target localization algorithm.Firstly,the fitness function about the target and scatterer locations is constructed with Time of Flight (TOF) differences.Then,the locations of the target and scatterers are searched jointly by Particle Swarm Optimization (PSO) and Angle of Arrivals (AOAs) that determines searching ranges.Secondly,the estimated locations of scatterers and TOF differences are used to estimate the target location.Finally,all target locations are clustered by using Affinity Propagation Clustering (APC),and a clustering criterion is proposed to eliminate big localization errors.The simulation results show that the proposed algorithm can achieve high localization accuracy with a single base station.关键词：indoor localization;multipath signal;particle swarm optimization;affinity propagation clustering103|62|3更新时间：2023-02-24
- 摘要：Based on the general application of chaotic synchronization control in many fields,such as secure communication and optimization,the synchronization of a class of drive-response systems is studied in this paper,including exponential synchronization and polynomial synchronization.One kind of cellular neural network with multiple proportional delays is used as the driving system and the other one without the proportional delay is used for the response system.Under the condition that the activation function satisfies the Lipschitz,by designing a proper delay-dependent controller,using the Lyapunov stability theory and some inequality analysis and discussing the stability of the error system,two criteria of the synchronization for the driven response systems are obtained;then two simulations are shown to verify the results.关键词：polynomial synchronization;exponential synchronization;proportional delays;cellular neural networks54|23|0更新时间：2023-02-24
- 摘要：In recent years,more and more attention has been paid to the study and research of noise suppression.This paper proposes a noise suppression deep learning strategy that deals with the presence of noise in training data by constructing a Noise Unaware Network (NUN) and Reliability Estimation Gate (REG).By evaluating the reliability of each sample and adjusting its weight during training,the influence of label noise on network training can be reduced.As the model is updated iteratively,the weight of clean data will gradually increase,while the weight of noise data will be suppressed.Experiments on multiple benchmark data sets demonstrate the effectiveness of the proposed deep learning strategy for noise suppression.关键词：deep learning;image classification;noise-suppression41|36|1更新时间：2023-02-24
- 摘要：In the X-ray CT imaging inspection system,due to the complicated structure of complex shaped workpieces,scattering,hardening and other phenomena are more serious.At some projection angles,due to the large difference in effective thickness in the ray transillumination direction,the fixed energy ray dose does not match the thickness,the quality of the projection data is poor,the reconstructed image obtained by the traditional CT reconstruction algorithm is of low quality,the edges are blurred,and the complete contour information of the workpiece cannot be obtained.In this paper,a CT imaging method for complex shaped workpieces based on contour prior constraints is studied.Firstly,the binocular stereo vision technology is used to obtain the contour information of the workpiece,and the registration of the a priori image is completed according to the spatial positional relationship between the binocular coordinate system and the CT coordinate system.The contour prior is then incorporated into the CT reconstruction process and contour constrained reconstruction is performed in conjunction with TV (Total-variation) regularization.The experimental results show that the proposed method can effectively suppress artifacts and noise,preserve the edges of reconstructed images,improve the quality of reconstructed images,and improve the reliability of defect detection of complex shaped workpieces.关键词：X-ray imaging;complex structural parts;contour constraint;prior information26|29|2更新时间：2023-02-24
- 摘要：The rapid development of machine learning makes itself one of the most effective tools in the data mining research community.However,the training of algorithm often needs a large amount of user data,which brings a great risk of privacy leakage to users.Due to the complex statistical characteristics and semantic richness of the data,traditional private data publishing methods usually sanitize original data too excessively to lead to low data availability and uselessness in data mining tasks.In this paper,a differential private data publishing method based on generative adversarial network (GAN) is proposed.The differential privacy of the GAN model is realized by adding carefully designed noise to the gradients during the training procedure,so that the GAN can generate unlimited synthetic data conforming to the original statistical characteristics without disclosing any privacy.Aiming at the problems of low quality synthetic data and slow convergence in the existing similar methods,several optimization strategies are designed to adjust the privacy budget allocation and reduce the overall noise scale.Moreover,we provide rigorous proof that the synthetic data satisfies the differential privacy.Comparisons with existing methods on public datasets show that the method proposed can generate private data with higher quality more efficiently,which is suitable for various data analysis tasks.关键词：differential privacy;generative adversarial network;private data publishing;synthetic data;data mining41|47|2更新时间：2023-02-24
- 摘要：Ciphertext-policy attribute-based encryption scheme is suitable for access control of ciphertext data in cloud environment.Most of the existing attribute-based encryption schemes supporting arithmetic span programs are key-policy schemes with large public parameter scale.By exploiting the double Dual Pair Vector Space(DPVS) technique,a ciphertext-policy attribute-based encryption scheme which has a fixed length of public parameters and supports arithmetic span programs is proposed in this paper.In this scheme,the ciphertext-related access control vector is combined with the random matrix,and the key-related attribute components are combined with the public parameters in the entropy expansion Lemma.The method is designed according to the form of ciphertext and key components given in the entropy expansion Lemma.Finally,the adaptive security of the scheme is proved based on the prime order bilinear entropy extension Lemma and k-Lin difficult assumption.Compared with the existing attribute-based encryption schemes which support arithmetic span programs,the new scheme has the advantages of ciphertext access policy,fixed length of public parameters and adaptive security.关键词：arithmetic span program;ciphertext-policy attribute-based encryption;bilinear entropy expansion;k-linear assumption;adaptively secure20|34|1更新时间：2023-02-24
- 摘要：Template attack is divided into two stages:template description and key recovery.For AES-128 algorithm,when the attacker only got 1000 energy traces,he would face two problems:one was that the template description would not be applicable,the other was that the correct key would not be recovered.To solve these problems,this paper constructed 9 templates for Hamming weight of S-box output value in the template description stage,and used 600 energy traces provided by panda 2018 data set to build the model;in the key recovery stage,this paper proposed the method of key advantage superposition,which only needed about 10 energy traces encrypted to distinguish the correct key.This method reduces the number of energy traces required in the template description stage and key recovery stage,lowered the difficulty of template attack,and improved the success rate of template attack.关键词：template attack;AES-128 algorithm;key advantage;Hamming weight model45|26|1更新时间：2023-02-24
- 摘要：As non-orthogonal multiple access technology can achieve higher system throughput,spectrum efficiency and energy efficiency than traditional orthogonal access technology,it has become a research hotspot of 5G multiple access technology.In this paper,a power allocation strategy based on Improved Particle Swarm Optimization (IPSO) is proposed to optimize the energy efficiency of NOMA downlink system.The Standard Particle Swarm Optimization (SPSO) is improved in three aspects,and the IPSO algorithm is used to solve the objective function to maximize the energy efficiency of the system.The simulation results show that at the optimal power allocation point,the IPSO algorithm can significantly improve the energy efficiency of the system.关键词：non-orthogonal multiple access;energy efficiency;power allocation;particle swarm optimization81|49|9更新时间：2023-02-24
- 摘要：Aiming at the fact that traditional control algorithm cannot continuously and accurately track the maximum power point under the condition of partial shading,a strategy for photovoltaic systems based on improved sheep behaviors optimization was proposed.Disturbance operator,reverse position and adaptive factor were introduced in sheep behaviors optimization to enhance the local development ability,and improve the optimization accuracy of the algorithm.The experimental results under the conditions of uniform illumination,partial shading and variable irradiation show that the improved sheep behaviors optimization can steadily track the maximum power point under different environmental conditions,and the algorithm has greater advantages in convergence time and convergence accuracy,verifying the feasibility of this algorithm in the maximum power point tracking control.关键词：photovoltaic system;sheep behaviors optimization;partial shading;maximum power point tracking26|37|1更新时间：2023-02-24
- 摘要：In recent years,the correlation filter based algorithm combined with deep features has received extensive attention.The period assumption of the training samples improves the computational efficiency,but also introduces the boundary effect,which limits the further improvement of the tracking performance.By exploring the deep feature representation ability,a new tracking framework is proposed.Since the deep features have good semantic information,the fifth layer convolution feature of VGG network is used to extract the spatially reliable region of the target,and the region information is introduced into the objective function to establish a spatial constraint model.Then iteratively solved by ADMM algorithm.In order to further improve the long-time tracking ability,a simple and effective method of occlusion detection is proposed.Experimental results show that the proposed algorithm outperforms most advanced algorithms in tracking accuracy and success rate.关键词：visual tracking;spatial regularization;deep features;correlation filter;model update61|44|7更新时间：2023-02-24
- 摘要：The finite element analysis model of micro-scale chip scale package(CSP) solder joints is set up and stress and strain analysis are performed under torsion load,and the verification experiment is also carried out.The influence of solder joint diameter,solder joint diameter,pad diameter and solder joint height on the torsion stress and strain of micro-scale CSP solder joints are analyzed.The solder joint material,solder joint diameter,pad diameter and solder joint height are selected as design variables;29 groups of solder joints with different levels are designed and simulated by response surface methodology.The regression equation of solder joint torsion stress and structural parameters is established.The structural parameters of solder joint are optimized based on the regression equation and genetic algorithm.The results show that the maximum torsion stress and strain occurs with the solder joints material of SAC305,the maximum solder joint torsion stress and strain decreases with the increase of solder joint diameter and pad diameter,and increases with the increase of solder joint height;the optimal combination of solder joint level is the solder joint material of SAC305,the solder joint diameter of 0.22mm,the pad diameter of 0.14 mm and the solder joint height of 0.14mm,and simulation verification result shows the maximum stress of the solder joint decreases by 3.7MPa after optimization.关键词：micro-scale solder joints;torsion load;stress and strain;response surface analysis;genetic algorithm7|24|0更新时间：2023-02-24
- 摘要：A fully customized digital signal processor for motor field-oriented control is proposed.The instruction set,memory model and interaction method with main-CPU is discussed.Fixing some of the operands in multi-op instructions allows a reduced instruction encoding length and therefore an economical code size.Background execution improves arithmetic and logic unit(ALU) computation parallelism.In this article,the instruction cycles of typical field-oriented control(FOC) algorithm is shown.The silicon implementation of the proposed digital signal processor (DSP) is provided.Taken the commonly used ARM Cortex-M0 processor and several main-stream DSPs as the comparison baseline,the experiment results demonstrate that the micro architecture and circuit implementation of this DSP is energy-efficient and competitive.This approach improves the processing performance at the cost of limited circuit resource.关键词：motor control;embedded system;field-oriented control(FOC);digital signal processor(DSP);microarchitecture;instruction set architecture(ISA)20|27|1更新时间：2023-02-24
- 摘要：Network representation learning aims to learn the low-dimensional dense real-valued vector of network information,which solves practical tasks such as link prediction,anomaly detection,and recommendation systems.Recently,network representation learning has made great progress.Most existing researches focus on static networks,while real network is dynamic all the time.This survey proposes state of the arts on representation learning of dynamic network.Firstly,it provides historical overview of representation learning in network,followed by the motivation and theoretical basis of dynamic network representation learning.Then comprehensive analysis of dynamic models is proposed,including matrix factorization,random-walk,deep learning,edge reconstruction based dynamic models,and gives the application scenarios of dynamic network embedding.Finally,research directions of representation learning in the future are summarized.Only when considering the temporal dynamics,structure and content can we truly reflect the evolution of the real network and make network representation learning more valuable.关键词：network representation learning;graph embedding;deep learning;graph convolutional networks66|72|6更新时间：2023-02-24
- 摘要：Local characteristic-scale decomposition (LCD) is an adaptive signal decomposition method proposed to overcome the shortcomings of mean curve construction in empirical mode decomposition (EMD) and has been applied to mechanical fault diagnosis.However,LCD also has the mode mixing problem that exists in EMD.Based on the construction of masking signal with uniform phase difference,the adaptive mask signal ensemble local characteristic-scale decomposition (AMSELCD) is proposed in this paper to adaptively decompose a complex signal into several intrinsic mode functions and a residue,which can effectively alleviate the mode mixing phenomenon of LCD.AMSELCD is compared with various existing methods for restraining mode mixing through simulation signal analysis and the results show the effectiveness and superiority of the proposed method.Finally,aiming at the modulation characteristics of fault signals of rolling bearing and rotor rubbing,the proposed AMSELCD method is applied to the fault diagnosis of rotor rubbing and rolling bearing,and the experimental comparison analysis results further verify the effectiveness and superiority of AMSELCD.关键词：empirical mode decomposition;local characteristic-scale decomposition;ensemble empirical mode decomposition;mode mixing;fault diagnosis36|29|2更新时间：2023-02-24
- 摘要：In order to harvest the acoustic energy in the environment efficiently,an acoustic energy harvesting system is proposed with piezoelectric transducer array,straight tube resonator and energy harvesting circuit.When the acoustic wave transmits into the straight tube resonator,the resonance standing wave is generated in the tube,and excites the piezoelectric transducers;then the acoustic energy is converted into electrical energy.An energy harvesting circuit is designed and analyzed theoretically with simulation.The relationship between the number of piezoelectric chips,the frequency of acoustic waves,the acoustic pressure level and the output voltage is studied with experiments.The output voltage and power are analyzed with the change of load resistance.Experimental results show that the proposed system can harvest the acoustic energy in different frequencies,and the harvested power achieve the maximum at the sound frequency of 96Hz.Without using the energy harvesting circuit,the maximum output AC voltage effective value is 12.9V and the maximum output AC power is 799μW at the incident acoustic pressure level of 110dB.When the proposed energy harvesting circuit is applied,the maximum output DC voltage is 64.2V and the maximum output DC power is 473μW.The acoustic energy harvesting system can not only harvest acoustic energy,but also can power the micro electronic systems such as wireless sensor nodes.关键词：acoustic energy;piezoelectric transducer array;straight tube resonator;energy harvesting circuit29|23|2更新时间：2023-02-24
- 摘要：In order to realize the energy-saving operation of medium and small power three-phase inverters at high switching frequency,a novel three-phase resonant DC link inverter topology is proposed for the first time.When the auxiliary circuit on the DC link of the inverter participates in the working process,the DC link voltage across the input end of the bridge arm can periodically form a zero voltage state,and the main switch and the auxiliary switch can achieve the zero-voltage soft-switching.When the high frequency metal oxide semiconductor field effect transistor is used as the switching device of the inverter,the zero-voltage soft switching can eliminate the capacitive turn-on loss of the MOSFET(Metal Oxide Semiconductor Field Effect Transistor),which is beneficial to optimizing the efficiency of the inverter.In this paper,the workflows of the circuit are analyzed.The experimental results on the 2.5kW prototype show that the switching devices are all at zero voltage soft-switching.Therefore,this topology has reference value for the development of high performance medium and small power three-phase inverters.关键词：inverter;DC link;auxiliary circuit;resonance;zero-voltage switching;switching loss27|63|0更新时间：2023-02-24
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0