最新刊期
卷 ， 期 4 ， 1997
- 摘要：We have studied the problem of blind source separation of instantaneous mix tures,which is to recover the unknown sources from observations without knowing the characteris tics of the transmission medium,by using third─and fourth─order cumulants.A class of criteria are presented,which state that for independent sources,it is sufficient for separation that there is no more than one Gaussian.lgorithms are also developed based on these criteria,whose efficiency is verified by experiments.关键词：blind source separation;cumulants;Instantaneous mixtures2|358|111更新时间：2025-12-08
- 摘要：A new fuzzy judgement approach named fuzzy tree（FT）is proposed to evaluate voice quality of communication network.First,the essential idea of FT is described.Next,the fuzzy membership functions of all factors affecting the voice quality are determined.And all the fact weights are also estimated by statistics.Finally,the voice quality of the communication network is evaluated by using the fuzzy tree.关键词：Fuzzy tree;communication network;quality2|94|19更新时间：2025-12-08
- 摘要：N#Burst/D/1/K is widely used in the study of traffic control in ATM networks.In small buffer capacity region,a minus exponential function can characterize the curve of the loss probability changing with buffer size in this qUeueing model.So the curve can be obtained directly using the loss probability without buffer and that with one buffer,thus making it very simple to analyze the loss probability with small buffer capacity.Simulation shows that this approach is very accurate.关键词：asynchronous transfer mode;Bursty sources;Cell loss2|29|2更新时间：2025-12-08
- 摘要：The capability of CLOS network to support connection requirements of conference is discussed in this paper.A scheme to combine an introducing network and a conference network is proposed.This scheme fits well to today’s network practice with the intensive use of the cross connection equipment. It is proved that, when the ports of conference network assigned to each conference are in order, the necessary and sufficient condition of conference rearrageable nonblocking is mn+ 1.An algorithm to select the routing set for each conference is constructed to realize the conference non-blocking and conference independent routing as well for the cases of m n+ 2.关键词：CLOS network;Conference connection;Conference independent routing;switching2|19|2更新时间：2025-12-08
- 摘要：In this paper a concept of multidimensional DS/SS communication systems is proposed. "Medium code" is defined which the system performance depends on.These systems are more generalized compared with other known M-ary DS/SS systems,which can broaden our mind to design more powerful DS/SS communication systems.In order to reduce the negative impact of multiple access and near-far interferences a trellis coded multidimensional DS/SS communication scheme is presented,and the numerical results of asyptotic error performance of the systems are given in the case of multiple access. In the end an overall conclusion in conjunction with some issues needed to be explored is drawn.关键词：medium code;multidimensional direct sequence spread spectuum（multidimensional DS/SS）;trellis coding2|29|0更新时间：2025-12-08
- 摘要：The performance for ATM multiplexer under integrated Service and with bursty traffic which is shaped is analyzed.It is proposed that the shaper output process can be approximately described as interrupted Bernoulli Process（IBP）. With an improved iterative algorithm,the cell loss probability and the average delay for ATM multiplexer are exactly calculated. The results show that by the choice of appropriate shaper parameters,the cell loss probability is greatly improved,moreover,the cell average delay is small. The correctness of the analyses is verified by the simulations.关键词：IBP;bursty traffic shaping;ATM multiplexer;iterative algorithm;cell loss probability2|21|0更新时间：2025-12-08
- 摘要：In separating concentrated signals,the technique of plane transformation[1～3]has multi-meaning. Sometimes the "Fake Characteristic Curved（FCC）on the plane which is similar to the true characteristic curve may be observed in plane transformation. This paper analyses the principle of multi-meaning, demonstrates the cause of the FCC. Meanwhile, and presents some methods of distinguishing and separating the FCC. The analyses of multi-meaning make the plane transformation more perfect and practical.关键词：Signal separation;Plane transformation;Multi-meaning;Fake characteristiccurve （FCC）2|83|25更新时间：2025-12-08
- 摘要：A time domain full wave method for extracting of the frequency-dependent equivalent circuit parameters of multiconductor interconnection lines is presented.The method is based upon the finite-difference time domain（FDTD）method.The circuit parameters extracted by this method can be inserted into a circuit simulation software to investigate the time domain response of high speed IC system with multiconductor interconnects. Because of the uncertain voltage and current definitions in full wave analysis,transformation among the circuit parameters upon different definitions of voltage and current is also given. The reliablity of the method is illustrated by its application to representative problems.关键词：electromagnetic modeling;Multiconductor interconnection lines;Frequency-de-pendent parameters;Time domain full wave;FDTD2|148|3更新时间：2025-12-08
- 摘要：In this paper,a thinning-like （TL） property of cellular neural networks（CNN）with opposite-sign templates and a non-unity gain output function is discovered and proved. An image thinning algorithm is realized by a multi-layer CNN,which operates with the TL property,with four local logic units （LLU）. An image edge detection is achieved by another parallel CNN with a LLU. Some simulation results are given.关键词：Cellular neural networks （CNN ）;Opposite-sign templates;Image thinning;image edge detection2|30|2更新时间：2025-12-08
- 摘要：When the attitude and height of the camera is changed by a small increment,a theoretical model of geometrically distorted scenes is estabilished,the registration accuracy of edge distance transformation matching method and its analytic expression is analyzed and derived. Many simulated experiments with real satellite images and aerial images demonstrate the validity of this proposed registration accuracy model,which provides a theoretical and practical basis for design of self-guided systems using image matching.关键词：image matching;distance transformation;Geometric distortion;Registration accuracy2|153|3更新时间：2025-12-08
- 摘要：In this paper,a new zerotree based wavelet image coding scheme is put forward.Prior to zerotree coding,all the wavelet coefficients are rearranged according to their energy estimation based on tree structure. As a result,bit allocation is optimized and PSNR is improved by 0.3-0. 5dB. Meanwhile,both encoding and decoding processes are about 10% less time-consuming,Furthermore,the code stream produced is embedded and able to support multirate image decoding.关键词：wavelet transform;Image compression;zerotree;Multirate2|74|34更新时间：2025-12-08
- 摘要：In this paper,we present a new algorithm,called adaptive oriented orthogonal projective decomposition （AOP）,to decompose a signal into a sum of Gauss functions. The elementary functions used in AOP are the dilations,modulations and translations of a normalized Gauss function. They can be adjusted to best match the original signal. We derive a Time-Frequency energy distribution, by adding the Wigner-Ville distribution of the selected elementary functions. Since the Gauss functions are well localized in the time-frequency domain,the AOP distribution has high resolution,no crossterm and no negative energy. It’s a powerful tool of time-frequency analysis.关键词：Time-frequency analysis;Projective decomposition;AOP decomposition2|274|76更新时间：2025-12-08
- 摘要：An Iterative Convex I-Projection(ICIP algorithm is designed to solve Maximum Entropy （ME） X-CT reconstruction problem. This difficult problem is divided into several relatively easy subproblems,so that the previous reconstruction problem is solved by means of alternative solving of these subproblems. The convergence of the algorithm is proven and at the same time the converging result is guaranteed to be ME solution. Computer simulation shows that the proposed algorithm is effective.关键词：Iterative convex I-projection;maximum entropy;X-CT reconstruction2|74|1更新时间：2025-12-08
- 摘要：In this paper,a novel coding algorithm is presented for still images using biorthogonal wavelet transform,and a new data structure named extended subtree is proposed to organize wavelet transform coefficients. Experimental results show that the new algorithm is very practical and with good performance.关键词：wavelet transform;image coding;entropy coding2|92|9更新时间：2025-12-08
- 摘要：On the basis of the general discrete Zak transform and the relationships between the continuous and discrete Gabor expansions,we present a method to construct the discrete Gabor expansion via tight Weyl-Heisenberg frame in the case of oversampled by an integer factor. The properties of the constructed discrete Gabor expansions are also given.关键词：Discrete Gabor expansion;Tight frame;Discrete Zak transform2|83|13更新时间：2025-12-08
- 摘要：When the sources are uncorrelated,the correlation matrix of output of a uniform array is Hermitian and Toeplitz. Therefore,the method of direction finding using ESPRIT with restricted minimum redundant linear arrays is presented in this paper. Compared with ESPRIT,theoretic analysis and simulations both show that the method leads to significantly improved performance,such as not requiring pairwise matched array with doublets,having higher utilization ratio of arrays’ aperture and higher angle resolution,and estimating more sources with fewer sensors.关键词：High resolution techniques;direction finding;Minimum redundant linear arrays2|120|17更新时间：2025-12-08
- 摘要：An algorithm of direction finding and array calibration,based on the Maximal Set of Nonredundant Cumulant （MSNC）,is presented in this paper. The algorithm utilizes all nonredundant information of the spatial fourth-order cumulants. It is insensitive to additive Gaussian noise （white or colored）and superior to conventional covariance - based DOA algorithms with increased tolerance,reduced estimation variances and enhanced resolvabilities. It also circumvents the iteration between direction finding and array calibration and provides robustness performance. Simulation experiments illustrate the potential performances of the proposed algorithm.关键词：DOA estimation;array calibration;Cumulants.2|146|14更新时间：2025-12-08
- 摘要：The concept of worst case source and worst case traffic for a policing algorithm is discussed. Emphasis is given to the importance of worst case traffic analysis in performance evaluation of a policing mechanism. Basic leaky bucket and buffered leaky bucket are analysed and compared for worst case traffic. A new algorithm of two stage leaky bucket is proposed. Simulation results show that the algorithm is superior to the buffered leaky bucket in worst case traffic and less cell buffers are needed.关键词：ATM;Policing;Worst case traffic;Two stage leaky bucket2|44|1更新时间：2025-12-08
- 摘要：A spatial null preprocessing algorithm is proposed,and it’s performance according to the criteria of the optimal resolution and Cramer-Rao lower bound has been studied. The method presented here and it’s application to the direction of arrival estimation with coherent sources are varified by the simulation results and experiments using real-world data.关键词：阵列;空间谱;预处理;零点;波束空间2|74|5更新时间：2025-12-08
- 摘要：By introducing concepts of cut,cut equivalence and cut complement, the convergence of associative memory neural networks based on outer product equstion learning rule under synchronous operation mode is studied. It is proved that the number of iterations required to achieve convergence is at most two.关键词：neural networks;Outer product equation rule;convergence;Cut;Cut equivalence2|33|2更新时间：2025-12-08
- 摘要：This paper investigates the implementation of the Bruun FFT on the TMS320C30. It presents effective implementation methods and techniques in the aspects of butterfly computation, loop control and data transfer. The resulting program runs faster than radix-2,radix-4 and any other FFT.关键词：fast Fourier transform;digital signal processor3|35|4更新时间：2025-12-08
- 摘要：In this paper,an efficient method is presented to improve the speed of minimum distortion encoding in vector quantization. By using a group of distortion measurement criterion and a presorted codebook,a large number of codewords will be rejected before computing MMSE for them. Using this method results in a considerable reduction in the number of operations. The result shows that the reduction of computation is up to 93% in the encoding time against full search method. Little precomputation and extra storage are required.关键词：Image coding;Vector quantization （VQ）;Minimum mean square error （MMSE）2|70|11更新时间：2025-12-08
- 摘要：A kind of large signal model for power factor correction,TL domain quasi-stationary model,is presented in this paper. Based on the model,the close loop system’s Lyapunov stability is analyzed. The global asymptotic stability can be reached with correctly selection of feedback function. Simulations of the system based on the model are provided with experimental results for verification.关键词：功率因数校正;大信号分析;稳定性;计算机模拟2|103|3更新时间：2025-12-08
- 摘要：In this paper,we obtained two necessary and sufficient conditions for existence of equilibrium porints to cellular neural networks with asymmetric template by matrix Ostrowski and Brauer dominant. Then the stability conditions of cellular neural networks are given by Ostrowski and Brauer theorem. A new investigation approach for cellular neural networks with asymmetric template is offered.关键词：Cellular neural networks;Asymmetric template;Complete stability;Equilibrium points2|28|1更新时间：2025-12-08
- 摘要：A novel waveform balance technique is presented for nonlinear circuit simulation in this paper. By applying the phasor concept and superposition principle of linear networks,analytical waveform balance equations have been established in the time domain. The equations are solved by Newton-Raphson method,and the Jacobian matrix is derived. Calculations show that the new method has good accuracy and convergence.关键词：Nonlinear circuit;Newton method;Waveform balance method2|211|0更新时间：2025-12-08
- 摘要：The wave grounded capacitor being taken as a basic building block,a new kind of current mode continuous time wave active filters using current mirrors and grounded capacitors is presented. The practical designed example demonstrates that the filters proposed have better passband characteristics than ones based on current integrators.关键词：Wave filters;current mode circuits;Integrated filters;Current mirrors2|52|8更新时间：2025-12-08
- 摘要：A test pattern generation approach for digital circuit is presented. The approach employs genetic algorithm to generate tests for faults. The gate level description of circuit under test （CUT） is translated into a nonlinear netwrok that can be computed easily,and an improved genetic algorithm is used to find out optimal solutions of energy function of the constrained network,so the test set of CUT is obtained. The approach has many merits,such as it can generate tests for all detectable faults including multi faults,possess the characteristics of parallel processing and can be implemented on multiprocessors.关键词：Digital circuit;test generation;genetic algorithm;expert system2|115|3更新时间：2025-12-08
- 摘要：A class of unconditionally secure authentication codes is constructed by constructing functions that satisfy some conditions.关键词：unconditionally secure;Authentication codes;Deceive probability2|86|3更新时间：2025-12-08
- 摘要：This paper presents a new digital signature algorithm（NDSA）,it is the improved algorithm of DSA proposed by NIST. In NDSA,the second component of the signatur is not k-1 （H（m ） +rx）modq but （k - rH （m ）x）modq. NDSA can be used to enhance the security of signature batch verification,construct multisignature scheme,blind signature scheme and multi-blind signature scheme.关键词：Digital signauture;message authentication;Crytographic protocol2|208|10更新时间：2025-12-08
- 摘要：In lower bit-rate speech coding systems,Codebook-Excited Linear Prediction（CELP）techniques are often used,where the stochastic codebook structure and its searching algorithms will directly affect the synthesized speech quality and real-time realization of speech coding systems.The paper proposes a new method for lower bit-rate speech coding-Wavelet-Excited Linear Predictive coding（WELP）,where it uses three wavelet codebooks instead of the stochastic codebook.WELP highly increases the computational speed and reduces the complexity. With the same quality of synthesized speech for 6. 6kb/s speech coder,WELP can increase the coding speed twice than that of CELP.关键词：Wavelet transform（WT）;Wavelet-excited linear predictive coding （WELP ）;linear prediction;Low bit-rate speech coding/decoding2|90|2更新时间：2025-12-08
- 摘要：By using modal matching, frequency response of frequency selective surfaces （FSS） with apertures is studied for zero thickness and infinite planar conducting sheet. The excitation is assumed to be a plane wave incident at any arbitrary angle. Based on Floquet theory,the solution is obtained by matching the tangential field components at the surface of the screen. Both single and double-FSS configurations are studied and the effects of multilayered dielectric substrate are taken into account. Then,the typical results of the influence of the lattice,element configurations and incidence angle on the performance of FSS consisting of arrays of cross-loop slots are summarized.关键词：Frequency-selective surfaces;Modal matching;moment method2|453|34更新时间：2025-12-08
- 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