最新刊期
卷 ， 期 10 ， 1996
- 摘要：In this paper, the relationships between IFS fractal-based source coding techniques and traditional approaches are studied systematically for the first time. Connections between arithmetic coding, LPC and IFS are presented through theoretical analysis. Experimental comparison of IFS compression and transform coding,which tests their difference, gives evidence to the suitability of each to act as a generic signal compresssion approach. The superiority that is unique to fractal based-algorithms is reported. Some important problems for the further improvement to IFS signal compression are also analyzed.关键词：source coding;Iterated function systems;fractal2|134|29更新时间：2025-12-08
- 摘要：The proper vertex-coloring, edge-coloring and total-coloring of a graph are systematically analysed with a new construction of the energy funtion for Hopfield nueral network ,the corresponding coloring alogrithms are introduced and the computer simulation as well as the coloring results are presented.关键词：Hopfield nueral network;Graphic vertex-coloring;Graphic edge-coloring;Graphic total-coloring;Coloring algorithm2|196|22更新时间：2025-12-08
- 摘要：This paper presents the new methods of suppressing CN-sequence so as to detect the subpixel target, point target and a few pixel target. First, the methods of accurately estimating the shift parameters between the successive frames are presented, and the equations of calculating 2-D subpixel shift parameters, dx and dy, are deduced. Then, to solve interpolated image, the equations of the FFT interpolation method in frequency field are emphatically deduced. Also, the interpolation control parameters are defined, and the method of interpolating and selecting points（MOISP） is presented. Therefore, a set of algorithms to suppress CN-sequence is set up. The algorithms have been crried out and the CN reJection rates are all more than 85 percent for the selected real-world infrared and television image sequences. The main components of surplus image are the additive uncorrelated noise, i. e. UCN-sequence, except a few target component.关键词：CN sequence;Subpixel shift parameters;The interpolated image;The method of interpolating and selecting points;The suppressed power of CN2|52|7更新时间：2025-12-08
-
Generalized Capacitance Matrix Method and Its Applications──Design of Combline and Interdigit Filter
摘要：A method of generalized capacitance matrix is presented in this paper, which can be used to calculate TEM transmission line comb-line filter and interdigital filter. The concepts of modal capacitance, fundamental and high order mode, mutual capcitance and load capacitance are introduced in the generalized capacitance matrix. So a complex structure can be solved by cascading of these matrices. We can get high flexibility of the optimal design, and the computing speed and accuracy are improved greatly.关键词：Generalized capacitance matrix;Modal capacitance;boundary element method;filter;optimal design2|217|6更新时间：2025-12-08 - 摘要：In this study on Genetic Algorithms（GAs）, Symmetric Code theory is developed.We define two kinds of Symmetric Codes, Symmetric Code Strings and their operative rules.Then the propertis of Symmetric Codes are given out. From the Symmetric Code theory, our Genetic Algorithm with Symmetric Codes（GASC） is proposed. In GASC, the following special techniques are used for the first time: Symmetric Codes, Good Members Protection, Hopeful Members Immigration and Partial Genetic Reservation etc. This genetic technigue can be applied to some system control,such as robot system. The simulation on its applications shows that Genetic Algorithm with Symmetric Codes can much overperforms general genetic algorithms and that Coding technique has important influence on genetic algorithms[7,8]. But this problem has been ignored before. Considered as a general-purpose search, optimization technique, GASC can also find its application in the areas of machine learning, neural network learning and problem resolution.关键词：genetic algorithms;Horizontal symmetric code string;Vertical symmetric code string;Partial genetic reservation2|93|37更新时间：2025-12-08
- 摘要：Projective invariants for five coplanar lines have found many important applications in computer vision. Since there are many invariants for five coplanar lines and these invariants are relative to order of the lines, it is very difficult process to match the invariants from diffierent views. In this paper, we first derive the relations among all the projective invariants for five coplanar lines by using cross-ratio; then, we divide these projective invariants into five classes, any two of which being independent of each other; upon this, we derive two joint projective invariants for five coplanar lines which are independent of each other and irrelative to the order of the lines.关键词：Projective invariants;cross ratio;Perspectively invariant recognition2|113|3更新时间：2025-12-08
- 摘要：In this paper, the least-square design problem of 2-D FIR digital filters with arbitrary frequency responses is studied. A closed-form solution for filter coefficients is obtained by minimizing the frequency-domain squared-error function and revealing some properties of the matrices related to the design problem. By using the derived closed-form expression, the impulse response of the designed filter can conveniently be evaluated from the specified frequency response,resulting in a significantly simplified design procedure and an extremely reduced computational burden.关键词：digital filters;Least-square design;2-D FIR filters2|96|6更新时间：2025-12-08
- 摘要：Fuzzy control is similar to artificial neural network control on many aspects. It is very important to exert their advantages. This paper uses the high-speed and concise properties of fuzzy control to make up the low-speed and complexities of artificial neural network. On the other hand, neural activation function makes it easy to deal with fuzzy control. Based on these advantages,a new fuzzy artificial neural network controller （FANNC） is designed. Its reasonableness is verified by simulation. The simulation results of FANNC are superior to literature[8]. Also, this paper shows that fuzzy control is a kind of special artificial neural network, they can be trained in the same way. It is very valuable for united dealing with both of them.关键词：fuzzy control;Fuzzy language;membership;Neural activation function;self-organizing;BP algorithm;artificial neural network2|130|2更新时间：2025-12-08
- 摘要：The modern cross-spectral estimation is an effective method to restrain colored noises. In this paper, the singular-value decomposition natures of an extended-order cross-correlation matrix associated with sinusoids are studied. With the results of the study, the Pisarenko and MUSIC methods for the cross-spectral estimation of the sinusoidal signal embedded in colored noise are proposed. The mainly advantages of the methods are that they don’t need any priori information about the colored noises and can work in very low SNR, which make the methods applicable in engineering. Finally, we compare the proposed methods to the Cadzow’s SVD method, used for the frequencies estimation of the sinusoidal signal in strong colored noise.关键词：colored noise;Cross-spectral estimation;Singular decomposition.2|756|81更新时间：2025-12-08
- 摘要：Multilayered neural networks are used to construct nonlinear learning control systems for a class of unknown nonlinear systems in a canonical form. An adaptive output tracking architecture is proposed using the three-layered neural networks which are trained by back-propagation algorithm. The closed-loop system is proved to be stable, with the output tracking error converging to the δ-neighbourhood of the origin.关键词：Multilayered feedforward neural network;BP algorithm;Output tracking control;Closed-loop stability2|102|3更新时间：2025-12-08
- 摘要：This paper proposes that the robust estimation method[1-4] be applied to a adaptive signal processing system. The theoritical analysis and simulation results show that the tolerance to the noise of the adaptive signal processing system is improved by the use of the robust estimation method.关键词：adaptive signal processing;Estimation theory;robust estimator2|126|11更新时间：2025-12-08
- 摘要：In view of objectively-existing forward and backward contextual dependent information in speech signal, in this paper we show that a conditional probability can be used to explicitly express these information in speech signal, which is called Forward and Backward Markov Contextual Dependences. And the Forward and Backward Hidden Markov Models, which are believed to contain these two Markov Contextual Dependences respectively, are also presented here. Our experimental results prove that we can obtain the relatively acceptable performance in speech recognition only using the Backward Markov Contextual Dependence. It is at least comparable with the recognition performance obtained from normal forward HMM with Forward Markov Contextual Dependence. In this paper, we also study the methods of simutaneously taking advantages of these two Markov Contextual Dependences of speech in isolated word recognition and continuous speech recognition respectively, and propose an effective algorithm in continuous speech recognition-Forward and Backward Mixed Bisected Search（FBMBS）, which can utilize the immediate results of Forward HMM based forward Viterbi search and Backward HMM based backward Viterbi search,so that we can conveniently combine these two Markov Contextual Dependences in continuous speech recognition at the same time. Our experiments show that the FBMBS algorithm really is consistently superior to that of using either of these two Contextual Dependences alone.关键词：speech recognition;continuous speech recognition;HMM model2|235|4更新时间：2025-12-08
- 摘要：In this paper, a method of estimating an upper bound of the entropy of printed Chinese is presented. A bound of 5. 17bits/character for the entropy is obtained by computing the entropy of the sample of Chinese corpus. The perplexity of several language models, which is a quantitative measurement for the ability of language models, is discussed. A new method of approximating high scale language model by the lower ones is also presented.关键词：entropy;perplexity;Stochastic Language Model3|1871|41更新时间：2025-12-08
- 摘要：The motivation of this study is to explore the ability of human visual phase discrimination for stereo vision. A very important issue is to understand how much information is carried by the phase information in human stereo interpretation, hence helping for creating new models and approaches which are efficient for interpreting natural scence. Based on the visual cue, we have designed a new approach of stereo vision utilized the wavelet theory. Unlike the area-based that is computationally expensive, and also the feature-based which generates only sparse disparity map, the approach uses the image phase as the matching primitives which can generate a dense disparity field efficiently. In order to deal with complicated natural scenes, the constraints controlling the surface discontinuity and occlusion are also introduced in our approach. Experiments have proved that the proposed approach can be adaptable for almost arbitrary images and resulting disparity surface is accurate and dense. From this study we could probably understand what role of phase information is in human stereo fusion and help developing appropriate computational approach for practices.关键词：Stereo matching;Phase discrimination;Wavelet theory;Visual mechanism2|213|40更新时间：2025-12-08
- 摘要：An experimental system of the digital HDTV source compression codec at bitrates less than 22Mb/s is presented. It is based on the MPEG-2 standard. The emphasis is put on the adaptive frame/field Discrete Cosine Transformation（DCT）, frame/field scanning, the design of quantizer, the controlling strategy of bit-rate and buffer and the layer structure of bit-stream. A bit-stream completely consistent with MPEG-2 standard is produced. The subjective and objective assesments of restored pictures provide satisfactory results.关键词：HDTV;MPEG-2;Adaptive frame/field DCT;Non-normal zig-zag scanning;Strategy of bit-rate and buffer control2|50|1更新时间：2025-12-08
- 摘要：This paper proposes a new structure of three dimensional（i. e. 3-D） filter applied to Luminance-Chrominance separation（i. e. Y-C separation as well known） in PAL TV signals.The 3-D filter consists of a 1-D horizontal filter and a 2-D vertical-temporal filter, where the 2-D filter can be decomposed into two 1-D filters along two new axis directions, which are oblained from suitable rotation transform from the old axis. It would be very useful for simplifying the design and implementation of 2-D filter. The new method behaves better in Y·C separation efficiency than the traditional ones. Moreover, it can avoid the trouble of motion-detection in the adaptive space filtering[1].关键词：Y·C Seperation;TV signal processing;Colour TV;video processing2|65|3更新时间：2025-12-08
- 摘要：An approach to fuzzy associative memory（FAM） system designing using genetic algorithm-GenFAM is presented. GenFAM requires no training data which are often used in adaptive FAM and Neural Network methods of FAM designing, and can get well designed FAM system. Finally, the application of this method is illustrated with an example.关键词：Genetic algorithm;Fuzzy associative memory;Gen FAM2|64|9更新时间：2025-12-08
- 摘要：In this paper, a composite state-space model is proposed for representing two-dimensional damped/undamped harmonics. Based on this, parameter estimations of two-dimensional harmonics are obtained by identifying parameters of the model. The new estimating method decomposes the original two-dimensional problem into several one-dimensional problems, but no pairing algorithm is requrired to yield two-dimensional parameters. Simulations are provided to show the efficiency of this method.关键词：multi-dimensional signal processing;Two-dimensional harmonics;parameter estimation;State space representation.2|69|0更新时间：2025-12-08
- 摘要：A ccording to the projective method by which Pless constructed（24,12,8） Golay code and （12, 6,6） Golay code from quaternary（6,3,4）linear block and ternary（4,2,3） linear code respectively, the binary（4n,n+2k,≥min（8, n,2d））linear code is constructed from quaternary（n,k,d）linear block code, and the ternary（3n,n+k,≥min（n,2d, 6））linear block code is constructed from ternary（n,k,d）linear code. The efficient maximum likelihood decoding algorithms are presented correspondingly for the new codes.关键词：Pless structure;Golay code;maximum likelihood decoding;Quaternary linear block code;Ternary linear block code2|99|1更新时间：2025-12-08
- 摘要：The error performance of continuous phase modulation （CPM）with optimal coherent detection depends on its minimum squared Euclidean distance（MSED）. In this paper,a reduced-state trellis for M-ary CPFSK is presented under the condition that the MSED is not reduced therefore the complexity of Viterbi decoding is reduced.关键词：continuous phase modulation;Optimal coherent detection;Viterbi decoding2|111|2更新时间：2025-12-08
- 摘要：Depth image of an object can be obtained directly with optical coherence tomography（OCT）. In this paper, a theoretical study on depth image information of the tissue structures by using OCT system was reported. Based on this study, the coherent transfer function（CTF）is derived and the depth-discrimination property is analyzed. In this system, structures with high spatial frequency are detected for image information, and low frequency in the image are contributed by nonlinear processing. This research will provide theoretical basis in quantiatively understanding the OCT measurements of an unknown structure.关键词：Optical coherence tomography;Coherent transfer function;Image formation2|228|29更新时间：2025-12-08
- 摘要：The security of MSR+DH scheme in [1] is analysed thoroughly, which is found to be insecure against the active attack of the operators in the base station to impersonate the legal users. A new authentication and key distribution scheme is designed using Rabin public-key cryptosystem,which is superior to the MSR+DH scheme in both security and complexity.关键词：Mobile communication network;authentication;key distribution;cryptanalysis2|122|20更新时间：2025-12-08
- 摘要：The conventional Hough transform cannot detect line contiguity and length of line segment. In this paper, we present an improved Hough transform algorithm which can compute the begining, end and length of the line segments. The algorithm can be parallelized and implemented in PRAM EREW computing model and in mesh connected processor array. Parallel algorithm for identifying and eliminating overlapping line segments is also presented.关键词：Hough transform;Straight line detection;parallel computing2|219|22更新时间：2025-12-08
- 摘要：In this paper, the decomposition method of circular class and equal modulus class is presented. The problem of statistical analysis of carrier phase estimation in M-QAM system is discussed. It is proved that the variance of carrier phase estimate is a direct ratio of square of carrier regular frequency difference and is dependent on the selfnoise besides the additive noise. Meanwhile,the result of this paper simplifies the analytical model of selfnoise.关键词：All digital receiver;Carrier phase recovery;Circular class;Equal modeulus Class2|106|2更新时间：2025-12-08
-
- 摘要：In this paper, the problem of optimal design of interpolation filter for all digital receiver is discussed. The measure formula of intersymbol interference is derived by using the cut off form of interpolation formula and its convergence analysis has been done. By using the measure criterion of intersymbol interference, the parameter optimization and robustness of timing jitter of interpolation filter for a few kinds of shaping functions has been studied.关键词：All digital receiver;Interpolation filter;Shaping filter;Timing jitter2|391|16更新时间：2025-12-08
- 摘要：插值处理与快速傅立叶变换相结合的时域频域网络参数转换技术赵劲松，高葆新，刘润生（清华大学电子工程系，国家教委博士学科点专项科研基金页目）在多导体传输线时域分析中为考虑色散、高次模等效应，需进行时域频域网络参数互换及其转换技术。快速傅立叶变换技术是最常...2|157|2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