最新刊期
卷 ， 期 5 ， 1991
- 摘要：The tracing subscript method of vectorization with array element as node offers different means of operating upon array elements, such as exchange, jump, drift, del-etion and fill. Compared with various methods of taking statement or array item as operated object, it not only strengthens the ability greatly to recognize vector-based parallelism, but also creates conditions for applying modern math tools to a larger area The comparison decisions about vectorism the paper has proved make it possible for a great number of loops to recognize whether they are vectorizable through simple math process only, thereby abandoning lengthy calculations in constructing statement graphs, or array-item graphs, or sequential hieranchies, with statements, or subscript variables, or array elements as nodes nespectively, This kind of decision method of vector-based parallelism is readily extended to other parallel modes.2|15|0更新时间：2025-12-08
- 摘要：Based on the dealing with the 3-D image reconstruction for seismics data, this paper proposes an algorithm which can not only reconstruct the 3-D surface of objects but also display the interior of them. The introduction of shading, transformation of control grid and appropriate assembly speeded the process of the algorithm. And the results showedgood sense of reality2|37|2更新时间：2025-12-08
- 摘要：In this paper the analyses for the input impedance of the probe-fed microstrip disk antenna with a diaphragm at the coaxial aperture have been made by means of mode matching theory and integral equation method. The theoretical results are in excellent agreement with experimental data. It is shown that the change of the gap capacitance will affect the resonance frequency and the input resistance at the resonance significantly when the dielectric substrate is electrically thick.2|119|0更新时间：2025-12-08
- 摘要：A set of complex source points are used to model an incident plane wave. The complex ray paraxial approximation is employed to calculate the far-zone scattering field from hollow pipes, and then the simulation and optimization of EM scattering from engine intakes are performed numerically. This method is also suitable for the EM simulation of other complicated targets.2|47|4更新时间：2025-12-08
- 摘要：This paper deals with the problem of adaptive displacement estimation in Digital Subtraction Imaging. Its mathematical model is given, in which it is considered as a problem of functional optimization. Based on the idea of finite element method, a family of models for correction of gray level variation is presented. New criterion functions and adaptive hierachical searching algorithms are developed. The experiment results in medical imaging （DSA） are reported.2|26|3更新时间：2025-12-08
- 摘要：The paper has derived an averege ambiguity function for random FM-CW radar signal and random binary-phase-coded CW radar signal. The function has "Thumbtack" forms, i. e. it has good reslution and precision of distance and velocity simultaneously, and good ability of antijams.2|239|26更新时间：2025-12-08
- 摘要：A finite-state vector quantization correlated encoding （FSVQ-CEC） strategy is presented in this paper. The rate-distortion performance of FSVQ-CEC for linear-correlated Gaussian observations is demonstrated, and is compared with the optimum performance theoretically attainable of correlated encoding and the performance by simple encoding which fails to capitalize on the correlation between sources. The FSVQ-CEC scheme is applied to an analysis-synthesis adaptive predictive speech encoding system so as to exploit the redundancy between vocal tract codevectors and excitational codevectors.2|37|1更新时间：2025-12-08
- 摘要：A public-key cryptosystem based on error correcting codes is presented. This public-key cryptosystem can perform signature, encrypting, error detecting and error correcting simultaneously. The security of this cryptosystem is enough to satisfy the user’s needs. The Computation for signature and encrypting is timesaving.2|122|3更新时间：2025-12-08
- 摘要：<正> 毫米波介于微波与光波之间,因此它具有两者的综合优点,如分辨率高,体积小,频谱宽波束窄抗干扰能力强,穿透能力强可以全天候工作等,海湾战争中美军彻底破坏伊拉克电子系统而自己的制导,侦察及通讯系统照样性能优良地工作就从一个侧面反映了毫米波的这些优点,民用方面,毫米波也有广阔的前景,如在移动电话,汽车防撞雷达,医疗等方面的应用,因此有人说,如果微波是四十年代早昼八,九点钟的太阳,那末,毫米波就是现在早晨八、九点钟的太阳,亚毫米波是五,六点钟的太阳。 国外毫米波的发展比我们早走一步,目前已经达到系统应用阶段,他们发展毫米波的经验是,充分借4|39|0更新时间：2025-12-08
- 摘要：Based on linear programming neural networks, a neural network approach to optimization in Hilbert space is proposed in this paper. The optimization problems are solved by setting up the associated linear programming circuit rather than by computation in a digital computer. Theoretical analysis and computer simulations show that the approach can solve the optimization problem within RC time constant （order of several hundred ns）, and has some advantages such as normal and simple structure and tolerance of inaccuraces in the conductance matrix.2|61|1更新时间：2025-12-08
- 摘要：<正> 国家863高技术计划智能计算机领域的重点课题——四达-863A语音文字系统于1991年8月9日在北京通过国家科委高技术司组织的鉴定。 四达-863A语音文字系统是一个面向实用的实时语音打字系统。该系统将音字流自动分词、拼音（近音）汉字转换和语音识别一起构成了较完善的通用语音输入系统。系统的语音识别率、音字转换水平及语音文字系统整体水平均处于国内外领先地位。在50dB噪声环境下,系统单音识别的首选正确率为87.0％,在有三个候选时的正确率为97.6％,在64dB噪声环境下首选正确串为86.9％,在有三个候选时的正确率为93.7％,四声正确率95.0％。对指定文2|26|0更新时间：2025-12-08
- 摘要：This paper proposes a new representation of three dimension objects——mathematical morphology representation. The properties of the representation and its inverse transformation are discussed. The algorithm is also realized.2|74|1更新时间：2025-12-08
- 摘要：This paper presents a fast spatially recursive algorithm for designing 2D adaptive LS filter of the noncausal and space-varying FIR type. The computational quantity of the algorithm is 15m3/2+16m MADPR （multiplications and divisions per recursion）, where m is the number of the filter weighting factors.2|46|0更新时间：2025-12-08
- 摘要：According to the number of left shifts producing interior key in the DES, this paper gives a new method for searching key space. From the view point of searching time, this method lessens the key space from 256 to 254.2|36|1更新时间：2025-12-08
- 摘要：A technique is presented for the measurement of complex permittivity of MIC substrates. The test frequencies range from 1 to 20 GHz. This technique covers determination of εr’ and tan5 of MIC substrates with εr’ in the range from 2 to 25 and with tan5 in the range from 3×10-4 to 1 × 10-2. All kinds of flexible and rigid materials of MIC substrates can be measured. An uncertainty of εr’ is less than ±1%, and that of tanδ is about ±10%. Nondestructive and automated measurements are made. The test circuit is simple, and the test speed is rapid. The technique is particularly suitable for industrial surveying of complex permittivity of MIC substrates.2|97|8更新时间：2025-12-08
- 摘要：This paper proposes an algorithm for Discrete Consine Transform （DCT-Ⅱ） of length p’ （p is an arbitrary positive integer）. The methods for reducing the time complexity are discussed in detail. The algorithm is different from that by converting DCT into DFT and using FFT. Combining with B. G. Lee’s prime-factor-decomposed algorithm, this algorithm can compute DCT-Ⅱ of arbitrary length.2|38|2更新时间：2025-12-08
- 摘要：This paper summarises and discusses the application of knowledge-based techniques to protocol engineering. First, we explain what is knowledge engineering and protocol engineering. Then, we discuss some main problems in the traditional protocol development pattern-life cycle pattern, and give the methods to solve those problems. We propose two kinds of methods for applying knowledge-based techniques to protocol engineering. That is, modification or revolution to life cycle pattern. Through the discussion, we point out: application of knowlede based techniques to protocol engineering will make the experiences of developers reusable and increase the dynamic feedback property of each development step. Consequently, the quality and efficiency of a protocol will be improved and the cost will be decreased. Moreover, this application will facilitate the maintenance of both protocol and development systems.2|84|1更新时间：2025-12-08
- 摘要：The process of a relaxation oscillation can be described by complex circle maps, which reduce to single inverse circle map when the falling time of the process in the oscillation is zero. The order of map is determined by the function form of the modulation signal. Supercritical behaviors of the oscillation ave studied experimentally and numerically. Two scaling laws, namely s（f） ∝f-δ and τ∝|f-fc |-γ are certified, in which for single circle map the exponent γ = 0.5 and does not change with the order of map, but for complex circle maps, there are two values in γ and both the scaling exponents S and γ2 increase when the order of the map is getting larger.2|33|2更新时间：2025-12-08
- 摘要：This paper presents the application of the combination analogy transformer circuits with nullators and norators. Thus, the direct analysis and calculations at different voltage levels can be conducted.2|93|2更新时间：2025-12-08
- 摘要：In this paper, a self consistent sub-donain decomposition method is performed to divide the defining region of process parameter super space, and an experimental design method is used to build the second order regression models of IC device characteristics with process parameters. The examples indicate that using regression model instead of process and device modelling is superior in process optimized design and device performance prediction.2|36|1更新时间：2025-12-08
- 摘要：Amorphous silicon films doped with transition metal Co were prepared by electron beam evaporation. The electrical and optical properties, the impurity effect on the defects in the films and the doping mechanism are studied using temperature dependence of conductivity, ESR technique and ultraviolet absorption. The impurity level lies about 0.13 eV above valence band EV At 480<T<290K, an analysis of conductivity data allows to reveal two conduction mechanisms: 1) at 480<T<330K, σ is due to conduction of holes in valence band. 2) at 330<T<290K, it is due to hoping conduction through an impurity band.2|35|1更新时间：2025-12-08
- 摘要：A new noncausal model which can be implemented recursively is peoposed in this paper to overcome the disadvantage of being unable to be implemented recursively. The corresponding estimation algorithm is also suggested and is an asymptotic method. Many experimental results confirmed the validity of our model. It can replace the common noncausal model to do image feature extraction, data co pression, etc.2|59|1更新时间：2025-12-08
- 摘要：SVD based spectrum estimation is of good performance, but the computational burden it heavy. This paper introduces a fast matrix eigenvalue decomposition algorithm which is several times faster than conventional one. The number of computation for SVO will be reduced if it is applied to forward-backward linear-prediction spectrum estimator.2|74|1更新时间：2025-12-08
- 摘要：A method of minimum mean square deconvolution is proposed and the mean square error of its solution deviating from the true solution is minimum. Therefore the effects of error data are effectively restrained.2|85|2更新时间：2025-12-08
- 摘要：Formulas used directly in calculating Schwarz functions by power series and of the retio linking accessory parameters of the functions are developed, with their applications being discussed. Finally, compact formulas concerning the transformation of regular polygons are given.2|39|0更新时间：2025-12-08
- 摘要：<正> 机电部第十二研究所是机械电子工业部直属的以真空、微波、激光电子技术为主的大型综合性研究所。 第十二研究所筹建于1956年,多年来主要从事真空微波器件系列产品的研究,近年来随着改革开放不断深入,以及真空、微波、激光电子技术的不断发展,其研究和应用范围不断扩大。目前,该所拥有2400余名职工,有1000余名中、高级工程技术人员及大批熟练的技术工人。近几年来数百名研究生,大中专毕业生来所工作,充实了所的各级科研队伍,具有综合的技术优势和人才优势。2|22|0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