Browsing by Issue Date, starting with "1995-01-13"

Now showing 1 - 20 of 87104

ELEMENT PERTURBATION PROBLEMS OF OPTIMUM SPANNING-TREES WITH 2-PARAMETER OBJECTIVES
(1995-01-13)
CHANG, YC
;
HSU, LH
;
資訊工程學系
;
Department of Computer Science
Let G = (V,E) be a graph. We associate with each edge e(i) is an element of E an ordered pair of rational numbers (a(i), b(i)). Let the weight of a spanning tree T, w(T), be defined as Sigma(ei is an element of T) a(i) + Pi(ei is an element of T) b(i). A spanning tree T in G is called a w-optimum spanning tree if w(T) greater than or equal to w(T') for all spanning trees T' in G. The function w is one instance in a class of two-parameter objectives. Hassin and Tamir proposed a unified approach for solving the class of two-parameter objective optimum spanning tree problems. Let s be an objective in the class and F-s(G) denote the weight of the s-optimum spanning tree of G. The element perturbation problem of the s-optimum spanning tree is to compute F-s(G - e(k)) for all e(k) is an element of E. With Hassin and Tamir's approach, let t(s)(p, q) be the complexity of computing the s-optimum spanning tree where p = V and q = E. In this paper, we present an approach to solve the element perturbation problem of the s-optimum spanning tree in t(s)(p, q).
STARTING DYNAMICS OF A CW PASSIVELY MODE-LOCKED PICOSECOND TI SAPPHIRE DDI LASER
(1995-01-15)
PU, NW
;
SHIEH, JM
;
LAI, YC
;
PAN, CL
;
交大名義發表
We show that, for a cw passively mode-locked picosecond Ti:sapphire/DDI laser, the first autocorrelation trace with negligible cw background occurs at a delay time of 20 mus, or 1600 round trips from the first relaxation-oscillation peak. The trace suggests that the pulse consists of a primary pulse as short as 4.4 ps and of small secondary pulses that form a much wider pedestal of the trace, each containing almost-equal-to 50% of the photon energy. Nearly transform-limited almost-equal-to5-ps-wide Gaussian pulses were observed at a delay time of 40 mus. After 45 mus, the optical spectrum broadened considerably, and the time-bandwidth product increased to 4 in the steady state (after 60 mus).
OPTIMAL PARAMERERS FOR CURING GRAPHITE-EPOXY COMPOSITE LAMINATES
(1995-01-15)
KAM, TY
;
LAI, FM
;
SHER, HF
;
交大名義發表
;
機械工程學系
The optimal curing parameters of a cure cycle for achieving maximum mechanical properties of Gr/ep composite laminates are determined via a multi-level optimization approach. Laminated composite specimens cured under various pressures, temperatures and times are tested with an Instron testing machine. Test results: are then used in the proposed multi-level optimization method to determine the optimal curing parameters. The effects of the curing parameters on the mechanical properties of the laminates are studied and discussed It was found that a cure cycle with two pressure-temperature-time steps can yield composite laminates with comparatively good material properties.
A NEW TECHNIQUE FOR REDUCING THE RESIDUAL-STRESS INDUCED BY WELDING IN TYPE-304 STAINLESS-STEEL
(1995-01-15)
LIN, YC
;
CHOU, CP
;
機械工程學系
;
Department of Mechanical Engineering
A new Technique named as ''parallel heat welding'' (PHW) is developed for the reduction of residual stress. It is performed by a pair of parallel heating torches attached to the welding torch on both sides as a movable heat source during welding. Autogenous gas tungsten are welding process was used in type 304 stainless steel. The experimental results showed that the maximum principal residual stress and parallel welding direction stress can be reduced by 21-32 % when the conventional welding(CW) process is replaced by the parallel heat welding process. The elevation of equilibrium temperature during welding process is a main mechanism for the reduction of residual stress in PHW process.
IMPROVEMENT OF SELECTIVITY DURING DIAMOND GROWTH UTILIZING A NEW PROCESS
(1995-01-15)
CHEN, CF
;
CHEN, SH
;
HONG, TM
;
TSAI, MH
;
交大名義發表
FUZZY SYSTEM RELIABILITY-ANALYSIS USING FUZZY NUMBER ARITHMETIC OPERATIONS (VOL 64, PG 31, 1994)
(1995-01-27)
CHEN, SM
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
;
Department of Computer Science
DESIGN OF EFFICIENT REGULAR ARRAYS FOR MATRIX MULTIPLICATION BY 2-STEP REGULARIZATION
(1995-02-01)
TSAY, JC
;
CHANG, PY
;
交大名義發表
;
工學院
;
National Chiao Tung University
A two-step regularization method in which first permutation sequences and then broadcast planes are selected is proposed to design various regular iterative algorithms for matrix multiplication. The regular iterative algorithms are then spacetime mapped to regular arrays, such as mesh, cylindrical, two-layered mesh, and orbital arrays. The proposed method can be used to design regular arrays with execution time of less than N (problem size).
ADAPTIVE DECISION-FEEDBACK EQUALIZATION FOR DIGITAL SATELLITE CHANNELS USING MULTILAYER NEURAL NETWORKS
(1995-02-01)
CHANG, PR
;
WANG, BC
;
交大名義發表
;
電信工程研究所
;
National Chiao Tung University
This paper introduces an adaptive decision feedback equalization using the multilayer perceptron structure of an Mary PSK signal through a TDMA satellite radio channel. The transmission is disturbed not only by intersymbol interference (ISI) and additive white Gaussian noise, but also by the nonlinearity of transmitter amplifiers. The conventional decision feedback equalizer (DFE) is not well-suited to detect the transmitted sequence, whereas the neural-based DFE is able to take into account the nonlinearities and therefore to detect the signal much better. Nevertheless, the applications of the traditional multilayer neural networks have been limited to real-valued signals. To overcome this difficulty, a neural-based DFE is proposed to deal with the complex PSK signal over the complex-valued nonlinear MPSK satellite channel without performing time consuming complex-valued back-propagation training algorithms, while maintaining almost the same computational complexity as the original real-valued training algorithm. Moreover, a modified backpropagation algorithm with better convergence properties is derived on the basis of delta-bar-delta rule. Simulation results for the equalization of QPSK satellite channels show that the neural-based DFE provides a superior bit error rate performance relative to the conventional mean square DFE, especially in poor signal-to-noise ratio conditions.
A FIELD-STUDY OF 3 COLLOCATED AMBIENT PM10 SAMPLERS
(1995-02-01)
TSAI, CJ
;
交大名義發表
;
環境工程研究所
;
National Chiao Tung University
;
Institute of Environmental Engineering
The performances of three ambient PM10 samplers were studied at three monitoring stations in Taiwan. It was found that differences in the daily measured PM10 concentrations of the SA 1200 and Wedding high-volume samplers are now within +/- 10% since the former now has a closer cut-point to the latter than in the earlier SA 321 A model. The Wedding beta gauge automatic sampler was found to be applicable in rainy and humid weather conditions in Taiwan. Its daily PM10 concentrations are typically within +/- 10% of those of the Wedding high-volume sampler. The particle loading effect of Wedding high-volume and beta gauge samplers was found to be important. To avoid sampling errors due to the loading effect with ambient PM10 samplers, they must be cleaned regularly at an interval depending on the ambient particulate level.
ISOLATED MANDARINE SYLLABLE RECOGNITION USING SEGMENTAL FEATURES
(1995-02-01)
CHANG, S
;
CHEN, SH
;
電信工程研究所
;
電信研究中心
;
Institute of Communications Engineering
A segment-based speech recognition scheme is proposed. The basic idea is to model explicitly the correlation among successive frames of speech signals by using features representing contours of spectral parameters. The speech signal of an utterance is regarded as a template formed by directly concatenating a sequence of acoustic segments. Each constituent acoustic segment is of variable length in nature and represented by a fixed dimensional feature vector formed by coefficients of discrete orthonormal polynomial expansions for approximating its spectral parameter contours. In the training, an automatic algorithm is proposed to generate several segment-based reference templates for each syllable class. In the testing, a frame-based dynamic programming procedure is employed to calculate the matching score of comparing the test utterance with each reference template. Performance of the proposed scheme was examined by simulations on multispeaker speech recognition for 408 highly confusing isolated Mandarin base-syllables. A recognition rate of 81.1% was achieved for the case using 5-segment, 8-reference template models with cepstral and delta-cepstral coefficients as recognition features. It is 4.5% higher than that of a well-modelled 12-state, 5-mixture CHMM method using cepstral, delta cepstral, and delta-delta cepstral coefficients.
SUPPRESSION OF BORON PENETRATION IN PMOS BY USING BRIDE GETTERING EFFECT IN POLY-SI GATE
(1995-02-01)
LIN, YH
;
LEE, CL
;
LEI, TF
;
CHAO, TS
;
電子工程學系及電子研究所
A new stacked poly-Si gate structure with a thin (similar to 20 Angstrom) oxide inserted in-between for p-channel metal-oxide semiconductor feild effect transistor (pMOSFET) application is proposed and demonstrated. Due to the gettering flourine effect of this thin oxide for BF2+-implanted poly-Si gate, the amount of fluorine in the gate oxide, consequently, the boron penetration enhanced by fluorine are reduced. The thicker or the more apart from the gate oxide of this gettering fluorine oxide is, the more effectively it can suppress the boron penetration. Moreover, this new structure improves more the electrical characteristics than the conventional and the stacked layer poly silicon gate structures.
MODELING NON-UNIFORM-SEDIMENT FLUVIAL PROCESS BY CHARACTERISTICS METHOD
(1995-02-01)
YEH, KC
;
LI, SJ
;
CHEN, WL
;
土木工程學系
;
Department of Civil Engineering
This paper presents a numerical model based on a multimode characteristics method for fully coupled simulation of water and sediment movement in mobile-bed alluvial channels with nonuniform bed materials. Along with the characteristics, the temporal evolutions of flow depth and velocity, bed elevation, and bed-material composition of a well-defined river reach with proper boundary conditions can be simultaneously determined. From the results of the test runs, the proposed model provides useful information about the interactions among disturbance waves of different orders of magnitude and their influence on the formation of flow field, bed topography, and bed-material composition. Such information cannot be easily obtained from finite-difference methods.
VERY-LOW TEMPERATURE DEPOSITION OF POLYCRYSTALLINE SI FILMS FABRICATED BY HYDROGEN DILUTION WITH ELECTRON-CYCLOTRON-RESONANCE CHEMICAL-VAPOR-DEPOSITION
(1995-02-01)
WANG, KC
;
CHENG, KL
;
JIANG, YL
;
YEW, TR
;
HWANG, HL
Characteristics of polycrystalline silicon films deposited both on SiO2 and Coming 7059 glass substrates are presented in this paper. The silicon films were deposited by a hydrogen dilution method using electron cyclotron resonance chemical vapor deposition at 250 degrees C without any thermal or laser annealing. The hydrogen dilution ratio was between 90% and 99%. The geometric configuration and surface morphology of polycrystalline silicon films were studied by atomic force microscopy. The largest grain size of the deposited silicon films; identified by plan-view transmission electron microscopy dark-field imaging, was about 1 mu m. From Raman spectrum, the crystalline fraction of polycrystalline silicon films was identified to be nearly 100%. The polycrystalline silicon was found to be preferentially [111]- and [110]-oriented, from the X-ray diffraction pattern.
RELIABILITY OPTIMIZATION OF DISTRIBUTED COMPUTING SYSTEMS SUBJECT TO CAPACITY CONSTRAINTS
(1995-02-01)
CHEN, RS
;
CHEN, DJ
;
YEH, YS
;
交大名義發表
;
資訊科學與工程研究所
In this paper, we propose a simple, easily programmed exact method for obtaining the optimal design of a distributed computing system in terms of maximizing reliability subject to memory capacity constraints. We assume that a given amount of resources are available for linking the distributed computing system. The method is based on the partial order relation. To speed up the procedure, some rules are proposed to indicate conditions under which certain vectors in the numerical ordering that do not satisfy the capacity constraints can be skipped over. Simulation results show that the proposed algorithm requires less time and space than exhaustive method.
CONFERENCE KEY DISTRIBUTION SCHEMES FOR SECURE DIGITAL MOBILE COMMUNICATIONS
(1995-02-01)
HWANG, MS
;
YANG, WP
;
資訊工程學系
;
Department of Computer Science
In this paper, we propose a new service for digital mobile communication systems, The service enables two or more users to hold a secure conference, Two requirements must be considered: privacy and authentication. Privacy involves ensuring that an eavesdropper cannot intercept the conversations of the parties holding the conference. Authentication involves ensuring that service is not obtained fraudulently in order to avoid usage charges. We present two new conference key distribution schemes for digital mobile communication systems. In these schemes, a group of users can generate a common secret key over a public channel so that they may hold a secure conference.
2-DIMENSIONAL BI-PERIODIC GRATING-COUPLED ONE-COLOR AND 2-COLOR QUANTUM-WELL INFRARED PHOTODETECTORS
(1995-02-01)
TSAI, KL
;
LEE, CP
;
TSANG, JS
;
CHEN, HR
;
CHANG, KH
Multiple quantum well infrared photodetectors (QWIP's) with two-dimensional bi-periodic grating couplers have been demonstrated. For one-color QWIP's, the response linewidths of such QWIP's are found to be much wider than those of conventional grating coupled QWIP's. The bi-periodic gratings have also been applied to two color QWIP's to separately optimize the absorption at two wavelength bands. The large difference in responsivities at the two wavelength bands for devices with single period gratings can be greatly reduced with the use of bi-periodic gratings.
PARAMAGNETIC MEISSNER EFFECT OF HIGH-TEMPERATURE GRANULAR SUPERCONDUCTORS - INTERPRETATION BY ANISOTROPIC AND ISOTROPIC MODELS
(1995-02-01)
CHEN, FH
;
HORNG, WC
;
HSU, HT
;
TSENG, TY
;
電子工程學系及電子研究所
The field-cooled magnetization of high-T-c superconducting ceramics measured in low magnetic fields exhibits the paramagnetic Meissner effect (PME), i.e., the diamagnetic signal initially increases with decrease in temperature but reaches a maximum at temperature T-d and later decreases with decrease in temperature. Even in some samples the signal is ultimately able to transform inversely into a paramagnetic regime once the sample is cooled below a temperature T-p as long as the applied field is sufficiently small. This PME has been observed in various high-T-c cuprates and is explained by disparate aspects. An anisotropic model, in which the granular superconductors are assumed to be ideally anisotropic, was first alternatively proposed in the present work so as to theoretically account for this effect. On the other hand, an isotropic model, suitable for granular superconductors with randomly oriented grains, was proposed to deal with the samples prepared by a conventional solid-state reaction method. The anomalous magnetization behavior in the present model was demonstrated to be the superposition of the diamagnetic signal, which occurs as a result of the intragranular shielding currents, over the paramagnetic one due to the induction of the intergranular component induced by these currents where the intergranular one behaved as the efective pinning centers. The PME was demonstrated by this model to exist parasitically in granular superconductors. This intergranular effect is therefore worthy of remark when evaluating the volume fraction of superconductivity for the samples from the Meissner signal, in particular, at a low magnetic field.
TRANSIENT SIMULATION OF EPROM WRITING CHARACTERISTICS, WITH A NOVEL HOT-ELECTRON INJECTION MODEL
(1995-02-01)
HUANG, CM
;
WANG, TH
;
交大名義發表
;
電子工程學系及電子研究所
;
National Chiao Tung University
A two-dimensional transient simulation of EPROM writing characteristics is presented. A Monte Carlo-based hot electron injection model which accounts for Fowler-Nordheim tunneling and thermionic emission has been included in the simulation. The simulated EPROM writing transient characteristics is compared favorably with experimental results for channel lengths down to 0.5 mu m. The importance of the two injection mechanisms, thermionic emission and quantum tunneling, is evaluated.
THE THERMAL-STABILITY OF OHMIC CONTACT TO N-TYPE INGAAS LAYER
(1995-02-01)
WU, JW
;
CHANG, CY
;
LIN, KC
;
CHANG, EY
;
CHEN, JS
The thermal stability of ohmic contact to n-type InGaAs layer is investigated. When Ni/Ge/Au is used as the contact metal, the characteristics of the ohmic contact are degraded after thermal treatment. The specific contact resistance of (Ni/Ge/Au)-InGaAs ohmic contact after annealing at 450 degrees C is about 15 times larger than that of as-deposited sample. This is due to the decomposition of InGaAs and the interdiffusion of Ga and Au.-A new phase of Au4In appears after annealing at 300 degrees C. While in the case of Ti/Pt/Au, Au does not penetrate into the InGaAs layer as revealed by secondary ion mass spectroscopy. The specific contact resistance of(Ti/Pt/Au)-InGaAs ohmic contact after annealing at 450 degrees C is eight times larger than that of as-deposited sample. Therefore, the thermal stability of(Ti/Pt/Au)-InGaAs ohmic contact is better than that of(Ni/Ge/Au)-InGaAs ohmic contact.
A FAST SEARCH ALGORITHM FOR VECTOR QUANTIZATION USING MEAN PYRAMIDS OF CODEWORDS
(1995-02-01)
LEE, CH
;
CHEN, LH
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
One of the most serious problems for vector quantization, especially for high dimensional vectors, is the high computational complexity of searching for the closest codeword in the codebook design and encoding phases. Although quantizing high dimensional vectors rather than low dimensional vectors results in better performance, the computation time needed for vector quantization grows exponentially with the vector dimension. This makes high dimensional vectors unsuitable for vector quantization. To overcome this problem, a fast search algorithm, under the assumption that the distortion is measured by the squared Euclidean distance, will be proposed. Using the mean pyramids of codewords, the algorithm can reject many codewords that are impossible matches and hence save a great deal of computation time. The algorithm is efficient for high dimensional codeword searches. Experimental results confirm the effectiveness of the proposed method.