Browsing by Issue Date, starting with "1994-08-01"

Now showing 1 - 20 of 88199

A NOVEL SYNTHESIS OF ORGANOIMIDO COMPLEX BY TRANSAMINATION REACTION - X-RAY CRYSTAL-STRUCTURE OF BIS-(ISOPROPYLIMIDO) COMPLEX OF TUNGSTEN(VI), [W(MU-NPRI)(NPRI)(NHPRI)2]2
(1994-08-01)
CHIU, HT
;
CHUANG, SH
;
LEE, GH
;
PENG, SM
;
應用化學系
The bis(isopropylimido) complex [W(mu-NPri)(NPri) (NHPri)(2)](2) was synthesized by reacting W-2(NBu(i))(2)(NHBu(i))(2) with excess (PrNH2)-N-i. The complex, characterized by X-ray crystallography, is a dimer of distorted trigonal bipyramids sharing an edge defined by the unsymmetrical bridging imido ligands. Exchange between the bridging and the terminal imido ligands were observed on the NMR timescale, Delta G double dagger = 14.5+/-1.0 kcal mol(-1). An equilibration process is proposed to account for the fluxional behaviour.
HETEROJUNCTION BIPOLAR-TRANSISTORS WITH EMITTER BARRIER LOWERED BY DELTA-DOPING
(1994-08-01)
CHEN, HR
;
HUANG, CH
;
CHANG, CY
;
LEE, CP
;
TSAI, KL
An Npn heterojunction bipolar transistor with a Si-delta-doped layer at the emitter-base hetero-interface is demonstrated. The Si delta-doped layer reduces the potential spike at the emitter junction. This design reserves the abruptness of the heterojunction, reduces electron barrier and increases the hole barrier simultaneously. Experimental results show that the HBT's turn-on characteristics are greatly improved while the current gain remains high. The offset voltages as low as 44 mV have been obtained. Very high current gains at very low collector current densities are obtained.
BUDGET MANAGEMENT OF NETWORK CAPACITY PLANNING BY SEARCHING CONSTRAINED RANGE AND DOMINANT SET
(1994-08-01)
YANG, CY
;
LIN, RY
;
交大名義發表
;
管理學院
;
National Chiao Tung University
A communication network's reliability, survivability, and interconnectivity are primarily based on the degree of interconnection between the existing nodes of the network. Enhancement of these characteristics can be obtained by adding direct communication links between nodes of the network. This process is generally subject to a budget constraint. From the service provider's perspective, enhancing the interconnectivity of heterogeneous networks is part of operations evolution. However, the interconnectivity or link enhancement problem, for a given budget, is NP-complete. Decisions by considering multiple criteria improve previous work and may search only the constrained range. The constrained range is determined by a dominant set of multiple criteria. A review of pertinent previous work, problem formulation, algorithm presentation, and discussion of improving the computation time with compromising the optimality by using the multiple-criteria constrained range are also provided.
MULTISTAGE PRE-CANDIDATE SELECTION IN HANDWRITTEN CHINESE CHARACTER-RECOGNITION SYSTEMS
(1994-08-01)
TUNG, CH
;
LEE, HJ
;
TSAI, JY
;
資訊工程學系
;
Department of Computer Science
This paper presents a model for reducing the processing time needed for recognition of handwritten Chinese characters via multi-stage pre-candidate selection. The features used in pre-candidate selection are ordered according to the reduction rates evaluated from a set of training characters. A multi-stage pre-candidate selection module is activated for each input character. In each stage, we use a single feature computed from the input character to eliminate impossible categories of characters. When the number of stages increases, the pre-candidate selection time will increase, while the candidate selection time will decrease. The total execution time will decrease when the increase in pre-candidate selection time is less than the decrease in candidate selection time. The total execution time can be predicted from the set of training characters. The number of stages of the pre-candidate selection module is selected to be that which results in the minimum total execution time. Experiments are performed on the CCL/HCCR1 database with 5401 categories of handwritten Chinese characters. The experimental results show that the proposed model can reduce the total execution time significantly without decreasing the precision of the candidate selection module.
NUMERICAL-SIMULATION OF THE TEMPERATURE-DEPENDENCE OF THE SIDEGATING EFFECT IN GAAS-MESFETS
(1994-08-01)
CHANG, SJ
;
LEE, CP
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
The temperature dependence of the sidegating effect in GaAs devices has been investigated by performing two-dimensional numerical simulations on realistic structures. The less sidegating and the higher sidegating threshold at higher temperatures are found to be caused by the temperature-dependent trapping properties of the deep traps and the difference in increase rate with temperature of the currents in various current paths. These results provide further support lo the sidegating picture in which Schottky contacts on the semi-insulating substrate play an important role.
AN IMPLEMENTABLE DISTRIBUTED STATE ESTIMATOR AND DISTRIBUTED BAD DATA-PROCESSING SCHEMES FOR ELECTRIC-POWER SYSTEMS
(1994-08-01)
LIN, SY
;
LIN, CH
;
交大名義發表
;
電控工程研究所
;
National Chiao Tung University
In this paper, we present an implementable distributed state estimator and distributed bad data detection and identification schemes for electric power systems. The system under consideration is partitioned into k areas, each of which is governed by a local control center that has at least one computer system. These decentralized computer systems are linked by a communication network. The distributed state estimator and distributed bad data processing schemes are designed to be carried out in this computer network. The proposed distributed state estimator possesses the following properties, which ensure its implementability: 1) it uses only the computers available in the decentralized local control centers, 2) it is executed in a small-scale computer communication network, 3) it can converge even if errors occasionally occur in the communication links, and 4) the decentralized computers need not be synchronous. Furthermore, our distributed bad data processing schemes can process the bad data in each area independently using the corresponding local computer system.
LOW-TEMPERATURE EPITAXY OF GAAS BY METALORGANIC CHEMICAL-VAPOR-DEPOSITION
(1994-08-01)
CHEN, WK
;
CHANG, CS
;
CHEN, WC
;
電子物理學系
;
光電工程學系
The low-temperature (LT) epitaxial growth of high quality GaAs films is achieved by using triethylgallium and tertiarybutylarsine as the precursors. Without any external work, the deposition can be conducted at a temperature as low as 350-degrees-C in a conventional metalorganic chemical vapor deposition system. The full width at half-maximum of 77 K photoluminescence and X-ray rocking curve for a sample grown at 425-degrees-C are 8.2 meV and 14 arcsec, respectively. Materials with good electrical properties were also obtained. A Schottky diode formed on the LT CaAs epilayer shows a barrier height of 0.83 eV, and a reverse saturation current of 1.1 X 10(-7) A/cm2, comparable to that of a normal GaAs Schottky diode.
DIGITAL TANLOCK LOOP FOR TRACKING PI/4-DQPSK SIGNALS IN DIGITAL CELLULAR RADIO
(1994-08-01)
WEI, CH
;
CHEN, WJ
;
電控工程研究所
;
電信研究中心
;
Institute of Electrical and Control Engineering
A digital tanlock loop (DTL) is a nonuniform-sampling digital phase-locked loop (DPLL) whose phase error detector, using the tan-1[.] function with in-phase and quadrature samples of incoming signal, has a linear characteristic with a period of 2pi. In this paper, we study the application of an N-phase DTL in digital cellular radio with pi/4-DQPSK modulation. The performance degradation due to fading effect is investigated in terms of bit error rate and steady state phase error variance by computer simulation. It is shown that, in the digital cellular radio system, the performances on both bit error rate and phase error variance using DTL are far better than that using the traditional digital N-phase I-Q loop.
FLOW MODELING IN TURBOFAN MIXING DUCT
(1994-08-01)
TSUI, YY
;
WU, PW
;
LIAO, CW
;
機械工程學系
;
Department of Mechanical Engineering
A computational procedure is described to study the mixing flow in a multilobe turbofan mixer. The predictions have been obtained using a finite volume method to solve the density-weighted lime-averaged Navier-Stokes equations. Turbulence is characterized by the k-epsilon eddy viscosity model. To fit the irregular boundaries of the flow field, the curvilinear nonorthogonal coordinates are employed. The robustness of the computational procedure is enhanced by making use of nonstaggered grids. Results shaw that the streamwise vortex generated at the exit of the lobes dominates the performance of the mixing process. Comparison with experimental data indicates that good predictions can be obtained provided that sufficient inlet conditions are given.
HIGH-LEVEL INJECTION INFLUENCE ON THE HIGH-FREQUENCY PERFORMANCE OF POLYCRYSTALLINE SILICON EMITTER BIPOLAR-TRANSISTORS
(1994-08-01)
CHYAN, YF
;
SZE, SM
;
CHANG, CY
;
CHIUEH, HM
;
REIF, R
This paper reports high-level injection influence on the high frequency performance in the polycrystalline silicon emitter bipolar transistors. Under high-level injection, the transconductance is not proportional to collector current. From the analysis, both the base resistance and the cutoff frequency are functions of terminal voltage, i.e. applied bias, and the maximum oscillation frequency is also a function of terminal voltage. By comparing these results obtained from the self-consistent high-level injection model considering injected charge storage (i.e. injected electron and hole concentrations comparable to doping concentrations in the base and emitter region) and base resistance with those results obtained from models which neglect the injected charge storage effect and/or the base resistance, it is found that both the injected charge storage effect and the base resistance must be considered to get accurate aforementioned device parameters.
EFFECT OF BORON DOPING ON THE STRUCTURAL-PROPERTIES OF POLYCRYSTALLINE SILICON FILMS GROWN AT REDUCED PRESSURES
(1994-08-01)
LIN, HC
;
LIN, HY
;
CHANG, CY
;
JUNG, TG
;
WANG, PJ
The deposition and properties of in situ boron-doped polycrystalline silicon (poly-Si) films grown at 550-degrees-C were investigated using an ultrahigh vacuum chemical vapor deposition system. It is observed that, if the doping level is high enough, the boron incorporation would significantly reduce the deposition rate, impede the grain growth, and degrade the crystallinity of the poly-Si films. We also found that the preferential adsorption of boron atoms on the SiO2 surface at the initial stage of deposition shortened the incubation time of deposition. The property of trapping states at the grain boundary is also examined, and a density of about 4.7 X 10(12) CM-2 is obtained for poly-Si films with a doping level less than 2.2 X 10(19) cm-3.
EFFECTS OF THE COOLING RATE ON THE MECHANICAL AND ELECTRICAL BEHAVIOR OF A 63SN/37PB SOLDER BUMP ON A METALLIZED SI SUBSTRATE
(1994-08-01)
CHIOU, BS
;
CHENG, JC
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
Flip-chip bonding technology, with solder bumps deposited on metal terminals on the chip, provides the highest functional densities of all the present bonding techniques. In this study, tin-lead solder bumps were screen printed onto a metallized Si substrate. Various cooling rates were employed after reflowing the solder bumps. The effect of the cooling rate on the microhardness, the adhesion strength and the electrical resistance of solder joints was studied and the mechanism explored. It was found that intermetallic compounds play important roles in both the mechanical and the electrical behaviour of the solder bumps.
ANOMALOUS HALL-EFFECT FROM VORTEX MOTION IN HIGH-TC SUPERCONDUCTORS
(1994-08-01)
CHEN, JL
;
YANG, TJ
;
電子物理學系
;
光電工程研究所
;
Department of Electrophysics
In this work, the unusual Seebeck effect is taken into consideration in explaining the possible origin of the anomalous Hall effect for high-T(c) superconductors. Combining Maki's theory of transport entropy and Tinkham's theory of resistive transition, we explain why the anomalous Hall effect can be observed in high-T(c) superconductors, but is absent in most conventional superconductors. The behavior of rho(xy) (H, T) in our theory is qualitatively consistent with experiments. In addition, our theory not only predicts that rho(xy) will become positive from rho(xy) < 0 when the temperature is decreased in constant magnetic field, but also predicts that \rho(xy)\ is-proportional-to rho(xx)2 in the region of pho(xy) < 0 and that the negative p(xy) will diminish with increasing defect concentration.
APPLICATIONS OF SELF-TUNING CONTROL ON INDUSTRIAL CNC MACHINES
(1994-08-01)
HSU, PL
;
HSIEH, MY
;
交大名義發表
;
電控工程研究所
;
National Chiao Tung University
Most successful applications of parameter adaptive control to obtain constant cutting forces in machining processes have been limited to laboratory environments. This paper presents a methodology for realizing efficient adaptive control algorithms in industrial CNC machines. Experimental results concerning the fixed gain integral control and variable gain MRAC show that the present set-up in a heavy-duty CNC machining center leads to predictable control behavior. Furthermore, two effective algorithms-a deadbeat self-tuning control with increased order and a PID-type self-tuning control based on the pole-placement method-are implemented to obtain more effective cutting control performance. In this study, repeatable and satisfactory experimental results show that the technique developed here for real manufacturing processes is feasible.
ROLE OF FLUORINE-ATOMS ON THE THERMAL-STABILITY OF THE SILICIDE SILICON STRUCTURE
(1994-08-01)
TSUI, BY
;
CHEN, MC
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
It has been known that thermal stability of the silicide/silicon system can be improved by fluorine incorporation. Two mechanisms have been proposed-the fluorine buffer model and the grain growth retardation model. This communication examines the validity of these two mechanisms. Fluorine ions are introduced into the PtSi/Si structure by implantation at various energies and to various dosages. If the fluorine ions are confined in the silicide layer, the grain growth is retarded, however, no improvement of the thermal stability can be obtained. On the other hand, if most of the implanted fluorine ions are located near the PtSi/Si interface, the grains of the fluorinated silicide film can grow to be larger than those of the unfluorinated silicide film without becoming discontinuous to the film. It is thus concluded that the fluorine accumulation at the interface, i.e., the fluorine buffer model, plays the dominant role on the thermal stability of the fluorinated PtSi/Si structure.
A REFLECTION-TYPE SUBSTRATE-MODE GRATING STRUCTURE FOR WAVELENGTH-DIVISION-MULTI-DEMULTIPLEXING
(1994-08-01)
TSAI, YK
;
HUANG, YT
;
SU, DC
;
電子工程學系及電子研究所
;
光電工程學系
We describe a technique to design and implement a wavelength-division-multi/demultiplexing device with a reflection substrate-mode grating pair in dichromated gelatin. This device is based on a diffraction volume grating pair, which provides and angularly dispersive function, to separate optical signals at different wavelengths through the dielectric substrate with total internal reflection. In this structure, the channel separation is enhanced by multiple reflections at substrate-air interfaces with a shorter longitudinal distance. Normally incident coupling and on-axis output imaging with this structure provide easier alignment with coupled optical fibers. The device with a polarization-insensitive high efficiency ( > 80 %) for 76-nm bandwidth has been fabricated.
MODE DESCRIPTION OF ROUTES TO CHAOS IN EXTERNAL-CAVITY COUPLED SEMICONDUCTOR-LASERS
(1994-08-01)
KAO, YH
;
WANG, NM
;
CHEN, HM
;
電信工程研究所
;
電信研究中心
The nonlinear behaviors of routes to chaos in semiconductor lasers with external optical feedback were explored in this work. It was indicated that the relaxation oscillation was the origin of the first nonlinear instability of optical intensity under a short delay. Three types of transition routes with quasi-periodic, subharmonic oscillation, and periodic doubling to optical chaos were distinguished in terms of the delay time normalized to the inverse of the relaxation oscillation frequency of a solitary laser diode. The linearized mode theory and small-signal response were confirmed as part of the phenomena in the transitions to chaos.
ANALYTICAL MODELING OF POLYCRYSTALLINE SILICON EMITTER BIPOLAR-TRANSISTORS UNDER HIGH-LEVEL INJECTION
(1994-08-01)
CHYAN, YF
;
CHANG, CY
;
SZE, SM
;
LIN, MJ
;
LIAO, K
We derived an analytical model, focused on injected charge storage, for electron current density J(n), hole current density J(ps), current gain beta, and forward transit time tau(f), for heavily doped base and emitter with non-uniform band structures. This model can apply in high-level injection regions before the onset of the Kirk effect. By comparing with those results obtained from Sukuki's low-level injection model, it is observed that a new high-level injection term, proportional to the square of the hole current density, appears in the injected hole concentration in the polycrystalline silicon emitter and this term vanishes in the limit of low-level injection. As bias increases, J(n) and J(ps) of high-level injection model are less than those of the low-level injection model. In addition. for low-level injection model, beta, tau(f), the base transit time tau(b), the crystalline silicon emitter transit time tau(se), and the polycrystalline silicon emitter transit time tau(pe) keep constant for all injection region. However, for high-level injection model, beta and tau(b) decreases as bias increases, while tau(se) and tau(pe) increase with increasing bias. Tau(f) reaches a minimum and then increases. We conclude that the high-level injection model offers significant information for device operations at various electron current levels.
ADAPTIVE PACKET EQUALIZATION FOR INDOOR RADIO CHANNEL USING MULTILAYER NEURAL NETWORKS
(1994-08-01)
CHANG, PR
;
YEH, BF
;
CHANG, CC
;
交大名義發表
;
電信工程研究所
This paper investigates the application of the multilayer perceptron structure to the packet-wise adaptive decision feedback equalization of a M-ary QAM signal through a TDMA indoor radio channel in the presence of intersymbol interference (ISI) and additive Gaussian noise. Since the multilayer neural networks are capable of producing complex decision regions with arbitrarily nonlinear boundaries, this would greatly improve the performance of conventional decision feedback equalizer (DFE) where the decision boundaries of conventional DFE are linear. However, the applications of the traditional multilayer neural networks have been limited to real-valued signals. To tackle this difficulty, a neural-based DFE is proposed to deal with the complex QAM signal over the complex-valued fading multipath radio 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, this neural-based DFE trained by packet-wise backpropagation algorithm would approach an ideal equalizer after receiving a sufficient number of packets. In this paper, another fast packet-wise training algorithm with better convergence properties is derived on the basis of a recursive least-squares (RLS) routine. Results show that the neural-based DFE trained by both algorithms provides a superior bit-error-rate performance relative to the conventional least mean square (LMS) DFE, especially in poor signal to noise ratio conditions.
MULTIPLE-ANGLE INCIDENT ELLIPSOMETRY MEASUREMENT ON LOW-PRESSURE CHEMICAL-VAPOR-DEPOSITED AMORPHOUS-SILICON AND POLYSILICON
(1994-08-01)
CHAO, TS
;
LEE, CL
;
LEI, TF
;
電子工程學系及電子研究所
;
奈米中心
Multiple-angle incident (MAI) ellipsometry is used to study the optical properties of both amorphous silicon and poly-Si after different heat-treatments. At first, the error sensitivity of MAI ellipsometry has been studied and the optimum conditions to achieve the best precision are obtained. The optical constants and thickness of amorphous silicon deposited by LPCVD at 550-degrees-C changed significantly after sample annealing. The optical constants of the amorphous silicon reached the values of single-crystalline silicon after annealing for both the high and low temperature treatments. A thickness shrinking phenomenon is observed when amorphous silicon is transformed to the polycrystalline form. This phenomenon is confirmed by cross-sectional TEM photography. The thickness shrinking ratio is 3.5 to 6.0%.