Browsing by Issue Date, starting with "1999-09-01"

Now showing 1 - 20 of 80362

Cost-effective VLSI architectures and buffer size optimization for full-search block matching algorithms
(1999-09-01)
Yeh, YH
;
Lee, CY
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
This paper presents two efficient very large scale integration (VLSI) architectures and buffer size optimization for full-search block matching algorithms. Starting from an overlapped data how of search area, both systolic- and semisystolic-array architectural solutions are derived. By means of exploiting stream memory banks, not only input/output (I/O) bandwidth can be minimized, but also processor element efficiency can be improved, In addition, the controller structure for both solutions are very straightforward, making them very suitable for VLSI implementation to meet computational requirements. Moreover, by exploring the dependency graph, we focus on the problem of reducing the internal buffer size under minimal I/O bandwidth constraint to derive guidelines on reducing redundant internal buffer as well as to achieve area-efficient VLSI architectures. Simulation results show that, for N = P = 16 (N is the reference block size and P is the search range), I/O bandwidth can be reduced by 2.4 times, while buffer size increases less than 38%. Two prototype chips for N = P = 16 have been designed and fabricated. Test results show that clock rate can be up to 90 MHz, implying that more than 87.9-K motion vectors per second can be achieved to meet real-time requirements specified in MPEG-2 MP@ML coding standard.
Bifurcation of fundamental Gaussian modes in Kerr-lens mode-locked lasers
(1999-09-01)
Wei, MD
;
Hsieh, WF
;
光電工程學系
;
Department of Photonics
We propose that more than one spatial fundamental Gaussian mode may occur in concentric and confocal resonators under persisting nonlinear effect. Extensively studying the influence of the resonator's parameters in a Kerr-lens mode-locked resonator, pitchfork bifurcation results in more symmetrical configurations, and saddle-node bifurcation appears as the symmetry being broken. From the properties of bifurcation, we suggest that the equal-arm and near-confocal resonator is suitable for the emergence of bistability in Kerr-lens mode-locked lasers. (C) 1999 Elsevier Science B.V. All rights reserved.
A high-temperature thermodynamic model for metalorganic vapor phase epitaxial growth of InGaN
(1999-09-01)
Ou, J
;
Pan, YC
;
Lee, WH
;
Shu, CK
;
Lin, HC
A thermodynamic analysis tailored with a high-temperature parameter was performed for metalorganic vapor phase epitaxial (MOVPE) growth of InxGa1-xN alloys. Accordingly, a series of InGaN samples were prepared under various input In flow rates and growth temperatures. Theoretical results indicate that the In solid concentration tends to increase with increasing In flow rate and with decreasing temperature, whereas the metal droplets form more easily on the surface at lower growth temperatures and higher In fluxes, which are in good agreement with our experimental data.
Instruction cache prefetching directed by branch prediction
(1999-09-01)
Chiu, JC
;
Shiu, RM
;
Chi, SA
;
Chung, CP
;
資訊科學與工程研究所
As the gap between processor speed and memory speed grow, so the performance penalty of instruction cache misses gets higher. Instruction cache prefetching is a technique to reduce this penalty. The prefetching methods determine the target line to be prefetched generally based on the current fetched line address. However, as the cache line becomes wider, it may contain multiple branches. This is a hurdle which must be overcome. The authors have developed a new instruction cache prefetching method in which the prefetch is directed by the prediction on branches, called branch instruction based (BIB) prefetching; in which the prefetch information is recorded in an extended BTB. Simulation results show that for commercial benchmarks, BIB prefetching outperforms traditional sequential prefetching by 7% and other prediction table based prefetching methods by 17% on average. As BTB designs become more sophisticated and achieve higher hit and accuracy ratios, BIB prefetching can achieve a higher level of performance.
Constructing perfect secret sharing schemes for general and uniform access structures
(1999-09-01)
Sun, HM
;
Shieh, SP
;
資訊工程學系
;
Department of Computer Science
A perfect secret sharing scheme allows a secret K to be shared among a set of participants in such a way that only qualified subsets of participants can recover the secret, and unqualified subsets of participants obtain no information regarding the secret. In this paper, we propose a construction of perfect secret sharing schemes with uniform, generalized access structures of rank 3. Compared with other constructions, our construction has some improved lower bounds on the information rate. In addition, we also generalize the construction to perfect secret sharing schemes with uniform, generalized access structures of constant rank.
Splitting and merging version spaces to learn disjunctive concepts
(1999-09-01)
Hong, TP
;
Tseng, SS
;
資訊工程學系
;
Department of Computer Science
We have modified the original version space strategy in order to learn disjunctive concepts incrementally and without saving past training instances. The algorithm time complexity is also analyzed, and its correctness is proven.
Effect of bottom electrode materials on the electrical and reliability characteristics of (Ba, Sr)TiO3 capacitors
(1999-09-01)
Tsai, MS
;
Sun, SC
;
Tseng, TY
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
The dielectric constant and the leakage current density of (Ba, Sr)TiO3 (BST) thin films deposited on various bottom electrode materials (Pt, Ir. IrO2/Ir, Ru, RuO2/Ru) before and after annealing in O-2 ambient were investigated. The improvement of crystallinity of BST films deposited on various bottom electrodes was observed after the postannealing process, The dielectric constant and leakage current of the films were also strongly dependent on the postannealing conditions. BST thin film deposited on Ir bottom electrode at 500 degrees C, after 700 degrees C annealing in O-2 for 20 min, has the dielectric constant of 593, a loss tangent of 0.019 at 100 kHz, a leakage current density of 1.9 x 10(-8) A/cm(2) at an electric field of 200 kV/cm with a delay time of 30 s, and a charge storage density of 53 fC/mu m(2) at an applied field of 100 kV/cm, The BST films deposited on Ir with post-annealing can obtain better dielectric properties than on other bottom electrodes in our experiments, And Ru electrode is unstable because the interdiffusion of Ru and Ti occurs at the interface between the BST and Ru after postannealing. The ten-year lifetime of time-dependent dielectric breakdown (TDDB) studies indicate that BST on Pt, Ir, IrO2/Ir, Ru, and RuO2/Ru have long lifetimes over ten years on operation at the voltage bias of 2 V.
草原上之小樹呀
(1999-09-01)
管管
Characterization and modeling of out-diffusion of manganese and zinc impurities from deep ultraviolet photoresist
(1999-09-01)
Wang, MY
;
Ko, FH
;
Wang, TK
;
Yang, CC
;
Huang, TY
The radioactive tracer technique was applied to investigate the out-diffusion of manganese and zinc impurities from deep-ultraviolet (DUV) photoresist. Two important process parameters, viz., baking temperature and the type of substrate (i.e., bare silicon, polysilicon, oxide, or nitride), were evaluated. Our results indicated that diffusion ratios were ail below 6%, irrespective of the substrate type and baking temperature. The substrate type did not appear to strongly affect the metallic impurity out-diffusion from DUV photoresist. However, solvent evaporation was found to have a significant effect on impurity diffusion. A new model, together with a new parameter, was proposed to describe the out-diffusion behavior of impurities from DUV photoresist. This model could explain the diffusion ratio of metallic impurities in photoresist layers under various baking conditions. The effectiveness of various wet cleaning recipes in removing metallic impurities such as manganese and zinc was also studied. It was found that (i) bath life due to temperature change can considerably affect the cleaning efficiency, and (ii) hot water immersion can effectively dissolve the impurities from the wafer surface. (C) 1999 The Electrochemical Society. S0013-4651(98)11-080-7. All rights reserved.
New direct velocity and acceleration feedforward tracking control in a retrofitted milling machine
(1999-09-01)
Chen, CS
;
Lee, AC
;
機械工程學系
;
Department of Mechanical Engineering
This paper proposes a new feedforward controller called the direct velocity and acceleration feedforward controller (DVAFF) to improve the tracking performance. The controller is adapted to both minimum phase system and nonminimum phase system and it can also take the practical driving capability of actuator and saturation of D/A converter into account by tuning their gains. Simulation and experiment results show that the proposed new DVAFF scheme can effectively improve the motion accuracy.
Micromechanics model for elastic stiffness of non-spherical granular assembly
(1999-09-01)
Dong, JJ
;
Pan, YW
;
土木工程學系
;
Department of Civil Engineering
This paper presents a micromechanics model for the elastic stiffness of a non-spherical granular assembly. The microstructural continuum model of ideal spherical assembly is extended for non-ideal assembly. The presented work takes the effects of gradation, shape, and preferred orientation into account by introducing a directional distribution function of branch-vector length. The microstructure of a granular assembly is described by the distributions of packing structure, branch-vector length, and particle number per unit volume. These distributions account or the random nature of a realistic granular material. The microfeatures relevant to the description of non-deal particle assembly are elaborated. The influences of various direction-dependent and direction-independent microfeatures on the elastic stiffness are demonstrated. Hypothetical non-ideal granular assemblies are used to study the effects of gradation, shape and preferred orientation. Based on the proposed model, the paper discusses the inherent anisotropy in a non-ideal granular assembly. The presented work also makes use of a generalized static hypothesis to estimate the contact-force distribution for specific microstructure and stress state. With the estimated contact-force and the Hertz-Mindlin contact theory, the elastic stiffness of a particulate assembly can be evaluated. Hence, the effects of geometric fabric and anisotropic stress state on the elastic stiffness can be deliberated. Consequently, the effects of geometric fabric and kinetic fabric of a natural granular material can be evaluated independently. It is shown that the proposed model can reasonably capture the phenomena of inherent anisotropy and stress-induced anisotropy of a non-spherical granular assembly under small strain. Copyright (C) 1999 John Wiley & Sons, Ltd.
A low-complexity frame synchronization and frequency offset compensation scheme for OFDM systems over fading channels
(1999-09-01)
Hsieh, MH
;
Wei, CH
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
dThis paper presents a fast low-complexity synchronization scheme for orthogonal frequency division multiplexing (OFDM) systems over fading channels. By utilizing the guard interval in OFDM signals, the Frame synchronization and the frequency offset estimation are considered simultaneously. The Implementation can be simplified by only using the sign bits of the in-phase and the quadrature components of the received OFDM signal for frame synchronization and frequency offset compensation. A frequency-offset independent frame synchronization algorithm is derived, and a low-complexity frequency offset estimator based on the synchronized correlator output is presented in this paper. Due to the subcarrier ambiguity in the guard-interval-based (GIB) frequency detector, the maximum correctable frequency range is limited to +/- 1/2 of the subcarrier spacing. In this paper, we also present a new frequency acquisition scheme that can solve the subcarrier ambiguity problem and extend the frequency acquisition range to nearly a half of the useful OFDM signal bandwidth.
Enhanced spatial model for landfill siting analysis
(1999-09-01)
Lin, HY
;
Kao, JJ
;
環境工程研究所
;
Institute of Environmental Engineering
A landfill siting analysis typically requires evaluating various rules, factors, constraints, and numerous spatial data. A modern geographical information system, although capable of rapidly processing a massive amount of spatial data, lacks the ability to locate an optimal site when compactness and other factors are simultaneously evaluated. A previously developed grid-based model could not be applied to resolve this inability for irregularly shaped spatial data. Therefore, an enhanced spatial siting model is proposed herein for general spatial data. A compactness index is applied to ensure the integrity of selected sites. Two case studies are presented to demonstrate the applicability of the proposed model. The proposed model and two models developed previously are compared in the first case study based on the single factor of land cost. A single factor model is perhaps quite unacceptable with respect to other factors. The second case study is thus presented to demonstrate the flexibility of the model for considering additional factors, land slope, and road network accessibility. Moreover, results obtained from various models and siting factors are compared and discussed.
Hyperbolic heat conduction in thin-film high T-c superconductors with interface thermal resistance
(1999-09-01)
Lor, WB
;
Chu, HS
;
機械工程學系
;
Department of Mechanical Engineering
This article numerically analyzes the hyperbolic heat conduction problem in the film and substrate composites under an imposed surface heat flux on the exterior film surface. The radiation heat flux model is employed to take account of the interface thermal resistance. The reflection and transmission occur at the contact surface of the dissimilar material which depends on the substrate properties and interface conditions. The interface resistance restricts the energy transmission across the interface and alerts the reflected and transmitted waves' strength. Neglecting the interface thermal resistance causes the temperature distribution in the film to be greatly underestimated. Moreover, the hyperbolic equation predicts significantly different results with those predicted by the parabolic equation at small time scales. The discrepancies between the solutions in an investigation of superconductor Y-Ba-Cu-O film depositions on several commonly used substrates are examined. (C) 1999 Elsevier Science Ltd. All rights reserved.
A comprehensive study of hot carrier stress-induced drain leakage current degradation in thin-oxide n-MOSFET's
(1999-09-01)
Wang, TH
;
Chiang, LP
;
Zous, NK
;
Hsu, CF
;
Huang, LY
The mechanisms and characteristics of hot carrier stress-induced drain leakage current degradation in thin-oxide n-MOSFET's are investigated. Both interface trap and oxide charge effects are analyzed. Various drain leakage current components at zero V-gs such as drain-to-source subthreshold leakage, band-to-band tunneling current, and interface trap-induced leakage are taken into account, The trap-assisted drain leakage mechanisms include charge sequential tunneling current, thermionic-field emission current,:and Shockley-Read-Hall generation current. The dependence of drain leakage current on supply voltage, temperature, and oxide thickness is characterized. Our result shows that the trap-assisted leakage may become a dominant drain leakage mechanism as supply voltage is reduced. In addition, a strong oxide thickness dependence of drain leakage degradation is observed. In ultra-thin gate oxide (30 Angstrom) n-MOSFET's, drain leakage current degradation is attributed mostly to interface trap creation, while in thicker oxide (53 Angstrom) devices, the drain leakage current exhibits two-stage degradation, a power law degradation rate in the initial stage due to interface trap generation, followed by an accelerated degradation rate in the second stage caused by oxide charge creation.
Matricial ranges of quadratic operators
(1999-09-01)
Tso, SH
;
Wu, PY
;
應用數學系
;
Department of Applied Mathematics
We show that if T is a quadratic operator on a Hilbert space, then (1) the numerical range of T is an (open or closed) elliptical disc (or its degenerate form) and (2) for every n greater than or equal to 1, the nth matricial range of T consists of n x n matrices whose numerical ranges are contained in the closure of the numerical range of T.
Effects of H-2 plasma treatment on low dielectric constant methylsilsesquioxane
(1999-09-01)
Chang, TC
;
Liu, PT
;
Mei, YJ
;
Mor, YS
;
Perng, TH
The quality of organic low-k methylsilsesquioxane (MSQ) film is degraded by the damage of oxygen plasma and hygroscopic behavior during photoresist stripping. In addition, the interaction between MSQ and copper is worth investigating. In this work, we have studied the H-2 plasma treatment to improve the quality and enhance the copper penetration resistance of MSQ, Experimental results show the leakage current of MSQ decreases as the H-2 plasma treatment time is increased. The dielectric constant of treated samples also remains constant (similar to 2.7). In addition, the copper diffusion resistance of MSQ film is significantly promoted. The H-2 plasma treatment can provide additional hydrogen to passivate the inner structure of porous MSQ film as well as reduce the probability of moisture uptake and interaction with Cu atoms. Therefore, the low-k dielectric properties of MSQ are significantly enhanced by H-2 plasma treatment. (C) 1999 American Vacuum Society. [S0734-211X(99)05105-7].
草原上之小樹呀
(1999-09-01)
管管
An incremental learning neural network for pattern classification
(1999-09-01)
Hung, CA
;
Lin, SF
;
電控工程研究所
;
Institute of Electrical and Control Engineering
A neural network architecture that incorporates a supervised mechanism into a fuzzy adaptive Hamming net (FAHN) is presented. The FAHN constructs hyper-rectangles that represent template weights in an unsupervised learning paradigm. Learning in the FAHN consists of creating and adjusting hyper-rectangles in feature space. By aggregating multiple hyper-rectangles into a single class, we can build a classifier, to be henceforth termed as a supervised fuzzy adaptive Hamming net (SFAHN), that discriminates between nonconvex and even discontinuous classes. The SFAHN can operate at a fast-learning rate in online (incremental) or offline (batch) applications, without becoming unstable. The performance of the SFAHN is tested on the Fisher iris data and on an online character recognition problem.
Priority call service on PCS networks
(1999-09-01)
Yang, WZ
;
Chang, MF
;
Lin, YB
;
資訊工程學系
;
Department of Computer Science
This paper studies the "priority call service" for personal communication services such as GSM cellular networks. With this service, a priority call requires is buffered in a waiting queue if no free channel is available. The buffered request can be served as soon as free channels become available. We develop an analytic model to evaluate the performance for the priority call service. In addition, computer simulations are conducted to verify the analytic model. The results indicate that priority calls experience very low new call blocking probability, as well as low call incompletion probability at the cost of increasing the force-termination probability.