Browsing by Author "Chang, YC"

Now showing 1 - 20 of 53

10 Gbit/s all-optical non-return to zero-return-to-zero data format conversion based on a backward dark-optical-comb injected semiconductor optical amplifier
(2006-05-15)
Lin, GR
;
Yu, KC
;
Chang, YC
;
光電工程學系
;
Department of Photonics
By using a semiconductor optical amplifier backward injected by a dark-optical-comb pulse train at 10 GHz, we demonstrate a 10 Gbit/s all-optical nonreturn-to-zero (NRZ) to return-to-zero (RZ) format conversion of an incoming optical pseudorandom binary sequence (PRBS) data strieam. Both the polarity and the wavelength of data are conserved during format conversion. Without any preamplification, the extinction ratio of degraded optical NRZ PRBS data is greatly improved from 7.13 to 13.6 dB after NRZ-to-RZ conversion. An ultralow bit-error rate of 10(-12) at a data rate as high as 10 Gbits/s is obtained with a received optical power of-18.3 dBm. The converted RZ PRBS data exhibit a negative power penalty of > 3.7 dB compared with the NRZ PRBS data at a bit-error rate of 10(-12). (c) 2006 Optical Society of America
10 Gbit/s tunable dual-wavelength nonreturn-to-zero-to-return-to-zero data-forma't transformer based on a non-direct-current-biased Fabry-Perot laser diode
(2005-05-20)
Chang, YC
;
Lin, GR
;
光電工程學系
;
Department of Photonics
We demonstrate the conversion of a nonreturn-to-zero (NRZ)-formatted electrical data stream into a wavelength-tunable return-to-zero (RZ)-formatted optical pulse code by externally seeding a synchronously sinusoidal-modulated Fabry-Perot laser diode (FPLD) with optical pseudorandom binary-sequence data at 10 Gbits/s (Gbps). The FPLD without a dc-biased current was modulated by use of a power-amplified sinusoidal wave signal (similar to 25.6 dBm) as an NRZ-to-RZ data-format transformer, which is regeneratively amplified by a closed-loop erbium-doped fiber amplifier. The gain switching and on-off keying operation of the FPLD is initiated under the seeding of self-feedback and extemal-injection signals. A maximum wavelength tuning range of 30 nm with a side-mode suppression ratio of greater than 36 dB is obtained. The power penalty of the NRZ-to-RZ data-format conversion at 10 Gbps is 1.5 dB. (c) 2005 Optical Society of America.
A 3.5-ps mode-locked semiconductor optical amplifier fiber laser generated by 60-ps backward optical dark pulse-train injection
(2005)
Chiu, IH
;
Liao, YS
;
Chang, YC
;
Lin, GR
;
光電工程學系
The relationship between the backward optical injection waveform and the mode-locked Pulse shape of semiconductor optical amplifier fiber laser (SOAFL) is studied. The SOA plays both the roles of a gain medium and an optically controlled modulator in this work. The injected optical comb-like bright and dark pulse-train with 60-ps pulsewidth was generated using a Mach-Zehnder intensity modulator (MZM) which DC-biased voltage of 1.7 V and 7.2 V, respectively. The backward injection of optical dark pulse-train results in a wide gain-depletion width (and a narrow gain window) within one modulation period, which is necessary for perfect mode-locking the SOAFL. In opposite, the backward injection of short optical pulse of bright optical pulse-train only causes a less pronounced gain-depletion effect. Such a broadened gain window can hardly initiate the mode-locking process. The backward comb-like dark pulse-train modulation IS Much easier to initiate harmonic mode-locking in the SOAFL than the bright pulse-train or sinusoidal-wave, wave injection, which generates pulsewidth as short as 15 ps at 1 GHz. After propagating through 195m-long dispersion-compensating fiber, the pulsewidth of the mode-locked SOAFL can be linearly compressed to 13.5 ps. The linewidth and time-bandwidth product of the compressed SOAFL pulses are 1.78 nm and 0.8, respectively. The pulsewidth can further be nonlinearly compressed by using a 4695m-long single-mode fiber. The shortest mode-locked SOAFL pulsewidth of 3.5 ps at repetition frequency of I GHz by using cross-gain modulation technique is reported for the first time.
A new method for fuzzy information retrieval based on fuzzy hierarchical clustering and fuzzy inference techniques
(2005-04-01)
Horng, YJ
;
Chen, SM
;
Chang, YC
;
Lee, CH
;
資訊工程學系
In this paper, we extend the work of Kraft et al to present a new method for fuzzy information retrieval based on fuzzy hierarchical clustering and fuzzy inference techniques. First, we present a fuzzy agglomerative hierarchical clustering algorithm for clustering documents and to get the document cluster centers of document clusters. Then, we present a method to construct fuzzy logic rules based on the document clusters and their document cluster centers. Finally, we apply the constructed fuzzy logic rules to modify the user's query for query expansion and to guide the information retrieval system to retrieve documents relevant to the user's request. The fuzzy logic rules can represent three kinds of fuzzy relationships (i.e., fuzzy positive association relationship, fuzzy specialization relationship and fuzzy generalization relationship) between index terms. The proposed fuzzy information retrieval method is more flexible and more intelligent than the existing methods due to the fact that it can expand users' queries for fuzzy information retrieval in a more effective manner.
A pulse-compressed and harmonic mode-locked SOA-EDFA laser link with suppressed phase and supermode noises
(2005)
Lin, GR
;
Chang, YC
;
Wu, MC
;
光電工程學系
;
Department of Photonics
An optically bandpass filtered SOA-EDFL laser link with suppressed phase noise of -110 dBc/Hz and improved supermode noise suppression ratio of 90 dB after harmonic mode-locking and fiber-based pulse compression (to 3.1 ps) are demonstrated.
A thermal-aware power management Soft-IP for platform-based SoC designs
(2004)
Chan, CH
;
Chang, YC
;
Ho, HC
;
Chiueh, H
;
電信工程研究所
A novel thermal-aware power management (TAPM) Software Intellectual Property (Soft-IP) for modem platform-based SoC designs is presented. This research proposes the system-level architecture of thermal-aware power management, which includes a power management bus (PMB), TAPM Soft-IP and interface circuitry for proposed PMB. Each component of proposed design is encapsulated into a Soft-IP. With above design, system architects are able to incorporate on-chip power-controls and sensors to achieve nominal power dissipation and ensure the targeting system working within specification. The design yields intricate control and optimal management with little system overhead and minimum hardware requirements, as well as provides the flexibility to support different management schemes. The proposed system and its components are designed, implemented and verified by a prototype chip, which was fabricated in a TSMC 0.25 mu m 1P5M standard CMOS technology through National Chip Implementation Center (CIC), Taiwan.
All-optical data format conversion in synchronously modulated single-mode Fabry-Perot laser diode using external injection-locking-induced nonlinear threshold reduction effect
(2005-06-01)
Lin, GR
;
Chang, YC
;
Lin, YH
;
Chen, JH
;
光電工程學系
High-speed all-optical data-format conversion and logic gate from a synchronously modulated Fabry-Wrot laser diode (FPLD) below threshold condition is demonstrated by using external injection-induced. nonlinear threshold reduction effect of the FPLD. A pulsed pseudoreturn-to-zero (PRZ) data with maximum extinction ratio of 14.5 dB is obtained by externally seeding the 9.953-Gb/s (OC-192) nonreturn-to-zero (NRZ) data stream to optically gain-switch the FPLD. At optimized radio frequency and optical injecting powers of 24 and 4 dBm, the root mean square timing jitter and the duty-cycle (pulsewidth) of the PRZ eye pattern are determined as 0.4 ps and 45% (44 ps), respectively. The bit-error rate (BER) can be as low as 10(-13) under the 4-dBm external injecting power. A 1.5-dB power penalty with the BER of < 10(-9) is measured at a bit rate of 9.953 Gb/s. Highest NRZ-to-PRZ conversion rate of 12.5 Gb/s is achieved by increasing biasing current of the FPLD.
All-optical NRZ-to-PRZ format transformer with an injection-locked Fabry-Perot laser diode at unlasing condition
(2004-09-20)
Chang, YC
;
Lin, YH
;
Chen, JH
;
Lin, GR
;
光電工程學系
By using an optical nonreturn-to-zero (NRZ) format data-stream to injection-lock an synchronously modulated Fabry-Perot laser diode at below threshold condition (without DC driving current), an output data-stream with pseudo-return-to-zero (PRZ) format can be generated at bit rate of up to 2.488 Gbit/s. Such an NRZ-to-PRZ format transformation is due to the injection-locking induced gain-switching of the FPLD with the incoming NRZ data. The PRZ data-stream with a maximum on/off extinction ratio of 12.2 dB is obtained under the optical injecting power of -2 dBm and the RF driving power of 24.4 dBm. The best side-mode suppression ratio of 40 dB and the lowest timing jitter of 0.4 ps for the PRZ data-stream are observed. A power penalty of 1.2 dB is measured at a bit-error rate of 10(-9) after NRZ-to-PRZ transformation. In application, the demonstration of an all-optical OR logic gate using the FPLD-based NRZ-to-PRZ transformer is also reported. (C) 2004 Optical Society of America.
All-optical pulse data generation in a semiconductor optical amplifier gain controlled by a reshaped optical clock injection
(2006-05-08)
Lin, GR
;
Chang, YC
;
Yu, KC
;
光電工程學系
;
Department of Photonics
Wavelength-maintained all-optical pulse data pattern transformation based on a modified cross-gain-modulation architecture in a strongly gain-depleted semiconductor optical amplifier (SOA) is investigated. Under a backward dark-optical-comb injection with 70% duty-cycle reshaping from the received data clock at 10 GHz, the incoming optical data stream is transformed into a pulse data stream with duty cycle, rms timing jitter, and conversion gain of 15%, 4 ps, and 3 dB, respectively. The high-pass filtering effect of the gain-saturated SOA greatly improves the extinction ratio of data stream by 8 dB and reduces its bit error rate to 10(-12) at -18 dBm. (c) 2006 American Institute of Physics.
An effective three-dimensional model for UHF radio propagation into multistory office buildings
(1999-01-20)
Tarng, JH
;
Chang, YC
;
Chen, CM
;
電信工程研究所
;
Institute of Communications Engineering
An effective three-dimensional (3-D) direct-transmitted model combined with a patched-wall model is proposed to explore UHF radio propagation into multistory buildings in a microcellular environment. The model is verified by comparing the computed path loss with the measured one of 1.7 and 2.44 GHz radio waves propagating at different sites. The comparison demonstrates that the model has reasonably accurate prediction when the number of floor levels between the transmitting and the receiving antennas is no more than four.
B*-trees: A new representation for non-slicing floorplans
(2000)
Chang, YC
;
Chang, YW
;
Wu, GM
;
Wu, SW
;
資訊工程學系
We present in this paper an efficient, flexible, and effective data structure, B*-trees, for non-slicing floorplans. B*-trees are based on ordered binary trees and the admissible placement presented in [1]. Inheriting from the nice properties of ordered binary trees, B*-trees are very easy for implementation and can perform the respective primitive tree operations search, insertion, and deletion in only O(1), O(1), and O(n) least O(n) time for each of these operations, where n is the number of modules. The correspondence between an admissible placement and its induced B*-tree is 1-to-1 (i.e., no redundancy); further, the transformation between them takes only linear time. Unlike other representations for non-slicing floorplans that need to construct constraint graphs for cost evaluation, in particular, the evaluation can be performed on B*-trees and their corresponding placements directly and incrementally. We further show the flexibility of B*-trees by exploring how to handle rotated, pre-placed, soft, and rectilinear modules. Experimental results on MCNC benchmarks show that the B*-tree representation runs about 4.5 times faster, consumes about 60% less memory, and results in smaller silicon area than the O-tree one [1]. We also develop a B*-tree based simulated annealing scheme for floorplan design; the scheme achieves near optimum area utilization even for rectilinear modules.
Bootstrap approach for estimating process quality yield with application to light emitting diodes
(2005)
Pearn, WL
;
Chang, YC
;
Wu, CW
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Process capability indices have been widely used by quality professionals for measuring process performance. Although process yield is the most common criterion used in the manufacturing industry for measuring process performance, a more advanced measurement formula Y-q, called quality yield index, has been proposed as an alternative measure of process performance. Quality yield can be viewed as the classical process yield minus the truncated expected relative process loss, within the specifications, which focuses on customer satisfaction. By taking customer loss into consideration, the advantage of using the quality-yield measure as process performance is that the formula can be applied to processes with arbitrary distributions. Unfortunately, statistical properties of the estimated Y-q are mathematically intractable. Therefore, capability testing cannot be performed. In this paper, a nonparametric but computer intensive method called bootstrap is used to obtain a lower confidence bound on quality yield for capability testing purposes. Simulation studies are conducted to examine the sampling distribution of the estimated Y-q. An application using the index Y-q for the light emitting diode manufacturing process is presented for illustration purposes.
CMP of low-k methylsilsesquiazane with oxygen plasma treatment for multilevel interconnect applications
(2004)
Chang, TC
;
Tsai, TM
;
Liu, PT
;
Chen, CW
;
Yan, ST
The effect of oxygen plasma treatment on methylsilsesquiazane (MSZ) dielectric was investigated for chemical mechanical planarization (CMP) process. Oxygen plasma treatment was implemented before CMP. Experimental results have shown that the polishing rate of MSZ film with O-2 plasma pretreatment is increased as much as two order of magnitude more than that of MSZ without O-2 plasma pretreatment. Moreover, the electrical properties of post-CMP MSZ are close to those of an as-cured MSZ. These results indicate that the modified surfaces resulting from O-2 plasma treatment increase the polishing rate of MSZ. After polishing, the MSZ film still maintains a low-k quality. (C) 2004 The Electrochemical Society.
CMP of ultra low-k material porous-polysilazane (PPSZ) for interconnect applications
(2004-01-30)
Chang, TC
;
Tsai, TM
;
Liu, PT
;
Chen, CW
;
Yan, ST
In this article, we investigated the impact of chemical mechanical polishing (CMP) on an ultra low dielectric constant (ultra low-k) material Porous-Polysilazane (PPSZ) with slurries of metal polishing during interconnect manufacture process. Since the CMP processing of metals such as TaN and Cu are inevitable steps for interconnect fabrication, we have utilized two types of slurries (marked as TaN and Cu slurries) to evaluate their effects on the dielectric properties of PPSZ films. Electrical and material analyses have shown surface planarity and dielectric properties of PPSZ films will not be degraded during these metal CMP processes. This indicates that the ultra low-k PPSZ films are promising for inter-level dielectric (ILD) applications in ultra large-scale integrated circuits (ULSI) technology. (C) 2003 Elsevier B.V. All rights reserved.
Comparison on the noise and jitter characteristics of harmonic injection-locked and mode-locked erbium-doped fiber lasers
(2005)
Wu, MC
;
Chang, YC
;
Lin, GR
;
光電工程學系
;
Department of Photonics
We compare the noise characteristics of optical pulses generated from an actively mode-locked (AML) erbium-doped fiber laser (EDFL) with a semiconductor optical amplifier and an injection-locked EDFL with a gain-switched Fabry-Perot laser diode (FPLD). The mode-locked EDFL pulse exhibits a phase noise of -110.1 dBc/Hz (at 1 MHz offset frequencies from the carrier), the timing jitter of 1.16 ps, and a supermode noise suppression ratio of 47.5 dB. The injection-locked EDFL pulse exhibits a phase noise of -121.1 dBc/Hz (at 1 MHz offset frequencies from the carrier), a timing jitter of 0.31 ps, and a supermode noise suppression ratio of 51 dB. It is demonstrated that the injection-locked EDFL with a gain-switched FPLD has lower noise characteristics than the AML-EDFL.
Dispatching in semiconductor fabs with machine-dedication features
(2006-04-01)
Wu, MC
;
Huang, YL
;
Chang, YC
;
Yang, KF
;
工業工程與管理學系
This research develops dispatching algorithms for a fab with machine-dedication characteristics. Machine-dedication, a new feature in a modern fab, has not been addressed in previous studies of dispatching. Three performance indices, including hit rate, mean cycle time, and throughput are of concern in dispatching. This research develops an algorithm, called LB-SA, based on a proposed simplification model of the process route. The line balance (LB) component aims to smooth the flow rate of the process route; and the starvation avoidance (SA) component aims to ensure that the bottleneck machine is not "starving" and has enough work-in-progress (WIP) to process all the time. Thirty dispatching algorithms, including the LB-SA algorithm, are compared by simulation. The LB-SA algorithm outperforms the other 29 algorithms both in terms of hit rate and mean cycle time, and is only slightly less than the best benchmark in throughput. Of the 29 other algorithms, one algorithm, called CR-SA, also performs very well. These two algorithms are both recommended for fabs with machine-dedication feature.
Distributional and inferential properties of the process loss indices
(2004-11-01)
Pearn, WL
;
Chang, YC
;
Wu, CW
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Johnson (1992) developed the process loss index L,, which is defined as the ratio of the expected quadratic loss to the square of half specification width. Tsui (1997) expressed the index L-e as L-e = L-pe + L-ot, which provides an uncontaminated separation between information concerning the potential relative expected loss (L-pe) and the relative off-target squared (L-ot), as the ratio of the process variance and the square of the half specification width, and the square of the ratio of the deviation of mean from the tat-get and the half specification width, respectively. In this paper, we consider these three loss function indices, and investigate the statistical properties of their natural estimators. For the three indices, we obtain their UMVUEs and MLEs, and compare the reliability of the two estimators based on the relative mean squared errors. In addition, we construct 90%, 95%, and 99%) upper confidence limits, and the maximum values of (L) over cap (c) for which the process is capable, 90%, 95%, and 99% of the time. The results obtained in this paper are useful to the practitioners in choosing good estimators and making reliable decisions on judging process capability.
Dynamic characteristics measurement system for optical scanning micromirror
(2000)
Chiou, JC
;
Lin, YC
;
Chang, YC
;
電控工程研究所
;
Institute of Electrical and Control Engineering
This paper describes the development of a novel, flexible, with appropriate accuracy dynamic characteristics measurement system for optical scanning micromirror. With the system, we can measure dynamic behaviors such as transient response, scan speed, scan angle, scan repeatability, and scan non-linearity of the scanning micromirror devices. Moreover, the optical system performances such as scan spot size and even scan spot intensity can also be obtained.
Effect of catalyst on growth behavior of carbon nanotube synthesizing by microwave heating thermal chemical vapor deposition process
(2001-05-01)
Chen, SC
;
Shih, LY
;
Chang, YC
;
Tu, GC
;
Lin, IN
A modified chemical vapor deposition process, which utilizes a susceptor to absorb the microwave and to self generate the heat, for heating up the Si substrate, was used for growing carbon nanotubes (CNTs). The advantage of such a process is that the deposition chamber can be maintained at around room temperature with only the substrates localized heated, such that the deposition temperature can be more precisely controlled. The influence of the pretreatment process for catalyst-coated Si substrates on growth behavior of CNTs and the related electron field emission properties were systematically examined. Among the form of catalyst used, the Fe(NO3)(3)-ethythersilicate mixture performs much better than the de sputtered Fe films. The higher the concentration of Fe species in the catalyst mixture, the denser the CNTs formed on the subs;rates, resulting in better held emission properties. Thus grown CNTs can be turned on at a very low field (E(0)congruent to0.78 V/mum), achieving a very large emission current density (J(e)congruent to 13 mA/cm(2)) at 5.5 V/mum applied field. (C) 2001 American Vacuum Society.
Family design of scroll compressors with optimization
(2006-07-01)
Tseng, CH
;
Chang, YC
;
機械工程學系
;
Department of Mechanical Engineering
This paper proposes a systematic design method for developing a family of scroll-type compressor (STC). Except of using the general design optimization model includes multi-variable, direct search, inequality constraints, both interactive session and discrete variable design optimization skills have also been employed in this study. A practical design case on the 5200-9800 W capacity range of the STC family using R22 refrigerant has been implemented, and to achieve a common share percentage of 80% for the major components of this STC series plus a coefficient of performance based on electrical power input (COPel) of over 3.02 for each specified capacity of this family. Comparisons between calculated and measured results show that the maximum deviation of cooling capacity and COPel are below 2.53% and 1.69%, respectively. (C) 2005 Elsevier Ltd. All rights reserved.