Browsing by Author "Lin, YM"

Now showing 1 - 19 of 19

A lossy substrate model for sub-100nm, super-100 GHz f(T) RF CMOS noise extraction and modeling
(2005)
Guo, JC
;
Lin, YM
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
A lossy substrate model is developed to accurately simulate the measured RF noise of 80 nm, super-100 GHz f(T) nMOSFETs. Substrate RLC network built in the model plays a key role responsible for the nonlinear frequency response of noise in 1 similar to 18 GHz regime, which didn't follow the typical thermal noise theory. Good match with the measured S-parameters, Y-parameters, and N-min before de-embedding proves the lossy substrate modell. The intrinsic RF noise can be extracted easily and precisely by the lossy substrate de-embedding using circuit simulation. The accuracy has been justified by good agreement in terms of I-d, g(m), Y-parameters, and fT under wide range of bias conditions and operating frequencies. The extracted intrinsic NFmin as low as 0.6 similar to 0.7dB at 10GHz indicates the advantages of super-100 GHz f(T) offered by the sub-100nm multi-finger nMOS. The frequency dependence of noise resistance R-n suggests the substrate RC coupling induced excess channel thermal noise apparent in 1 similar to 10 GHz regime. The study provides useful guideline for low noise and low power design by using sub-100nm RF CMOS technology.
A new lossy substrate de-embedding method for sub-100 nm RF CMOS noise extraction and modeling
(2006-02-01)
Guo, JC
;
Lin, YM
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
A new equivalent circuit method is proposed in this paper to de-embed the lossy substrate and lossy pads' parasitics; from the measured RF noise of multifinger MOSFETs with aggressive gate length scaling down to 80 run. A new RLC network model is subsequently developed to simulate the lossy substrate and lossy pad effect. Good agreement has been realized between the measurement and simulation in terms of S-parameters and four noise parameters, NFmin,, (minimum noise figure), R, (noise resistance), Re(Y-sopt), and Im(Y-sopt) for the sub-100-nm RF nMOS devices. The intrinsic NFmin extracted by the new de-embedding method reveal that NFmin at 10 GHz can be suppressed to below 0.8 dB for the 80-nm nMOS attributed to the advancement of f(T) to 100-GHz level and the effectively reduced gate resistance by multifinger structure.
A novel optical label swapping technique using erasable optical single-sideband subcarrier label
(2000-08-01)
Lin, YM
;
Way, WI
;
Chang, GK
;
電信工程研究所
;
Institute of Communications Engineering
To support optical MultiProtocol Label Switching (MPLS) technology in an optical network, we have demonstrated subcarrier label swapping in a 96.2-km, three-node experiment, by using optical single-sideband modulation and an optical notch filter. At each intermediate switching node, an old subcarrier label can be suppressed by 25 dB, while the burst-mode 2.5-Gb/s payload experienced only 2-dB power loss.
A physical model for hole direct tunneling current in P+ poly-gate PMOSFETs with ultrathin gate oxides
(2000-11-01)
Yang, KN
;
Huang, HT
;
Chang, MC
;
Chu, CM
;
Chen, YS
A model of the hole direct tunneling gate current accounting for heavy and light hole's subbands in the quantized inversion layer is built explicitly. This model comprises four key physical parameters: inversion layer charge density, hole impact frequency on SiO2/Si interface, WKB transmission probability, and reflection correction factor. With the effective hole mass m(oxh) = 0.5m(o) for parabolic dispersion relationship in the oxide, experimental reproduction without any parameter adjustment is consistently achieved in p(+) poly-gate pMOSFETs with 1.23, 1.85, and 2.16 nm gate oxide thicknesses. The proposed model can thereby serve as a promising characterization means of direct tunnel oxides. In particular, it is calculated that the secondary subbands and beyond, although occupying few holes, indeed contribute substantially to the direct tunneling conduction due to effective lo,vcr barrier heights, and are prevailing over the first subbands for reducing oxide field down below 1 MV/cm.
A physical model for the hysteresis phenomenon of the ultrathin ZrO2 film
(2002-10-01)
Wang, JC
;
Chiao, SH
;
Lee, CL
;
Lei, TF
;
Lin, YM
This work studies and presents an inner-interface trapping physical model for the ultra-thin (effective oxide thickness=15 Angstrom) zirconium oxide (ZrO2) film to explain its hysteresis phenomenon. The shift of the capacitance-voltage characteristics swept from accumulation to inversion and then swept back with light illumination is about 110 mV, which is larger than the shift without light illumination (similar to45 mV). The mobile ion effect is obviated using bias-temperature stress measurement. The proposed model successfully explains not only the phenomenon but also the thickness effect for the capacitance-voltage characteristics and the different turn-around voltages of the current density-voltage characteristics of the zirconium dielectrics. (C) 2002 American Institute of Physics.
Characterization and modeling of edge direct tunneling (EDT) leakage in ultrathin gate oxide MOSFETs
(2001-06-01)
Yang, KN
;
Huang, HT
;
Chen, MJ
;
Lin, YM
;
Yu, MC
This paper examines the edge direct tunneling (EDT) of electron from n(+) polysilicon to underlying n-type drain extension in off-state n-channel MOSFET's having ultrathin gate oxide thicknesses (1.4-2.4 mm), It is found that for thinner oxide thicknesses, electron EDT is more pronounced over the conventional gate-induced-drain-leakage (GIDL), bulk band-to-band tunneling (BTBT), and gate-to-substrate tunneling, and as a result, the induced gate and drain leakage is better measured per unit gate width. A physical model is for the first time derived for the oxide field fox at the gate edge by accounting for electron subband in the quantized accumulation polysilicon surface. This model relates fox to the gate-to-drain voltage, oxide thickness, and doping concentration of drain extension, Once fox is known, an existing DT model readily reproduces EDT I-V consistently and the tunneling path size extracted falls adequately within the gate-to-drain overlap region. The ultimate oxide thickness limit due to EDT is projected as well.
Design and implementation of ultra-low latency optical label switching for packet-switched WDM networks
(2000-12-01)
Meagher, B
;
Chang, GK
;
Ellinas, G
;
Lin, YM
;
Xin, W
An ultra-low latency, high throughput internet protocol(IP) over wavelength division multiplexing (WDM) packet switching technology for next-generation internet (NGI) applications has been designed and demonstrated. This method overcomes limitations of conventional optical packet switching, which require buffering of packets and synchronization of bits, rind optical burst switching methods that require estimation of delays at each node and for each path. An optical label switching technique was developed to realize flexible bandwidth-on-demand packet transport on a reconfigurable WDM network. The aim was to design a network with simplified protocol stacks, scalability, and data transparency, This network will enable the NGI users to send their data applications at gigabit/second access speed with low and predictable latency (<1 sec per switch node), with a system capacity of beyond multi-Tb/s. Packet forwarding utilizes WDM optical headers that are carried in-band on the same wavelength and modulated out-of-hand in the frequency domain.
Edge hole direct tunneling in off-state ultrathin gate oxide p-channel MOSFETs
(2000)
Yang, KN
;
Huang, HT
;
Chen, MJ
;
Lin, YM
;
Yu, MC
This paper examines the edge direct tunneling (EDT) of hole from p(+) polysilicon to underlying p-type drain extension in off-state p-channel MOSFETs having ultrathin gate oxide thicknesses (1.2 - 2.2 nm). It is found that for thinner oxide thicknesses, hole EDT is more pronounced over the conventional GIDL and gate-to-channel tunneling, and as a result, the induced gate and drain leakage is better measured per unit gate width. A physical model accounting for heavy and light hole's subbands in the quantized accumulation polysilicon surface is built explicitly. This model consistently reproduces EDT I-V and the tunneling path size extracted falls adequately within the gate-to-drain overlap region. The ultimate oxide thickness limit due to EDT is projected as well.
Edge hole direct Tunneling leakage in ultrathin gate oxide p-channel MOSFETs
(2001-12-01)
Yang, KN
;
Huang, HT
;
Chen, MJ
;
Lin, YM
;
Yu, MC
This paper examines the edge direct tunneling (EDT) of holes from p(+) polysilicon to underlying p-type drain extensions in off-state p-channel MOSFETs having ultrathin gate oxides that are 1.2 nm-2.2 nm thick. It is for the first time found that for thinner oxides, hole EDT is more pronounced than both conventional gate-induced drain leakage (GIDL) and gate-to-channel tunneling. As a result, the induced gate and drain leakage is more accurately measured per unit gate width. Terminal currents versus input voltage are measured from a CMOS inverter with gate oxide thickness T-OX = 1.23 nm, exhibiting the impact of EDT in two standby modes. For the first time, a physical model is derived for the oxide field E-OX at the gate edge by accounting for the heavy and light holes' subbands in the quantized accumulation polysilicon surface. This model relates E-OX to the gate-to-drain voltage, oxide thickness, and doping concentration of the drain extension. Once E-OX is known, an existing direct tunneling (DT) model consistently reproduces EDT current-voltage (I-V), and the tunneling path size extracted falls adequately within the gate-to-drain overlap region. The ultimate oxide thickness limit due to hole EDT is projected.
High-efficiency wide-band SOA-based wavelength converters by using dual-pumped four-wave mixing and an assist beam
(2004-08-01)
Hsu, DZ
;
Lee, SL
;
Gong, PM
;
Lin, YM
;
Lee, SSW
Wavelength conversion with a high efficiency and wide conversion range can easily be realized by applying an assist beam of short wavelength to a dual-pumped semiconductor optical amplifier (SOA). Experimental results reveal that an assist beam can improve the conversion efficiency and signal-to-background ratio by up to 8.5 and 4.1 dB, respectively, when the SOA is biased at the current that makes the SOA transparent to the assist beam. The efficiency can exceed -10 dB over an 80-nm wavelength conversion range using a 23-dB gain SOA. The assist beam increases the system power budget because it simultaneously improves the conversion efficiency and reduces the receiver power penalty. The power penalties for up-converting and down-converting 10-Gb/s signals to 25 nm of wavelength range are less than 0.5 and 2.5 dB, respectively.
Improvement of water-related hot-carrier reliability by optimizing the plasma-enhanced tetra-ethoxysilane deposition process
(1997-08-01)
Lin, YM
;
Jang, SM
;
Yu, CH
;
Lei, TF
;
電子工程學系及電子研究所
As-deposited, steam-exposed O-3-tetra-ethoxysilane (TEOS)-based intermetallic dielectric were characterized according to the LR-absorption spectra. The plasma-enhanced-TEOS films investigated here were formed by mixed-frequency plasma processes and differentiated by reaction oxidizers and/or plasma powers. Increasing stress and nitrogen or oxygen concentrations improved the moisture resistance of PE-TEOS oxide films. In addition to the moisture resistance of PE-TEOS films, another factor affecting hot-carrier robustness is the maximum thickness of the underlayer for O-3-TEOS films, which is limited by their conformity The integrated intermetallic dielectrics were evaluated by constant-current and hot-carrier stressing. In obtaining the best device reliability, a trade-off exists between moisture resistance, gapfilling capacity of the PE-TEOS underlayer, and plasma damage. Our results indicate that an O-2-rich, PE-TEOS film with a mechanical stress of 3 x 10(9) dyn/cm(2) is the optimum.
Improvement of water-related hot-carrier reliability by optimizing the plasma-enhanced tetra-ethoxysilane deposition process
(1997-08-01)
Lin, YM
;
Jang, SM
;
Yu, CH
;
Lei, TF
;
電子工程學系及電子研究所
As-deposited, steam-exposed O-3-tetra-ethoxysilane (TEOS)-based intermetallic dielectric were characterized according to the LR-absorption spectra. The plasma-enhanced-TEOS films investigated here were formed by mixed-frequency plasma processes and differentiated by reaction oxidizers and/or plasma powers. Increasing stress and nitrogen or oxygen concentrations improved the moisture resistance of PE-TEOS oxide films. In addition to the moisture resistance of PE-TEOS films, another factor affecting hot-carrier robustness is the maximum thickness of the underlayer for O-3-TEOS films, which is limited by their conformity The integrated intermetallic dielectrics were evaluated by constant-current and hot-carrier stressing. In obtaining the best device reliability, a trade-off exists between moisture resistance, gapfilling capacity of the PE-TEOS underlayer, and plasma damage. Our results indicate that an O-2-rich, PE-TEOS film with a mechanical stress of 3 x 10(9) dyn/cm(2) is the optimum.
Plasma charging damage and water-related hot-carrier reliability in the deposition of plasma-enhanced tetraethylorthosilicate oxide
(1997-07-01)
Lin, YM
;
Jang, SM
;
Yu, CH
;
Lei, TF
;
Chen, JY
This paper presents plasma-enhanced tetraethylorthosilicate oxide process (PETEOS) induced hot-carrier reliability associated with an intermetallic dielectric formed by depositing oxide films from PETEOS, ozone-TEOS, and spin-on glass (SOG), where a single or dual-frequency PETEOS film is deposited as an underlayer followed by an ozone-TEOS deposition for gap filling and a SOG coating for planarization. The impact of PETEOS process conditions such as low-and high-frequency plasma power and PETEOS film thickness on hot-carrier reliability are studied. It is observed that increasing the low- and/or high-frequency plasma power can improve the film quality and moisture-related hot carrier immunity, but at the expense of plasma charging damage. This plasma charging damage accumulates gradually and finally saturates as the deposition of oxide from PETEOS proceeds. In general, a thick PETEOS layer is desired for improving the moisture-related hot carrier immunity. As a result, there is a trade-off between the plasma charging damage and the moisture-related hot carrier immunity. The plasma charging damage is dominant in large devices and the moisture-related hot-carrier immunity is dominant in small devices.
Plasma charging damage and water-related hot-carrier reliability in the deposition of plasma-enhanced tetraethylorthosilicate oxide
(1997-07-01)
Lin, YM
;
Jang, SM
;
Yu, CH
;
Lei, TF
;
Chen, JY
This paper presents plasma-enhanced tetraethylorthosilicate oxide process (PETEOS) induced hot-carrier reliability associated with an intermetallic dielectric formed by depositing oxide films from PETEOS, ozone-TEOS, and spin-on glass (SOG), where a single or dual-frequency PETEOS film is deposited as an underlayer followed by an ozone-TEOS deposition for gap filling and a SOG coating for planarization. The impact of PETEOS process conditions such as low-and high-frequency plasma power and PETEOS film thickness on hot-carrier reliability are studied. It is observed that increasing the low- and/or high-frequency plasma power can improve the film quality and moisture-related hot carrier immunity, but at the expense of plasma charging damage. This plasma charging damage accumulates gradually and finally saturates as the deposition of oxide from PETEOS proceeds. In general, a thick PETEOS layer is desired for improving the moisture-related hot carrier immunity. As a result, there is a trade-off between the plasma charging damage and the moisture-related hot carrier immunity. The plasma charging damage is dominant in large devices and the moisture-related hot-carrier immunity is dominant in small devices.
QoS contention control for optical coarse packet switched IP-over-WDM network
(2005)
Yuang, MC
;
Tien, PL
;
Shih, J
;
Lee, SSW
;
Lin, YM
In this paper, we first introduce an Optical Coarse Packet Switching (OCPS) paradigm. In principle, OCPS advocates the enforcement of traffic engineering and control to realize bandwidth-on-demand on sub-wavelength basis while circumventing optical packet switching limitations. Based on OCPS, we have constructed an experimental optical IP-over-WDM network testbed, referred to as OPSINET. In the paper, we present the architectural design of OPSINET and its Hybrid Contention Control (HCC) mechanism aiming to satisfy various classes of loss QoS guarantees. The mechanism preventively adopts appropriate burst sizes for different traffic classes during packet burstification at ingress routers. It also reactively performs prioritized contention resolution at OLSRs in the presence of contention. Experimental results show that HCC invariantly achieves low loss probability as a result of burstification-based traffic shaping. Moreover, it provides satisfied loss guarantees for high-priority classes while incurring minimal loss degradation for lower-priority classes.
The characteristics of polysilicon oxide grown on amorphous silicon deposited from disilane
(1997-08-01)
Lin, YM
;
Lei, TF
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
The characteristics of polysilicon oxide (polyoxide) grown on amorphous silicon deposited from disilane as the silicon source are reported on. The rate of amorphous silicon deposition from disilane is comparable to that of polysilicon deposition from silane. Moreover, the obtained polyoxide has the desirable electrical characteristics of lower leakage current and higher breakdown held than those of polyoxide grown on polysilicon deposition from silane, urgently needed for nonvolatile memory application. At the same time! the grown polysilicon oxides have lower electron trapping rates and larger charge-to-breakdown (Q(bd)), both attributable to their smoother polyoxide/polysilicon-1 interface than those of polyoxides grown on polysilicon deposition from silane.
The feasibility study of transporting IS-95CDMA signals over HFC networks
(1999-06-01)
Lin, YM
;
Way, WI
;
電信工程研究所
;
Institute of Communications Engineering
This letter studies the feasibility of transmitting IS-95 code-division multiple-access (CDMA) wireless signals in hybrid-fiber-coax (HFC) networks, Capacity estimation of an HFC optical node-based CDMA cellular systems using a return-path channel model is provided. In addition to multiple-access interference in IS-95 CDMA networks, the return-path channel model takes into account ingress noise, impulsive noise, and microreflections in a coaxial cable transmission system; and laser diode clipping-induced intermodulation noise in an optical fiber link.
The feasibility study of transporting wireless CDMA signals over HFC networks
(1998)
Lin, YM
;
Way, WI
;
電信工程研究所
;
Institute of Communications Engineering
Wireless access in hybrid fiber/coax (HFC) networks can integrate two major telecommunication infrastructure, i.e., wireless and HFC networks. This paper studies the technical feasibility of integrating IS-95 code-division-multiple-access (CDMA) wireless networks and HFC networks. Capacity estimation of an HFC optical node-based CDMA cellular systems using a return-path channel model is provided. In addition to multiple access interference in IS-95 CDMA networks, the return-path channel model takes into account ingress noise, impulsive noise, and micro-reflections in a coaxial cable transmission system, and laser diode clipping-induced intermodulation noise in an optical fiber link. Furthermore, in our Monte Carlo simulation, we have assumed slow flat fading for radio propagation, and ideal timing recovery and power control.
Using superimposed ASK label in a 10-GB/s multihop all-optical label swapping system
(2004-02-01)
Lin, YM
;
Yuang, MC
;
Lee, SL
;
Way, WI
;
資訊工程學系
A novel optical-label-swapping technique is proposed, experimentally verified, and theoretically analyzed in this paper. The technique superimposes a low-speed amplitude-shift-keying (ASK) label on top of a high-speed de-balanced-line-coded ASK payload. A multihop long-distance transmission experiment using a recirculating loop has been successfully demonstrated, and an experimental record is set when compared with other optical-label-swapping techniques.