• Proceedings of the Korean Vacuum Society Conference
• /
• 2015.08a
• /
• pp.160-160
• /
• 2015

InGaZnO (IGZO) thin-film transistors (TFTs) are very promising due to their potential use in high performance display backplane [1]. However, the stability of IGZO TFTs under the various stresses has been issued for the practical IGZO applications [2]. Up to now, many researchers have studied to understand the sub-gap density of states (DOS) as the root cause of instability [3]. Nomura et al. reported that these deep defects are located in the surface layer of the IGZO channel [4]. Also, Kim et al. reported that the interfacial traps can be affected by different RF-power during RF magnetron sputtering process [5]. It is well known that these trap states can influence on the performances and stabilities of IGZO TFTs. Nevertheless, it has not been reported how these defect states are created during conventional RF magnetron sputtering. In general, during conventional RF magnetron sputtering process, negative oxygen ions (NOI) can be generated by electron attachment in oxygen atom near target surface and accelerated up to few hundreds eV by self-bias of RF magnetron sputter; the high energy bombardment of NOIs generates bulk defects in oxide thin films [6-10] and can change the defect states of IGZO thin film. In this study, we have confirmed that the NOIs accelerated by the self-bias were one of the dominant causes of instability in IGZO TFTs when the channel layer was deposited by conventional RF magnetron sputtering system. Finally, we will introduce our novel technology named as Magnetic Field Shielded Sputtering (MFSS) process [9-10] to eliminate the NOI bombardment effects and present how much to be improved the instability of IGZO TFTs by this new deposition method.

• Proceedings of the Korea Contents Association Conference
• /
• 2007.11a
• /
• pp.586-588
• /
• 2007

The Moore's law states that the number of transistors on a chip doubles about every 18 months. And it was reported that the network speed has been doubled about every 9 months. This indicates that computing power and network is no longer the obstacles for the high performance applications requiring terabits networks. We believe that the application motivates the network and vice versa. This presentation will introduce the GLORIAD which is the first ring network connecting six countries around the world and provides scientists with advanced networking tools that improve a communications and data exchange. The GLORIAD trans-Pacific link started its service on August 1, 2005. Since then, there has been remarkable demonstrations were performed through major conferences like Supercomputing Conference. This paper will introduce the global collaborative works on demonstrations of VMT, high energy physics, SDSS and HD video transmission during SC'06 in Tampa, FL.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.6A
• /
• pp.697-704
• /
• 2004

This paper describes the design and fabrication of an MMIC variable gain LNA for 5GHz wireless LAN applications, using 0.5${\mu}{\textrm}{m}$ gate length GaAs MESFET transistors. The advantages of high gain and low noise performance of E-MESFETS and excellent linear performance of D-MESFETS are combined as a cascode topology in this design. Behavioral model equations are derived from the MESFET nonlinear current voltage characteristics by using Turlington's asymptote method in a cascode configuration. Using the behavioral model equations, a 4${\times}$50${\mu}{\textrm}{m}$ E-MESFET as a common source amplifier and a 2${\times}$50${\mu}{\textrm}{m}$ D-MESFET as a common gate amplifier are determined for the cascode amplifier. The fabricated variable gain LNA shows a noise figure of 2.4dB, variable gain range of more than 17dB, IIP3 of -4.8dBm at 4.9GHz, and power consumption of 12.8mW.

• KIISE Transactions on Computing Practices
• /
• v.20 no.9
• /
• pp.507-512
• /
• 2014

Recently, Distributed computing processing begins using both CPU(Central processing unit) and GPU(Graphic processing unit) to improve the performance to overcome darksilicon problem which cannot use all of the transistors because of the electric power limitation. There is an integrated graphics processor that CPU and GPU share memory and Last level cache(LLC). But, There is no LLC access rules between CPU and GPU, so if GPU and CPU processes run together at the same time, performance of both processes gets worse because of the contention on the LLC. This Paper gives evidence to prove the need of the Cache Partitioning and is mentioned about the cache partitioning design using page coloring to allocate the L3 Cache space only for the GPU process to guarantee GPU process performance.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.38 no.3
• /
• pp.211-219
• /
• 2001

This paper describes a read-only CMOS transponder IC for RFID applications. A full-wave rectifier implemented using NMOS transistors supplies the transponder with a dc supply voltage using the magnetic field generated from a reader. A 64-bit ROM has been designed for a data memory. Front-end impedance modulation and Manchester coding are used for transmitting the data from the transponder memory to the reader. A new damping circuit which has almost constant damping rate under the variations of the distance between the transponder and the reader has been employed for impedance modulation. The designed circuit has been fabricated using a 0.65${\mu}{\textrm}{m}$2-poly, 2-metal CMOS process. Die area is 0.9mm$\times$0.4mm. Measurement results show that it has a constant damping rate of around 20~25% and a data transmission rate of 3.9kbps at a 125KHz RF carrier. The power required for reading operation is about 100㎼. The measured reading distance is around 7cm.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.36C no.5
• /
• pp.1-12
• /
• 1999

A new algorithm of constructing binary decision diagram(BDD) for design verification of switch-level circuits is proposed in this paper. In the switch-level circuit, functions are characterized by serial and parallel connections of switches and the final logic values may have high-impedance and unstable states in addition to the logic values of 0 and 1. We extend the BDD to represent functions of switch-level circuits as acyclic graphs so called switch-level binary decision diagram (SLBDD). The function representation of the graph is in the worst case, exponential to the number of inputs. Thus, the ordering of decision variables plays a major role in graph sizes. Under the existence of pass-transistors and domino-logic of precharging circuitry, we also propose an input ordering algorithm for the efficiency in graph sizes. We conducted several experiments on various benchmark circuits and the results show that our algorithm is efficient enough to apply to functional simulation, power estimation, and fault-simulation of switch-level design.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.37 no.2
• /
• pp.29-36
• /
• 2000

The principle and design of two-stage CMOS operational amplifier with rail-to-rail input and class-AB output stage is presented. The rail-to-rail input stage shows almost constant transconductance independent of the common mode input voltage range in global transistor operation region. This new technique does not make use of accurate current-voltage relationship of MOS transistors. Hence it was achieved by using simple linear relationship of currents. The simulated transconductance variation using SPICE is less the 4.3%. The proposed global two-stage opamp can operate both in strong inversion and in weak inversion. Class AB output stage proposed also has a full output voltage swing and a well-defined quiescent current that does not depend on power supply voltage. Since feedback class- AB control is used, it is expected that this output stage can be operating in extremely low voltage. The variation of DC-gain and unity-gain frequency is each 4.2% and 12%, respectively.

• Proceedings of the Korea Association of Crystal Growth Conference
• /
• 1996.06a
• /
• pp.422-448
• /
• 1996

Low-temperature epitaxial growth of Si and SiGe layers of Si is one of the important processes for the fabrication of the high-speed Si-based heterostructure devices such as heterojunction bipolar transistors. Low-temperature growth ensures the abrupt compositional and doping concentration profiles for future novel devices. Especially in SiGe epitaxy, low-temperature growth is a prerequisite for two-dimensional growth mode for the growth of thin, uniform layers. UHV-ECRCVD is a new growth technique for Si and SiGe epilayers and it is possible to grow epilayers at even lower temperatures than conventional CVD's. SiH and GeH and dopant gases are dissociated by an ECR plasma in an ultrahigh vacuum growth chamber. In situ hydrogen plasma cleaning of the Si native oxide before the epitaxial growth is successfully developed in UHV-ECRCVD. Structural quality of the epilayers are examined by reflection high energy electron diffraction, transmission electron microscopy, Nomarski microscope and atomic force microscope. Device-quality Si and SiGe epilayers are successfully grown at temperatures lower than 600℃ after proper optimization of process parameters such as temperature, total pressure, partial pressures of input gases, plasma power, and substrate dc bias. Dopant incorporation and activation for B in Si and SiGe are studied by secondary ion mass spectrometry and spreading resistance profilometry. Silicon p-n homojunction diodes are fabricated from in situ doped Si layers. I-V characteristics of the diodes shows that the ideality factor is 1.2, implying that the low-temperature silicon epilayers grown by UHV-ECRCVD is truly of device-quality.

• 전자공학회논문지 IE
• /
• v.46 no.1
• /
• pp.16-22
• /
• 2009

Integrated circuit of a pulse-type neuron for Hodgkin-Huxley model is implemented in a $0.5{\mu}m$ 1 poly 2 metal CMOS technology. Proposed pulse-type neuron model consist of input stage with summing function and pulse generating block which make neuron pulse above threshold value. Pulse generating circuit consist of several transistors, capacitors and negative resistor with a charge supply function. SPICE simulation results show that neuron pulse is generated above threshold current of 70 nA. Measurements of the fabricated pulse type neuron chip in condition of 5 V power supply are shown and compared with the simulated results.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.40 no.3
• /
• pp.33-42
• /
• 2003

This paper describes an efficient design of an asynchronous 16-bit divider using the NST (new Svoboda-Tung) algorithm. The divider is designed to reduce power consumption by using the asynchronous design scheme in which the division operation is performed only when it is requested. The divider consists of three blocks, i.e. pre-scale block, iteration step block, and on-the-fly converter block using asynchronous pipeline structure. The pre-scale block is designed using a new subtracter to have small area and high performance. The iteration step block consists of an asynchronous ring structure with 4 division steps for area reduction. In other to reduce hardware overhead, the part related to critical path is designed by a dual-rail circuit, and the other part is done by a single-rail circuit in the ring structure. The on-the-fly converter block is designed for high performance using the on-the-fly algorithm that enables parallel operation with iteration step block. The design results with 0.6${\mu}{\textrm}{m}$ CMOS process show that the divider consists of 12,956 transistors with 1,480 $\times$1,200${\mu}{\textrm}{m}$$^2$area and average-case delay is 41.7㎱.