본 논문에서는 이동통신환경에서 데이터 전송시 발생되는 열화를 줄이기 위한 다이버시티(diversity)기법 중 일반적인 블록 인터리버의 성능을 분석하고 보다 나은 성능을 얻기 위하여 새로운 병렬식 크로스 인터리버(Parallel cross interleaver)를 제안한다. Upband IS-95에서 사용되고 있는 블록 인터리버는 384(16×24)의 크기인 반면에 병렬식 크로스 인터리버를 적용하는 경우 216(6×6×2+12×12)크기에서도 약 1.8dB정도 향상되었다.

The error performance of 16 Square-QAM signal with Optimum Threshold Detection(OTD) under the consideration of AWGN and Rician fading was analyzed for various value of Rician fading depth K. And error performance of 16-QAM signal with OTD, which considered in AWGN and Rician fading, was compared with that of 16-QAM signal with OTD, which considered in Rician fading only. And BCH coding is adopted. From the results, the error perfomance evlauated by proposed OTD was superior to that of 16-QAM signal with OTD, which considered in Rician fading without AWGN, in low CNR.

This paper introuces a new efficient method of synchronization acquisition which is the most important element in DS-CDMA system using the phase offset of binary code. This approach uses the binary code function which can easily estimate the phase offset from the received spreading waveforms which respect to the receiver-stored replica of the spreading code. This paper proposes the initial acquisition model with repeat error control device that is good for perfomance. The hardware is implemented by TMS320c30

본 논문에서는 광대역 CDMA용으로 제안되고 있는 유한체 GF(28) 상의 3중 오류정정 (47, 41) Reed-Solomon 복호기를 설계하였다. 복호법으로는 오류정정 능력이 비교적 작은 경우 매우 효율적인 직접복호법을 이용하였다. 설계된 복호기는 복호지연이 매우 짧으며 기존의 복호기보다 훨씬 간단한 하드웨어로 구현할 수 있는 장점을 가지고 있다.

In this study, the efficient channel coding method combined with CS-ACELP is proposed. The same convolutional coder and Viterbi decoder of COMA mobile communication system is used as channel coder. To make the best available use of limited channel coding redundancy, unequal error protection of punctured convolutional coder is used for variable reate allocation. But, the overall code rate is given by 2. The performance of proposed coder is analyzed and simulated in a Rayleigh fading channel. Experimental results show that the objective and subjective speech quality of variable rate channel coding methods are superior to those of non-variable channel coding method.

Many of the currently used PN code acquisition algorithms detect the phase of the incoming PN signal on the basis of ML estimation principle and utilize statistics grounded in taking inner products. By showing that any set of 2n-1 PN sequences arising in SSRG or MSRG (those typically used in IS'95 implementations) configuration constitutes a linearly independent set and that the number of candidate PN sequences has to equal the dimension of the span of the candidate PN sequences, we prove that the lowerbounding computational complexity involved in any PN code acquistion, utilizing (only) inner product computations at each stage of acquisition, corresponds precisely to those, such as double dwell acquistion circuitries, currently used.

비정치제도 위성통신링크에서는 지표상의 단말기, 지구국 등에 대한 상대적인 위성의 빠른 이동으로 인하여 도플러천이가 발생하며, 이는 이동위성통신 시스템 수신단의 성능을 약화시킨다. 본 논문은 비정지궤도 위성을 이용하는 통신링크 상에서 발생하는 도플러천이의 특성을 분석한다. 분석된 결과는 도플러천이에 대한 시뮬레이션 모델과 보상 알고리즘의 개발에 활용활 수 있을 것이다.

In this paper, 60 GHz indoor broadband wireless channels are measured with various configurations in a typical coffice environment. Mesurements are taken at nine positions of the room and the base-station antenna is placed either at the center or at an edge of the measurement room, and the remote-station antenna is either sharp beam or broad beam type. The rms delay spread(RDS) and normalized received power (NRP) are estimated from the measurements. Bit error rate simulations are performed using impulse responses for two measurement positions with QPSK/DQPSK OFDM modulation. Using sharp beam antenna results in superior performances than using broad beam antenna in terms of both bit error rates (BER) and NRP penalty. Also, placing the bese-station antenna at the center is superior to placing it at an edge in terms of BER and NRP.

Signal Attenuation in the design of satellite channel is due mostly to rain. In this paper, the rain rate characteristics based on the recent year(1994-1996) data for Changwon-Masan were approximated to Moupfouma distribution, and the parameters of distribution were calculated from a simple conversion method of 1-h rate to 1-min rain rate data. From rain rate characteristics, the rain attenuation was estimated and compared with CCR model, Global model and SAM model.

Due to the increasing demand to multimedia services over wireless communication network, LMDS(Local Multipoint Distribution Service) has been proposed as an alternate approach for broadband wireless access network. But LMDS, Which uses the mmwavelength radio spectrum, has been limited for unknown microwave frequency characteristics. In this paper, we analyze the parameters related to the system design on 28GHz frequency: Propagation path loss and rainfall attenuation. Based on these analytical results, the proper cell size in Korea is determined. We also analyze the bit error rate and frame loss rate for various kinds of digital modulation schemes(QAM 4, 16, 64, 256). Finally, we apply to ㅍ service to evaluate a call blocking rate, efficiency and channel capacity.

Adaptive array antenas have emerged as a useful technique to enhance the cell capacity of mobile communications. In this research, to analyze the noncoherent CDMA systems employing adaptive array antennas, we modeled the transmitting signal of CDMA systems using M-ary orthogonal modulation. And we induced the conditional probability density function about the decision variable, the output of 2D-RAKE receiver and mean symbol error prabability through statistical analysis about MAI(Multiple Access Interference), SI(Self Interference) and Noise. Also, we analyzed the charateristics of adaptive array antenna for noncoherent CDMA systems using M-ary orthogonal modulation according to the distance between the array elements, doppler frequency and AOS(Angle of Spread).

In this paper, we introduce the notion of intelligent communication by introducing a novel intelligent receiver model. This receiver is continually evolving and learns and improves in performance as it compiles its experience over time. In digital communication context, in a typical training mode, it jearns the concept of "1" as is deteriorated by arbitrary (not necessarily additive as is typically assumed) disturbance and /or modulation. After learning "1", in test mode, it classifies the received signal "1" and "0" almost completely. The intelligent receiver as implemented is grounded on the recently introduced Stochastic Morphological Sampling Theorem(SMST), a distribution-free result which gives theoretical bounds on the sample complexity(training size) needed for the required performance parameters such as accuracy($\varepsilon$) and confidence($\delta$). Based on this theorem, we demonstrate --almost irrespective of channel and modulation model-- the number of samples needed to learn the concept of "1" is not too "large" and the resulting universal receiver structure, that corresponding to classical Nearest Neighbor rule in Pattern Recognition Theory, is trivial. We check the surprising efficiency and validity of this model through some simple simulations. and validity of this model through some simple simulations.

In this paper, the modified Usage Parameter Control(UPC) method based on Dual Leaky Bucket algorithm is proposed to improve the perfor mance of multimedia synchronization and to control burst traffic. In the proposed algorithm, Synchronous cells are by passed to the 2nd stage Leaky Bucket (LB) of the DLB. for overloaded traffic condition, the violated cells are saved in a temporary buffer and they are retransmitted when the 2nd stage bucket is empty.

통신망의 물리적 기반이 광케이블로 대치됨에 따라 광케이블 선로의 품질유지 및 체계적인 관리에 대한 요구도 점차 증대되고 있다. 한국통신은 광케이블 선로의 유지보수 체계를 확립하고 자동화 하기 위하여 광선로 운용 감시 시스템(FLOMS:Fiber Line Operation & Monitor System, 이하 FLOMS라 한다.)을 개발하였다. FLOMS는 광케이블 시설에 대한 체계적인 관리 및 광케이블 선로 특성시험 자동호를 통한 선로의 이상 유무를 체크하고 보고할 수 있다. 광케이블 유지보수 요원은 FLOMS 시스템을 이용하여, 운용센터에서 관리대상 전화국의 광케이블 선로들을 원격시험 할 수 있으며 기존의 광전송장치 운용관리 시스템과 연동함으로써 전송분야 운용관리의 통합관리가 가능하게 되었다. 이 시스템을 적용함으로써 광케이블 선로 고장위치의 신속한 파악에 의한 공장시간 단축, 광케이블 선로 특성의 주기적인 측정에 의한 예방보전, 광케이블 심선 정보의 전산화 관리에 의한 업무효율의 향상을 이룩하였다.

In this paper, we analyzed SSL(Secure Socket Layer) protocol, symmentric algorithm is borken by choosen attack and known attack. This protocol prevents SSL form reusing a key and make new key without computational load of public-key algorithm.

This paper presents a method and implementation of the parallel distributed sample scrambler(DSS) in the cell-based ATM transmission environment. In the serial processing, it requires very high speed clock because the processing clock of the serial DSS is equal with the data transmission speed. In this paper, we develop a conversion method of the serial SRG(shift register generator) to 8bit parallel realization. In this case, it has a sample data processing problem which is a character of DSS. So, a theory of correction time movement is presented to solve this problem. We has developed a ASIC using this algorithm and verified the recommendation of ITU-T, I.432.

The synchronous stream cipher has the advantage that one bit error in the ciphertext only affects the corresponding bit in the plaintext, but it requires the perfect synchronization between encryptor and decryptor. For synchronization, a periodic resynchronization has been used in many applications. In this paper, we propose the periodic resynchronization scheme for radio secure communication and evaluate the performances according to the period of sync pattern and session key under radio channel environment having 10-2~ 10-6 BER.

This paper present a new priority control mechanism which is to balance the cell loss rate by measuring, in real time, the number of discarded cells in the queuing system with a different loss priority for each class of service such that each class of service meets its cell loss rate requirements. And, to reduce the delay rates we modified existing cell scheduling scheme. Throughout the computer simulation, the existing methods and proposed scheme is compared with respect to cell loss rate and average delay time. In the result of simulation, the proposed scheme have more improved performance than the other schemes with respect to cell loss rate and average delay time.

We consider connection-oriented wireless cellular networks. These networks employ dedicated radio channels for the transmission of signaling information. A forward signaling channel is a common signaling channel assigned to carry the multiplexed stream of paging and channel allocation(virtual circuit allocation) packets from a base station to mobile stations. The delay levels experienced by paging and channel allocation packets have serious effect on the utilization level of the limited radio channel capacity. While a slotted mode operation is used to reduce the power consumption level at mobile stations, it may induce an increase in packet delay levels. In this paper, we thus consider a multiplexing scheme for paging and channel allocation packets under which slots are dynamically allocated for the paging packet transmission. For this dynamic scheme, we develop an analytical method for deriving the delay characteristics exhibited by paging and channel allocation packets, and investigate the effect of network parameters on the delay level by using this method.

본 논문에서는 위치 등록 빈도의 증가를 해결할 수 있는 클러스터 개념을 이용한 위치 등록 프로토콜의 성능을 분석하고, 클러스터 크기와 시간의 흐름에 따른 위치 등록 확률을 계산하여 클러스터 크기와 위치 등록율의 관계를 정량적으로 분석하였으며, 실제로 육각형 셀 구조에 적용하여 위치 등록 확률을 클러스터의 크기에 따라 구해 보고 시뮬레이션을 통해 이를 검증하였다.

Because ATM was originally designed for the optical fiber environment with bit error rate(BER) of 10-11, it is difficult to maintain ATM cell extraction capability in wireless environment where BER ranges from 10-6 to 10-3. Therefore, it must be proposed the algorithm of ATM cell extraction in wereless environment. In this paper, the frame structure and synchronization algorithm satisfyling the above condition are explained, and the new ASIC implementation method of this algorithm is proposed. The known method using shift register needs so many gates that it is not suitable for ASIC implementation. But in the proposed method, a considerable reduction in gate count can be achieved by using random access memory.

Multiple types of information in a multimedia service are delivered though multiple virtual connections on ATM network, while each virtual connection may be controlled independently. A multimedia service requires an associative relationship among multiple information streams to provide required harmonization. There may be required additional traffic descriptors to guarantee the required harmonization among multiple information streams in a multimedia service. For buffering of large bandwidth information stream(e.g., video), extremely large buffer size is necessary, but this approach should not be efficient way to compensate a severely delayed cells/blocks experienced at network. The best way to solve this problem will be minimization of relative-delayed-transfer of cells/blocks to application processes through ATM network control. To minimize a delayed transfer the mapping between relative delay parameter(i.e., associative Group QoS parameters) and per-VC traffic descriptor will be necessary. This paper is present additional functions and parameters to interpret the mapping between relative delay parameters(i.e., associative Group QoS parameters) and per-VC traffic descriptors in ATM API for multimedia application services.

컴퓨터 통신에서 실시간 멀티미디어 응용은 보장된 통신 서비스 성능을 요구하므로 이를 위해서는 트래픽이 버스트할 때 망 내에서 자원을 예약하여 접속시 피크율을 기반으로 망의 이용율을 높인다. 본 논문에서 결정론적 지연 범위을 이용하여 모든 접속에서 피크 율의 합이 링크 속도 보다 더 커질 때 버스트 상태의 트래픽에 대한 링크를 넘어 보장될 수 있음을 보인다. 또한 트래픽이 버스트할 때 새로운 결과는 연결수락된 접 속의 수를 증가시키므로 이러한 결과는 스위치에서 속도가 제어된 서비스 규율을 사용함으로써 단일 스위치로부터 망의 이용율을 개선할 수 있다.

In this paper we derive the pdf for the envelope of the received signal in Nakagami fading environment. Then, using this pdf we analyze the performance of Noncoherent FSK and DPSK in Nakagami fading channel. For ASK it is difficult to find the threshold value in Nakagami fading we confine our discussion in the special case of m=1, which is in Rayleigh fading.

The intermodulation distortion(IMD) due to laser diode nonlinearity of an asynhcronous direct sequence code diviion multiple access(DS/CDMA) system in wireless-optical transmission system model is analzed. A third order polynomial is used to represent laser diode nonlinearity. In DS/CDMA system, only one harmonic of the third-order intermodulation term falls on the signal frequency band and influences the system performance characteristics. To cancel multi-user interference and nonlinear distortion in a DS/CDMA wireless-optical transmission system model, the simple transversal filter structure with N-taps of (N-1) tap delay of 1 chip time delay line is used. It is necessary to select an optimal modulation index that provides a maximum signal-to-noise ratio and the results are useful for CDMA system design.

차세대 이동통신 시스템을 위해 활발히 연구되고 있는 MC-CDMA 방식은 OFDM과 DS-CDMA의 장점을 결합하여 심벌간 간섭과 페이딩에 강하고 IFFT/FFT를 이용하여 효과적으로 변조/복조부를 구현할 수 있다는 장점을 갖는다. 하지만, 송신 신호가 다중 레벨 진폭 특성을 갖기 때문에 손신기에서 사용되는 고출력 증폭기의 포화 특성으로 인해 심각한 비선형 왜곡이 발생한다. 본 논문에서는 OFDM에서 이러한 비선형 왜곡의 보상을 위해 우리가 제안한 고정점 반복 기반의 사전왜곡기를 MC-CDMA 시스템에 적용하고, 이의 우수한 성능을 부가성 백색 가우시안 잡음 채널에서 64-FFT/IFFT을 이용하여 변조 및 복조를 수행하는 MC-CDMA 시스템에 대한 모의실험을 통해 확인하였다.

Wireless communication systems, in particular, must operate in a crowded electro-magnetic environmnet where in-band undesired signals are treated as noise by the receiver. These interfering signals are often random but not Gaussian Due to nongaussian noise, the distribution of the observables cannot be specified by a finite set of parameters; instead r-dimensioal sample space (pure noise samples) is equiprobably partitioned into a finite number of disjointed regions using quantiles and a vector quantizer based on training samples. If we assume that the detected symbols are correct, then we can observe the pure noise samples during the training and transmitting mode. The algorithm proposed is based on a piecewise approximation to a regression function based on quantities and conditional partition moments which are estimated by a RMSA (Robbins-Monro Stochastic Approximation) algorithm. In this paper, we develop a diversity combiner with modified detector, called Non-Linear Detector, and the receiver has a differential phase detector in each diversity branch and at the combiner each detector output is proportional to the second power of the envelope of branches. Monte-Carlo simulations were used as means of generating the system performance.

An analysis of the relationship between a linear amplifier chain and an analog-to-digital converter(ADC) in a digital microwave widevand receiver, with respect to sensitivity and dynamic range issues, is presented. The effects of gain, third-order intermodulation products and ADC characteristics on the performance of the receiver are illustrated and design criteria for the linear amplifier chain given a specified ADC are developed. A computer program is used to calculate theretical receiver performance based on gain and third-order intermodulation product selections. Simulated results are also presented and compared with theoretical values.

This paper highlights new IP network design and internetworking scenarios overlayed on the existing ATM technology network. It describes the proper network configurations to solve the IP/ATM internetworking probrems on the AISN(Advanced Information Superhighway Network). The paper outlines that IP network can be constructed on ATM backbone network. The paper looks at how new design technology is to enhance the AISN in Korea. The paper concludes that new IP network design is applied to AISN and there is enough possibility that AISN can be developed to efficient NGI(Next Generation Internet).

Standardized representations and data codes are crucial for the exchange of information between systems, organizations, and people in the telecommunications operations area. ITU-T M.1400 recommends a unified representation for the international network routes which includes such location data elements as Town A, Suffix Code which was designed for the representation of domestic network locations to meet the ITU-T recommendation. This paper suggests an expanded building ID code for the representation of the international network locations.

본 논문에서는 인터넷텔레포니 인터페이스(Internet Telphone Interface)를 제작하고 자체 개발된 PC-to-PC 방식의 인터넷 전화 소프트웨어인 Realphone을 이용하여 Phone-to-PC 방식의 인터넷 전화를 구현하였다. 실험을 위하여 인터넷텔레포니 인터페이스를 제작하였으며, 펜티엄 PC 2대와 일반 상용화 되어있는 전화기를 사용하였다. 통화는 목원대학교 학내전화와 목원대학교내 LAN으로 연결된 PC사이에 로컬(Local)상황과 국내 교육망으로 연결된 목원대학교와 서울대학교 사이에 트래픽이 서로 다른 새벽(Unload), 오전(Mid-load), 오후(Load)에 이루어졌다. 실험을 통하여 대형 게이트웨이의 도움없이 간단한 인터페이스만으로 Phone-to-PC 방식 인터넷전화의 통화가 가능함을 확인하였다.

In order to control the flow of traffics in ATM networks and optimize the usage of network resources, an efficient control mechanism is necessary to cope with congestion and prevent the degradation of network performance caused by congestion. This paper proposes a new UPC(Usage Parameter Control) mechanism that varies the token generation rate and the buffer threshold of leaky bucket by using a Neural Network controller observing input buffers and token pools, thus achieving the improvement of performance. Simulation results show that the proposed adaptive algorithm uses of network resources efficiently and satisfies QoS for the various kinds of traffics.

A imput buffer typed ATM switch and an appropriate cell-scheduling algorithm are necessary for avoiding output blocking and internal blocking respectively. The algorithm determining a set of non-blocking data cells from the queues can greatly affect on the switch's throughput as well as the behavior of the queues. In this paper bit pattern optimization combined with the Token method in presented in order to improve the performance of ATM switch. The digital Hopfield neural cell scheduler is designed and used for the maximum numbers of cells in real-time

In this paper, we propose a new consolidation algorithm for point-to-multipoint ABR flow control and compare the performance of the proposed algorithms with that of an existing consolidation algorithm. We first investigate problems of the existing algorithm in consolidating backward RM cells and propose two new consolidation algorithms to resolve these problems. Also, we investigate the performance of the proposed algorithms by simulation. As results, we show that the proposed algorithms exhibit a better performance than the existing algorithm in terms of ACR, queue length, utilization, and fairness.

An L-band omnidirectional circular polarized wave antenna is designed and evaluated for transmitting/receiving of vehicle data. Conventional linear polarized wave antenna can not reveive clearly all of the vehicle data which come from the wide driving test range on account of distortion. To overcome this problem, an omnidirectional circular polarized wave antenna is required for the design, first of all, the characteristics, design principle and theoretical background of circular polarized wave with a little signal loss have been reviewed. The designed antenna characteristics an analysed and compared to the desired ones. Our results demonstrate that the strength of vehicle data is flat enough over the full test range using this new antenna.

In this paper, a microstrip antenna is designed using a rectangular patch. To find characteristics of the antenna, computer simulations of the rectangular single microstrip patch antenna are performed with changing width. And we compared the result of computer simulation with the experimental value. Through the results, we found that the 3-D FDTD method is an effective method for designing microstrip patch antenna. According to simulation the resonant point has been found it in the frequency received from GPS satellite. It is thought that make it match by adjusting the feedpoint.

A synchronization is very important element not only wire communication but also wireless communication. A synchronous oscillator(SO) is a network which synchronizes, tracks, filter, amplifies and divides (if necessary) in a single process. The coherent phase synchronous oscillator(CPSO) is created by adding two external loops to the SO. The CPSO ratains all virtues of a SO while providing coherency throughout the tracking range. This paper has applied a clock recovery of GPS signal using CPSO.

구형도파관의 넓은 면에 축방향 슬롯을 파고 또 각각의 슬롯을 축방향으로 배열하였 때의 원거리 방사패턴을 구하는 과정을 보였으며 측정된 결과와 계산결과를 비교하였다. 방사전자계는 슬롯의 표면에 유기되는 등가의 자기전류로부터 계산할 수 있으며 임의의 개수를 가지는 배열안테나의 경우에는 각각의 슬롯이 자기 어드미턴스와 상호결합에 의한 영향을 받기 때문에 원하는 방사패턴을 얻기 위해서는 상호 어드미턴스를 고려하여 슬롯의 길이와 오프셋을 고려하여야 한다. 슬롯에서의 전계분포 해석은 모멘트법(method of moment)을 사용하여 해석하였다. 본 연구에서는 직선편파특성을 가지는 슬롯 배열 안테나의 설계를 위하여 슬롯의 자기 및 상호 어드미턴스를 구하고, 반복적인 수치해석 과정을 통하여 입력임피던스의 최적화 방법에 대하여 논하였다.

The analysis of a coaxial cable with periodic slots as a leaky-wave antenna is considered. The mode matching method and method of moments with Galerkin testing procedure is then used to obtain the determinantal equation for the unknown leaky-wave propagating constant. Both technigues allow the fields to be obtained in any region, and the radiation characteristics of the structure. By resorting to a suitable numerical techniques, it is possible to calculate the leaky-wave propagating constants and the radiation patterns. Numerical results demonstrate the validity of the technique.

광대역 주파수 상에서 외부로부터 입사하는 전자파를 산란시키지 않고 유전체 내부에서 흡수시키기 위해 유전체를 다층으로 배열하여 전파 흡수체를 최적 설계하고자 한다. 수직 및 여러 각도로 전파가 입사하는 경우 각 유전층의 두께, 유전 상수, 손실 탄젠트 등의 설계변수를 유전자 앨거리즘을 이용하여 최적화한다. 다극함수에서의 부분 초기화 유전자 앨거리즘의 성능 향상을 위해 부분 초기화율, 부분 초기화 시점, 스케일 인자의 변화에 따른 성능을 비교, 개선하여 흡수체 설계에 적용하였다.

This paper describes the characteristics of CT-2 radio wave propagation at 900MHz range. Typical cell coverage of CT-2 based on picocellular design provides a 200m radius. The five sites of urban regions of Kwang Ju city were chosen to characterize propagation in various environments. On the basis of cell coverage data measured with field strength meter, average 75% of telepoint service was retained and it was lack 25% of overall gain. In some sites telepoint service was limited to two of four directions around a station. If specially designed collinear antenna with specific directivities are used in these shadow area, improved results will be acquired.

A 3D-ray tracing using triangular ray tubes for predicting propagation in indoor environments is presented. Employed ray tracing scheme needs no reception sphere often suffered from how to assign the correct radius as a touching ray on the receiver. To verify the developed codes path loss for a rectangular corridor has been computed, measured, and compared with those by image methods, all shows good agreement to each other. Discussions are made on the path loss fluctuations along the distance in a rectangular corridor having a conducting knife.

The surface wave launcher from the dielectric coated coaxial cable to dielectric slab is investigated. FDTD method using local subcell and contour-path model is applied to determine the fine geometrical features. The reflection coefficient in coaxial cable region is found using extract algorithm. In this paper, two structures are presented as coaxial slot surface wave launcher. One structure has a vertical launching angle, and the other has an arbitrary launching angle. The numerical results show that a certain launching angle is minimized the reflection coefficient.

In high frequency environment such as microcell, it is requried to adopt a new method different from the empirical channel modeling applied in macrocell. In this thesis, measurement for residential area in Sinrimdong and the result of the simulation are compared and analyzed. Simulation technique utilizes 3D ray launching method to consider some factors which were not included in the ray tracing method. Propagation environment in the urban area shows the typical channel characterisitics of the propagation environment. Using 3D ray launching method, pathloss of the wave and delay characteristics of the signal with respect to the height of the transmitter is researched. If the transmitter is located on the top of the building, radiowave experiences diffraction on the rooftop and it influences the total received signal strength.

본 연구에서는 단층구조 평면형 슬롯 도파관 배열 안테나에 있어서 전력분배기로 이용되는 유도성 포스트가 있는 도파관 π-결합의 특성과 각 포트로 나누어지는 전력의 주파수 특성을 고찰하였다. π-결합은 한 개의 결합창을 통해서 두 개의 도파관으로 전력을 분배하여 결합된 도파관들을 동위상·동진폭으로 급전시킬 수 있다. 전력분배 특성은 Galerkin법을 사용해서 해석했고 유도성 포스트를 붙임으로써 π-결합으로부터의 반사는 -27dB이하로 억제할 수 있음을 확인하였다.

A three-waveguide coupler consisted of a central guide and two side guides is theoretically explored by using modal transmission-line theory. The numerical results reveal that the coupling length Lc is different from that calculated by coupled-mode theory and the difference increases gradually as the guiding modes increase.

In this paper, we analyzed Microwave-DC conversion efficiency for the rectennas and it's position change. Rectenna consist of a two major parts, receiveing antenna and rectifying circuits. We made two types of 2.45C rectennas which the dipole and the patch antenna. Rectifying circuit is a GaAs-schottky diode with a large forward current and reverse breakdown voltage. The results of RF-DC conversion efficiency for two rectennas, patch type has 75.6% efficiency with 400$\Omega$ load resistor and dipole type has 69.75% efficiency with 360$\Omega$ load resistor. When the rectennas has optimal load resistor, Rectenna efficiency shows of $\pm10%$ at $70^{\circ}$~$110^{\circ}$ position.

본 논문에서는 DAVIC 표준안에 근거하여 주파수 대역폭이 하향스트림 120MHz와 상향스트림 60MHz로 디지털 LMDS 서비스 망이 구축되었을 때, 하향스트림과 상향스트림 구조 및 프레임 관계에 대하여 기술하고, 이를 기반으로 하여 LMDS 상에서 멀티미디어 서비스 제공시 음성:전용회선:데이터를 10%:30%:60% 비율로 가정하고 서비스별 수용 가능한 가입자 채널 용량을 산출하였다.

LMDS시스템을 이용하여 다양한 멀티미디어 서비스를 제공하기 위한 기지국의 위치 선정시 필요한 몇 개의 셀 모델에 대하여 동일채널간섭 현상에 대하여 분석하여 국내의 무선 멀티미디어 통신시스템의 기지국 치국을 위한 참고자료로 활용될 수 있도록 하였다.

This paper describes the procedure of getting information about auto-tuning of PID regulator by the injection of high step input, called limited input, during a transient time of control. The key point is that system identification and control could be continuously executed. This means that the system information obtained by limited input despite of system uncertainty can be continuously applied to the PI regulator. Simulation and experiment result of brushless DC motor system having monotone increasing step response demonstrate the usefulness of proposed auto-tuning algorithm.

Brushless servomotor systems offer a great advantae in unmanned factories where a great number of servomotors are employed, because of its easy maintenance charateristics and controllability. This paper propose a sinusoidal type brushless DC motor speed servo system which has high efficiency and usability in industrial field as described above. And this servo system is realized by a new Auto-Tuning PI control algorithm and verified it's practical potential of implementation by self-designed driver system. In particular, DSP(digital signal processor) is adopted in this study as controller and sensor signal processor owing to their fast computational capability and suitable architecture.

A detection filter assigns a specific direction to the response with respect to each fault, by which it can detect the occurrence of the several faults. The separability of a detection filter can be determined by the orthogonality among these directions. In this paper, we define the separability of a detection filter as the orthogonality of the directions in output space, and present it mathematically by using conditions number. An algorithm to determine the optimal sensor gain to maximize separability is proposed and applied to the PWR nuclear reactor model.

In this paper, we design the $H_{\infty}$ controllers satisfying robust stability and performance for underwater vehicle. The underwater vehicle has computations delay time and input delay. In addition, there exist parameter uncertainties by the roll motion coefficient error, buoyance error, and gravity error. We design the $H_{\infty}$ controllers using model-matching method and check the performance of the proposed controller by nonlinear simulation which includes time delay model, sensor error model, and actuator model.

This paper presents the Mixed H2/H$\infty$ controller design method with the regional pole placements for underwater vehicle. Since the small and light underwater vehicle is sensitive to disturbances and parameter uncertainties, we design the controller which guarantees robustness against time-delays, parameter uncertainties and disturbances. The LMI(linear matrix inequality) formulations for pole placements in specific regions and H2 and H$\infty$ performances are reviewed. The desired controller can be obtained by solving these LMIs.

The control objective of steam generator water level in the secondary circuit of a nuclear power plant is to regulate the water level at the desired set point. The dynamics of steam generators is non-linear in nature. The task of modelling such plant is very difficult and especially so when plant operating conditions change frequently. In these reasons, conventional PI gains over all pover range will not work efficiently and a manual control is generally used in low power operation. Therefore the robust H$\infty$ controller design method should be required. In this paper, we design the robust H$\infty$ controller for water level control of steam generator using a mixed H$\infty$ optimization with model-matching method. Firstly we choose the desired model that has good disturbance rejection performance. Secondly we design a stabilizing controller to keep the model-matching error small and also provide sufficiently large stability margin against additive perturbations of the nominal plant.

In this paper, we propose the design method of fuzzy robust H$\infty$ controller for the uncertain nonlinear discete-time systems with time delay. First, we represent a nonlinear plant with a modified T-S(Takagi-Sugeno) fuzzy model. Then design method utilizing the concept of PDC (parallel distributed compensation) is employed. For the modified T-S fuzzy model with uncertainty and delay, the sufficient condition of the quadratic stabilization with an H$\infty$ norm bound is presented in terms of Lyapunov stability theory and fuzzy robust H$\infty$ controller design method is given by LMI(linear matrix inequality) approach. Also an illustrative example is given to demonstrate the result of the proposed method.

In this paper, we proposed a adaptive controller which need not adjustment of the scale factor. We have verified by experimental results that applying this proposed adaptive control algorithm to a hydraulic servo system is very satisfactory.

This paper presents the detection of internal combustion engine's multiple misfire. The primary cause of air pollution by vehicles is imperfect conbustion of fuel. The CARB(California Air Resources Board) have imposed regulations for the detection of misfiring in automotive engines. The OBD-II regulations requir that misfire should be monitored by the diagnostic system, and that the goal of OBD-II is to alert the driver to the presence of a malfunction of the emission control system. Present invention based upon measurements of engine roughness as derived from crankshaft angular velocity measurements with special signal processing method. Crankshaft angular velocity signals are processed by walsh-fourier transform. Experimental work confims that it's possible to apply walsh-fourier transform for the detection of multiple misfires in no-load idle and road testing.

Most of time delay existing in industrial processes can be modeled as an input time-delay which causes a lot of difficulties. To implement a time-delay free control, loop transfer function is recovered through the direct state feedback. For the proper state feedback, and observer is designed based upon the factorization approach which is known as one of the most powerful tool for the design of various control systems. The design scheme is parctically applied for the control of a DC motro which suffers from input time-delay. Simulational results demonstrate that this new algorithm guarantess the stability of the time-delayed system, as well as performance improvement compared to the conventional PI control.

A predictive controller is designed based upon stochastic methods for compensation of network time delay which caused by the spatial separation between controllers and actuators. Current commands are generated by using time varying probability functions which can be defined according to the values of previous control inputs and actual outputs. To demonstrate the effect of this control methodology, simulation experiments are performed. The results show that even an unstabilized system by a long time delay can be stabilized with this predictive controller.

In this paper, a special kind of timer-controlled network is analyzed. A timer-controlled network with an immediate response scheme is modeled and analyzed. It is assumed that each station on the network has single buffer and individual finite number of maximum retransmission limit. The physical meanings of derived results are briefly explained, which prove that the results completely coincide with the physical operations of that kind of network. It is also shown by computer simulations that derived analysis results are valid.

This paper suggests two data exchange protocols (DEPs) using shared memory. They support reliable data exchanges among components which are located in the same rack and use the same backplane bus and shared memory.

이동로봇이 움직이는 물체를 추적하는 기능을 갖도록 하기 위해서는 움직이는 물체의 존재를 판단하고 위치 및 속도벡터를 측정해야 한다. 본 논문은 단시간에 이동 물체의 위치와 속도벡터 추정하여 실시간 추적이 가능하도록 하기 위해 초음파센서를 이용한 가상의 초음파영상 해석방법을 제안한다. 초음파 영상은 초음파센서의 매회 측정값을 누적하여 구성하며 거리축과 시간축을 갖는 2차원 영상으로 표현된다. 영상에서 측정된 물체가 만드는 선분의 기울기에 따라 이동하는 물체와 정지한 물체를 분리하고 속도를 측정한다. 제안된 방법은 시뮬레이션을 통해 타당성을 입증하였다.

로봇 손이 자율적으로 동작하는데 있어서 촉각 센서는 매우 중요한 역할을 한다. 지금까지 로봇 손가락을 위한 광촉감 센서 융합 시스템들이 제안된 바 있으나 대부분 복잡한 시스템 구조로 인한 물리적인 한계로 소형화가 어렵다는 점이 커다란 문제였다. 본 논문에서는 복잡한 구성의 힘/역각 센서를 대치할 수 있는 3점 힘센서를 구성하여 힘백터를 측정하는 기법을 제안한다. 이 센서는 힘/역각 센서보다 간단한 구성으로 경제적으로 제작할 수 있다는 장점이 있으며 소형화 할 수 있기 때문에 로봇 손을 정밀 작업 등에 응용할 수 있게 할 것이다.

This paper propose a prefilter type inverse control algorithm using chaotic neural networks. Since the chaotic neural networks show robust characteristics in approximation and adaptive learning for nonlinear dynamic system, the chaotic neural networks are suitable for controlling robotic manipulators. The structure of the proposed prefilter type controller compensate velocity of the PD controller. To estimate the proposed controller, we implemented to the Cartesian space control of three-axis PUMA robot and compared the final result with recurrent neural network(RNN) controller.

Unequal wheel-radius causes odometry errors which may be increased unbounded. This paper deals with the practical method for modification of wheel-radius through experiments. This can increase the robot's odometric accuracy. Experimental results are presented that show improvement of odometric accuracy.

Ultrasonic sensors are becoming indispensable components in every sector of automation equipments due to many advantages. But the main purposes of the noncontact sensing device are rather narrowly confined within object detection and distance measurement. To widen the realm of the applications to object recognition, ultrasonic sensors need to improve the recognition resolution to a certain amount. To resolve the problem of spatial resolution restriction, an increased number of the sensors in the forms of a linear array or 2-dimensional array of the sensor has been used. Also better resolution has been obtained by shifting the array in several steps using mechanical actuators. For an object recognition using ultrasonic sensors, measurements of distance, shift, oblique angle in certain ranges should be obtained. But a little attention has been paid to the measurement of angles. In this paper we propose a practical method for an object angular value detection in addition to distance measurement in ultrasonic sensor array system with little additional hardware burden. Using the established measurement look-up table for the variations of distance, shift, angle and transmitter voltages for each sensor characteristics, a set of different return echo signals for adjacent receivers are processed to provide enhanced angular value reading for an object.

The paper provides an overview of those requirements. A detailed description is given of the construction of the new channel code, called 24/25 code, that compiles with the given constraints and involves only a minor drawback in terms of the overhead needs. The servo position information is recorded as low frequency componets, pilot tracking tones, which are embedded in the recorded stream of binary digits. Pilot tracking Tones are used to derive head position reference information in camcorders and DVCRs. A simple pilot tone encoder has been designed by using a new approach, "2 path precoder". Owing to this method, the hardware size can be significantly reduced. the correctness of the method has been verified by theoretical analysis and by extensive simulation.imulation.

In this paper, a RS(Reed-Solomon) docoder is designed in the transform domain instead of most time domain. The transform RS decoder have simpler structure for error-correction procedure but because of his larger chip area, the time domain RS decoder is popular currently. To solve this proplem, the nomal basis representation and the conjugate property is utilized. Therefore the chip area can be reduced for the stucture of syndrome delay, nomalization and inverse transform circuit. These modified structures have been implemented using VHDL and synthesized on 0.8${\mu}{\textrm}{m}$ CMOS technology. The results have been compared with other structure for chip area and performance.

Viterbi docoder is a maximum likelihood decoding method for convolution coding used in satellite and mobile communications. In this paper, a Viterbi decoder with constraint length of K=7, 3-soft decision and traceback depth of $\Gamma$=96 for convolution code is implemented using VHDL. The hardware size of designed decoder is reduced by 4 bit pre-traceback in the survivor memory.

Trees are the underlying structure for divide-and-conquer algorithms and the graphs that provide the solution spaces for NP-complete problems. Complete binary trees are the basic structure among trees. Therefore, if complete binary trees can be embedded in binomial trees, the algorithms which are provided by complete binary trees can be performed efficiently on the interconnection networks which have binomial trees as their subgraphs or in which binomial trees can be embedded easily. In this paper, we present dilation 2 embedding of complete binary trees in binomial trees.

In this paper, we propose a matrix hypercube graph as a new topology for parallel computer and analyze its characteristics of the network parameters, such as degree, routing and diameter. N-dimensional matrix hypercube graph MH(2,n) contains 22n vertices and has relatively lower degree and smaller diameter than well-known hypercube graph. The matrix hypercube graph MH(2,n) and the hypercube graph Q2n have the same number of vertices. In terms of the network cost, defined as the product of the degree and diameter, the former has n2 while the latter has 4n2. In other words, it means that matrix hypercube graph MH(2,n) is better than hypercube graph Q2n with respect to the network cost.

A honeycomb network is considered as a multiprocessor interconnection network. The network cost defined as the product of degree and diameter is better for honeycomb networks than for other types of meshes. Parallel paths offer us bypasses when the networks have faulty nodes or edges. In this paper, we present a method to construct a parallel path in honeycomb network, and show that the fault diameter for the honeycomb network is d+8, where d is diameter of the honeycomb network.

Now, the world is faced with radical changing in the new rvolution which is followed by speading and utilization of internet. The domain names which are currently used on the internet are unique throgh out the world. Therefore, U.S government positively intervenes in the management of the internet domain name for the effective implementation. The thesis descrbes research of the korean domain name system which allows us to use korea character into current internet domain name system and implements the architecture of the korean domain name system by analysis the structure and behavior of domain name server.

GILS is decentralized collection of locators and associated information services used by the public either directly or through intermediaries to find information. This standard specifies how electronic network searches can be experessed and how results are returned. GILS is an innovative network approach to assist user in locating information resoruces. This thesis proposes architecture of GILS. First, we study fundamental implementation technologies by the way of GILS and Z39.50 protocol. And then, after surveying GILs related technologies we introduce web-based GILS implementation. This study failed to implement GILS in commercial sector, but have an impliction to raise further research issue.

In this paper, we psropose a virtual information desk using VRML and internet. We review about VRML which can construct a virtual world on the internet. Next, we explain a basic concept of a web-based virtual information desk using VRML and javascript. Implementation details and advantages of the developed system are described also.

본 논문은 한국어 단어의 의미 영역 정보를 저장하기 위한 시소러스를 설게하고 이를 구축하기 위한 시스템을 설계하였다. 기존에 TOP_DOWN이나 BOTTOM_UP 방식을 이용하는 경우 각각 비 객관성과 작업 속도의 문제와 비구조성, 비일관성의 문제를 안고 있어 이를 혼합하여 어의문을 이용하여 객관성을 유지하면서도 기본모델을 이용하여 비구조성과 비일관성의 문제를 해결하고 있다. 또한 그 동안 필요성이 증가되었으나 작업을 하지 못했던 가장 큰 이유인 단어의 방대성으로 기인한 작업속도의 문제해결하기 위하여 C/S 모델을 적용하여 다수의 입력자들에 의해 동시 입력을 가능케 함으로써 작업 속도의 향상을 이루었다.

본 연구에서는 마스크의 폭을 각각 0.3.mu.m , 0.1.mu.m으로 설정하여 boron이 도핑된 실리콘 기판 위에 arsenic을 주입 시켰을때, 열처리 전과 각각 inert, dry, wet 산화분위기에서 열처리 한 후의 도핑 농도 프로파일과 수직 깊이와 측면으로의 퍼짐을 시뮬레이션하여 2차원적으로 관찰하였다. 마스크 폭을 축소시킴으로 인해 이온의 입사 방향으로 더 깊이 침투됨이 관찰되었으며, 더욱이 측면으로도 퍼짐이 관찰되었다. 열처리 전과 열처리 후의 비교에서도 농도프로파일이 보다 더 기판 내에서 넓게 분포됨이 관찰되었다.

본 연구는 boron-doped 실리콘 기판에 heavy 인온인 비소를 먼저 이온 주입시키고 비소의 주입에 의해 실리콘 표면이 손상된 영역에 다시 인을 이온 주입시켰을때, 인의 확산을 관찰하기 위해 microtec 시뮬레이터를 통해 모의공정실험을 실행하였다. 손상된 비정질의 실리콘 기판에서 열처리 전과 inet(N/sub 2/) 분위기에서 인은 느리게 확산을 하였다. 그렇지만 dry와 wt oxidation 열처리 분위기에서는 의의 확산 속도가 증가됨 (OED:oxidation-enhanced idfusion)을 관찰되었다. 실리콘 기판에서 인의 확산을 관찰하기 위해 ICECREM 시뮬레이터를 사용하여 앞의 경우와 동일하게 먼저 비소를 주입하여 실리콘 표면에 손상을 입히고 그 다음 공정에서 인을 주입하였을 때, 열처리 전과 inet, dry 산화분위기에서는 비정질의 실리콘 기판에 이온 주입한 경우와 동일하게 의의 확산 속도가 증가하였지만, wet 산화분위기에서는 오히려 dry 산화분위기에서 보다 확산이 늦어짐이 관찰되었다.

In this theory, we explained the impossibility of the motion or miving of the hole that has been recognized to be a carrier, by giving some fundamental reasons. We treated energy gap and impurity concentration, in p- and n-type region, as functions of the mobility that is one of te factors which determine current quantity, and analyzed the primary electron theory as a carrier by introducing 2 hypotheses.

p형 실리콘 기판위에 100.angs.의 초기 산화막을 성장시킨 후 붕소(B)와 인(P)을 1MeV 이온주입 에너지로 4.dec. tilting하여 붕소의 도즈량은 1*10/녀ㅔ 13/[cm/sup -2/]까지, 인은 1*10/sup 13/[cm/sup -2]로부터 1*10/sup 14/[cm/sup -2/] 까지 변화시키며 이온 주입하였다. 이온주입 후 RTA 로서 열처리 하였으며, 열처리 시간은 10초에서 40초까지,열처리 온도를 1000.deg.C에서 1100.deg.C까지 변화하였다. 이후 기파낸의 불순물의 프로파일 및 미세 결함의 분포를 분석하기 위하여, SIMS, SRP, XTEM 분석을 실시하였고, 이를 monte-carlo 모ㅓ델로서 시뮬레이션하여 비교하였다. SIMS 분석 결과 열처리 온도와 시간이 증가할수록 접합깊이가 증가하였고, 프로파일이 넓어짐을 볼수 있다. SRP 측정에서 붕소는 주해거리 (Rp)값은 1.8.mu.m~1.9.mu.m, 인의 경우는 1.1.mu.m~1.2.mu.m의 주행거리 (Rp) 값이 나타났다. XTEM 분석결과 붕소의 경우 열처리에 전후에도 결함을 볼수 없었고, 인의 경우 열처리 이후에 실리콘 결정내부에 있던 산소(O)와 인(P)우너자의 pinning효과에 의해 전위다이폴을 형성하여 표면근처로 성장함을 볼수 있었다.

We have calculated ultra-low energy silicon-self ion implantations and silicon damages through classical molecular dynamics simulation using empirical potentials. We tested whether the recently developed environment-dependent interatomic ptential (EDIP) was suitable for ultra low ion implantation simulation, and found that point defects formation energies were in good agrrement with other theoretical calculations, but the calculated vacancy migration energy was overestimated. The number of isolated defects that are produced by collision cascades are onlya few of the total number of defects, and fmost of the damages are concentrated into amorphous-like pockets.

Plasma etching of photoresist needs high etch rate, good uniformity and rae, good uniformity and low damage in low cost. ICP asher is expected to satisfy these requriement for next eneration semiconductor devices. ICPsimulator has been used to design the ashing chamber to redcue the development time and cost, and its results have been verified by QMS, OES and langmuir probe measurments. Plasma characteristics are monitored in terms of RF power and chamber pressure.

Ultra-thin gate oxide in MOS devices are subjected to high-field stress during device operation, which degrades the oxide and exentually causes dielectric breakdown. In this paper, we investigate the electrical properties of ultra-thin nitrided oxide (NO) and reoxidized nitrided oxide(ONO) films that are considered to be promising candidates for replacing conventional silicon dioxide film in ULSI level integration. We study vriations of I-V characteristics due to F-N tunneling, and time-dependent dielectric breakdown (TDDB) of thin layer NO and ONO depending on nitridation and reoxidation condition, and compare with thermal $SiO_{2}$. From the measurement results, we find that these NO and ONO thin films are strongly depending on its condition and that optimized reoxided nitrided oxides (ONO) films show superior dielectric characteristics, and breakdown-to-change ( $Q_{bd}$ ) performance over the NO films, while maintaining a similar electric field dependence compared to NO layer.

차세대 CMOS 구조에서 래치업 최소화를 위하여 고에너지 이온주입을 이용한 retrograde well 과 매몰층의 최적 공정 설계 변수 값들을 설정하였다. 본 논문에서는 두 가지의 모듸 모델 구조를 제안하고 silvaco 틀에 의한 시뮬레이션 결과를 비교 분석하엿다. 첫 번째 모델은 매몰층과 retrograde well을 조합한 구조이며, p+ injection trigger current가 600.mu.A/.mu.m 이상의 결과를 얻었고, 두번째 모델은 twin retrograde well을 이용하여 p+ injection 유지전류가 2500.mu.A/.mu.m이상의 결과를 얻었다. 시뮬레이션 결과, 두 모델 모두 도즈량이 많을수록 래치업 면역 특성이 좋아짐을 보았다. 시뮬레이션 조건에서 두 모델 모두 n+/p+ 간격은 2..mu.m 로 고정하였다.

In this paper, electrical characteristics of an n-channel Si MOSFET with L$_{s}$=0.6.mu.m under optical illumination are charaterized on wafer. Energetic photons with .gamma.=830nm, hv=1.494eV, P$_{opt}$=300mW are injected near the drain junction, the most photoresponsive region in the device, via optical fiber. We observed significantly increased drain current and transconductance, which is considered to be useful for the implementation of OEICs on silicon substrate, under optical control with P$_{opt}$=300mW. Optical power-dependent physical mechanisms responsible for the variation of electrical characteristics under optical input are also reported.d.d.d.

소신호 해석을 위해 사용되는 여러가지 방법을 비교하고 sinusoidal steady state analysis (S/sup 3/A) 방법을 이용해 반도체 소자를 소신호 해석한다. 소신호 행렬의 풀이방법으로 메모리 소비량이 적고 고주파수에서 행렬 연산으로 인한 잡음성분이 저은 전진해법을 사용한다. MEDICI에 의한 모의실험 결과와 비교하여 10GHz 이하의 주파수 영역에서는 비슷하였으나 10GHz~100GHz의 주파수영역에서 MEDICI에 비하여 정확한 결과를 보였다.

2차원 혼합 모드 소자-회로 시뮬레이터를 이용한 과도상태 해석의 알고리즘을 제시한다. 1변수 muller 및 regular falsi법을 회로의 절점 전압과 분기(branch) 전류를 계산하는데 적용하였다. 제안된 알고리즘의 정확도와 유호성을 검증하기 위해 PN 다이오드의 양극(anode)에 저항이 직렬로 연결된 회로의 모의실험을 수행한 결과, MEDICI의 모의실험 결과에 비해 과도상태에서 전류 및 전압 특성은 각각 0.06%, 0.2% 오차 범위 한도 내에서 일치함을 보였다.

25.deg. C 에서 120.deg. C까지 온도를 증가시키면서 hot carrier effect에 의한 nMOSFET의 degradation을 drain current와 transconductance의 변화를 통해 알아보았다. 온도가 증가할수록 hot carrier에 의한 degradation 이 전체적으로 줄어드는 것을 볼수 있었다. stress를 가한 후 reverse mode로 측정하였는데 saturation 영역보다 linear 영역에서 drain current의 degradation이 크게 나탔으며 온도가 증가할수록 이러한 경향이 유지되면서 degradation이 감소하였다. transconductance는 linear 영역과 saturation 영역에서 각각 측정하였는데 온도가 증가할수록 linear 영역의 degradation이 더많이 감소하였다.

A hybrid-integrated SAW(surface acoustic wave) douplexer including transmitter(Tx) filter, receiver(Rx) filter an dantenna terminal has been studied. The prototype duplxer was designed for CDMA system application in the 800 MHz-band. Ladder type SAW filters consisting of one port SAW resonators wer integrated on epoxy substrate to demonstrate the feasiblity of antenna duplexer. In order to perform the duplexing function, a strip line for impedance transformaton with length of .gamma./4 was inserted at rx path. The fabricted duplexer has an insertion loss of 3.0 dB at Tx path and 3.7 dB at Rx path. The attenuation has more than 40dB.

본 논문에서는 muller법과 regular falsi법에 의한 CMOS 반전 증폭 회로를 해석하는 방법을 제안한다. Muller법과 regular falsi법을 이용하여 회로의 절점전압과 branch 전류를 예측하였고 회로의 출력 절점에서 KCL을 만족하도록 하였다. CMOS 반전 증폭 회로의 모의실험을 수행한 결과 MEDICI에 사용된 결합법에 비해 전압특성과 전류특성은 각각 5%와 5.4%의 최대상대오차를 보였다.

Platinum thin film which hardly form volatile compounds with any reactive gas at normal process temperature was etched in inductive coupled BCl$_{3}$/Cl$_{2}$ plasma. The etch rate of platinum thin film increased with increasing Cl$_{2}$/(Cl$_{2}$ + BCl$_{3}$) ratio. That reasoned increasing of ion current density.

In this paper, the microwave dielectric properties of the yPb($Fe_{2/3}W_{1/3}$)$O_{3}$ - (1-y)($Pb_{x}Ca_{1-x}$)$ZrO_{3}$ system were investigated. It was found that an addition of Pb($Fe_{2/3}W_{1/3}$)$O_{3}$ promotes the grain growth, the solid solution between the Pb-based perovskite and the Ca-based perovskite and the decreasing of the sintering temperature. Especially, we found that 0.05Pb($Fe_{2/3}W_{1/3}$)$O_{3}$-0.95($Pb_{x}Ca_{1-x}$)$ZrO_{3}$ ceramic has not only lower sintering temperature but also better dielectric properties than those of the (Pb, Ca)$ZrO_{3}$ systems.

The microwave dielectric properties and sintering properties of the xCaTi $O_{3}$-yMgTi $O_{3}$-z(L $i_{1}$2/N $d_{1}$2/)Ti $O_{3}$ systems were investigated for the development of microwave dielectric materials. Dielectric constants decreased with increasing the amounts of MgTi $O_{3}$ and increased with reducing the amounts of MgTi $O_{3}$ and increased with reducing (L $i_{1}$2/N $d_{1}$2/)Ti $O_{3}$ contents. The microscopic structures were mixed phase with a little second phase. We found some compositions with .tau.$_{f}$ changed from positive to negative value with increasing MgTi $O_{3}$ or (L $i_{1}$2/N $d_{1}$2/)Ti $O_{3}$ contents. The microscopic strucstures were mixed phase with a little second phase. We found some compositions with .tau.$_{f}$ =0ppm/.deg.C. these compmsitions exhibited the stable dielectric properties on the various sintering temperatures.res.

In this paper, HBT's with lower phase noise and passive elements, such as resistors, capacitors and inductors, for resonance and impedance matching networks are designed, fabricated, tested, and carefully analysed, respectively, and then, they are integrated for the design and fabrication of functional X-band oscillators with lower phase noise. Epi-wafers for HBT's with the structure of graded $Al_{x}$G $a_{1-x}$ As emitter and C-doped base layer of 700.angs. thick were used to specially emphasize the improvement of $f_{T}$ and $f_{max}$, and the lowering of phase noise, in design aspects. At the test frequencies of 12GHz, capacitances of MIM capacitors, spiral inductor, and resistances are 0.5~10pF, 0.4~11.06nH, and 20~1,380.ohm., respectively. The emitter size of HBY's for the X-band MMIC oscillators is 3*10u $m^{2}$, and find chip size is 0.9*0.9m $m^{2}$..EX>.

The effect of the heat treatment of the AZO transparent conductive film prepared by rf magnetron sputtering was investigated. The variations of the electrical and optical properties with heat treatment ambient and temperature were studied. After the heat treatment in air above 300.deg. C, the resistivity of AZO films increased by 1 to 8 orders of magnitude. However, no significant change in the AZO films after th eheat treatment in vacuum was not observed.

NH$_{3}$ gas sensitive ZnO:Al, In thin films were prepared by the heat treatment following continuous deposition of very thin In layer and ZnO:Al layer to obtain the modified surface morphology for good sensitivity. Dependence of the structural electrical and optical properties of them on heat treatment temperature was investigated by x-ray diffraction, SEM, 4-point probe method and spectrophotometer.

In this paper, the etching studies for n-GaN were carried out using the wet chemical, the photo-enhanced-chemical, and the electro-chemical etching methods. The experimental results show that n-GaN is etched in diluted NaOH solution at room temperture and the etched thickness of NaOH and electron concentrations. Te etching rate of n-GaN samples with n.simeq.1*10$^{19}$ cm$^{-3}$ were used to compare the photo-enhanced-chemical etching with the electrochemical etching methods. The removed thickness was 680.angs./25min by the electrochemical etching methods. The removed thickness was 680 .angs./25min by the electrochemical etching method ad 784.angs./25min by the photoenhanced-chemical etching method. The patterns are 100.mu.m*100.mu.m rectangulars covered with SiO$_{2}$film. It is shown that the profile of etched side-wall of the pattern is vertical without dependance of the n-GaN orientations.

Recently, InGaP/GaAs HBTs have been much interested as a potential replacement for AlGaAs/GaAs HBTs because of their superior device and material properties. In this paper, DC characteristics of InGaP/GaAs HBTs and the temperature dependance as well as the reliability were investigated comparing with AlGaAs/GaAs HBTs. As a results InGaP/GaAs HBTs produced the superior performance to AlGaAs/GaAs HBTs.

Due to the large conduction band discontinuity between emitter base, OmGaAs HBT has an advantge to enable the hot electrons to inject into the base. In this paper, InAlGaAs/InGaAs HBT with the various emitter junction gradings and the modified collectors are simulated and analyzed by HMC(hybrid monte carlo) simulator in order to find a optimal structure for the shortest transit time. A minium base transit time (.tau.$_{b}$ ) of 0.21 ps was obtained for HBT with the grading layer, which is parabolically graded from x=1.0 to x=0.5. The minimum collector transit time (.tau.$_{c}$ ) of 0.31ps was found when the collector was modified by inserting p$^{[-10]}$ and p$^{+}$ layers. Thus HBT in combination with the emitter grading and the modified collector layer showed the cut-off frequency (f$_{T}$) of 183GHz.z.z.

BCl/sub 3/ 및 Cl/sub 2/ 반응가스를 사용하여 RIE 장치로 GaN의 건식식각을 연구하였다. RF 전력, 반응가스의 유량 및 반응가스의 혼합비 등의 변화에 따른 최적의 식각공정 조건 및 결합특성을 연구하였다. RF 전력에 따른 GaN의 식각율은 챔버압력 25mTorr, BCl/sub 3/ 유량 40 sccm의 조건에서 RF 전력이 100W일때 17nm/min을 얻었다. BCl/sub 3/의 유량에 따른 식각율은 RF 전력 100W 챔버압력 20mTorr, Cl/sub 2/ 유량 5sccm의 조건에서 BCl/sub 3/ 유량이 40 sccm일때 65nm/min을 얻었다. Cl/sub 2//BCl/sub 3/ 혼합가스 비율에 따른 식각율은 Cl/sub 2/ 유량을 5sccm으로 고정하고 BCl/sub 3/ 유량을 변화시켰을때 RF 전력 100W 및 챔버압력 20mTorr의 조건에서 혼합비가 0.25일때 50nm/min을 얻었다. RF 전력에 따른 PR의 식각율은 챔버압력 25mTorr, Cl/sub 2/ 유량 0 sccm 및 BCl/sub 3/ 유량 40 sccm의 조건에서 RF 전려이 100W일때 15nm/min을 얻었다. 또한, 챔버압력 20mTorr, Cl/sub 2/ 유량 5 sccm 및 BCl/sub 3/ 유량 20sccm의 조건에서 RF 전력이 100W 일때 82nm/min을 얻었다.

This paper presents a dAC calibration technique for high resolution and monolithic operation. The calibration technique consists of basic source, current memory cell (C.M) and current substrator. Current memory supplies the error current to basic source. Current substrator extracts the error current from the main source. It is simple and needs no special calibration period. The proposed current cell has high calibration performance and guarantees 100MHz operation.

In this paper the formation mechanism of tail current is analyzed and its effect on GTO turn-off performance is given. The conclusion is that the large tail current will considerably increase the turn-off loss $E_{off}$ and cause the re-triggering during GTO's off-switching, therefore the best design criterion is that the tail current of power GTO must be as low as possible.w as possible.

In this paper are presented a newly proposed 3D monte carlo (MC) damage model for the dynamic simulation in order to more accurately and consistently predict the implant-induced point defect distributions of the various ions in crystalline silicon. This model was applied to phosphorus implants for the ULSI CMOS technology developement. In additon, a newly applied 3D-trajectory split method has been implemented into our model to reduce the statistical fluctuations of the implanted impurity and the defect profiles in the relatively large implanted area as compared to 1D or 2D simulations. Also, an empirical electronic energy loss model is proposed for phosphorus and silicon implants. The 3D formations of the amorphous region and the ultra-shallow junction around the implanted region could be predicted by using our model, TRICSI(Transport ions into crystal-silicon).

In this work, single-poly EEPROM has been designed and fabricated by using standard 0.8.mu.m CMOS process. The initial threshold voltage was aobut 0.8V but it increased ot about 6.5V after programming at Vds=11.5V and Vcg=6.5V. After erasing devices at Vs=14.2V, the threshold voltage decreased to about 1.5V. The programming time and erasing trime wree about 6ms. and 100ms. respectively. The erasing time can be reduced by applying a series of shorter erase pulse s instead of a long single erase pulse.

The impact of hot carrier induced gate leakage current on the refresh time of memory devices has been examined. The maximum allowable supply voltage for cell transistor has been determined form the degradation of the refresh time. The desing guideline for cell capacitors and refresh circuits has been suggested.

Hot carrier induceed the performance degradation of sense amplifier circuit in DRAM has been measured and analyzed using 0.8.mu.m CMOS process. Simulation and experimental results show that the degradation of the MOS devices affects the decrease of the half-Vcc, voltage gain and the increase of the sensing voltage gain and the increase of the sensing voltage. The dominant degradation mechanism is the capacitance imblance in the bit-line pair. We carried out the spice simulation to investigate the degradation of the sense amplifier circuit.

We studied phosphorus doping effect on the epitaxial growth of $Si_{1-x}Ge_{x}$ film with high Ge fraction on Si substates at 550.deg. C by LPCVD. In a low $Ph_{3}$ partial pressure region such as below 1.25 mPa, the phosphorus dopant concentration increased linearly with increasing $PH_{3}$ partial pressure while the deposition rate and the Ge fraction were constant. In a higher $PH_{3}$ partial pressure region, the phosphorus dopant concentration and the deposition rate decreased, while the Ge fraction slightly increased. The deposition arate and the Ge fraction increased with increasing $GeH_{4}$ partial pressure while the phophours dopant concentration decreased. But the increasing rate of Ge fraction with incrasing $PH_{3}$ partial pressure was reduced at a high $GeH_{4}$ partial pressure. According to test results, it suggests that high surface coverage of phosphorus atoms suppress both the $SiH_{4}$ adsorption/reasction and the $GeH_{4}$ adsorption/reaction on the surfaces, and the effect is more stronger on $SiH_{4}$ than on $GeH_{4}$. In a higher $PH_{3}$ partial pressure region, the epitaxial growth is largely controlled by surface coverage effect of phosphorus atoms. The phosphorus surface coverage was slimited at a high $GeH_{4}$ partial pressure because adsorbed Ge atoms effectively suppresses the adsorption of phosphorus atoms.

This paper is concerned with researching latch-up immune CMOS structure was performed. By the simulation results, the connecting layer had more effect than the buried layer to latch-up immune. When the connecting layer was the dose 1*10$^{14}$ /cm$^{2}$ and the energy 500KeV, the trigger current was more 0.6mA/.mu.m and the trigger voltage was 6V. The more the connecting layer dose was lower, the more the latch-up immune characteristics was butter.

본 논문은 triple-well과 twin-well에서의 고에너지 이온주입 에너지와 도즈량 변화에 따른 래치업 특성을 비교하였다. 공정시뮬레이터인 ATHENA로 소자를 제작하고 도핑프로파일 형태와 구조를 조사한 후, 래치업 특성은 소자 시뮬레이터인 ATLAS를 이용하였다. triple-well 공정이 마스크 스텝수를 줄이고, 이온주입 후 열처리시간을 단축하며 별도의 열처리 공정없이 도핑르로파일을 넓은 형태로 분포시켜 래치업 면역특성이 매우 좋은 결과를 얻었다.

We have studied the oxidatio nrte of $Si_{1-x}Ge_{x}$ epitaxial layer grown by MBE(molecular beam epitaxy). Oxidation were performed at 700.deg. C, 800.deg. C, 900.deg. C, and 1000.deg. C. After the oxidation, the results of AES(auger electron spectroscopy) showed that Ge was completely rejected out of the oxide and pile up at $SiO_{2}/$Si_{1-x}Ge_{x}$ interface. It is shown that the presence of Ge at the $SiO_{2}$/$Si_{1-x}Ge_{x}$ interface changes the dry oxidation rate. The dry oxidation rate was equal to that of pure Si regardless of Ge mole fraction at 700.deg. C and 800.deg.C, while it was decreased at both 900.deg. C and 1000.deg.C as the Ge mole fraction was increased. The ry oxidation rates were reduced for heavy Ge concentration, and large oxidation time. In the parabolic growth region of $Si_{1-x}Ge_{x}$ oxidation, The parabolic rate constant are decreased due to the presence of Ge-rich layer. After the longer oxidation at the 1000.deg.C, AES showed that Ge peak distribution at the $SiO_{2}$/$Si_{1-x}Ge_{x}$ interface reduced by interdiffusion of silicon and germanium.

The electrical properties of $Si_{1-x}Ge_{x}$ samples have been investigated. The sample structures were grown by MBE (molecular geam epitaxy) with Ge mole-fraction of x=0.0, x=0.05, x=0.1, and x=0.2. To examine the influence of the thermal processing, the $O_{2}$ and N$_{2}$ process were performed at 800[.deg. C] and 900[.deg. C], respectively. After this thermal process, hall measurements have been done over a wide range of the ambient temperature between 320[.deg. K] and 10[.deg. K] to find the temperature dependence using the comparessed-He gas system. The Ge-rich layer has been formed at the $SiO_{2}$/SiGe interface and it has an effect on the hall mobility. And it has been found that hall mobility was increased by the $N_{2}$ annealing process comparing with dry oxidation process at both 800[.deg.C] and900[.deg. C].

단 채널 MOSFET 소자의 드레인 전압-드레인 전류 특성을 예측하기 위해서 caughey-thomas 이동도 모델을 수치적으로 구현하는 방법을 제안한다. 구현된 caughey-thomas 모델의 정확한 특성을 검증하기 위해서 0.5[.mu.m]의 설계규칙을 가즌 ASIC용 공정으로 n-MOSFET과 p-MOSFET을 제작하였다. 전자 및 정공의 포화속도 값이 각각 6.2*10/sup 6/[cm/sec] 과 1.034*10/sup 7/[cm/sec]인 경우에 채널길이가 0.5[.mu.m] 이상인 n-MOSFET과 p-MOSFET의 드레인 전압-드레인 전류특성의 모의실험 결과는 측정값에 비하여 10% 이내의 상대오차를 보였다.

Self-aligned AlGaAs/GaAs HBTs with the emitter area of 1.5*10.mu.$m^{2}$ were fabricated usng $WN_{x}$ as emitter metal. Their DC and RF characteristics were investigated. The common emitter current gain was 45 at $J_{c}$ = 6*$10^{4}$A/$cm^{2}$. From the Gummel plot, the ideality factors of $I_{c}$ and $I_{B}$ were 1.18 and 1.70, respectively. Emitter and base resistance were extracted from voltage drop region in gummel plot, and their values were 5.3.ohm. and 38.2.ohm.. The extrapolated $f_{T}$ = 72GHz and $f_{max}$ = 81GHz were obtained at $V_{CE}$ = 2V.

For porous silicon layer to be used as active layer in various devices, it is necessary to be formed locally along with a designed pattern on the wafer. However, there is still no suitable masking layer to withstand against the high concentration of HF for a time of some minutes up to some hours during the anodic process effectively. In this work, we investigated the property of selectivity between p$^{+}$ and n layers to form localized porous silicon even without a mask by the difference of the anodic I-V characteristics on the doping level and doping type. The width of the pattern made in the sample was 2mm, and the formed porous silicon layer was observed by SEM to see the morphology on the cross section below the surface. As the results, it was found that the selectivity was reasonable for the pattern size over 100.mu.m.m.

InGaAsP/InP ridge waveguide(RWG) distributed feedback(DFB) lasers (LD) operating at 1.55.mu.m were fabricated by a two step MOCVD process. The CW laser threshold and the slope efficiency from these lasers are 22.5mA and 0.127mW/MA perfacet at room temperature. Single longitudinal mode operation with side mode suppression of more than 30dB is obtained at 5mW. The temperature sensitivity and characteristic temperature (T$_{0}$) from DFB lasers were obtained 1.0.angs./.deg. C and 50.3K between 20.deg. C and 70.deg. C, respectively.y.

ULSI 소자에서 폴리사이드 구조를 사용하고 dual 게이트 산화막에 대한 공정 특성을 최적화 하는 2스텝 게이트 산화막의 형성 공정에 관한 연구를 하였다. 이러한 특성의 측정은 HP4145B 파라메터 분석기와 C-V meter 그리고 multi-frequency LcR meter를 사용하여 2스텝 산화막의 공정 방법과 cleaning에 따른 게이트 사화막의 공정 특성에 대한 연관 관계로 연구하였다. I-V 특성 면에서는 G$_{ox}$ 80.angs.의 경우 base 210.angs.의 경우에서는 dual 210.angs.의 특성이 base 210.angs.에 비하여 상대적으로 열화된 특성을 나타내었다. CCST 결과에서는 G$_{OX}$ 80.angs.과 210.angs.에서 dual 게이트 산화막의 cleaning 방법으로 piranha cl'n 과 SCl cl'n 방법에서 우수한 결과를 얻을 수 있었다. 또한 게이트 전압의 벼화량에 대한 결과에서는 dual 산화막의 경우 초기상태에서는 호울포획 현상이 나타나다가 이후에는 전자포획 현상이 나타나는 결과를 얻었다.

본 연구에서는 정보통신 시스템 민 전자 제어 시스템, atuomobile 저자부문의 핵심부품으로 사용가능한 HV-IC영 BCD 공정 및 소자를 설계하였다. 60V 이상의 bipolr, 20V급 HV-CMOS 소자기술을 one-chip에 구현하는 고내압 BCD 소자구조를 제안, 설계하고 시뮬레이션을 통하여 고안된 소자구조를 검증하여 최적화된 공정 및 소자 변수를 추출하였다.

In this study, oxide(TEOS) CMPs were carried out for various head pressures. Table and head speeds are fixed at 25 RPM. Head pressures are 5, 7.5, 10, 12.5 PSI, and under these conditions, 1,587, 1,631, 2,556, 2,871.agns./min of oxide (TEOS) removal rates and 14.7, 18.5, 9.52, 7.9% of uniformities are obtained, respectively. Also, these experiments for local and global planarizations were done using the patterned 4" wafers. These conditions are applicable to STI(shallow trench isolation) structures and planarizations for sub-half micron lithography.aphy.

During driving an automobile, drivers push the clutch and transmit gear by their feeling. At the automobile which have the manually operated gear box by using apart designed equipment, for get the semi-auto function, a lot of research to realize the system which have passive transmission and automotive transmission has developed. In this paper, basically using one chip .mu.-comcontrolled the clutch of automobile by means of LPF, F/V converter and TR driver circuit which is designed. In the past, it was controlled the motor by operating relay which is int eh output system to control the clutch. This system some noises is made by on and off the relay, lessen the motor speed and causes many error due to consume the large power on driving the motor. In this paper, to solve these problems the output TR is operated by using one chip .mu.-com decrease the electic power consumption and designed the actuator to control the clutch by measns of vacuum.

Due to the improvement of PC performance and windows-based OS, PC-based control system is gradually taking over industrial control in the industry site, replacing the closed proprietary syste. PC-based system gives the avantages of ease-of-use, cost-effectiveness, simplicity of maintenace and the flexibility to design I/O system using standard technology. While there are many real time software soltions, most of hardware solutions are closed, vendor-depednent, proprietary systems for large scale plants. In this paper, we suggest the method to configure PC-based I/O system based on standard technology which gives the avantages to use off-the-shelf products for system integration.

본 논문에서는 요즘 많은 관심 속에서 활발히 연구되고 있는 WLL(wireless local loop)에서 이동국 (mobile station)과 FWT(fixed wireles sterminal)를 이용한 고정국 (RSU. Radio subscriber unit) 시스템의 구성을 제안한다. 이동국은 FWT로부터 분리하여 휴대사용할 수 있다는 점에서 일반적인 WLL 시스템의 고정국과 차이가 있다. 본 논문에서는 FWT의 하드웨어와 소프트웨어의 구성에 대하여 기술한다.

This paper presents a new panel structure and the associated driving method for TFT-LCDs, which is called the alternate-column driving (ACD). The ACD allows us to implement the dot-inversion driving with much lower power consumption than in conventional methods. This paper describes the ACD and analyzes the image quality and power consumption of the ACD.

본 연구에서는 방대한 양의 데이터를 실시간으로 처리하기 위한 병렬 고속 디지털 신호처리시스템을 제안한다. 시스템의 성능을 평가할 수 있는 확률적인 분석방법을 제시하며, FFT 와 같이 보드간 또는 프로세서간 통신부담이 많은 알고리즘과 행렬연산과 같이 통신부담이 적은 알고리즘에 적용하여 본다. 제안한 시스템의 다양한 구성에 대하여 두 가지 알고리듬의 성능을 확률적 방법으로 평가하였으며, 그 결과는 알고리즘 분석에 듸한 성능수치와 근접함을 확인하였다. FFT는 프로세서 개수가 증가해도 보드수가 많아지면 성능이 감소하였으며, 행렬연산은 프로세서 개수에 비례하여 시스템의 성능이 선형적으로 증가함을 확인하였다.

Performance of AMLCD pixels depends on the electrical characteristics of thin film transistor switches. The high quality of LCD can be obtained by minimizing the process and device variations of TFT. The effect of process and device factors on poly-Si TFT characteristics are calculated by ATHENA and ATLAS, and the optimized design windows based on statistical design of experiment are suggested for high performance 20 inch LCD are suggested for high performance 20 inch LCD monitors.

본 연구는 영상소자의 감도를 향상 시키기 위하여 전압 분배 회로를 가진 감지 회로를 설계하였다. 전압분배 회로는 NMOSFET과 Poly 저항의 두 경우로 설계하였으며 감지 회로에 흐르는 전류는 전압 분배 회로를 NMOSFET으로 설게하였을때가 Poly 저항으로 구성한 경우보다 적게 흐르며 감도 특성도 좋은 것으로 나타났다. 또한 poly 저항보다 NMOSFET을 이용한 전압 분배 회로가 동작 주파수에 따른 특성이 우수하였다.

This paper proposed a symmetric feedback current enhancement circuit with 1.5V power supply to design a 3$^{rd}$ order butterworth low pass filter. The proposed filter designed on 0.8.mu.m CMOS n-well double poly/double metal process simulated in HSPICE composed of the 3dB frequency enhancement circuit and the unity-gain frequency enhancement circuit. The simulation result on the design filter shows the badnwith of 25MHz, phase of 92.6 .deg. and power consumption of 0.3mW..

This paper describes a 3.3V-65MHz 12BIT CMOS current-mode DAC designed with a 8 MSB current matirx stage and a 4 LSB binary weighting stage. The linearity errors caused by a voltage drop of the ground line and a threshold voltage mismatch of transistors have been reduced by the symmetrical routing method with ground line and the tree structure bias circuit, respectively. In order to realize a low glitch energy, a cascode current switch ahs been employed. The simulation results of the designed DAC show a coversion rate of 65MHz, a powr dissipation of 71.7mW, a DNL of .+-.0.2LSB and an INL of .+-.0.8LSB with a single powr supply of 3.3V for a CMOS 0.6.mu.m n-well technology.

본 논문에서는 아날로그 증폭기의 특성 개선을 통해 아날로그 신호처리 시스템의 동작속도를 향상시키기 위해서 2단 연산증폭기 (two-stage opamp:TSO)의 주파수 응답 특성과 이득을 개선하기 위한 회로를 설계하고 시물레이션을 통해서 설계된 회로의 우수성을 증명하였다.

This paper describes new sense amplifier with fast sensing delay time of 0.54ns and 32kb CMOS embedded SRAM with 4.67ns access time for a 3-V powr supply. It was achieved using the sense amplifier with multiple point sensing scheme and high speed bit-line scheme. The sense amplifier saves 25% of the power dissipation compared with the conventional one while maintaining a very short sensing delay. The SRAM uses 0.5.mu.m double-polysilicon and triplemetal CMOS process technology. A die size is 1.78mm*2.13mm.

In this paper, a new scheme of a boosted voltage generator (BVG) is designed for low powr DRAM's. The designed BVG can supply stable $V_{pp}$ using a new circuit operting method. This method controls charge pumping capability by switching the supply voltage and ring oscillator frequency of driving circuit, so the BVG can save area and reduce the powr dissipation during $V_{pp}$ maintaining period. The charge pumping circuit of the BVG suffers no $V_{T}$ loss and is to be applicable to low-voltage DRAM's. $V_{pp}$ level detecting circuit can detect constant value of $V_{pp}$ against temperature variation. The level of $V_{pp}$ varies -0.55%~0.098% during its maintaining period. Charge pumping circuit can make $V_{pp}$ level up to 2.95V with $V_{cc}$ =1.5V. The degecting level of $V_{pp}$ level detecting circuit changes -0.34% ~ 0.01% as temperature varies from -20 to 80.deg. C. The powr dissipation during V.$_{pp}$ maintaining period is 4.1mW.W.1mW.

In this paper, a set of novel self-timed latches are introduced and analyzed. These latches have no back-to-back connection as in conventional self-timed latch, and both inverting and noninerting outputs are evaluated simultaneously leading to thigher oepating frequencies. Power consumption of these latches ar ealso comparable to or less than that of conventional circuits. Novel type of cross-coupled inverter used in the proosed circuits implements static operatin without signal fighting with the main driver during signal transition. Proposed latches ar tested using a 0.6.mu.m triple-poly triple-metal n-well CMOS technology. The resutls indicates that proposed active-low sefl-timed latch (ALSTL) improves speed by 14-34% over conventional NAND SR latch, while in active-high self-timed latch (AHSTL) the improvements are 15-35% with less power as compared with corresponding NORA SR latch. These novel latches have been successfully implemented in a high-speed synchronous DRAM (SDRAM).

This paper describes the development of 5.15ns 32kb asynchronous CMOS SRAM using 0.6.mu.m CMOS technology. The proposed high speed embedded SRAM is realized with optimized control circuit and sense amplifier at a power supply of 3V. Using proposed control circuit, the delay time from address input to wordline 'on' is reduced by 33% and mismatch-insensitive sense amplifier can sense a small difference of bit-line voltage fast and stably.

This paper presents a new reference voltage generator(RVG) for advanced DRAM's. The proposed RVG with three temperature coefficient is independent of temperature variation, and supply voltage. This is used to shorten leakage current using the boosted sense ground(BSG). This circuit is designed in a 0.8.mu.m nwell CMOS, double-polysilicon, double-metal technology. The simulation resutls in jindependent temperature and supply voltage. In hspice simulation results, temperature dependency of RVG is 130.mu.V/.deg. C and supply voltage dependency is .+-.0.91%, $V_{cc}$ =3.3V.+-. 0.5V.3.3V.+-. 0.5V.

다치결정도 (multiple-valued decision diagram : MDD) 와 순서화된 다치결정도 (ordered MDD:OMDD)는 다치논리함수의 표현에 폭넓게 사용된다. p치 n변수인 경우 p/sup (n-1)/으로 증가하는 노드의 수는 ROMDD(reduced OMDD)를 사용하여 현저하게 감소시킬 수 있다. 그러나 다치와 다변수의 경우에는 더욱 많은 공정을 수반하게 된다. 이러한 단점을 보완하기 위해 honghai jiang이 제안한 2치시스템에서의 input implict/output explicit 관계를 갖는 OVAG(output value array graph)를 사용하여 다치논리함수를 표현한다. 그리고 MDD 표현이 어려운 상황에서 MOVAG(multi OVAG)를 사용하여 보다 쉽게 출력값을 배열하는 그래프를 이끌어 낼 수 있다. 본 논문에서는 MOVAG의 구성방법과 회로에서 MOVAG로으 변환에 대한 알고리듬을 제안하였고, 알고리듬에 의한 결과를 MDD와 비교하여 노드수 감소에 따르는 처리속도가 개선됨을 검증하였다.

본 논문에서는 노드들간의 입출력 관계가 트리형태로 주어진 경우에 이 관계를 수식으로 해석하여 최소화시키고 이를 회로로 구현하는 새로운 알고리즘을 제안한다. nakagima 등에 의해 제안된 알고리듬은 트리의 특성을 갖는 노드들의 관계를 2치논리에 근거하여 회로로 구현하였으나, 이러한 기법은 일반적인 형태로 주어진 트리구조에 대한 해석이 충분치 못하므로, 일반화된 회로의 구성에 많은 제약을 가지고 있다. 이러한 문제점에 대하여 본 논문에서는 트리구조를 갖는 노즈들의 전체적인 입출력관계를 수식으로 정리하여 최소화된 회로설계 알고리즘을 제안하고 예를 들어 이를 검증한다.

In this paper, we presetn the WWW (world-wide-web) base dtelePACS, which uses the communication network as asymmetric satellite data communication system (ASDCS). Web-based telePACS is imple-mented with JAVA language. Through the internet, web-based TelePACS can make it possible for remote hospital or doctor to access PACS data and information easily and cheaply. The ASDCS uses receive-only satellite links for data delivery and PSTN (Public switched telephone Network) modem or N-ISDN (narrow band integrate services digital network) for communication. The satellite communication linking shows the very high-seed performance such as 10-30 times faster than the modem linking.To solve the speed limits of internet, JPEG and wavelet compression methods were adapted in WWW-based TelePACS and were evaluated by PSNR (peak signal-to-noise ratio) and by radiological experts that 10:1 or less JPEG and wavelet compression may be acceptable for diagnostic purposes. Consequently, we get the conslusion that our system is suitable for tele-radiology and telemedicine world widely.

This paper evaluated autonomic nervous system function in 23 patients with syncope and a positive tilt test result, 21 with a negative test result, and 19 healthy controls. Indexes of heart rate variability were measured during supine resting, immediately afte rtilt-up, standing rsting, immediately before syncope and immediately after tiltdown. There were no significant differences among the groups in any of the indexes of heart rate varability over the 24-hour holter recordings. In patients with a positive tilt result, tilting gaused a decrease in low-frequency (LF) immediately before syncope and incsrease in high-frequency (HF) bands immediately before. In patients with a negative tilt result, tilting caused a decrease in low-frequency (LF) immediately before syncope and decrease in high-frequency (HF) bands immediately before, different from positive tilt results. Our findings showed that patients with vasovagal syncope have no chronic differences from normal subjects in autonomic nervous system activity, but that these patients respond differently to the orthostatic stimulus.

In this paper, we implemented the electornic moving aid system for safe walking of the blind. An obstacle detecting of each sector used ultrasound and a distance measurement used time of flight. The alarm is designed to have a sound and a tactile function that can be selected on an user's convenience. This system can detect and obstacle of upward, forward, downward and optimally warn to the blind with vibration, beep sound by appling warning algorithm on object detection. Experimental testing and performance evaluation have been successfully carried out with a prototype cane, and the experiment shows the capability of the function to detect unknown objects within an assigned distance, under knees, over head height, and crushed puddles.

In this paper, we devised pulse rate diagnosis for provide basic index of cold-hot diagnosis. The system consist of database part and pulse rate detection part for detection pulse wave, respiration and ECG. The database is constructed windwos95 platform using DAO(data access object). Search algorithm used ISAM algorithm. The database consist of one's information and medical report for a subject and detected pulse wave.

In this paper, we devised pulse detection transducer that has non-differential characteristics for pulse detection on chongu arterial. The transducer consist of load cell and driving electronic circuits. Load cell consist of cantilever and two metal film strain gauge. The pressure signal from chongu artery is delivered to load cell using artery rider that attached to cantilever. Therefore the pressure pulse signal can obtain by the developed transducer. As the results of experiment, the developed transducer has a good linearity at pressure to voltage conversion and acan detect non-differential pulse signal from chongu artery.

In this paper, we psresent the design strategy of the assistive communication system for the mute and speech disorder to be able to communicate with people under the restricted circumstances. First, it should be used conventiently in daily life and a specific domain should be set accoridng to the circumstance. Second, proper words in each domain should be selected according to the physical and mental ability of users. This paper seeks to highlight some of the issues involved in using semantic compaction in adynamic environment. (e.g. noun and verb explorable system). In additon, symbols are used in this system in order to make users recognize clearly the meaning of a words. The sentences are compounded from symbols. The compounded sentences are shown on the display monitor in letter,s and synthesized naturally in voice in order to be understood easily.

To obtain 3-D topography of EEG records, we propose a new method based on the polygon mapping technique. The method has the low complexity to calculate the interpolation of the EEG records on the scalp and maintains the high resolution topography because the polygon technique performs the interpolation at the only vertexes of each polygon. We implemented the topographic system with 3D barycentric, 3D polynomial and spherical spline algorithms in a personal computer.

현재의 언어자동인식은 목소리의 음성 정보와 얼굴영상의 화상정보를 병행처리하여 초기의 음성 정보만을 처리한 경우보다 높은 인식률을 가져왔다. 이러한 연구는 언어의 인식에서 뿐만 아니라, 신체가 부자연스러운 장애니에게는 간단한 의미 전달의 수단이나 각종 재활보조기기의 제어 등에도 매우 유용하게 활용될 수 있다. 음성정보처리를 배제한 영상정보처리 면에서, 본 논문의 대상이 되는 장애인은 신체의 부자연스러움으로 인해 처리되는 얼굴의 영상이 정상인과 같은 바른 자세가 되지 못하고, 기울어진다든지 틀어져서 기존의 방법들로는 정상인보다 현저히 낮은 입모양의 인식률을 보인다. 본 논문에서는 기존의 방법들과는 달리 청색 표식을 이용한 색상불할법과 라벨링을 통해 입 모양을 인식하는 방법을 제안하였으며, 이는 장애인과 같은 기울어진 얼굴영상에 대해서도 정상인과 같은 입 모양 인식률을 나타내었다.

One of the main techniques for diagnosing heart disease is by examining the electrocardiogram(ECG). The earlier noise reduction techniques can not effectively cancellation complex noise from the noisy ECG such powrline interference, baseline drift, muscle artifact. In this paper, we performed the extrude noise from and recovering the ECG signal using wavelets transform that has recently been applying to various fields.

본 논문의 목적은 한국어 모음의 음성 신호에 대하여 2차 및 3차 비선형서의 존재에 대한 정량적인 분석을 수행함에 있다. 음성 신호의 비선형성을 분석하기 위하여 표본화한 음성 신호에 대하여 bicoherence 및 tricoherence를 측정하였다. 실험 결과에 의하면 한국어 모음의 음성 신호의 발생과정에 상당히 강한 2차 및 3차 비선형성이 존재함을 알수 있었다. 특히 음성신호의 3차 비선형성에 대한 연구는 처음 수행되는 것으로 음성 신호 분석에 있어서 매우 중요한 결과로 사료된다.

In this paper, a speech enhancement method based on phonemic properties and masking effect is propsoed. It is a modified type of spectral subtraction wherein the spectral sharpening process is exploited in unvoiced state considering the phonemic properties. The masking threshold is used to remove the residual noise. The proposed spectral subtraction shows similar performance as that of the classical spectral subtraction method in view of the SNR. But by the prposed scheme, the unvoiced sound region is shown to exhibit relatively less signal distortion in the enhanced speech.

In a large number of words and the continued speech recognition system using a phoneme as teh recognition unit, it is necessary to segment processing. In this paper, a normalized AMDF new method. The suggested parameter represents a degree of sharpness at valley point. This method can detect the speech segment between the steady state and transient region to the continued speech without a prior information of speech signal.

The major objectives of speech coding include high compression ratio for transmission in the band limited channel, high synthesized speech quality in terms of the intelligibility and the naturalness and fast processing speed. In general, speech coding methods are classified into the following three categories: the wavelform coding, the source coding and the hybird coding. In this paper, we proposed a new waveform coding method using PSOLA(pitch-synchronous overlap add) technique. First, we fixed one basic waveform per pitch and measured the formant similarity between basic and neighbor waveform. Second, if the similairy satisfied threshold values, we compress the neighbor waveform per pitch and then store or transmit. When the comparession is about 45%, we obtained about 4 in MOS.

윈도우즈 95와 같은 멀티미디어 환경 하에서 개인 신분 확인을 위한 방법은 비밀번호를 키보드로 입력받는 것이었으나, 본 논문에서는 음성을 이용하는 방법으로 기존의 방법이 기준패턴의 시간에 따라 변하는 특성을 보상하지 못한다는 단점을 보완하는 방법이다. 즉, 이를 위해 음성신호의 특징인 기본주파수와 제1포만트의 비율을 이용하여 기준패턴을 형성화하는 방법에 관한 것이다. 제안한 방법으로 실험한 결과, 98%의 전체 인식율을 얻게 되었고, 윈도우즈 환경에서 비밀번호 사용 대신 음성 사용에 대한 가능성을 보여 주었다.

음성부호화의 주요목적은 대역 제한된 전송 대역폭에 전송을 하기위한 음성압축, 명료성과 자연성을 유지하는 고음질 음성합성, 그리고 처리 속도등의 요소에 따라 달라진다. 일반적으로 음성 부호화 방법은 파형 부호화범, 신호원 부화화법, 그리고 혼성 부호화법으로 나누어질 수 있다. 이러한 방법으로 전송되어진 음성은 다시 합성을 하게되는데, 이때 고음질을 유지할 수 있는 PSOLA법을 사용하였다. 본 논문에서 제안한 방법으로 전송되어진 음성은 다시 합성을 하게되는데, 이때 고음질에 유지 할 수 있는 PSOLA법을 사용하였다. 본 논문에서 제안한 방법은 기존의 PSOLA 합성법에서 사용되어지는 hanning 윈도우가 음성이 갖는 golttal wave shape의 특성에 적합하지 않다는 것을 이용하여 기존의 hanning 윈도우가 아닌 비대칭성을 가진 새로운 형태의 비대칭 윈도우(asymmetry window)를 제안하였다. 비대칭 윈도우의 형태는 위도우를 중심으로 왼쪽편은 기울기가 심하고, 오른쪽은 기울기가 완만하여 음성의 기울기에 적합한 웨이팅을 갖는 형태이다. 제안한 비대칭 윈도우를 사용하여 PSOLA 합성을 하였을 경우 SNR 2~3dB 정도 향상되었음을 알 수 있다.

본 논문은 웨이브렛 패킷에 기초한 직교 다중화 변복조 방식을 소개하고 특히 시스템 설계자 입장에서 전송 신호의 특성을 시간-주파수 공간에서 신호 파형을 설계하고 채널 특성에 맞게 설계할 수 있는 궂를 제공하는 WPM(wavelet packet modulation) 방식을 기술하였다. 직교 기저 함수 집합을 만들어 시간주파수 공간을 임의적으로 partitioning하고 간섭 잡음 재철에 더잘 적응할 수 있는 구조를 찾는 방법을 소개하였고 튜닝 알고리듬의 실험적인 결과를 가지고 WPM변조 방식의 간섭 잡음에 대한 우수한 성능을 갖음을 보였다.

NLMS algorithm is degraded by the ambient noises and the near-end speech signals. In this paper, a robust acoustic echo cancellation algorithm is proposed. To enhance the echo cancellation performance, the step size of the proposed algorithm is normalized by the sum o fthe power of the reference signals and the primary signals. as results of comparing the excess mean square errors, it is shown that the proosed algorithm can enhance the performance of cancelling the echo signals. Some experiments, which is used multimedia personal computer, are carried out. As results of experiments, the proposed algorithm shows better performance than conventional ones.

In this study, we derived stop-and-go normalized DD, dual-mode normalized sato, dual-mode NCMA blind equalization algorithm for complex data. and then, the convergence characteristics of the proposed SG-NDD, dual-mode NSato blind equalization algorithms are compared with those of SG-DD, dual-mode sato algorithm. In genral, the normalized blind equalization algorithms have better convergence characteristics than the conventional algorithms.

본 논문은 음향반향제거기에서 동시통화구간을 검출하고 처리하는 알고리듬에 대하여 논하고 있다. 새로운 동시통화처리 알고리듬은 송신신호와 잔류신호간의 상호 상관계수를 이용하여 동시통화 여부 및 구간을 검출하고 이 구간에서는 반향제거기의 계수적응을 중지하도록 하는 것이다. 본 알고리듬을 사용함으로서 동시통화구간의 검출에 소요되는 계산량이 대폭 감소할 뿐만 아니라 비정상상태의 원인이 동시통화에 의한 것인지 아니면 반항경로응답의 변화에 의한 것인지를 구별할 수가 있다. 모의실험을 수행하여 본 알고리듬과 기존의 것과의 성능을 비교, 검토하였다. 실험 결과 몬 알고리듬을 적용하므로서 동시통화 후에도 계수가 발산하지한ㅎ고 안정적으로 정상상태를 유지함을 확인하였다.

This paper proposes a simple method to measure system's performance in target tracking problems. Essentially employing the Cramer-Rao lower bound (CRLB) on trakcing accuracy, an algorithm of predicting system's performance under various scenarios is developed. The input data is a collection of measurements over time fromsensors embedded in gaussian noise. The target of interest may not maneuver over the processing time interval while the own ship observing platform may maneuver in an arbitrary fashion. Th eproposed approach is demonstrated and discussed through simulation results.

This paper proposes a new wide-band beamforming which has guassian weighting. For the prupose of measurement this beam is formed by using all sensors toghether, not octave-by-octave. Weighting is applied to each sensor before time-delay compensation as a frequency-dependent function. As the resutls of the simulation of the proosed algorithm for nested linear array having 17 sensors for each octave, it is confirmed that beam can be formed with all sensors together and uniform directivity index can be achieved by proposed algorithm.

소나 시스템과 같이 방대한 양의 연산을 요구하는 고속 신호처리기를 구현하기 위해서는 상용 DSP 칩의 병렬 처리방법은 필요 불가결하다. 본 논문에서는 TI사의 TMS320C40을 이용한 병렬 신호 처리 시스템을 소개한다. TI사의 TMS320C40을 이용한 소나 시스템 신호처리부의 기본 모델을 제시하고, TI에서 제공하는 FFT구현 소스의 분석을 통한 연산의 수학적인 모델을 제시하고 이를 근거로 제안된 모델의 성능을 분석하였다.

본 논문에서는 원자로의 몸통부위와 상,하부 헤드에 존재하는 금속파편에 의한 충격위치 예측을 위한 알고리즘에 금속파편의 질량을 동시에 판별하는 프로그램을 접목시켜 금속파편의 충격위치와 질량을 동시에 판별할 수 있는 통합 환경 LPMS(loose parts monitoring system)에 관한 연구를 수행하였다. 또한 모의실험을 통하여 본 연구에서 제안된 통합 환경 LPMS 알고리즘이 금속파편의 위치와 질량 예측을 함에 있어서 우수한 성능을 보임을 확인하였다.

FM 문자다중방송의 일종인 DARC를 이용한 교통정보의 실시간 처리 시스템은 지능형 교통체계 (ITS)에서 필수적인 요소가 될 것이다. 본 논문에서는 다양한 교통정보 중 운전자가 필요로 하는 정보만을 선택하여, 신속.정확하게 제공하기 위한 방안으로 FM 문자다중방송(DARC)과 위치정보를 이용하는 방안을 제시하였다. 또한, 교통정보의 전달형태로 음성을 사용함으로써 운전중의 안전과 정보전달 과정에서의 효율성을 극대화하였다.

This paper proposes a moving image compression algorithm using subband coding that divides and processes the differential image of each moving image frame into subband frequency. This method decomposes the original image with each subband and performs DCT. The moving blocks are detected from each of wubband using threshold value which is select from the variance of difference between frames. then they are performed SBDCT and huffman coding in order to reduce the data. As the result of simulation, we confirmed that this method improved the blocking effect of reconstructed image in low bit rate.

Since the region segmentation at high resolution contains most of viable semantic object contours in an image, the bottom-up approach for image segmentation is appropriate for the application such as MPEG-4 which needs to preserve semantic object contours. However, the conventioal region merging methods, that follow the region segmentation, have poor performance in keeping low-contrast semantic object contours. In this paper, we propose an image segmentation algorithm based on classified region merging. The algorithm pre-segments an image with a large number of small regions, and also classifies it into several classes having similar gradient characteristics. Then regions only in the same class are merged according to the boundary weakness or statisticsal similarity. The simulation result shows that the proposed image segmentation preserves semantic object contours very well even with a small number of regions.

본 논문에서는 초저속 동영상 부호화에 관한 것으로, 움직임 추정 효율을 개선하기 위해 분할된 영역별로 움직임 정보를 추정하여 부호화를 수행할 뿐만 아니라 분할된 영역중 의미있는 부분을 선택적으로 부호화할 수 있는 영역분할 기반 영상부호화 기법에 관한 것이다. 첫째로, 움직임 추정은 분할된 영역 정보를 이용한 가변 탐색 영역 설정을 통해 전역 탐색 움직임 추정시 소모되는 많은 연산량을 줄이고, 둘째로, 움직임 추정 후 추정된 움직임 정보를 이용해 영역의 재분할 과정을 통해 분할된 영역별로 움직임 정보를 부호화 함으로써 개선된 부호화 효율을 보이며, 셋째로, 분할된 영역 중 얼굴과 같은 의미를 갖는 영역을 선택적으로 부호화하고 전송하기 위한 인식 기법을 제안하고자 한다.

In this paper, w epropose a fast fractal coding method based on correlation coefficients of subblocks in input image. In the proposed method, domain pool is selected based on correlation analysis of input image and the isometry transform for each block is chosen based on the IFS method. To investigate the performance of the proposed method, we compared image quality and encoding time with full search PIFS method and jacquin's PIFS method. Experimental results show that proposed method yields nearly the same performance in PSNR, and its encoding time is reduced for images size of 512*512 compared with full search PIFS method and jacquin's PIFS method.

Future land remote sensing satellite systems will kikely be constrained in terms of communication band-width. To alleviate this limitation, the data must be compressed. Image data obtained from satellite exhibit a high degree of spatial and spectral correlations that must be properly exploited. In this paper we propose multispectral image data compression using wavelet transform and selective predicted vector quantization. Th eproposed method is based on accuratly predicting other band from reference band and reducing bit rate through threshold map. we can achieve better compression effeciency than conventional methods.

In this paper, we proposed a lwo bit rate subband coding with adaptive vector quantization using the correlation between motion vector and block energy in subband. In this method, the difference between the input signal and the motion compensated interframe prediction signal is decomposed into several narrow bands using quadrature mirror filter (QMF) structure. The subband signals are then quantized by adaptive vector quantizers. In the codebook generating process, each classified region closer to the block value in the same region after the classification of region by the magnitude of motion vector and the variance values of subband block. Because codebook is genrated considering energy distribution of each region classified by motion vector and variance of subband block, this technique gives a very good visual quality at low bit rate coding.

본 논문에서는 객체기반의 스테레오 동영상 부호화기를 제안한다. 제안된 부호화기는 움직임, 변이, 그리교 밝기 정보를 이용하여 객체를 추출하고 객체의 움직임과 변이 정보를 예측하여 부호화함으로써 기존의 블록기반 방법에 비해 정보량을 줄일 수 있었다. 또한 새롭게 발생되는 영역과 기존 영역의 소멸 및 변형은 부호화 효율을 떨어뜨리게 되는데, 이를 해결하기 위해 새로운 객체갱신 방법이 제안되었다. 실험을 통하여 제안된 객체기반 부호화기는 기존의 블록기반 부호화기에서 나타나는 블록화 및 모스키토 현상을 제거할 수 있었고 복원화질과 부호호율을 개선시킬 수 있었다.

Measurement of optical flow is a core problem of matching through the analysis of image sequences. In this paper, horn's and anandan's techniques are analyzed to derive a better technique for matching. Experimental results show that Horn's technique has low accuracy in measuring the velocity of optical flow while anandan's technique has poor performaance for diverging images. Based upon this observation, a new technique for the measurement of optical flow is proposed.

Matching two images is an essential step for many computer vision applications. A new approach to the scale and rotation invariant scene matching is presented. A set of andidate parameters are hypthesized by mapping the angular difference and a new distance measure to the hough space and by detecting maximally consistent points. The proposed method is shown to be much faster than the conventinal one where the relaxation process is repeated until convergence, while providing robust matching performance, without a priori information on the geometrical transformation parameters.

For camera calibration, this paper describes two steps to camera constants and camera parameters. The former is the radial distortion of lens, image center and focal length etc.. The latter is translation, rotation etc.. Camera calibration use tsai's algorithm. In this paper, the solutions are introduced into overdetermined system as matching points that are acquired from two CCD and measured object depth information.

스테레오 영상으로부터 3차원 거리 정보를 추출하는 일은 수학적으로 불량 설정 문제(ill-posed problem)이다. 본 논문은 스테레오 정합을 정칙화(regularization)와 최소화(minimization) 문제로 설정하여 변이 (disparity)를 구한다. 최소화할 에너지 함수로 기울기 벡터(gradient vector)를 사용하여 촤우 카메라의 차이에 대응하고, 영상에 존재하는 깊이 불연속점(depth discontinuties)을 찾아내어 이를 보존하면서 정칙화를 실행하는 스테레오 정합 알고리즘을 제시한다. 다양한 스테레오 영상에 대한 실험 결과를 함께 나타내었다.

본 논문에서는 스테레오 정합에서 불연속성과 폐색영역을 고려하고 다중 계층 구조를 사용한 relaxation 알고리듬을 이용하여 좌우 영상간의 최적의 정합점을 찾는 스테레오 정합을 구현하였다. 잡음을 제거하기 위해 베이시안 필터링 방법을 사용하였다. 좌우 영상에서 얻은 변이(disparity)를 이용하여 좌우 폐색영역(occlusion region)을 구한 후, 다시 최적의 변이를 얻기 위해 내삼과정을 통해 양쪽의 정보를 상호 보완하는 병렬적인 relaxation 방법을 구현한다. 구현한 방법은 다양한 영상에 적용하였다.

In this paper, a new algorithm is proposed that the template matching is used to devide SOFM (self-organizig feature map) for fast learning and to extract features for considering input data types. In order to verify the superoprity of the proposed algorithm, applied to the recognition of handwritten numerals. Templates of handwritten numerals are created by a line of external-contact.

본 논문에서는 지능형 교통 정보처리 시스템의 구축을 위해 기존의 불완전한 영상분할 기법을 후처리 과정에서 자동차의 특징과 context 정보로 구성된 지식기반에 의한 전문가 시스템으로 보완함으로써 자연적인 영상에서도 정확한 영상분할이 가능한 컴퓨터 시각장치 알고리듬을 제안하였다. 다양한 환경에서 구축한 3000여개의 영상 데이터베이스 중 100여개의 영상에 대해 위의 차량검출 기법을 적용한 결과, 영상의 대비도가 높을 경우 대부분의 영상에서 약 10%의 오차 범위 내에서 정확한 영역검출이 가능하였으며, 교통량의 최적제어를 위한 지능적 신호체계 구축의 자동화에 사용될 수 있다.

We propose the dejittering method against the jitter originated from the cell losses in ATM network when CBR traffic is transfereed on AAL5. Cell numbering algong with maintaining a timer at receier side introduces 0.669 times peak-to-peak PDV of the AAL5 PDU to that in ITU-T AAL5 standard, and the AAL5 user-HDTV decoder-receives the same or more error-free transport packets in the proposed algorithm than those in the ITU-T AAL5 standard for the same network simulation environment.

In this paper, we propose a VLSI architecture for implementing a crecently proposed fast block matching algorithm, which is called the HSBMA3S. The proposed architecture consists of a systolic array based basic unit and two shift register arrays. And it covers a search range of -32 ~+31. By using a basic unit repeatedly, we can redcue the number of gates. To implement the basic unit, we can select one among various conventional systolic arrays by trading-off between speed and hardware cost. In this paper, the architecture for the basic unit is selected so that the hardware cost can be minimized. The proposed architecture is fast enough for low bit-rate applications (frame size of 352x288, 30 frames/sec) and can be implemented by less than 20,000 gates. Moreover, by simply modifying the basic unit, the architecture can be used for the higher bit-rate application of the frame size of 720*480 and 30 frames/sec.

A circular image in a space don't appear as an exact circular image and appear as an oval in image buffer because distortion of camera lens, number of horizontal pixel of CCD photographing element and unmber of horizontal pixel of image buffer are not in accordance. By using familiar 3-D coordinate, know as circle's diameter, and cicle's center, you correct a pin-hole camera and get an exact circle with reprojection a circle into image buffer, according to a perspective.

흉부 단순 X선 사진에 있어서 폐의 간질성 음영의 특징을 컴퓨터를 이용하여 해석하는 시스템을 구축하여 그 정확도를 검토하였다. 이 시스템에서는 X선 사진을 디지털화 해서 관심영역을 설정하여, 4방향 라플라시안 -가우시안 형상필터를 행하여 2치화 한후, 계속해서 선상음영 판정처리 및 선상음영 차분처리를 햐해 간질성 음영의 특징을 나타내는 몇가지 물리적인 지표를 계산하였다. 82명의 정상예와 이상예에 대하여 이들 물리적 지표의 ROC 해석을 행하여 간질성 폐질환 변화의 양호한 검출결과를 얻었다.

ITU-T recommends T.120 for multimedia conferencing data protocols. In this paper, generic model of application protocols is shown for guaranteeting internetworking between application protocols in application users. And using generic conference control and multipoint communication service, multipoint binary file transfer application protocol is shown throughout application protocol generic model on public session environment.

In this paper, we propose a new method for content-based retrieval system using wavelet transform and correlation, which has were used in signal processing and image compressing. The matching method is used not perfect matching but similar matching. Used feature vector is the lowest frequency(LL) itself, energy value, and edge information of 4-layer, after computng a 4-layer 2-D fast wavelet transform on image. By the proosed algorithm, we got the result that was faste rand more accurate than the traditional algorithm. Because used feature vector was compressed 256:1 over original image, retrieval speed was highly improved. By using correlation, moving object with size variation was reterieved without additional feature information.

Under the worldwide booming internet environment, there has been increasing demand for various multimedia services. Especially the demand for effective video services has been rapidly increased. In this paper, we describe the implementation of video service system for internet use based on H/263 video compression technique and UDP socket on the TCP/IP environment. In addition, by using the plug-in-play technique, the implemented system improved user interface for correct retrieval and easy usage.

In the area of the multimedia computer, the demand for the image processing engineer is large. However, there are not many engineers capable of developing practical systems. To teach practical image processing techniques, we need an intergrated education systems which can efficiently present the image processing theory and, at the same time, provide interactive experiments for the theory presented. In this paper, we propose an integrated education systems for the image processing. It consists of the theory presentation systems and the experiment systems. The theory presentation systems uspports multimedia display functions and HTML. It is tightly integrated with the experiment systems which developed based on the integrated image processing algorithm development system called Hello-vision.

Recently, image and voice processing part is using wavelet transform. We propose thining algorithm using wavelet tranform. Wavelet transform consists of low frequency and high frequency in the spatial and frequency domain. After the wavelet decomposition, more than 90 percents of energy are contained in lowest frequency band. Therefor, for images with large difference of gray value between foreground and background like character images, we can more accurately in the lowest frequency band. Lowest frequency band has wavelet transform significant coefficient(WTS) that is required for the thinning algorithm we proposed Paper [3][5][7][8] can not separate consonants and vowels of korean characters. Becuase korean characters have structural feature. This paper can separate consonants and vowels. Simulation executed low frequency image and data compression can reduce 1/4$^{n}$ with level n. we can redcue time complexity 3/8.

Recognizing a human part becomes very important for applications which are based on the interaction between computers and their users. In this paper, we design and implement a system which recognizes and tracks a human head using a sequence of images. Difference images are used to easily extract feature vectors from images with very complex backgrounds. A human bhead is represented with an ellipse and recognized by searching for a maximum value from preprocessed gradient images. The method is developed by considering the fact that the tracking system should be real-time. The designed system not only shows an excellent performance for the normal up-right position of the head, but also for the cases of 360.deg. rotated head position, occluded images of heads, and tilted head positions.

본 논문은 연속적인 수화 동영상에 있어서 가장 의미가 있는 손 영역을 검출하는 방법에 대한 연구이다. 부자연스러운 센서 등을 피하고 사용자 친화성과 범용성 및 실시간 구현을 도모하기 위하여 흑백 동영상의 처리가 요구된다. 농아자의 수화를 인식하기 위해 전철로서, 복잡한 배경으로부터 손 영역의 검출은 매우 중요하다. 정확한 손 영역의 검출로 수화 인식률의 개선과 영상처리의 고속화가 가능하다. 제안한 방법은 인접한 프레임간의 차 영상에 대하여 엔트로피를 측정하여 배경영상과 분리를 행한 후, 손 영역에 대한 추적을 수행하였다.

A new motion estimation algorithm with lower computational complexity and good image quality when compared to the FBMA will be presented in this paper. In the proposed method, by considering the relation between neighboring blocks, the searching area in the algorithm is adjustable according to mean absolute difference of the block. By the computer simulation the computation amount of the proposed than that of the FBMA and the good result of the PSNR can be attained.

본 논문에서는 CCCD 카메라로부터 획득한 영상시퀀스들에서 인접한 두 영상 사이의 차영상과 얼굴이 가지는 컬러정보를 이용하여 분리한 얼굴 영역에서 양쪽 눈과 입의 위치좌표를 특징점으로 이용하여 입력된 얼굴에 자동정합된 변형가능 템플리트(deformable template)를 가지고 연속된 다음 프레임에서 얼굴 전체를 트래킹하는 알고리듬을 제안한다. 실제 입력영상의 얼굴 영역과 변형 가능 템플리트의 차이를 비교하기 위해 텍스쳐 매핑(Texture mapping)을 도입하여 트래킹의 정확도를 살펴본다.

Level set method is used for extracting a shape within an image. This method is based on the ideas developed by osher and sethian to model propagating solid/liquid interfaces moving with curvature-dependent speeds. In this paper, we propose the new generalized formulation and stopping function of evolution for more accurate extraction of shape. Our method shows good results in some cases compared with the conventional ones.

This paper present an efficient techniques for visual inspection of taped electronic parts, suitable for real time implementation. The main environments of developed system are IBM-compatible personal computer, frame grabber, digital input-output board. It is connected to the programmable logic controller unit of the taping machine in real time. Using a queuing structure, operator or extractor machine can remove easily the defect one from production line. Also, we design a new illumination system for sacquring shape and subface features of object. Therefore, it redue pre-processing step and processing time.

In this paper, we propose an efficient multi-view images coding algorithm which finds the optimal texture from the restricted number of multi-view images. The X-Y plane of the normalized object space is divided into triangular patches. The depth value of the node is determined by applying the block based disparity compensation method and then the texture of the each patch is extracted by applying the affine transformation patch is extracted by applying the affine transformation based disparity compensation method to the multi-view images. We restricted the number of images contributed to determining the texture comapred to traditional methods which use all the multi-view images in the texture extraction. Experimental results show that the SNR of images encoded by the proposed algorithm is better than that of imaes encoded by the traditional method by the amount about 0.2dB for the test sets of multi-view images called dragon, kid, city and santa. The recovered images from the encoded data by the proposed method show the better visual images than the recovered images from the encoded data by the traditional methods.

물체의 3차원적인 정보를 복원하는 일은 그 정보의 일련된 이용에 있어서 중요한 문제이다. 이를 위해 여러가지 방법들이 연구되고 있으며, 그 중 shape from focus(SFF) 방법은 영상의 초점이 맞는 렌즈의 위치를 찾아내어 렌즈 공식에 의해 초점이 맞는 부분의 거리 정보를 구할 수 있다. 기존의 이 방법은 초점이 맞았는지의 정도를 계산하기 위한 focus measure 값들을 카메라의 광학축에 수직인 단순한 평면으로 가정하여 그 합이 최대가 되는 위치를 찾아내었다. 이를 개선하기 위해서 focused image surface(FIS) 개념이 연구되었고 그로 인해 더욱 나아진 결과를 얻었다. 물체의 FIS는 카메라 렌즈에 의해 초점이 맞게된 물체의 점들의 집합으로 이루어진 공간상의 면이다. 기하광학에 의해 물체의 모양과 FIS 상이에는 일대일 대응 관계가 있고 FIS의 형태를 구하는것이 결국은 물체의 모양을 복원하는것이다. FIS 개념을 처음 적용할 때는 물체의 모양이 부분적으로 영상 탐지기(image detector)와 같은 평면으로 가정하여 3차원 공간상에서 가능한 모든 방향의 평면에 대한 focus measure를 구하여 그 값이 최대가 되는 렌즈의 위치를 구하였다. 그러나 이러한 방법은 focus measure의 합이 정사각형의 윈도우에서 계산되기 때문에 곡면으로 이루어진 실제 물체에서는 오차르 ㄹ가지게 된다. 본 논문에서는 이와는 달이 평면이 아닌 곡면에 대한 focus measure의 합이 최대가 되는 렌즈의 위치를 구하여 이전의 방법들 보다 정확한 복원이 가능함을 보인다.

An image segmentation approach based on the fuzzy worm searching and MIN-MAx clusterng algorithm is proposed in this paper. This algorithm deals with fuzzy worm value and min-max node at a gross scene level, which investigates the edge information including fuzzy worm action. But current segmentation methods based edge extraction methods generally need the mask information for the algebraic model, and take long run times at mask operation, wheras the proposed algorithm has single operation ccording to active searching of fuzzy worms. In addition, we also genetic min-max clustering using genetic algorithm to complete clustering and fuzyz searching on grey-histogram of image for the optimum solution, which can automatically determine the size of rnages and has both strong robust and speedy calculation. The simulation results showed that the proposed algorithm adaptively divided the quantized images in histogram region and performed single searching methods, significantly alleviating the increase of the computational load and the memory requirements.

In this paper we consider strongly polynomial variations of the auction algorithm for the single origin/all destinations shortest path problem. These variations are based on the idea of graph reduction, that is, deleting unnecessary arcs of the graph by using certain bounds naturally obtained in the course of the algorithm. We study the structure of the reduced graph and we exploit this structure to obtain algorithm with O(n min{m, nlogn}) and O(n$^{2}$) running time.

칩의 집적도에 비레한 테스트 문제의 원초적인 해결은 VHDL등으로 기술되는 상위기능 수준에서부터 고려되어야 한다. 본 논문에서는 상위수준의 기능정보에서 테스트점을 삽입 제어흐름(control flow)를 변경하여 고집적 회로의 고장점검도를 증진시키는 기술을 소개한다. while 푸프와 if-then-else 제어문에 AND 및 OR 타입 등의 테스점을 삽입하여 내부 신호의 조정도를 최적화시킨다. 랜덤패턴 시뮬레이션을 벤치마크 회로에 적용 각 변수의 조정도를 산출하여 테스트점의 종류 및 삽입할 위치를 결정하였다. 본 연구에서 제안하는 상대적 랜덤도에 의하여 VHDL 코드에 단일 테스트점을 삽입 합성한 결과 게이트 수준회로에 대한 고장점검도가 최대 30% 까지 증진됨을 알 수 있었다.

IEEE1149.1 바운다리스캔은 칩과 칩간의 연결선상에서 발생가능한 지연고장을 점검 할 수 없게 설계되어있다. 칩에서 패턴을 주입하는 UpdateDR과 연결선을 통해서 전달된 결과 값을 관측하는 captureDR간의 간격이 ITCK가 되도록 UPdaeDR을 변경하는 기술보다 동작속도 및 추가영역면에서 최적임을 보여준다.

본 논무은 HLS에서 CPLD 조건식을 고려한 RTL바인딩 기술로서 HDL로 기술된 회로의 스케쥴링을 한후 모듈 연산 간격을 고려하여 합당한 모듈을 선택하고 스케쥴링과 할당을 수행한 후 주어진 조건식에 맞도록 CPLD를 선정한다. 또한 할당된 결과의 모듈을 CPLD 내부의 CLB의 크기를 고려하여 부울식을 분할하고 최적의 CLB를 사용하여 회로를 구현할 수 있도록 binding 알고리즘을 제안하였다.

This thesis proposes a new heuristic algorithm of integrated allocation and binding for high level synthesis. The proposed algorithm simultaneously allocates binds functional units, interconnections and registers by considering interdependency between operations and storage elements in each control step, in order to share registers and interconnections connected to functional units, as much as possible. This thesis shows the effectiveness of the algorithm by comparing the results of out experiments with those of existing system.

This paper presents BIST (Built-in self-test) techniques for detection of path delay faults in digital circuits. In the proosed BIST schemes, the shift registers make possible to concurrently generate and compact the latched test data. Therefore the test time is reduced efficiently. By reordering the elements of th shifte register based on the information of the hamming distance of each memory elements in CUt, it is possible to increase the number of path delay faults detected robustly/non-robustly. Experimental results for ISCAS'89 benchmark circuits show the efficiency of the proposed BIST techniques.

REcent research about current testing($\textrm{I}_{DDQ}$ testing) has been emphasizing that $\textrm{I}_{DDQ}$ testing in addition to the logical voltage testing is necessary to increase the fault coverage. The $\textrm{I}_{DDQ}$. testing can detect physical faults other than the classical stuck-at type fault, which affect reliability. One of the most critical issues in the $\textrm{I}_{DDQ}$ testing is to insert a built-in current sensor (BICS) that can detect abnormal static currents from the power supply or to the ground. This paper presents a new BICS for internal current testing for large CMOS logic circuits. The proposed BICS uses a single phase clock to minimize the hardware overhead. It detects faulty current flowing and converts it into a corresponding logic voltage level to make converts it into a corresponding logic voltage level to make it possible to use the conventional voltage testing techniqeus. By using current mirroring technique, the proposed BICS can work at very high speed. Because the proposed BICS almost does not affects normal operation of CUT(circuit under test), it can be used to a very large circuit without circuit partitioning. By altenating the operational modes, a circuit can be $\textrm{I}_{DDQ}$-tested as a kind of self-testing fashion by using the proposed BICS.

Hardware formal verification involves the use of analytical techniques to prove that the implementation of a system confroms to the specification. The specification could be a set of properties that the system must have or it could be an alternative representation of the system behavior. We can represent our behavioral specification to be written in VHDL coding. In this paper, we proposed a new hardware design verification method. For theis method, we assumed that a verification pattern already exists and try to make an algorithm to find a place where a design error occurred. This method uses an hierarchical approach by making control flow graph(CFG) hierarchically. From the simulation, this method was turned out to be very effective that all the assumed design errors could be detected.

The high complexity of digital circuits has changed the digital circuits design mehtods from schemeatic-based to hardware description languages like VHDL, verilog that make hardware faults become more hard to detect. Thus test generation to detect hardware defects is very important part of the design. But most of the test generation methods are gate-level based. In this paper new high-level test generation method to detect stuck-at-faults on gate level is described. This test generation method is independent of synthesis results and reduce the time and efforts for test generation.

This paper addresses a characterization of fault effects in asynchronous circuits. A characterization has been performed on races caused by a single stuck-at faults (SSAF). The faults sometimes lead to races in faulty circuits, which prevent faults from observing and the circuit is insufficiently tested. To identify those obstacles, we have proposed non-detectable single stuck-at fault(NDSSAF) conditions and proposed an algorithm to find them. In the help of the proposed methodology, the asynchronous circuits can be fully SSAF testable.

멀티미디어 응용분야에 사용하는 데이터는 매우 크다. 이러한 대용량의 데이터를 관리하기 위해서 압축은 반듯이 필요하다. 본 논문에서는 허프만 부호화, 복호호 과정의 불규칙한 정렬구조와, 검색과정을 없애서 고속의 허프만 부호화 및 복호화가 가능한 허프만 코덱의 하드웨어 구조를 제안하였다.

This paper describes a bus interface unit which is used in a 32bit high-performance multimedia RISC CPU including DSP unit. The main idea adopted in designing is that the bus interface unit enables the processor to provide on-chip functions for controlling memory and peripheral devices, including RAS-cAS multiplexing, DRAM refresh and parity generation and checking. The number of bus cycles used for a memory or I/O access is also defined by the processor, thus, no external bus controllers are required. All memories and peripheral devices can be connected directly, pin to pin, without any glue logic. That is the key point of the design.

본 연구에서는 칼라 CCD 센서를 제어하여, shading과 .gamma. correction 된 데이터를 읽어 들여, 이를 이진레벨 데이터로 바꾼후, 원래의 다치레벨 또는 이진레벨 데이터를 SCSI나 DMA I/F를 통해 전달하는 ASIC을 설계하였다. 본 ASIC에서는 이진화를 위하여 문자 모드에서는 simple threshold와 LAT(local adaptive threshold) 알고리즘을, 그림모드에서는 stucki error diffusion 알고리즘을 적용하였다. 그리고, 구성은 CCD센서 제어블락, 스텝 모타 제어제어블락, 이미지 축소블락, 데이터 이진화 블락, 그리고 DATA I/F 블락 등으로 이루어져 있다. 또한 사용된 technology는 삼성 0.5um CMOS standard cell이며, 크기는 45K gates(내부 메모리 제외)이고, 160QFP package로 구현되었다. ㅎㅁㅅㄷㄴ (soqn apa

The wide spread of multimedia system demands a large viewin gdesply device which can inform a message to many peoples in open area. This paper is about the design, simulating and testing of a large viewing LED pannel control ASIC(application specific integrated circuit). This LED pannel control ASIC runs on 16 bit microprocessor MC68EC000 and has following functions:16 line interlaced LED pannel controller, memory controller, 16 channel priority inerrupt controller, 2 channel direct memory access controller, 2 channel 12 bit clock and timer, 2 channel infrared remocon receiver, 2 channel RS-232C with 16byte FIFO, IBM PC/AT compatible keyboard interface, battery backuped real time clock, ISA bus controller, battery backuped 256 byte SRAM and watech dog timer. The 0.6micron CMOS sea of gate is used to design the ASIC in amount of about 39,000 gates.

산술연산을 수행하는 가산기는 ALU(arithmetic logic unit)의 성능을 좌우하는데 매우 중요한 역할을 하며, 어떠한 캐리 생성 방식을 사용하는냐에 따라 그 성능이 결정될 수 있다. RCA(Ripple carry adder)는 간단하고, 쉬운 설게로 널리 사용되자만, 캐리의 전파지연 문제로 인해 고속의 가산기 응용에의 부적합하다. 또한, CLA(carry lookahead adder)방식의 가산기는 캐리의 지연시간이 가산기의 단수와 무관하므로, 연산속도를 높일 수 있는 장점이 있지만 더하고자 하는 bit의 수가 클수록 회로가 매우 복잡해지는 큰 단점을 가지고 있다. 따라서, 본 논문에서는 간단하면서도 성능이 우수한 64bit 가산기를 설계하고 시뮬레이션을 통하여 설계된 회로의 우수성을 증명하였다.

One of the most difficult problem of designing VLSI is a mixed-circuit design, that is to design circuit containing both analog parts and digital parts. Digital to analog converter and analog to digital converter is a typical case. Especially it can be a serious problem when mixed circuit are put into a large digital circuit like microcontroller. However nowadays this problem is settled by separating analog circuit parts outside the IC. This technique is based on converting a digital signal into a pulse sequence. Then an analog signal is obtained by averaging this pulse sequence at the external low-pass filter. An anlog to digital converter is designed using a stochastic logic instead of a traditional PWM (pulse-width modulation) signal and ins implemente dusing FPGa. Stochastic pulse sequence can be made as a simple circuits and moreover can be mathematically processed by simple circuits -AND gates. The spectral property of stochastic pulse sequence method is better than that of PWM method. So it make easy to design a external low-pass filter. This technique has important advantages, especially the reduction of the ADC cost.

본 연구는 ring oscillator 및 이의 delay까지도 계수화 함으롯 고정밀 pulse modulation 을 시키기 위한 것이다. 특히 ring oscillator는 온ㄷ, 전압 및 공정에 따라 주파수 변화폭이 심하므로 이에대한 보정 방법과 높은 발진으로 인해 실리콘(반도체) 제작시의 열문제를 최소화 하는 방안에 대해서도 언급하였다. 본 논문은 삼성반도체 0.6um CMOS standar cell을 사용하여 구현 했으며, 레이져 프린터의 해상도 향상을 위해 gray image와 text에 대한 적용 예로서 화상의 미세수평이도오가 화소의 크기를 제어하는 예를 보였다. 또한 향후에는 칼라 프린터에 있어서 인쇄화상의 품질 향상을 위한 정밀 보정에도 응용할수 있음을 보였다.

본 연구에서는 MML(merged memory logic)구조를 갖는 스캔라인 컨버터를 설계하여 제안한다. 비월주사 방식인 TV 비디오 신호를 FIFO 메모리에 저장하여 순차주사방식인 VGA 비디오 시모호 변환하는 주사선 변환기를 MML 개념으로 설계하였다. MML 회로는 VHDL로 설계하여 V-system으로 시뮬레이션을 수행하고 altera FPGA에 구현한 후, TV 비디오 신호를 PC 모니터로 보기 위한 외장형 tV 수신 시스템에 적용하여 성능을 검증했다. MML 개념으로 설계된 컨버터는 system-on-a-chip 설계의 첫 단계로 메모리와 로직부분으로 구성된 일반적인 컨버터보다 효율적인 시스템 설계를 할 수 있다.

An offset compensation scheme for rail-to-rail CMOS op-amps with complementary input stages is presented. Two auxiliary amplifiers are used to compensate for the offsets of NMOS and PMOS differnetial input stages, and ping-pong control is employed for continuous-time operation. Simulation and measurements resutls show that offsets are reduced about 20 times by this scheme.

In this paper two quick boosting charge pump circuits for high-speed EEPROM memory are proposed. In order to improve initial charge transfer efficiency, one uses weighted capacitors where each stage has different clock coupling capacitance, and the other uses a multi-path structure at the first stage. SPICE simulation results show that these charge pumps have improve drising-time characteristics, but their $V_{DD}$ mean currents are increased a little compared with conventioanl charge pumps. The rising time upt o 15V of the proposed charge pumps is 3 times faster than that of dickson's pump at the cost of 1.5 tiems more $V_{DD}$ mean current.rrent.

In this paper we simulated 0.30um NMOS transitor to analysis hot carrier degradation depend on As, As+P, P LDD structure. As a result we obtained As+P LDD structure was good hot carrier immunity. Also we find that hog carrier life time improved a sincresing P dose due to P dose helps in grading the nLDD junction. However As-only junction was poor due to junction high peak position located near the surface.

A new transmission line parameter extraction method of iC interconnects on silicon substrate is presented. To extract the acurate parameters, the silicon substrate effects were taken into account. Since the electromagnetic fields under the silicon substrate are propagated with slow wave mode, effective dielectric constant and different ground plane with the multi-layer dielectric structures were employed for inductance and capacitance matrix determination. Then accurate signal transients simulation were performed with HSPICE by using the parameters. It was shown that the simulation resutls has an excellent agreement with TDR/TDT measurements.

In this thesis, we present efficient techniques to reduce the switching activity in a CMOS combinational logic network based on local logic transforms. But this techniques is not appropriate in the view of testability because of deteriorating the random pattern testability of a circuit. This thesis proposes a circuit design method having two operation modes. For the sake of power dissipation(normal operation mode), a gate output switches as rarely as possible, implying highly skewed signal probabilities for 1 or 0. On the other hand, at test mode, signals have probabilities of being 1 or 0 approaching 0.5, so it is possible to exact both stuck-at faults on the wire. Therefore, the goals of synthesis for low power and random pattern testability are achieved. The hardware overhead sof proposed design method are only one primary input for mode selection and AND/OR gate for each redundant connection.

본 논문에서는 리드 피치가 15mil인 208핀 QFP칩을 검사하는 고해상도의 비젼 모델 시스템을 개발하였다. 전체적인 하드웨어와 고해상도에 적합한 알고리즘 및 이를 실현하는 프로그램을 개발하였으며, 특히 칩 리드 검출을 위한 새로운 알고리즘을 제안하였다. 본 논문에서는 제안한 방법의 특징은 조명의 간섭에 대해 강하며 검사 속도 및 정밀도가 향상되었다.

We introduce a high level synthesis methodlogoy for automatic synthesis of asynchronous circuits form a language based on CSP. The input language is a high level concurrent algorithmic specification that can model complex concurrent control flow, logical and arithmetic computation and communications between them. This specification is translated into petri net which has actions. These actions are refined to synthesize the controllers and to allocate the data resources. We use the automatic synthesis through signal transition graphs (STGs) that allows to take advantage of logic synthsis methods to optimize the circuit.

We present a method for initialization of asynchronous circuits using binary decision space representation. From state transition graph(STG) which is given as a specification a circuit, the BDD is generated to solve the state space explosion problem which is caused by concurrecy of STG. We first step, we construct the necessary informaton as a form of K-map from BDD, then find an initial state on the K-map by assignment of don't care assignment.

A new evolutionary programming technique using multiple subpopulations with completely differnt evolution mechanisms is propsed to solve the optimization problems. Three subpopulations, each has different evolution charcteristics and uses different EP algorithms such as SAEP, AEP and FEP, are cooperating with synergy effect in which it increases the possibility to quickly find the global optimum of continuous optimization problems. Subpopulations evolve in differnt manner and the interaction among these leads to global minimum quickly.

Through fuzzy logic controller is very useful to many areas, it is difficult to build up the rule-base by experience and trial-error. So, effective self-tuning fuzzy controller for the position control of ball and beam is designed. In this paper, we developed the neural network control system with fuzzy-neuron which conducts the adjustment process for the parameters to satisfy have nonlinear property of the ball and beam system. The proposed algorithm is based on a fuzzy logic control system using a neural network learinign algorithm which is a back-propagation algorithm. This system learn membership functions with input variables. The purpose of the design is to control the position of the ball along the track by manipulating the angualr position of the serve. As a result, it is concluded that the neural network control system with fuzzy-neuron is more effective than the conventional fuzzy system.

In this paper, we describes PID-neural network controller for the rotary inverted pendulum. PID control is applied to many fields but has some problems in nonlinear system due to a variation of parameter. So, we should desing the controller which is adjusted PI parameters by the neural network which is learned by backpropagation algorithm. And we show that on-line control is possible through the PID-neural network controller. The angle of the pendulum is controlled and then the position of the rotating arm is also controlled to maintain with in the set point. Measurement of the pendulum angle is obtained using a potentionmeter. The objective of the experiment is to design a PID-neural network control system that positions the arm as well as maintains the ivnerted pendulum vertical. Finally, we describe the actual experiment system and confirm the experimental results.

In this paper, we propose an identification method for nonlinear systems. In order to identify the nonlinear system parameters, we are represented the linearization from the nonlinear system, and use a genetic algorithm(GA). The parameters are coded into binary string and searched by GA. The simulation results show the effectiveness of the proposed approach.

In this paper, a new connection admission controller using neural network is presented. The controller measures traffic flow, cell loss rate, and cell delay periodically. Using those measured information, it learns the distributions of traffics of each traffic. Also the proposed controller is able to measure and manage the delays that source traffics experience through the network by using DWRR multiplexer with buffers dedicated to each traffic source. Experimental result show that the heterogeneous traffic sources with various QoS requirement.

In this paper, to provide the robustness of character recognition, we propose a recognition method using the dilated boundary curve feature which has the invariance characteristics for the shift, scale, and rotation changes of character pattern. And its some characteristics and effectieness are evaluated through the experiments for both the english alphabets and the numeral digits. The feature vector is represented by the fourier descriptor for a boundary curve of the dilated character pattern which is generated by the circular mask dilation method, and is used for a nearest neighbort classifier(NNC) or a nearest neighbor mean classifier(NNMC). These the processing time and the recognition rate, and take also the robustness of recognition for both some internal noise and partial corruption of an image pattern.

In this paper, we propose a modified elman neural network structure for nonlinear system identification. The proposed structure is that all of network output feed back into hidden units and output units. Learning algorithm is standard back-propagation algorithm. The simulation showed the effectiveness of using the modified elman neural network structure in the nonlinear system identification.

In this paper, we investigate a system identification with a single neuron. The identifed performance of multi-perceptron is compared with proposed identification model and simulted.

In this paper, we implemented a sign language communication system between korean and japanese using 3D model. Unitl now, we have developed the system through 2D image. The system, however, has some drawbacks base don the limitations of 2D model. Especially it is not comprehensive in the sign language gesture including forward or backward movements becasue it can only display a front view. Therefore, we employed a 3D model for the representation of sign language. The proposed communication system si implemented with windows 95 visual C++ 5.0 imported open inventor library on pentium 233 system. Experimented results show that the 3D system can solve the problems in 2D representation.

In this paper we propose a method of representing emotional experessions and lip shapes for sign language communication using 3-dimensional model. At first we employ the action units (AU) of facial action coding system(FACS) to display all shapes. Then we define 11 basic lip shapes and sounding times of each components in a syllable in order to synthesize the lip shapes more precisely for korean characters. Experimental results show that the proposed method could be used efficiently for the sign language image communication between different languages.

The HMM-Net is an architecture for a neural network that implements a hidden markov model (HMM). The architecture is developed for the purpose of combining the discriminant power of neural networks with the time-domain modeling capability of HMMs. Criteria used for learning HMM-Net classifiers are maximum likelihood(ML) and minimization of mean squared error(MMSE). In this paper we propose an efficient learning method of HMM_Net classifiers using a ML-MMSE hybrid criterion and report the results of an experimental study comparing the performance of HMM_Net classifiers trained by the gradient descent algorithm with the above criteria. Experimental results for the isolated numeric digits from /0/ to /9/ show that the performance of the proposed method is better than the others in the repects of learning and recognition rates.

본 논문에서는 고밀도 잡음에 효과적인 WFM(weighted fuzzy mean) 필터를 이용하여 잡음게거능력 향상과 정보 손실을 최소화할 수 있도록 WFM의 성능을 개선하기 위하여 게층적 구조의 DBNN에 퍼지알고리즘을 적용하여 구현하였으며, 모의실험을 통하여 성능을 비교, 분석하였다. 모의실험 결과 boats의 영상에 대해서 MSE를 비교 분석한 결과 가중 FDNN이 우수함을 확인하였다.

본 논문은 조명의 전체 혹은 국부적인 변화에 견실한 특성을 갖는 동작검출 기법에 관한 것이다. 피사체의 움직임과 조명이 변화의 특성을 파악하기 위해 본 논문에서는 두 영상의 상관계수와 영상의 차이를 종합적으로 이용하여 움직임 여부를 판단할 수 있는 지표를 산정하는 기법을 제시한다.

오늘날 항만에서의 컨테이너 처리를 하지 못하고 있는 가운데 자동화 처리를 위해서는 식별자의 인식이 가장 큰 문제점을 해결하기 위해 항만에서 컨테이너의 식별자 인식에 신경망 알고리즘의 하나인 back-propagation을 사용하여 기존의 식별자 인식 방법보다 신속하고 정확한 처리가 가능하도록 구현하였다.

This paper proposed that a high power AC-DC buck converter topology of high efficiency using loss-less snubber operates with four chopper connecting a number of parallel circuit. To improve these, a large number of soft switching topologies included a resonant circuit have been proposed. And, some simulative resutls on computer is included to confirm the validity of the analytical results. The partial resonant circuit makes use of a inductor using step-down and a condenser of loss-less snubber. The result is that the switching loss is very low and the efficiency of system is high. And the snubber condenser used in partial resonant circuit makes charging energy regenerated at input power source of rresonant operation. The proposed conversion system is deemed the most suitable for high power applications where the power switching devices are used.

Switched reluctance motor(SRM) has the advantages of simple structure, low rotor inertia, and high poer rate per unit volume. However, position sensor isessential in SRM in order to synchronize the phase excitation to the rotor position. The position sensors increase the cost of drive system, and tend to reduce system reliability. This paper investigtes the speed control of sensorless SRM. The proposed system consists of position detection circuit, dwell angle controller, digital logic commutator, PI speed controller and 4-phase inverter. The performances in the proposed system are verified through the experiment.

In this paper, a three phase boost converter that operates with unity power factor and sinusodial input currents is presented. The current control of the converter is based on the space vector strategy with fixed switching frequency and the input current tracks the reference current within one sampling time interval. Space vector strategy for current control was materialized as a digital control method by using DSP. By using this control strategy low ripples in the output voltage, low harmonics in the input current and fast dynamic responses are achieved with a small capacitance in the dc link.

In this paper, single-phase PWM AC to AC converter that operates with unit powr factor and sinusidal input line currents is presented. The currentcontrol of this converter is based on the predicted current control method with fixed switching frequency and the line currents track to reference currents within sampling time interval.