In this paper, variable threshold detection with weighted pulse code modulation-encoded signals transmitted over Gaussian channels has been investigated. Each bit in the \ulcornerlaw PCM word is weighted according to its significance in the transmitter. It the output falls into the erasure zone, the regenerated sample replaced by interpolation or prediction. To overall system signal to noise ratio for BPSK/PCM speech signals of this technique has been found. When the input signal level was -17 db, the gains in overall signal s/n compared to weighted PCM and variable threshold detection were 5 db and 3 db, respectively. Computer simulation was performed generating signals by computer. The simulation was in resonable agreement with our theoretical prediction.

In calculating resonant frequencies of a microstrip resonator the effect of the stray field at resonator edges has been neglected in most publications. The formula for resonant frequencies therefore provided inaccurate results. In the paper the effective dimension and the dynamic dielectic constant of a resonator are defined and obtained for more accurate resonant frequencies and qualty factor by including the effect of stray fields. Theoretical results obtained from the new formula show good agreement with experimental results. In addition, the effect on resonant modes of the distance and coupling angle between the microstrip line and the resonator is also observed.

Analysis and design procedures for both symmetrical and non-symmetrical open-circuited interdigital multiple coupled microstrip line structures for applications as wide-band DC blocks/filters have been presented. The design equations, as is the case of other microstrip structures, are based on a simplified TEM model. The experimental results are in good agreement with the theoretically predicted ones.

Approximation of a Cartesian straight line motion with linear interpolation in the joint space has many desirable advantages and applications. But inappropriate determination of the corresponding subtravelling and transition times makes such joint-trajectories violate the input torque/force constraints. An approach that can overcome this difficult and yield the joint trajectories utilizing the allowable maximum input torque/force is established in this paper. The effectiveness of these results is demonstrated by using a three-joint revolute manipulator.

The capacitive transducers, which are used of the principle of capacitance changes by means of the variation of distance between electrodes, are usually utilized for measurement of small displacements. But the measurement ranges of this method are limited. In this study, the measurement ranges are extended by using the variation method of the surface area of electrodes. Thus the capacitive transducer and signal conditioner, which are having the capability of measurement range of 25mm and resolution of 0.1\ulcorner, have developed, and its environmental characteristics are also tested.

A method of on-line recognition of Korean characters (Hangeul) by augmented conterxt free grammar is described in this paper. Syntactic analysis with context free grammar oftern has ambiguity. Insufficient description of relations among Hangrul sub-patterns causes this ambiguity can be determined through repetition of experiments. Flexible syntactic analysis is executed by adapting the condition to the (advice)part of augmented context free grammar. The ratio of correct recognition of this method is more than 99%.

In this paper, a tightly-coupled multiprocessor system is implemented with four processing elements based on MC68000 CPU, a common menory (128KB), and a single time-shared bus. The multi-tasking operating system, MTOS, is modified so that the multiprocessor system can support multitasking and multiprocessing. The performance of the proposed system is evaluated by stochastic Petri Net system modeling. The efficiency and the processing power are simulated for various load factors and up to 16 PEs. By running benchmark programs, such as quicksort, FFT, and matrix-multiplication, the speed of parallel processing is compared with that of a single processor.

This paper describes the design and fabrication of the polysilicon selfaligned bipolar transistor with 1.6\ulcorner epitaxy and SWAMI isolation technologies. This transistor has two levels of polysilicon. Also emitter and adjacent edge of polysilicon base contact of this PSA device are defined by the same mask, and emitter feature size is 2x4 \ulcorner. DC characteristic of the fabricated transistor was evaluated and analyzed for the SPICE input parameters. The minimum propagation delay time per gate of 330 ps at 1mW was obtained with 41 stage CML ring oscillator.

In this paper, the SPICE 2.G6 JFET subroutine and other related subroutines are modified for circuit simulation of GaAs MESFET IC's. The hyperbolic tangent model is used for the drain current-voltage characteristics of GaAs MESFET's and derived channel-conductance and drain-conductance model from the above current model are implemented into small-signal model of GaAs MESFET's. And, device capacitance model which consider after-pinch-off state are modified, and device charge model for SPICE 2G.6 are proposed. The result of modification is shown to be suitable for GaAs circuit simulator, showing good agreement with experimetal results. Forthermore the DC convergence of this paper is better than that of SPICE 2.G JFET subroutine. GaAs MESFET model in this paper is applied for both depletion mode GaAs MESFET and enhancement-mode GaAs MESFET without difficulty.

1\ulcorner gate MESFETs are fabricated on MOCVD and VPE grown GaAs wafers using photolithography, chemical wet etching and lift-off techniques. DC characteristics such as Vt, Gm, Rs, etc. are studied and active channel parameters of MESFET(a, n, Leff, \ulcorner)are analyzed for 1-4 \ulcorner gate FETs and 100\ulcorner FAT FET. The correlation between DC data and active channel parameters are experimentally analyzed. The measured transconductance and low-field mobility in the active channel for the 1\ulcorner gate MESFET made on MOCVD wafer are 67mS/mm and 2980cm\ulcornerVs respectively.

High value sheet resistance (Rs' 1K-33K\ulcorner/) polysilicon resistors were fabricated using double ion implantation with boron as the major dopant and phosphorus compensation. It is observed that Rs sensitivity to the net doping concentration is decreased by one order of magnitude compared to the conventional (boron implanted)polysilicon resistors. The temperature co-efficient of resistance (TCR) measured between 25\ulcorner and 125\ulcorner shows equivalent values to those of non-compensated resistors for the same Rs. A qualitative electrical conductiion mechanism for compensated polysilicon resistor is proposed, based on the existing grain boundary charge trapping theory.

Gate capacitances have been measured directly on small-geometry MOSFET's with the drain voltage as a parameter for various channel lengths and for p and n channel types and the characteristics have been compared with each other. The influence of 'hot carrier effect' of short channel devices on capaciatance has been compared with long channel devices. The results show that gate capacitance characteristics of short channel device deviate from those of long channel device. The accuracy of the measurement system is less than a few femto Farad, and the minimum geometry (W/L) of device for which reliable measurement can be obtained is 6/3.

This pper presents a method for constructing multiple-valued logic sequential machines based on Galois field. First, we assign all elements in GF(2**m) to bit codes using mathematical properties of GF(2**m). Then, we realized the sequencial machine circuits with and withoutm feed-back. 1) Sequential machines with feed-back are constructed by using only MUX from state-transition diagram expressing the information of sequential machines. 2) Sequential machines without feed-back are constructed by following steps. First, we assigned states in state-transition disgram to state bit codes, then obtained state function and predecessor table explaining the relationship between present states and previous states. Next, we obtained next-state function from state function and predecessor table. Finally we realized the circuit using MUX and decoder.

A new Block Truncation Coding(BTC) technique reducing the bit-plane and using differential pulse code modulation (DPCM) is proposed and compared with the conventional BTC methods. A new technique decides whether the subblock can be approximated to be uniform or not. If the subblock can be approximated to be uniform(merge mode), we transmit only the gray-level informantion. It not (split mode), we transmity both the bit-plane and the gray-level information. DPCM method is proposed to the encoding of gray-level information when the subblock can be approximated to be uniform. Also modified quantization method is presented to the encoding of gray-level information when the subblock is not uniform. This technique shows the results of coding 256 level images at the average data rate of about 0.75 bits/pel.

This paper presents a new two-stage edge detection algorithm which can give thin edges and can find edges between low contrast regions. In the first stage, the potential edge points are obtained using the proposed edge operator which is motivated by the concept of the E filter. And in the second stage, the real edge points adre detected by comparing the potential edge values with the lically adaptive threshold values of the corresponding potential edge points and by testing whether the potential edge points are zero-crossing points or not. It is shown quantitatively that the proposed algorithm can find edges effectively in an image with impulse noise, and also the results of applying the proposed algorithm to real images are presented.

This paper describes an implementation of the absolute moment block truncation coding(AMBTC) in real time for the moving picture data compression. We have realized a system composed of the encoder and decoder, and operated it using an NTSC TV signal. The encoder consists of a 4-1line buffer memory and a data processing block. Besides, there are signal conditioner and a control signal generator. Experimental results show that the quality of the processed image with a data rate of 10Mb/s is slightly degraded, but not objectionable, comparing data rate of 80Mb/s.

An local adaptive image segmentatin algorithm for local feature detection and effective clustering of unimodal histogram shape are proposed. Local adaptive difference image and its histogram are obtained from the input image. The parameters are derived from the histogram and used for the segmentation based on relaxatin process. The results showed effective region segmentation and good noise cleaning for the ocular fundus fluorescein angiogram which has low contrast and unimodal histogram.

When digital image signals are transmitted or stored, they may be usually degraded by impulsive noise such as BSC noise. Though median filtering is a very effective method to reduce the impulsive noise, it brings non-negligible distortion after filtering. Several algorithms have been proposed to reduce such a distortion, but their reconstructed image quality are inadequate in some cases and they have a difficulty in real-time processing. In this paper, an effective filtering algorithm which can not only reduce the noise effectively but also preserve the edges well and lessen the distortion greatly, is presented. The proposed algorithm is an adaptive algorithm of median filter using local statistics, based on the characteristics of human eyes. The adaptive algorithm results shwo performance improvement of up to 3-4 dB over the nonadaptive one.

Holographic diffraction gratings which are the core element of the optical instruments such as a spectrophotometer have been fabricated using the photoresist as a recording material. A 488nm line from an argon laser is used in making the gratings. Transmission type surface-relief phase grating and reflection type which is fabsricated by coating the aluminum on the transmission type with thickness 2000\ulcornerare fabricated, the spatical frequency of which are 1200 lines/mm. Diffraction efficiency, scattered light and wave-front aberratin are measured to test the performance of the developed gratings. A marimum diffraction efficiency is given when light is incident at the Bragg angle. Theoretical efficiency is about 80%, but measured efficiency of the transmission type and reflection type is 50% and 45%, respectively.

An exact wave-optics analysis of wave propagation in thin-film optical waveguide using gyrotropic materials as the substrate or film of the guide is presented for the first time. Based on the Maxwell's equations and the boundary conditions of the guide, the field composition and the boundary conditions of the guide, the field composition and the phase velocity for the eigenmodes of the guide are determined. The field patterns of the guided waves are shown for the eigenmodes of the guides. The present analysos allows a new interpretation in the mode conversion of the thin-film optical waveguides.

This paper proposes a new pattern recognition system using Radon transform and analyzes the performances of the system for given input patterns. The proposed system uses many optical parts in order to utilize the high speed characteristics of light and processes a signal easily by transforming 2-D image into a 1-D signal to increase flexibility. The squared Mahalanobis distance obtained from means and standard deviations of the features for the given input patterns is used for discrimination. As a result, this system represents a better recognition rate than any other systems using the same input patterns.

This paper deals with the CMOS PLA Generator using CHISEL language. The program which plots a CMOS PLA alphanumeric layout automatically according to desired input functions and output functions has been developed. The program consists of procedures. These procedure are drawing a stick diagram with input data, converting any design rule, plotting a physical layout at IBM PC-AT with CIF input data. Physical layout information of dynamic CMOS PLA is stored CIF form. The CMOS PLA Generator is written in CHISEL which is layout language VLSI design tools and run on a VAX11/750 running UNIX.

In this paper, a design method for sequential circuit which is easy to have Built-in Self Test is kproposed using the functional advantages of multifunctional BILBO and LSSD. To achieve the hardware reduction, it is designed that a multifunctional BILBO has double operational functions of NLFSR and LFSR, when neccessary, and that test signal could be used as an input-output signal in the same line. By applying the proposed multifunctional BILBO to the sequential PLA, the test patterns and the additional circuit could be reduced in test operation and the propagation delay is vanished in normal operation, as we expected. Above them, the partitioned method for large scale sequential circuit is also suggested and it is observed that test patterns and additional circuit in them reduced by this method.

This paper presents a speaker dependent isolated digit recognition algorithm using the Mellin transform. Since the Mellin transform converts a scale information into a phase information, attempts have been made to utilize this scale invariance property of the Mellin transform in order to alleviate a time-normalization procedure required for a speech recognition. It has been found that good results can be obtained by taking the Mellin transform to the features such as a ZCR, log energy, normalized autocorrelation coefficients, first predictor coefficient and normalized prediction error. We employed a difference function for evaluating a similarity between two patterns. When the proposed algorithm was tested on Korean digit words, a recognition rate of 83.3% was obtained. The recognition accuracy is not compatible with the other technique such as LPC distance however, it is believed that the Mellin transform can effectively perform the time-normalization processing for the speech recognition.

An image coding technique based on a segmentation, which utilizes a simplified description of regions composing an image, is investigated in this paper. The proposed coding technique consists of 3 stages: segmentation, contour coding. In this paper, emphasis was given to texture coding in order to improve a quality of an image. Split-and-merge method was employed for a segmentation. In the texture coding, a linear predictive coding(LPC), along with approximation technique based on a two-dimensional polynomial function was used to encode texture components. Depending on a size of region and a mean square error between an original and a reconstructed image, appropriate texture coding techniques were determined. A computer simulation on natural images indicates that an acceptable image quality at a compression ratio as high as 15-25 could be obtained. In comparison with a discrete cosine transform coding technique, which is the most typical coding technique in the first-generation coding, the proposed scheme leads to a better quality at compression ratio higher than 15-20.