In speech coding scheme, because information bits have different error sensitivities over channel errors, the channel coder for combining with speech coding should be realized by the variable coding rate considering the bit importance of speech information bits. In realizing the 4 kbps channel coder for 12kbps speech, this paper have chosen the channel coding method by analyzing the hard-decision post-decoding error rate of RCPC(Rate Compatible Punctured Convolutional) codes and bit error sensitivity of 12 kbps speech. Under the coherent QPSK and Rayleigh fading channel, the performance analysis has showed that 10dB gain was obtained in speech SEGSNR by 4-level uneuqal error protection, which was compared with the caseof no channel coding at 7dB channel SNR.

Recently, four speech coding techniques, namely, SBC-APCM(sub-band coding adaptive PCM), RPE-LPC(regualr pulse excitation linear predictive codec), MPE-LTP(multi-pulse excited long-term prediction) and CELP (code-excited linear prediction) are proposed for digital mobile radio applications. However, a performance comparison of these coders in the Rayleigh fading environment has not been made yet. In this paper, the performances of the four spech coders in the random bit error and burst error environment are investigated. For the channel coding of SBC-APCM, RPE-LPC and MPE-LTP, the sensitivity of output bit stream is measured and a bit selective forward error correction is provided acording to the measured bit sensitivity. And for an attempt to improve the performance of CELP, an optimum quantizer is applied for transmitting scalar quantities in CELP. However, an improvement over the conventional approach is found to be negligible. For the channel coding of CELP, Reed-Solomon code, Golay code, convolutional code of rate 1/2 shows the best performance. Finally, from the simulation results, it is concluded that CELP is the best candidate for digital mobile radio and is followed by MPE-LTP, SBC-APCM and RPE-LPC.

In order to realize an efficient and stable X-band harmonic generator, a 100 MHz frequency multiplier, an impulse generator using SRD(Step Recovery Diode) module, and a narrow-band bandpass waveguide filter have been designed and tested. By properly combining these devices an X-band harmonic generator has been realized. The output power of the harmonic generator was measured to be -1.5 dBm at 9 GHz which is the 90th harmonic of the 100MHz input. The power fluctuation of the harmonic generator due to temperature variation was observed to the about 0.15 dB during 24 hours of 4\ulcorner temperature variation.

Zone-melting recrystallization (ZMR) system using two graphite heaters(a stationary sheet and a narrow movable bar) was constructed and implemented in recrystallization op Si films on insulating layers. The recystallized Si films were examined by Nomarski contrast optical microscopy after Dash etching, transmission electron diffraction pattern, and x-ray diffraction. With optimum conditions of process parameters(input powers of the bottom and upper heater, scanning speed of the upper heater, and the gap between sample and upper heater), the recrystallized Si layer has a (100) texture, but contains many subboundaries. The subgrains are misoriented by < 0.5\ulcorner and the average spacing between subboundaries is about 25\ulcorner.

In this paper the characteristics of submicron gate GaAs MESFET's have been studied using a particle model which takes into account the hot-electron transport phenomena, i.e., the velocity overshoot. \ulcornervalley(<000> direction), L valley (<111>direction), X valley (<100>direction) as the GaAs conduction energy band and optical phonon, acoustic phonon, equivalent intervalley, nonequivalent intervalley scattering as the scattering models, have been considered in this simulation. And the GaAs material and the device simulation have been done by determination of the free flight time, scattering mechanism and scattering angle according to Monte-Carlo algorithm which makes use of a particle model. As a result of the particle simulation, firstly the electron distribution, the potential energy distribution and the situation of electron displacement in 0.6 \ulcorner gate length device have been obtained. Secondly, the cutoff frequency, obtained by this method, is k47GHz which is in good agreement with the calculated result of theory. And the current-voltage characteristics curve which takes account of the buffer layer effect has been obtained. Lastly it has been verified that parasitic current at the buffer layer can be analyzed using channel depth modulation.

The current conduction and dielectric breakdown properties of oxide grown on phosphorous-doped polysilicon have been investigated by means of the ramped I-V measurements. The effective barrier heights of polyoxide grown by different silicon deposition and oxidation conditions were calculated from the Fowler-Nordheim tunneling characteristic. The average critical fields were also obtained for each film. From the results, the high temperature oxided polyoxide grown on amorphous silicon film shows superior electrical characteristics comparing to the other films.

Considering channel length modulation, we proposed a GaAs MESFET model for circuit simulator. In existing M.S. Shur's model, two different models are used according to pinch-off voltage of devices. One model for both type of devices was proposed. In this model we introduced weighted switching function(WSF) based on channel length modulation. This proposed model showed better accuracy comparing with existing single law model and complete velocity saturation model.

Alloyed ohmic contact properties of Au-Zn/Au, Au-Be/Au,Au-Zn/Cr/Au, and Au-Be/Cr/Au metal system to p-InP were investigated. Optimum alloying conditions were obtained at the annealing temperature of 425\ulcorner for all the metal systems using a rapid thermal annealing system. Surface AES analysis and auger depth profiling were done for each metal system annealed at the optimum conditions. Outdiffusions of In and P from the InP substrate were found in the metal systems without Cr intermediate layer. Also, small amount of In. P and Cr were detected at the surface in the case of Au-Zn/Cr/Au system, while there were occured no outdiffusion of In, P, and Cr for Au-Be/Cr/Au system. The best surface morpholoty and specific contact resistivity of 4.5x 10**-5 \ulcornercm\ulcornerhave been obtained in this Au-Be/Cr/Au material system alloyed at 425\ulcorner for 60 second.

A silicon pressure sensor made of a full bridge of diffused resistors was designed and fabricated using semiconductor integrated circuit process. Thin diaphragms with 30\ulcorner thickness were obtained using anisotropic wet chemical etching technique. Our device showed strong temperature dependence. Compensation networks are used to compensate for the temperature dependence of the pressure sensor. The bridge supply voltage having positive temperature coefficient by compensation networks was utilized against the negative temperature coefficient of bridge output voltage. The sensitivity fluctuation of pressure sensor before temperature compensation was -1700 ppm/\ulcorner, while it reduced to -710ppm\ulcorner with temperature compensation. Our result shows that the we could develop accurate and reliable pressure sensor over a wide temperature range(-20\ulcorner~50\ulcorner).

A dynamic relocation algorithm for non-deterministic process graph in distributed computer systems is proposed. A method is represented for determining the optimal policy for processing a process tree. A general database query request is modelled by a process tree which represent a set of subprocesses together with their precedence relationship. The process allocation model is based on operating cost which is a function fo selection of site for processing operation, data reduction function and file size. By using expected values of parameters for non-deterministic process tree, the process graph and optimal policy that yield minimum operating cost are determined. As process is relocated according to threshold value and new information of parameters after the execution of low level process for non-deterministic process graph, the assigned state that approximate to optiaml solution is obtained. The proposed algorihtm is heuristic By performing algorithm for sample problems, it is shown that the proposed algorithm is good in obtaining optimal solution.

This paper describes an efficient method of texture image segmentation based on conditional 1-dimensional histograms. We consider the multi-dimensional histogram, and it is projected into each axis in order to obtain conditional 1-dimensional histograms. And we extract uniform regions by iteratively applying the peak-valley detection method to conditional 1-dimensional histograms. In view of the amount of memory and computation time, the proposed method is superior to the conventional method which uses the multi-dimensional histogram. By applying the proposed method to the artificial and natural texture images some desirable results are obtained.

To represent shape characteristics of digital closed curve, many algorithms, mainly based on local properties, have been proposed. In this paper, we propose a new algorithm for detecting local curvature maxima which reflects context, i.e., structural or surrounding regional characteristics. The algorithm does not require the value of k as an input parameter which is the major problem in k-curvature method in digital curve, but calculates it at each point depening on the context. The algorithm has been applied to two dimensional image boundaries. The efficiency of the algorithm is addressed by comparing the result of existing contest dependent algorithm.

This study evaluates the performance of an optical receiver for binary phase shift keying (BPSK) signals in the presence of short noise originating from the photo diode and phase noise of the optical source. The case of using I.O. hybrid compare with the fiber optic hybrid to mix received optical signal and laser local oscillator signal. The impact of these noise is minimized if loop natural frequency and power split ratio between data and carrier recovery branch are choosen optimally. Then it is obtained that required laser linewidt to achieve a BER of 10**-9. The results are the same except theat in case of using the fiber optic hybrid the required optical power is twice as much as the I.O. hybrid.

One of the confronting problems in using optical fibers under radiation environments is producing of color centers in optical materials due to nuclear radiation. These centers increase transmission loss by absorbing propagating light. In this study, the radiation effects on optical fiber are studied theoretically. Also, optical attenuation induced by \ulcorner-ray irradiation from Co**60 for single mode and multimode optical fibers is measured at the optical wavelength of 0.85\ulcorner and 1.3\ulcorner, and the results are analyzed. Gammaray is irradiated for 5hours at the rate of 300rads/min, which is corresponding to 90 krads of integrated dose. In case of multimode optical fibers, the induced loss at 0.85\ulcorner wavelength has been twice higher than that at 1.3\ulcorner. The loss in multimode fibers has been significantly larger by 7-20 times than that in single mode fibers, dependently on fiber materials at 1.3\ulcorner.

In this paper, a design method of optical FSK transmitter is described using LD-FM circuit model. In the design of optical FSK transmitter, an optimum bias current was chosen using LD-FM circuit model, and an unequalized FM transfer function was determined at this current. The equalizers that can make this transfer function uniform were designed by use of a simple passive network. For the designed optical FSK transmitter, the pulse-transient and small-signal frequency-deviation responses were simulated and discussed.

A software tool implementing cascaded SCF (Switched Capacitor Filter) is developed. A comprehensive procedure is described for the design of cascaded SCF, and some design considerations are discussed. In program several user-selectable filter approximation techniques, and two optimising methods which reduce the total capacitance are included. Analysis of the partial positive feedback and the effect of clock frequency to total capacitance are presented in detail. Two pratical examples are given to prove its validity in reducing the total capacitance. Developed tool can be used for general SCF generator.

This paper describes a new gate array layout system, HAN-LACAD-G(HANyang LAyout CAD system for Gate array). HAN-LACAD-G is composed of placer, global router, detailed router, and output processor. In placement design, initial placement is performed by repetitive clustering and min-cut partitioning followed by placement improvement using the concept of pairwise interchange. In global routing phase, pins are assigned in each channel considering the routing congestion estimation and overflows in feedthroughs are restricted. For the detailed routing, we use layer and three layer channel routing techniques. Layout results are displayed graphically and modified interactively by the user using the layout editor.

Hidden Markov models(HMM's) have been known to be useful representation for speech signal and are used in a wide variety of speech systems. For speech recognition applications, it is desirable to incorporate durational information of states in model which correspond to phonetic duration of speech segments. In this paper we propose duration-dependent HMM's that include durational information of states appropriately for the left-to-right model. Reestimation formulae for the parameters of the proposed model are derived and their convergence is verified. Finally, the performance of the proposed models is verified by applying to an isolated word, speaker independent speech recognition system.