We derive a back-propagation learning algorithm of fuzzy neural networks using fuzzy operations, which preserves the shapes of fuzzy numbers, in order to utilize fuzzy if-then rules as well as numerical data in the learning of neural networks for classification problems and for fuzzy control problems. By introducing the shape preseving fuzzy operation into a neural network, the proposed network simplifies fuzzy arithmetic operations of fuzzy numbers with exact result in learning the network. And we illustrate our approach by computer simulations on numerical examples.

This paper describes the design of a variable rate QPSK demodulator for digital satellite TV system. This true variable-rate demodulator employs a unique architecture to realize an all digital synchronization and detection algorithm. Data-flow based design approach enabled a seamless transition from high level design optimization to physical layout. The demodulator has been integrated with Viterbi decoder, de-interleaver, and Ree-Solomon decoder to make a single chip Digital Video Broadcast (DVB) receiver. The receiver IC has been fabricated with a 0.5mm CMOS TLM process and proved fully functional in a real-world set-up.

For practical application of the concept of chaos, we propose a chaotic state machine as a sequential system. Chaotic state machine which is suggested and implemented in this paper has chaotic motions relying on the dynamics only through the deterministic chaos function. Also, we present and verify that the properties of chaotic state machine is equal to the characteristics of chaos.

Free-space optical interconnection networks can be classified into two types, space variant and space invariant, according to the degree of space variance. In terms of physical implementations, the degree of space variance can be interpreted as the degree of sharing beam steering optics among the nodes of a given network. This implies that all nodes in a totally space-invariant network can share a single beam steering optics to realize the given network topology, whereas, in a totally space variant network, each node requires a distinct beam steering optics. However, space invariant networks require mechanisms for distinguishing the origins of incoming signals detected at the node since several signals may arrive at the same time if the node degree of the network is greater than one. This paper presents a signal source encoding scheme for distinguishing incoming signals efficiently, in terms of the number of detectors at each node or the number of unique wavelengths. The proposed scheme is solved by developing a new parallel genetic algorithm called distributed asynchronous genetic algorithm (DAGA). Using the DAGA, we solved signal distinction schemes for various network sizes of several topologies such as hypercube, the mesh, and the de Brujin.

We study the decentralized stabilization problem of linear time-invariant large-scale interconnected systems with delays without any system structure. We obtain sufficient stability conditions for interconnected systems which are equivalent to disturbance attenuation of some scaled system. A decentralized output-feedback controller is obtained using standard H$\infty$ control theory. The obtained controller is delay-independent. We also obtain an observer for the interconnected system.

In this paper, we consider the problem of designing H$\infty$ state feedback controller for the generalized time systems with delayed states and control inputs in continuous and discrete time cases, respectively. The generalized time delay system problems are solved on the basis of LMI(linear matrix inequality) technique considering time delays. The sufficient condition for the existence of controller and H$\infty$ state feedback controller design methods are presented. Also, using some changes of variables and Schur complements, the obtained sufficient condition can be rewritten as a LMI form in terms of transformed variables. The propose controller design method can be extended into the problem of robust H$\infty$ state feedback controller design method easily.

In practice, there may be various defects in an insulating system, so that the PD signals can be produced from these defects simultaneously. Regarding these situations, we have to discriminate the type of defect as well as determine whether the PD occurs or not. In this paper, some analysis results of the PD signals from multi-defects insulating system were presented. We measure the PD signals by using three kind of electrode system, IEC(b), Needle-Plane and mixed electrodes. To simulate multi-defect systems, we combined to electrode systems and apply test voltage simultaneously. Neural network, statistical analysis methods were tried, and the possibilities and limitations of each method were clarified.

A simple and robust digital current control technique for a permanent magnet (PM) synchronous motor under the parameter variations is presented. Among the various current control schemes for an inverter-fed PM synchronous motor drive, the predictive control is known to give a superior performance. This scheme, however, requires the full knowledge of machine parameters and operating conditions, and cannot give a satisfactory response under the parameter mismatch. To overcome such a limitation, the disturbances caused by the parameter variations will be estimated by using a disturbance observer theory and used for the computation of the reference voltages by a feedforward control. Thus, the steady-state control performance can be significantly improved with a relatively simple control algorithm, while retaining the good characteristics of the predictive control. The proposed control scheme is implemented on a PM synchronous motor using the software of DSP TMS320C30 and the effectiveness is verified through the comparative simulations and experiments.

In this paper, we present a nonlinear feedback linearization-full order observer/sliding mode controller (NFL-FOO/SMC), to obtain smmoth control as a linearized controller in a linear system (or to cancel the nonlinearity in a nonlinear system), and to solve the problem of the unmeasurable state variables as in the conventional SMC. The proposed controller is obtained by combining the nonlinear feedback linearization-sliding mode control (NFL-SMC) with the full order observer (FOO)and eliminates the need to measure all the state variables in the traditional SMC. The proposed controller is applied to the nonlinear power system stabilizer (PSS) for damping oscillations in a power system. The effectiveness of the proposed controller is verified by the nonlinear time-domain simulations in case of a 3-cycle line-to-ground fault and in case of the parameter variation for the AVR gain K\ulcorner and for the inertia moment M.

In this paper, the standard Dole, Glover, Khargoneker, and Francis (abbr. : DGKF 1989) H\ulcorner controller (H\ulcornerC) is extended to the nonlinear feedback linearization-H\ulcorner/sliding mode controller (NFL-H\ulcorner/SMC), to tackle the problem of the unmeasurable state variables as in the conventional SMC, to obtain smooth control as the linearized controller in a linear system, and to improve the time-domain performance under a worst scenario. The proposed controller is obtained by combining the H\ulcorner estimator with the nonlinear feedback linearization-sliding mode controller (NFL-SMC) and it does not need to measure all the state variables as in the traditional SMC. The proposed controller is applied as a nonlinear power system stabilizer (PSS) for the improvement of the power system damping characteristics of an single machine infinite bus system (SMIBS) connected through a double circuit line. The effectiveness of the proposed controller is verified by nonlinear time-domain simulation in case of a 3-cycle line-to-ground fault and in case of the parameter variations for the AVR gain K\ulcorner and for the inertia moment M.

This paper is concerned with an efficient generation of stereoscopic views for complex virtual environments by exploiting frame coherence in visibility. The basic idea is to keep visible polygons throughout the rendering process. P-buffer, a buffer of image size, holds the id of the visible polygon for each pixel. This contrasts to the frame buffer and the Z-buffer which hold the color information and the depth information, respectively. For the generation of a consecutive image, the position and the orientation of the visible polygons in the current view are updated according to the viewer's movements, and re-rendered on the current image under the assumption that, when the viewer moves slightly, the visibility of polygons remains unchanged. In the case of stereoscopic views, it may not introduce much difficulty when we render the right(left) image using visible polygons on the (right) image only, The less difference in two images is, the easier the matching becomes in perceiving depth. Some psychophysical experiments have been conducted to support this claim. The computational complexity for generating a fight(left) image from the previous left(right) image is bounded by the size of image space, and accordingly. It is somewhat independent of the complexity of the 3-D scene.

It is becoming apparent that stereo matching algorithms need much information from high level cognitive processes. Otherwise, conventional algorithms based on bottom-up control alone are susceptible to local minima. We introduce a system that consists of two levels. A lower level, using a usual matching method, is based upon the local neighborhood and a second level, that can integrate the partial information, is aimed at contextual matching. Conceptually, the introduction of bottom-up and top-down feedback loop to the usual matching algorithm improves the overall performance. For this purpose, we model the image attributes using a Markov random field (MRF) and thereupon derive a maximum a posteriori (MAP) estimate. The energy equation, corresponding to the estimate, efficiently represents the natural constraints such as occlusion and the partial informations from the other levels. In addition to recognition, we derive a training method that can determine the system informations from the other levels. In addition to recognition, we derive a training method that can determine the system parameters automatically. As an experiment, we test the algorithms using random dot stereograms (RDS) as well as natural scenes. It is proven that the overall recognition error is drastically reduced by the introduction of contextual matching.

In the circumstances we want to deal with, a transmission channel is limited and a global motion can happen by camera movement, and also there exists a region-of-interest(ROI) which is more important than background. So very low bit rate coding algorithm is required and processing of global motion must be considered. Also ROI must be reconstructed with required quality after decoding because of its importance. But the existing methods such as H.261, H.263 can not reconstruct ROIs with high quality because they do not consider the fact that ROIs are more important than background. So a new coding scheme is proposed that describes a method for encoding image sequences distinguishing bits between ROI and background. Experimental results show that the suggested algorithm performs well especially in the circumstances where background changes and the area of ROI is small enough compared with that of background.

When size and shape of moving object have been changed, a correlator often accumulates walk-off error. A success of correlation-based tracking largely depends on choosing suitable window size and position and thus transferring the proper reference image to the next frame. For this, we propose the Adaptive Window Algorithm with Four-Direction Sizing Factors (AWA-FSF) for fast adjusting a reference region to enhance reliability of correlation-based image tracking in complex cluttered environments. Since the AWA-FSF is capable of adjusting a reference image size more rapidly and properly, we can minimize the influence of complex background and clutter. In addition, we can finely tune the center point of the reference image repeatedly after main tracking process. Thus we have increased stability and reliability of correlation-based image tracking. We tested performance of the AWA-FSF using 45 real image sequences made of over 3400 images and had the satisfied results for most of them.

Piezoelectric pressure sensors and pyroelectric infrared detectors based on ZnO thin film have been integrated with GaAs metal-semiconductor field effect transistor (MESFET) amplifiers. Surface micromachining techniques have been applied in a GaAs MESFET process to form both microsensors and electronic circuits. The on-chip integration of microsensors such as pressure sensors and infrared detectors with GaAs integrated circuits is attractive because of the higher operating temperature up to 200 oC for GaAs devices compared to 125 oC for silicon devices and radiation hardness for infrared imaging applications. The microsensors incorporate a 1${\mu}$m-thick sputtered ZnO capacitor supported by a 2${\mu}$m-thick aluminum membrane formed on a semi-insulating GaAs substrate. The piezoelectric pressure sensor of an area 80${\times}$80 ${\mu}$m2 designed for use as a miniature microphone exhibits 2.99${\mu}$V/${\mu}$ bar sensitivity at 400Hz. The voltage responsivity and the detectivity of a single infrared detector of an area 80${\times}$80 $\mu\textrm{m}$2 is 700 V/W and 6${\times}$108cm$.$ Hz/W at 10Hz respectively, and the time constant of the sensor with the amplifying circuit is 53 ms. Circuits using 4${\mu}$m-gate GaAs MESFETs are fabricated in planar, direct ion-implanted process. The measured transconductance of a 4${\mu}$m-gate GaAs MESFET is 25.6 mS/mm and 12.4 mS/mm at 27 oC and 200oC, respectively. A differential amplifier whose voltage gain in 33.7 dB using 4${\mu}$m gate GaAs MESFETs is fabricated for high selectivity to the physical variable being sensed.

Novel fabrication process of vertical spring for micro mirror array is proposed. The proposed fabrication process adopts a shadow evaporation process using shielding screen structure on top of the sacrificial layer. The 50${\times}$50 micro mirror arrays are fabricated using the proposed process and ceramic packaged. The static and dynamic characteristics of mirror are measured. The mirror plate touches substrate at 16V and the response time of about 16.8 ${\mu}\textrm{s}$. The resonant frequency of mirror is 16kHz. The spring thickness is calculated from static characteristic to be 1075${\AA}$.

A DSP-based robust nonlinear speed control of a permanent magnet synchronous motor(PMSM) which is robust to unknown parameter variations and speed measurement error is presented. The model reference adaptive system(MRAS) based adaptation mechanisms for the estimation of slowly varying parameters are derived using the Lyapunov stability theory. For the disturbances or quickly varying parameters. a quasi-linearized and decoupled model including the influence of parameter variations and speed measurement error on the nonlinear speed control of a PMSM is derived. Based on this model, a boundary layer integral sliding mode controller to improve the robustness and performance of the nonlinear speed control of a PMSM is designed and compared with the conventional controller. To show the validity of the proposed control scheme, simulations and experimental works are carried out and compared with the conventional control scheme.

This paper describes the fabrication and characterization of dual-polysilicon gated surface channel 0.1$\mu\textrm{m}$ CMOSFETs using BF2 and arsenic as channel dopants. We have used and LDD structure and 40${\AA}$ gate oxide as an insulator. To suppress short channel effects down to 0.1$\mu\textrm{m}$ channel length, shallow source/drain extensions implemented by low energy implantation and SSR(Super Steep Retrograde) channel structure were used. The threshold voltages of fabricated CMOSFETs are 0.6V. The maximum transconductance of nMOSFET is 315${\mu}$S/$\mu\textrm{m}$, and that of pMOSFET is 156 ${\mu}$S/$\mu\textrm{m}$. The drain saturation current of 418 ${\mu}$A/$\mu\textrm{m}$, 187${\mu}$A/$\mu\textrm{m}$ are obtained. Subthreshold swing is 85mV/dec and 88mV/dec, respectively. DIBL(Drain Induced Barrier Lowering) is below 100mV. In the device with 2000${\AA}$ thick gate polysilicon, depletion in polysilicon near the gate oxide results in an increase of equivalent gate oxide thickness and degradation of device characteristics. The gate delay time is measured to be 336psec at operation voltage of 2V.

This paper introduces an adaptive algorithm determining the measurement-track association problem in multi-target tracking(MTT). We model the target and measurement relationships with mean field theory and then define a MAP estimate for the optimal association. Based on this model, we introduce an energy function defined over the measurement space, that incorporates the natural constraints for target tracking. To find the minimizer of the energy function, we derived a new adaptive algorithm by introducing the Lagrange multipliers and local dual theory. Through the experiments, we show that this algorithm is stable and works well in general environments. Also the advantages of the new algorithm over other algorithms are discussed.

We investigate a Leaky Bucket(LB) scheme with a threshold in the data buffer, where leaky rate changes depending on the contents of data buffer. We use the fluid flow model for the analysis of the LB scheme with a threshold. We model the bursty input source as markov modulated fluid flow(MMFF) As performance measures we obtain loss probability and mean delay. We present some numerical results to show the effects of the level of a threshold, the rate of token generation, the size of token pool, and the size the data buffer on the performances of the LB scheme with a threshold.