Search

Two-dimensional analysis of off-state drain current-drain voltage characteristics in AlGaN/GaN high electron mobility transistors is performed; where two cases with a single passivation layer (SiN) and double passivation layers (thin SiN and high-k dielectric) are compared. It is shown that in the case of double passivation layers, the breakdown voltage is enhanced significantly when comparing at the same insulator thickness. This is because in this case the electric field around the drain edge of gate is weakened. Also, in the case of double passivation layers, the breakdown voltage is shown to become higher when the relative permittivity of the second passivation layer becomes higher.

The purposes of this study were to obtain geographical and geological information before the dyke of the Isahaya Land Reclamation Project is opened, and to evaluate the natural recovery in the excavated area in Isahaya Bay. The presence of no new sediment within the ditches suggests that the natural return to the original geography is difficult after being artificially modified. However, the presence of many fish in the ditches suggests that the hollows are not necessarily a cause of the occurrence of low oxygen water, and that the hollows do not need any special measures to recovery geographically.

Nitrogen oxides (NOx) are one of the air pollutants that cause acid rain. Although utilization of catalysts is a typical established method of treatment to remove NOx, the catalyst is so expensive that the cost of treatment is high. The efficient treatments using pulsed power, which utilize the pulse streamer discharges, have been studied. We have brought the nanosecond pulsed power to the NOx treatment. More efficient treatment is expected with the high electric fields with a nanosecond pulse width. In using the nanosecond pulsed power, the issue of design and modification of reactors configuration becomes significant. The matching of reactor to the generator output is strict and the behaviors of discharges in the reactors as an electrical circuit are complicated. The reactors with coaxial electrodes were basically adopted and the removal ratios and the efficiencies were evaluated on various reactor configurations, in order to optimize the reactors configuration. The number, connecting method of the reactor chambers and of gas tubing was investigated to improve treatment efficiency in the experiment.

Adaptiveness is one of the target functions of research on autonomous robotic systems. However, the adaptiveness realized in the robotic research so far is quite limited and specific to sample problems and systems configurations. On the other hand, all the animals from primitive ones to insects or mammals have commonly the adaptiveness to behave in an unexpected environment. Such adaptive behaviors are the intelligent sensory-motor functions, and most essential and indispensable ones for animals to survive.

This paper is concerned with an input output relation of linear systems when using the PTI (perfect translation invariance) complex wavelet packet transform. In general, the linear relation between input-output is guaranteed by the following two conditions: (I)the mother wavelet has a better frequency resolution, and (2)the real and imaginary parts in the mother wavelet consist of a Hilbert transform pair. The PTI complex wavelet satisfying the above conditions has been used for demonstrating he linear relation. In this paper, a PTI complex wavelet packet transform is applied for improving the frequency resolution. The validity of our approach is shown through a numerical experiment.

We propose a novel interference canceller with an automatic phase control (APC) circuit that compensates for the phase error of interference signals. The proposed APC circuit estimates and compensates for the phase rotation of interference signals, enabling the proposed interference canceller with APC to achieve high BER performance. We use a maximum likelihood sequence estimation (MLSE) interference canceller for fixed wireless access (FWA) co-channel interferences. Degradations of signal replica accuracy incur a BER performance penalty. The novel APC circuit, which uses a simple estimation algorithm, can effectively compensate for frequency offset and phase rotation. Computer simulations show that our proposed interference canceller improves BER performance.

Blind source separation (BSS) is an approach to estimate original source signals only by observed signals through multiple sensors without any prior knowledge about their mixture process. The BSS is currently expected to be applied to such a broad field as preprocessing of speech recognition, analysis of bio-signals, etc. In this paper, we focus on underdetermined BSS (i.e., BSS with more sources than sensors). In underdetermined BSS, there are many estimation algorithms based on sparse representation of source signals. The accuracy of decomposition of source signals depends strongly on sparsity hypothesis. For this purpose, it is effective to preprocess observed signals by using a linear transformation with respect to the base so as to make it more sparse representation of signals. We now propose an application of time-frequency analysis to preprocessing of sound source signals using the RI-spline wavelet transform. The effectiveness of the proposed approach is demonstrated through a computer simulation using real data.

In CMOS-based smart sensors, it is particularly important for the sensor's high performance to be compatible with advanced CMOS technology. The paper introduces a split-drain MAGFET with a native substrate structure that can improve the signal-to-offset deviation ratio by a factor of four compared with that of a normal MAGFET made on a p-well substrate. The native type MAGFET is incorporated into a CMOS rotary encoder based on magnetic sensor arrays, and the resulting angle detection accuracy is improved by a factor of three when compared to the conventional structure. As a result, the proposed technology is an effective way to fabricate a low-cost and high-performance sensor chip.

Temperature stability is one of the major concerns of sensors for practical applications. This paper presents a method for improving the temperature stability of CMOS magnetic rotary encoders. The main factor of the offset drift is the piezoresistive effect due to the stress induced by packaging. A connection of four MAGFET elements offers a good offset drift cancellation capability while keeping the sensitivity. Using the proposed method, the offset drift is reduced from 0.4% without the cancellation to 0.15% with cancellation under the temperature range of 5degC to 80degC

With the emergence of the so called Lab-on-a- Chip (LOC), various chemical processes such as transport, reaction, separation, purification and detection are now integrated on glass and plastic substrates of with dimensions of several centimeters. Further development of LOC requires new microfluidic technology to process liquid droplets of a very small size without requiring complicated pumps and tubes. In this paper, we report two types of ultrasonic micromanipulators for use in liquid droplet processing. The first type of manipulator uses a single phase unidirectional interdigital transducer on a piezoelectric Lamb waveguide. The second type of manipulator uses wedge type transducers on an isotropic glass plate. Experiments using a frequency of about 6MHz confirm one-dimensional and two-dimensional transportation of water droplets.

Blind source separation (BSS) is an approach to estimate original source signals only by measurements through multiple sensors without any prior knowledge about their mixture process. BSS is currently applied to a broad range of fields, such as pre-processing of speech recognition, analysis of bio-signals, etc. In BSS of convolutive mixtures, the signals must be expanded into time-frequency domains. In this case, permutation between frequencies becomes indefinite. This paper proposes a method for solving the indefiniteness problem by using directional peaks in addition to directional nulls in directivity patterns. This method makes it possible to combine the beamforming and subspace methods and to improve the accuracy in separatability of BSS. The effectiveness of the proposed method is demonstrated by computer simulation.

The penetration of embedded computing is happening in almost every area, from low price consumer electronics to high-end aerospace and avionics environments. In today's world, automobiles, airplanes, home appliances, medical devices, cell phones, network processors, handhelds, digital audio/video, imaging, game consoles, graphics and other system-on-a-chip (SoCs) utilize embedded computers. Embedded computers, even sophisticated ones, have been used in academic and commercial products and systems, for over twenty years. However, only recently has embedded computing moved from craft to an emerging and very promising discipline. The workshop is to bring together computer scientists, industrial engineers and researchers to discuss and exchange experimental or theoretical results, novel designs, work-in-progress, experience, case studies, and trend-setting ideas in the area of embedded computing. The workshop received large number of submissions representing many different countries. All submissions were carefully reviewed by committee members. In order to allocate as many papers as possible and also keep the high quality of the workshop, we finally decided to accept 12 papers for oral technical presentation. We believe all of these papers and topics will not only provide novel ideas, new results, work in progress, and state-of-the-art techniques in this field, but also stimulate future research activities in the area of embedded computing. The program for this workshop is the result of hard and excellent work of many people, including authors, external reviewers, and program committee members. We would like to express our sincere appreciation to all of them for their cooperation in completing the workshop program under a very tight schedule. Last but not least, we thank Fusun Ozguner and Tim Pinkston, ICPP-2006 Workshop Co-Chairs for helping and encouraging the inclusion of IWEC-06 in ICPP-06.

A single-chip CMOS magnetic rotary encoder system based on magnetic pattern analysis is presented. Magnetic sensor arrays are integrated with ADCs and DSP circuits and fabricated in a standard 0.25 /spl mu/m CMOS technology. The chip detects the rotation angle within an error of /spl plusmn/0.36 degrees (10b per rotation).

The main contribution of this work is to present several hardware implementations of an "n choose k" counter (C (n, k) counter for short), which lists all n-bit numbers with (n-k) 0's and k 1's. We first present concepts of C (n, k) counters and their efficient implementations on an FPGA. We then go on to evaluate their performance in terms of the number of used slices and the clock frequency for the Xilinx VirtexII family FPGA XC2V3000-4. The best implementations run in 100 MHz for n=16, and in 40 MHz for n=1024.

Model predictive controllers have been widely used in process industries as an advanced control technology. Among them, generalized predictive controller (GPC) has been proposed by Clarke, el al. Especially, since the GPC is designed based on multi-step prediction, it is well known that the GPC is effective for systems with ambiguous time-delays and/or time-variant time-delays. On the other hand, most real systems have uncertainty, and self-tuning controllers and robust controllers are proposed to cope with such uncertainties. According to the self-tuning control scheme, the stability is not necessarily guaranteed in connection with the convergence of estimated parameters. Moreover, in robust control, the control system becomes quite conservative, and tracking properties are not good. In this paper, a new design scheme of robust self-tuning 2-degree-of-freedom GPC. That is, the closed-loop controller is designed so that the robust stability is satisfied, and the pre-filter is designed to improve the transient response in an on-line manner. According to the newly proposed control scheme, the transient property for the set point response can be improved keeping the robust stability. Finally, the proposed scheme is numerically evaluate on two simulation examples.

PID control schemes have been widely used in most process systems represented by chemical processes for a long time. However, it is still a very important problem how to determine or tune the PID parameters, because these parameter have a great influence on the stability and the performance of the control system. On the other hand, it is well known that the generalized predictive controller(GPC) is effective for systems with ambiguous time-delays and/or time-variant time-delays. Then, the PID parameters adjusting method based on GPC has been proposed. In this paper, a new tuning algorithm of PID parameters based on GPC is proposed, which is composed of an adaptive switching scheme and some robust PID sub-controllers switched by the former according to a certain property of the controlled object. Finally, the proposed control scheme is investigated on a numerical simulation example.

Before the era of information and communication technology society, both of a company, a university, and a sight-seeing city could keep their benefits with aids of an intellectual property law system and of physical barrier. Now, contents of a university or a sightseeing city, e.g. lecture notes or pictures of cultural assets can easily flow through Internet. Also results of intellectual activity, e.g. patented inventions can freely flow and be able to be realized after the time limit of patent protection. Usage of a brand is a powerful method to keep benefits in such an environment. If they could take advantage of using brand efficiently, free flows of those intellectual results through Internet could assist to increase benefits.

In this paper, we propose a method for precise camera calibration which conducts feature point extraction and camera parameter estimation iteratively. Many of conventional researches on camera calibration have focused on how to calculate camera parameters using data obtained from input images, that is, location of feature points. However, these input images suffer from distortions caused by perspective and lens imperfections. In our proposed method, at the beginning of the procedure, projective transformation matrices between image planes and a calibration target and lens distortion parameters are approximately estimated. These parameters are used in order to reduce the influence of distortions in input images. After the removal of distortions, feature points in processed images are localized precisely and used to update/calculate the projective transformation matrices, the lens distortion parameters and intrinsic camera parameters. These procedures are iterated until they converge and this iteration results in precise estimation of the parameters. The effectiveness of the proposed method has been recognized through experiments using synthesized data and real images.

We have newly developed SiGe power-bipolar-transistors with optimum base and emitter structures. The presented SiGe transistor allows fast switching operation of less than 100 ns, while the low on-state voltage drop of 0.2 V has been achieved at a current density of 113 A/cm/sup 2/. The optimized transistor with poly-Si emitter, minimized SiGe layer, and interconnect structure, enables very high current gain of /spl sim/200 at a high current region exceeding 100 A/cm/sup 2/. The presented SiGe transistors are effective in lower power dissipation in switching circuits.

In this paper, we propose a robust method to estimate planar regions using sequential stereo images for visual navigation of an autonomous vehicle. The proposed method estimates 2D projective transformations dynamically, which represent planes in space, for both stereo images and sequential images. This can be done robustly by utilizing sequential information, i.e. previous estimation of both the projective transformations and the corresponding planar region. In addition, a method for preventing misdetection due to textureless areas is proposed. The experimental results, using sequential stereo images taken from a moving vehicle, have shown that the proposed method can work robustly even in the conditions of undulation of the road and rolling and pitching of the vehicle.

Recently, evolutionary calculation has been widely used in design problems. Using this method, intelligent control is expected to be constructed adaptable to disturbances. In this paper, a routing algorithm for multi mobile robots in transportation is presented. This algorithm is based on the Ant System and extended to the routing of multi mobile robots in transportation. The proposed algorithm is applied to two kinds of transportation problems, multi vehicle transportation and routing for multi Auto Guided Vehicles (AGV) in a semiconductor fabrication bay.

One of the typical nonlinear system is a pendulum. Cause of difficulties of controlling a pendulum is mismatching of number between inputs and outputs, that is to say, single input and double outputs. The feature of a pendulum is nonlinear as to angle and linear as to cart position. Another point is that state equations do not have terms of position of a cart. In addition, the angle of a pendulum must satisfy the dynamic equation at horizontal line. In order to swing up a pendulum of these properties, first, we define a reference angle of a pendulum which satisfies the dynamic equation at horizontal line. After the pendulum goes up around upright position, we change the curve of reference angle so that the pendulum standstill at the origin. In other word, when the pendulum system becomes a linear time-invariant system, we define the ideal curves of both angle and position which will converge to the origin. The angle of the pendulum is controlled by SFC (simulator-based foresight control) all the way which can be applicable for a nonaffine nonlinear system.

Recently, Ce-doped fluoride crystals have been studied for UV solid-state laser materials. Ce:YLiF/sub 4/ crystals are known as an ideal matrix in which a Y/sup 3+/ ion can be replaced by a Ce/sup 3+/ ion. Ce:LuLiF/sub 4/ has been reported as a new and more attractive UV laser host because of its reduced solarization effects. In this presentation, we report the growth of Ce:YLiF/sub 4/ and Ce:LuLiF/sub 4/ single crystals by the Czochralski method, as well as the optical properties of the grown crystals. The crystals were grown in a RF-heated Czochralski furnace equipped with high vacuum.

Two different types of HEMT (high-electron-mobility transistor) monolithic low-noise amplifiers (LNAs) using AlGaAs/GaAs and pseudomorphic InGaAs/GaAs materials have been developed and have demonstrated excellent performance at 60 GHz. These monolithic LNAs have achieved noise figures of 5 dB, as well as associated gains of 11 dB. These two-stage circuits both utilize 0.2*60- mu m HEMT devices for both bandpass and low-pass circuit topologies. Noise figures as low as 4.5 dB have been observed for single-stage MMIC- (monolithic-microwave-integrated-circuit) implemented LNAs, and gains in excess of 20 dB have been observed for three-stage versions of this amplifier with a 5-dB noise figure in the V band. This result represents the state-of-the-art monolithic HEMT amplifier performance for AlGaAs and pseudomorphic InGaAs materials. This MMIC amplifier can occupy about less than one-third the size of existing MIC hybrid V-band LNAs. This represents a significant size reduction and cost saving due to repeatable circuit performance with monolithic technology. The chip sizes are both 1.6*2.7 mm for these two-stage amplifiers. >

Q-band monolithic amplifiers using CPW (coplanar waveguides), GCPW (ground coplanar waveguides) and MS (microstrip) structures were designed and RF-evaluated. At 44 GHz, the CPW amplifier exhibited a gain of 7.5 dB, while the GCPW amplifier exhibited a gain of 7.0 dB. These results compare very favourable with the MS amplifier, which has a gain of about 7.5 dB. It is concluded that these results are encouraging and show the great promise of future millimeter-wave monolithic microwave integrated circuit implementation using this structure. >

A high-density multipin connector that utilizes a miniature compliant press-in pin connection applied to the smaller plated-through hole to connect the pin and the backpanel is discussed. The procedure for designing the connection is described. Numerical elastic-plastic deformation analysis is used. A miniature press-in pin has been fabricated using the results of the analysis, and its mechanical performance is examined. Reliability tests confirmed that the contact resistance characteristics are sufficiently stable. The connector is applicable to equipment used in electrical communication systems. >

Five different types of high-electron-mobility transistor (HEMT) low-noise amplifiers (LNAs) utilizing AlGaAs/GaAs and pseudomorphic InGaAs/GaAs materials have been designed for 60-GHz military electronic warfare and communication applications and have demonstrated excellent performance. The best noise figure achieved, was 4.5 dB NF with an associated 4.5-dB gain at 60 GHz for a one-stage 60-GHz MMIC (monolithic microwave integrated circuit) LNA. Also, two- and three stage MMIC LNAs with higher gains were fabricated with noise figures of 5.0 and 5.3 dB, with gains of 11 and 15 dB, respectively, at 60 GHz. The highest gain achieved in V-band was over 20 dB for a three-stage version exhibiting noise figures close to 5 dB for an AlGaAs version. Both bandpass and quasi-low-pass topologies were realized for these V-band MMIC amplifiers. >

The electrical properties of CnTCNQ Langmuir-Blodgett films sandwiched between aluminum electrodes were studied using thermally stimulated electrical measurements, and both the effect of film structure and the influence of aluminum native oxide are assessed. Change in induced charge on an electrode ( Delta q/sub 2/) was measured while heating a sample. The relation between Delta q/sub 2/ and the number of deposited layers is in good agreement with a simple model based on a double layers dielectric system. Direction of the polarization in a film as deposited was determined using a photomeasurement. It was also found that Schottky conduction is the dominant mechanism for a C/sub 15/TCNQ LB film with seven monolayers. >

The authors present the systematic design, fabrication, and evaluation of a distributed wideband monolithic mixer using dual-gate HEMTs (high-electron-mobility transistors). The circuit was fabricated on an Al-GaAs-GaAs heterostructure wafer. The measurement results show that the mixer operates from 5 to 18 GHz with a conversion loss from 3 to 5 dB. This is believed to be the first reported wideband distributed mixer using dual-gate HEMT devices. The mixer chips developed are key components for EW (electronic warfare) applications. >

A Q-band high-electron-mobility transistor (HEMT) amplifier was designed and fabricated. This three-stage circuit utilizes 0.2- mu m*60- mu m micron HEMT devices. The amplifier has a measured gain of greater than 10 dB from 42 to 47.5 GHz and a peak gain of 16 dB at 44.5 GHz. >

A printed-circuit-board connector meeting the requirements of fast pulse-signal transmission in the gigabits-per-second range has been developed. Flexible printed-circuit film with a microstrip line configuration is used for the connector leads. The electrical design procedure for the connector first focuses on reducing the reflection noise and then considers other types of noise. A packaging system using the connector is shown to have excellent electrical characteristics: the maximum pulse-signal speed is 1.5 Gb/s (NRZ signal), and crosstalk on the nearest leads is below 5% (for a 300-ps pulse rise and fall time). The speed is about 3-5 times as great as for conventional connectors. >

An accurate distributed model of monolithic metal-insulator-metal (MIM) capacitors has been developed for computer-aided design (CAD) of millimeter-wave monolithic integrated circuits at V-band. The model is based on experimental measurements on microstrip ring resonators with and without the air-bridges and capacitors. The effects due to the air-bridge discontinuity, as well as dielectric and ohmic losses are taken into account. The calculated resonant frequencies are in good agreement with the experiments. This capacitor model reduces discrepancy in the resonant frequency by more than 30% compared with that used in commercial available programs. The distributed model of the MIM capacitor was utilized in the simulation of the performance of a V-band two-stage low-noise high-electron-mobility transistor (HEMT) amplifier. Computer simulations obtained using this model are in good agreement with the measured performance. >

The two-port S-parameters of 0.1- mu m gate length AlGaAs/GaAs high-electron-mobility transistors (HEMTs) at W-band (75 to 115-GHz) frequencies were found using a specially constructed six-port network analyzer. In conjunction with the millimeter-wave measurements, cryogenic microwave characterizations were also performed on 0.25- mu m-gate-length HEMTs (made with similar materials and fabrication technology as the 0.1- mu m HEMTs) to determine the extent to which the frequency range could be increased by cooling. A gas helium refrigerator was used to cool the transistors down to 50 K, and the S-parameters were measured from 2 to 18 GHz with an automatic network analyzer. The maximum available gains of a typical HEMT at room temperature and 50 K are shown. Small-signal transistor models were derived from the measured S-parameters to determine the projected maximum unity-gain frequency, f/sub max/. For this transistor, f/sub max/ increased from 33 to 45 GHz on cooling. >

A novel approach to developing a battery state of charge indicator for an electric vehicle is discussed. An approach to determining the state of charge by combing the available data of discharge characteristics of a lead-acid battery with an identification technique based on the battery mathematical model is presented. The basis for the approach is to introduce a nonlinear battery mathematical model, and to estimate the battery state of charge by least squares estimation. This approach was realized by a digital signal processor and its validity was confirmed by simulation tests using data measured during urban driving and road tests. >

A visual sense system for robots which can measure and recognize 3-D objects is proposed. This robot vision system is capable of recognizing machine parts moving on a line so that an automated transferring and assembly system with high flexibility and no dependence on external peripheral equipment can be realized. This system possesses two remarkable features. One is the measuring technique, which can effectively remove the false 3-D information due to mirror reflection using laser slit beams from two directions. The other is the technique of carrying out the recognition of polygonal surfaces by p-type Fourier descriptor, area and perimeter, and, further, the recognition of objects based on aspect graphs, using multistage neural networks. As compared with the conventional recognition method, the speed and accuracy of recognition rose a step. >

Extensions of the results of model reference adaptive control to nonlinear and/or time-varying systems have rarely been achieved. For this purpose, the authors present an alternative method of exact model matching (EMM) based on a computing predictor network with assigned damping characteristics under prior knowledge of the plant parameters. A direct method for identifying physical parameters based on small parameter changes of a simulator, called the SANQ method, is introduced for overcoming the parameters' uncertainties. Several sets of simulation and experimental results are given to demonstrate the validity of the present EMM/identification method. >

When perceiving an object humans do not receive stimuli passively, but observe the object actively to obtain useful information for recognizing it. A neural network model which realizes such active perception by utilizing neural dynamics is proposed. The model consists of a sensory unit, a representation field, and recognition units. The model sequentially collects local information of the presented figure, gradually constructs its internal image, and recognizes it based on the constructed internal image. An intentional observation mechanism was constructed by which the model can selectively observe informative parts of the figure for constructing the internal image. Several computer simulation results are described. The model is discussed from the viewpoint of Bayesian inference and information theory. >

For studying the mechanism of the brain, the synthetic approach is effective. The synthetic approach is to conjecture the mechanism of the target through constructing its model. Some 20 models of the brain were constructed for this study. One of them is described. The model includes the faculties of perception, memory, and action. Having these three faculties enables the model to realize highly intellectual behavior or self-organizing ability that cannot be realized by a model having just one faculty. The model was realized in the form of a robot which organizes purposive behavior by itself. This robot forms effective behavioral patterns to achieve the purpose through trial and error. >

The image restoration problems that arise in early vision are formulated as ill-posed inverse problems. Through regularization, they should be well-posedly solved. This is an approach to exact modeling for constraints on visual systems. The problem of modeling for constraints can be regarded as that of estimating the mean of the marginal conditional distribution of a random function based on prior information. First, the basic models with non-symmetric half plane causality for noisy and blurred image are introduced. Secondly, the parameters of the image model can be estimated from distorted images using the adaptive identification technique based on multiple edge models. Thirdly, after reformulating the image restoration problem as a class of Bayesian estimation problems, this can be well-posedly solved using a 2-D Kalman filter-type algorithm with edge-adaptation. The algorithm makes it possible to optimize a stochastic image model from the standpoint of a Bayesian information criterion. Lastly, a restoration example is given to demonstrate the feasibility and validity of the authors' approach. >

Extensions of the results of the model reference adaptive control (MRAC) to nonlinear systems have been attempted. In particular, design methods of MRAC for nonlinear mechanical systems have been studied with much interest. For this purpose, the authors propose a design method for model following control (MFC) as a first step toward MRAC for nonlinear mechanical systems affine in the control. First, they consider Hirschorn's algorithm concerning the invertibility of nonlinear control systems. However, it is difficult to apply the algorithm directly to the control objective due to the small number of input channels and no matching conditions. Secondly, they propose an alternative design method for MFC, based on a kind of nonlinear model matching techniques which make it possible to assign any damping characteristics. Thirdly, they consider an example of application to the inverted pendulum to demonstrate the feasibility and efficiency of the proposed control design strategy. Finally, they investigate two extentions of the present method to both the case of modeling error and of nonaffine type control.

Continuous-wave (CW) light output powers of 87 mW at room temperature and 30 mW at 60/spl deg/C have been obtained for ZnCdSe/ZnSSe/ZnMgSSe single quantum-well (SQW) separate-confinement heterostructure (SCH) laser diodes grown on a GaAs substrate by molecular beam epitaxy (MBE).

The effect of device layout on thermal impedance of thermally-shunted HBTs was investigated. A direct comparison of thermally shunted devices and standard airbridge devices is made. Changes in thermal resistance of up to 67% were observed. While thermal resistance remains sensitive to emitter element placement in thermally shunted devices, variations in the location of thermal shunt landings had little effect. These results provide a basis for optimizing thermally-shunted devices.

We have to synthesize a control signal to achieve an ideal control performance taking advantage of ZIR (zero input response) and ZSR (zero state response). First, we define ideal transient dynamics. Then we construct a simulator based on a physical model of a nonlinear plant, and identify not only parameters but also initial states of state variables. Since we can foresee the dynamics of the nonlinear plant output with information of the internal states of the nonlinear simulator, an ideal control performance is achieved by a simulator-based controller taking advantage of ZIR and ZSR.

The authors constructed a haptic recognition system which discriminates feel of touch based on the principles of sensory integration and attentional perception. The system is equipped with several sensors and can push and rub the object's surface with several values of force and speed. It integrates the sensory information iteratively by selecting appropriate sensors and measurement conditions according to the proceeding of the recognition. The algorithms of sensory integration and of attentional perception are realized by Bayes inference and by an iterative experimental design based on an information criterion, respectively. The experimental result shows that the system can discern a subtle difference in feel of touch. It is also proved that the system selects appropriate sensors and conditions according to the situation, that is, the attentional perception algorithm realizes good recognition accuracy with fewer observations. In addition, it is shown that the characteristics used in the system correspond well to those human beings utilize in haptic perception. These results suggest that the constructed system is a faithful model for the human haptic mechanism. >

Everything on the earth, from chemical compounds to living things, consist of about 100 kinds of chemical elements. Moreover, they are considered to have self-organized. In order to elucidate the mechanism of such self-organization we made a simple model. When self-organization occurs, there must be some information of the structures. We call the information "system description". Our model has 5 kinds of elements. If the elements are assembled to a system according to a system description, the system will have a specific function. When two system descriptions mate with each other accompanied by mutations. A new system appears. We devised the format of "system descriptions" and the mechanism of producing the structure corresponding to each system description. In our experiment several initial systems repeated mating and mutation during some generations to produce many kinds of systems each of which has its specific function. If natural selection is added to this model, it will be a model for evolution. >

Describes a method of compressor valve diagnosis in which a neural network was applied to analyze valve plate sound. A neural network approach to pattern recognition was considered. The system reduces the load of neural networks in function representation and function learning by using appropriate preprocessing methods, for instance the Mahalanobis generalized distance. The resulting system requires only a small-scale neural network, is capable of extracting the invariant features, and attains an extremely high recognition accuracy. >

A Q-band balanced resistive high-electron-mobility transistor (HEMT) mixer has been developed for high-level integration of millimeter-wave receivers. The mixer consists of two AlGaAs/GaAs HEMTs, a CPW/slot-line LO (local oscillator) balun, and an active IF balun. This mixer can be easily integrated with the RF, LO, and IF HEMT amplifiers on one chip because it uses the HEMT as the mixing device. This mixer does not require backside and via-hole processes since it is realized with CPW (coplanar waveguide) circuitry. This unique feature increases the circuit yield and shortens the processing time. The mixer downconverts the 42-46 GHz RF to a 2.3-3.2 GHz IF. The conversion loss is less than 8 dB over the entire RF frequency range with a LO drive of 14 dBm. The minimum conversion loss is 4 dB from a RF at 43.0 GHz and a LO at 40.25 GHz. The Q-band mixer also achieves a 10.3 dBm output third-order intermodulation intercept point and a 1.5 dBm output 1 dB compression point. >

A Q-band balanced diode mixer developed using AlGaAs/GaAs HEMTs (high electron-mobility transistor) as diodes, a CPW (coplanar waveguide) ratrace hybrid, and a lumped-element, low-pass filter is presented. The mixer can be easily integrated with the RF, LO (local oscillator) and IF HEMT amplifiers on one chip because it uses HEMT as a mixer diode. Furthermore, the mixer does not require backside and via-hole process and has small size (1.4 mm*1.5 mm). Therefore, it has good RF circuit yield. The mixer downconverts the 41-48-GHz RF to a 0.5-3.5-GHz IF. Without DC bias, it shows 9.4-dB conversion loss for RF at 42 GHz, with a LO drive of 11 dBm at 44.45 GHz. The mixer is the first monolithic CPW mixer for operation at Q-band frequencies. >