本記事に「抄録」はありません。

抄録全体を表示

In this work, the effects of source and drain resistances (R_S, R_D) on device characteristics are investigated for sub-20nm MOSFETs. The current driving capability is calculated for several structures such as planar bulk, SOI, and Multi gate MOSFETs by using the ITRS data. It is found that the degradation of the drain currents due to R_S and R_D becomes significant as the gate lengths are scale down to sub-20nm region. In order to investigate the effects of R_S and R_D on the device parameters such as the channel length modulation coefficient λ and the saturation drain current I_DSAT, the drain currents are simulated by using the circuit simulation. The intrinsic MOSFET model parameters were extracted from the experimental I_D-V_D characteristic of 20nm nMOSFET. The source and drain resistances are changed from 0 to 100 ohm. It is found that the degradation of I_DSAT due to R_S and R_D shows the linear gate voltage dependence. For the long channel MOSFET, the degradation of λ shows the linear gate voltage dependence. On the contrary, for the short channel MOSFET, the degradation of λ shows the little gate voltage dependence.

抄録全体を表示

It has been thought that silicon (Si) is a material inapplicable to active optical devices, such as light-emitting diode (LED), because of its indirect electronic band structure. However, Prof. Kensho Okamoto, who is one of the authors, found the fact that commercially available Si photodiodes emitted infra-red light, and he developed an original and unique two-way optical communication system in which each Si photodiode (Si-PD) played two roles of light-emitting diode and light-receiver.
In this study, we observed directly infra-red light emission from a PN Si-PD and a PIN Si-PD using a infra-red camera and measured the emission spectra of the Si-PDs. Further, we succeeded in sending and receiving of light signal at room temperature between two PN Si-PDs, between two PIN Si-PDs, and also between a Si-PD and an InGaAs photodiode
In this paper, we introduce our novel sending and receiving of light signal system using Si-PD as LED.

抄録全体を表示

The electrodeposition coating process, which is a polymer film deposition method using water electrolysis, is widely used for automobile body primers. Recently this process is being used in the insulating polymer films deposition for the microelectromechanical system (MEMS) or micro electric components. However, this process has difficulty in depositing polymer film on complex shapes and non-conductive surfaces. In this paper, we demonstrate that epoxy resin thin films used extensively as insulating polymer films were successfully deposited using the electroless chemical reaction in aqueous solution on a non-conductive surface and high aspect glass tube. The substrates catalyzed using a commercialized three-step Sn/Ag/Pd activation process were immersed in the reaction solution containing water-soluble resin and NO₃^- ion, reducing agent (DMAB). The pH near the substrate rose when NO₃^- was reduced by released electrons from DMAB. Water-soluble resin combined with OH^-; hence, polymer thin film was deposited by the electroless deposition reaction. By FE-SEM and FT-IR measurement, it was clear that the conformal and dense epoxy resin films were deposited. Using the present method, epoxy films could be deposited on the surface of a high aspect ratio glass tube 50 mm in length and φ3 in inner diameter. These films had high insulation resistivity of 10⁸∼10¹¹Ωm with applied voltage of 250 V.

抄録全体を表示

A stochastic flash analog-to-digital converter (SF-ADC) utilizing device mismatch is designed using a 65-nm CMOS process. Since the proposed SF-ADC uses thresholds determined by the input-referred comparator offsets, the large input-referred offsets are allowed. The quantization error and non-linearity of SF-ADC are demonstrated, and the input range is enlarged by using non-linearity reduction technique. At 1.6GS/s sampling, the designed ADC achieves 34.7dB SFDR and 29.0dB SNDR without any calibration circuits despite the large input-referred offset of 102mV. At this conversion speed, it consumes 134mW with a 1.2-V power supply.

抄録全体を表示

A voltage source inverter with an LCL filter is often used for a utility interface to control its output current to a grid side because of its harmonic reduction advantages. The integral compensator is often used to reduce the steady-state errors. However, there is always a control delay due to sinusoidal variations. This paper proposes a digital sinusoidal compensator which is based on the internal model principle to realize a fast sinusoidal response with no delay. It is based on the internal model principle to realize a response with no deviation for a periodic sinusoidal reference input. It has a simple numerator and a denominator z²-2zcosωT +1 of a transfer function which is equal to the z function of a sinusoidal waveform of the angular frequency ω and the sample time T. Compensator and feedback gains of the inverter are determined by the dead-beat or the optimal control principle. The proposed method is investigated for performances and it is validated through simulation and experimental results by a DSP control system.

抄録全体を表示

In many engineering fields, distance to targets is very important. General distance measurement method uses a time delay between transmitted and reflected waves, but it is difficult to estimate the short distance. On the other hand, the method using phase interference to measure the short distance has been known in the field of microwave radar. Therefore, we have proposed the distance estimation method based on interference between transmitted and reflected audible sound, which can measure the distance between microphone and target with one microphone and one loudspeaker. However, the distance measurement performance of the proposed method was degraded by additional noise in noisy environment. In this paper, the acoustic distance measurement method based on the interference between the transmitted wave and the reflected waves is expanded to a noise-robust method in a noisy environment. In order to measure distance in a low-signal-to-noise ratio (low-SNR) environment, we introduce a concept from synchronous addition and spectral subtraction to the proposed method. Evaluation experiments in a real environment reveal that the proposed method was valid and effective because the proposed method could measure distance in a low-SNR environment.

抄録全体を表示

In this paper, we consider parameter estimation of the state variables and unknown parameters of Induction Motors (IMs) using nonlinear filters. Simultaneous estimation is the most general method for sensorless controlled IMs, and at present, by the advance of computer processors, nonlinear filters have been applied to various occasions, so we describe the method for applying nonlinear filters to Induction Motors model, and consider its estimate performance by simulations. Simulation results showed that nonlinear filters have more accuracy estimate performance than the adaptive observer, and the excellent noise immunity.

抄録全体を表示

In recent years remarkable results have been achieved in the field of object recognition. Recognition performance of more than 90% are not rare anymore leading to the conclusion of an application beyond scientific fields. However, such a high performance is often a result of unrealistic constraints of the images to be recognized and the use-cases which are only applicable in controlled laboratory environments. In this paper we propose a system working even under difficult conditions and achieving higher recognition performance as compared to other state-of-the-art systems.

抄録全体を表示

Recently, graphic boards have higher performance with development of 3DCG and movie processing than CPU, and widely used with progress of computer entertainment. Implementation of the General-purpose computing on GPU (GPGPU) become more easier by the integrated development environment, CUDA distributed by NVIDIA. GPU has dozens or a hundred arithmetic circuits, whose allocations are controlled by CUDA. In the previous researches, the implementation of the neural network using GPGPU have been studied, however the learning of networks was not mentioned because the GPU performance is low in conditional processing whereas high in linear algebra processing. Therefore we have proposed two methods. At first, a whole network is implemented as a thread, and some networks are taught in parallel to shorten the time necessary to find the optimal weight coefficients. Secondly, this paper introduces parallelization in the neural network structure, that is, the calculation of neurons in the same layers can be paralleled. And the processes to teach for same network with different patterns are independent also. As a result, the second method is 20 times faster than CPU, and compared with the first proposed method, that is about 6 times faster.

抄録全体を表示

In recent years, Augmented Reality (AR) becomes the focus of attention as a technology for obtaining some information. Most AR systems have used some markers to display any information. However, since markerless AR systems can be used intuitively, they are researched actively.
In this paper, we study about the AR system based on the hand recognition as a markerless system. The reason why we used a hand is that we don't need to prepare or carry on any tool, and can easy to watch any information on a hand. Taehee Lee et al. introduced HandyAR which employ the hand recognition. In this method, there are some problems. At first, when the method learns skin model, it needs many learning data. This is because it uses RGB color model. Secondly, it is difficult to get hand because hand images which are extracted by the skin model have noise. Thirdly, the method uses finger positions to estimate the hand coordinate. Therefore it is sensitive to finger state. We improved HandyAR to solve these problems. Experimental result showed that our method had higher performance than conventional methods.

抄録全体を表示

Super-resolution is a process for obtaining high quality and high resolution images from a set of or only one low-resolution image. The most practical one for image super-resolution is reconstruction-based method, which minimizes the difference between observed low-resolution images and the estimation for high resolution image. Therein, interpolation step plays a key role for the estimated high resolution image quality. Usually, the conventional bilinear or bicubic methods are used in the reconstruction-based super-resolution. However, these conventional interpolations generally lead to blurring on edge regions and need more time for convergence in reconstruction-based super-resolution method. Therefore, this paper propose a gradient based edge preserving interpolation method, which can reduce not only artifact noise but also blurring near edge regions in the estimated high resolution image. Furthermore, our proposed interpolation method can also solve large complexity and time-consuming problem in the recently developed New Edge-directed interpolation, which usually can achieve sharp edge in the high resolution reconstructed image. Experiments validate that our proposed interpolation method for image super-resolution is more effective than the conventional interpolation ones.

抄録全体を表示

This paper proposes a new RTL power estimation method based on power contour model. This model realizes an accurate and efficient power estimation. The experimental results show that the max error is reduced by 56.19%, and RMS error is also reduced by 3.41% from the conventional table-based RTL power estimation.

抄録全体を表示

Specific object recognition methods based on exact matching of feature vectors are known as one of methods which can achieve high recognition performance for large-scale 3D specific object recognition. Since there are few common 3D object datasets whose size is sufficient to explore the effect of difference of object dataset composition and the effect of increasing number of objects for recognition, these effects for specific object recognition methods based on exact matching of feature vectors are discussed insufficiently. The number of objects in famous datasets (e.g., COIL-100) is around 100. Therefore, in this research, we prepare the dataset of 1002 3D objects by ourselves. In this paper, we will discuss the effect of dataset differences, which are based on object structure, texture and the number of objects, for those methods such as the method based on the Bloomier filter and the method based on a hash table with this dataset in addition to COIL-100.

抄録全体を表示

In this paper the authors propose a classification method of inquiry e-mails for describing FAQ (Frequently Asked Questions) and automatic setting mechanism of judgment thresholds. In this method, a dictionary used for classification of inquiries is generated and updated automatically by statistical information of characteristic words in clusters, and inquiries are classified correctly to each proper cluster by using the dictionary. Threshold values are automatically set by using statistical information.

抄録全体を表示

The authors inject NO into a diesel engine intake and investigate the reduction of NOx generation at the combustion chamber at first. The result shows that 20 - 30% of the injected NO (0.225 ∼ 0.72 slm) is reduced. Discussion through the calculation of the extended Zeldovich mechanism suggests that the reduction is mainly attributed to the region where equivalence ratio ranges in 1.1 - 1.5 and that the reaction between NO and hydrocarbon species slightly contributes to the NO reduction. Moreover, another experimental result shows that the injection of NO slightly improves specific fuel consumption, e.g. 0.4% at NO injection of 0.72 slm for intake airflow of 285 slm. Calculation of reaction enthalpy of NO reduction and CO oxidation considerably meets the experimental results on the change in fuel consumption.

抄録全体を表示

Recently, side-channel attacks have become a serious problem. These attacks estimate secret keys of cryptography circuits embedded in hardware. In particular, the most threatening side-channel attacks are differential power analysis and correlation power analysis that use the correlation between information processing and power consumption, which are related to secret keys in cryptography circuits. Therefore, new measures are required to prevent confidential information in cryptography circuits from being leaked to side-channel information, such as power consumption. When designing preventive measures, resistance to side-channel attacks, i.e., tamper resistance, must be evaluated. This study proposes a new simulation method by which tamper resistance can be verified in the algorithm and architecture design phases. Experimental results show the validity of the proposed simulation method.

抄録全体を表示

A unified design method for multiband filters using a frequency mapping approach in the classical filter design is proposed. It has previously been reported that dual band bandpass and band elimination mapping approaches are obtained by applying the bandpass or band elimination mapping twice. In this study, we examine the variation of the filter characteristics by repeating the frequency mapping, and we have found that the frequency mapping for power-of-two bands can be obtained for the filter design. We have also confirmed that not only the mapping for the power-of-two bands but the mapping for arbitrary bands including odd number bands can be obtained. The present study has demonstrated that the filter design method by using the proposed frequency mapping is very effective for the multiband filter design.

抄録全体を表示

Cortical dipole imaging is one of the spatial enhancement techniques from the scalp electroencephalogram. We investigated the dipole imaging for multiple signal sources under time-varying non-uniform noise conditions. The effects of incorporating statistical information of noise into the spatiotemporal inverse filter were examined by computer simulations and experimental studies in three sphere volume conductor model. The parametric projection filter that incorporated with noise covariance was applied to the inverse problem of EEG measurements. The noise covariance matrix was estimated by applying independent component analysis to the scalp potentials. The spatial filter was expanded to apply to the time-varying non-uniform noise conditions such as eye blink artifact. Moreover, multiple dipole distributions were introduced to extract and to visualize individual signal sources. The proposed imaging technique was applied to human experimental data of visual evoked potentials. We obtained reasonable results that coincide to physiological knowledge.

抄録全体を表示

In this paper, we propose a novel method for extending bandwidth of ECG measuring device without deteriorating the baseline tolerance for body motion. The proposed method is realized by synthesizing a two-stage analog forward filter with two different corner frequencies and a digital inverse filter having a corner frequency identical with the higher one of the analog filter. We applied the method to bed-type capacitive ECG sensor which can detect electrocardiographic potential non-intrusively and indirectly. The results demonstrated that the proposed method could precisely recover low-frequency components like T-wave. Furthermore, we confirmed that QTc (corrected QT interval) could be estimated from the recovered wave and that the QTc correlated 0.81 on the average with that obtained from Lead II ECG for seven subjects. These results indicate that the proposed method is useful to screening of long QT syndrome by combining the method with the bed-type capacitive ECG sensor.

抄録全体を表示

One of physical indices for evaluating the image quality of digital radiography is contrast-to-noise ratio (CNR). The CNR was reported to simulate the human perception. Thus, the CNR is a variable index for evaluating detectability for digital radiography. However, it is difficult to evaluate the CNR in non-uniform image area. In this study, we have devised a new method for measuring CNR in non-uniform image area on digital radiography using top-hat transform. Furthermore, we investigated the radiation dose dependence of CNR in the digital image of an anthropomorphic breast phantom containing a simulated calcification. We have shown that the CNR in non-uniform image area can be analyzed with high accuracy by our devised method for measuring CNR. In the radiation dose dependence of CNR, CNR increased with mAs values and reduced depending on the increase in the concentration of glandular in the anthropomorphic breast phantom.

抄録全体を表示

Transplantation of stem cell-derived neurons into the damaged neuronal tissue is one of the promising methods for the recovery of neuronal functions. However, little is known about the signal transmission between transplanted stem cell-derived neuronal networks and the host brain tissue during recovery. In this study, we developed microelectrode-array (MEA) substrates with multiple micro-chambers for cell cultures. Each micro-chamber consisted of 4 cell-plating spots with interconnecting conduits fabricated by polydimethylsiloxane (PDMS) lithography. The conduits had the tunneling structures, the height of which was approximately 6 μm, through which only neurites could grow into adjacent cell-plating spots. We cultured mouse cortical neurons in the subset of cell-plating spots, and stem cell-derived neurons were injected into the residual spots 7 days after starting incubation of primary cultures. Intracellular calcium transients as well as electrical activity were detected from both primary cultured-cells. And extracellular electrical activities were recorded from both primary cultured-cells and stem cell-derived neurons in the co-cultures. These results suggested that our MEA-based co-culture system would be a useful tool for in vitro monitoring of transplantation processes of stem cell-derived neurons into host neuronal tissue.

抄録全体を表示

In this paper, we propose a new design method of frequency dependent servo system for the positioning of flexible structures. In order to reduce the sensor costs and to suppress spillover phenomena which are caused by ignored elastic modes of the plant, we introduce a pre-filter into an input which possesses characteristics of notch filter. The proposed method enables to design specified characteristics of the pre-filter and to assign poles to desirable locations of a closed loop system exactly. As a result, it enables to solve trade-off problem between suppression and transient. Some results of simulation and experiment are given in order to verify the usefulness of our method.

抄録全体を表示

In the development of a wearable robot assisting paraplegic walking, prevention of falls is important for practical applications. Considering fall prevention, it will be efficient to measure ground reaction force. We developed a measurement system of ground reaction force which consists of small 3-axis force sensors. Measured data, such as ground reaction force and center of pressure (COP), was applied to fall prevention control. In this paper, we propose a control method to adjust the timing of swing motion besed on zero-moment-point (ZMP) and to support weight shift. We evaluate effectiveness of the proposed method for prevention of falls according to measurement experiments of walking.

抄録全体を表示

Histogram Equalization is a traditional image improvement method for enhancing image contrast. However, this image contrast improvement is unsuitable for some consumer electronic applications since mean brightness of output image is different from that of the original image. Nevertheless, when the average brightness of the output image is close to the mean brightness of the original image, its contrast is also decreased thereby reducing efficiency of contrast improvement algorithm. In this paper, we propose a method that not only preserves the mean brightness but also improves the contrast of the output image. By separating the Histogram into Sub-histograms or Multi-peak histogram based on local minima, each sub-histogram will be spread out in a maximum dynamic range based on their gray level interval proportions. Subsequently, the conventional histogram equalization is independently applied to sub-histogram of each peak. To preserve the mean brightness, positions of local minima will be finely adjusted by shifting them within a certain gray scale interval. During each adjustment, sub-histograms of the two local minima which connected to the new local minimum point are re-equalized again. Finally, standard deviation will be normalized in order to improve contrast of the output image.

抄録全体を表示

The paper proposes a new system that can detect suspicious actions such as a car break-in and surroundings in an open space parking, based on image processing. The proposed system focuses on three points of “order”, “time”, and “location” of human actions. The proposed system has the following features: it 1) deals time series data flow, 2) estimates human actions and the location, 3) extracts suspicious action detection rules automatically, 4) detects suspicious actions using the suspicious score. We carried out experiments using real image sequences. As a result, we obtained about 7.8% higher estimation rate than the conventional system.

抄録全体を表示

The process of “image segmentation and extracting remarkable regions” is an important research subject for the image understanding. However, an algorithm based on the global features is hardly found. The requisite of such an image segmentation algorism is to reduce as much as possible the over segmentation and over unification. We developed an algorithm using the multidimensional convex hull based on the density as the global feature. In the concrete, we propose a new algorithm in which regions are expanded according to the statistics of the region such as the mean value, standard deviation, maximum value and minimum value of pixel location, brightness and color elements and the statistics are updated. We also introduced a new concept of conspicuity degree and applied it to the various 21 images to examine the effectiveness. The remarkable object regions, which were extracted by the presented system, highly coincided with those which were pointed by the sixty four subjects who attended the psychological experiment.

抄録全体を表示

A method for designing wide-band complex impedance-matching circuits using nonuniform transmission lines is presented. The proposed method is based on genetic algorithms and cubic spline interpolation. It is shown that various wide-band matching circuit design problems can be solved without increasing theoretical and computational complexity. Design examples are performed, and the proposed method is demonstrated to be effective for designing wide-band matching circuits.

抄録全体を表示

In examination with flexible cystoscope, medical doctors need to observe whole inner of the bladder which is broadly divided into seven parts. However, since five of the parts have no difference in shape and color, sometimes it is difficult for medical doctors to confirm that all the parts were completely observed. As the preprocessing of a system for thorough observation, we have proposed a system for estimating handling techniques of flexible cystoscope. However, the system could not estimate the handlings against noisy images obtained from the cystoscope. In this paper, we propose a method of correcting outliers in the noisy images using time-series optical flows.

抄録全体を表示