An automatic air-cooling incubator is proposed to replace the manual water-cooling blanket to control the brain tissue temperature for brain hypothermia treatment. Its feasibility is theoretically discussed as follows: First, an adult patient with the cooling incubator is modeled as a linear dynamical patient-incubator biothermal system. The patient is represented by an 18-compartment structure and described by its state equations. The air-cooling incubator provides almost same cooling effect as the water-cooling blanket, if a light breeze of speed around 3 m/s is circulated in the incubator. Then, in order to control the brain temperature automatically, an adaptive-optimal control algorithm is adopted, while the patient-blanket therapeutic system is considered as a reference model. Finally, the brain temperature of the patient-incubator biothermal system is controlled to follow up the given reference temperature course, in which an adaptive algorithm is confirmed useful for unknown environmental change and/or metabolic rate change of the patient in the incubating system. Thus, the present work ensures the development of the automatic air-cooling incubator for a better temperature regulation of the brain hypothermia treatment in ICU.

View full abstract

I propose a novel locally adaptive smoothing algorithm for noise reduction filtering without losing contour. Using stochastic feature of speckle pattern, this filter works as speckle suppression in ultrasonic images. The medical ultrasonic image is widely used because of many merits such as machinery simplicity, low cost, real-time observation, and high safety without radiation. However, the ultrasonic image has a defect of image quality caused by speckle pattern noise. Therefore, speckle suppression filtering is important and indispensable. In this paper, the proposed locally adaptive filter for speckle suppression is evaluated by simulations. I also attempt to automatic segmentation of the superficial peripheral vein image obtained by a high-frequency ultrasonic equipment.

View full abstract

We developed an ultrasound reflectivity computed tomography(CT) system with only small number of transmitter/receiver transducers for the carotid diagnosis. To this end, a ring array system with 8 element transducers was fabricated with center frequency 10MHz. The present CT technique assures the image with high resolution in the order of 0.1mm over the entire imaging region. In addition to that, the reduction of the number of transducer element enables the fast data acquisition and the lower cost hardware system. With appropriate combinations of the dual and/or separate transmitter and receiver pairs, 21 backscattering projection data were observed. To comply with the ill-posed inverse problem due to the lack of the angular view data, conjugate gradient method was applied for the solution of the matrix observation equation. Both experiment and simulation examinations were made using the cylindrical shell object as a carotid phantom. As a result, it is demonstrated that the image with greater resolution compared with the conventional B-mode method can be obtained by using only 8 element transducers.

View full abstract

The purpose of this study is to develop the tactile display which consists of a single vibrator by using the modulated vibration. To achieve this purpose, we must evaluate the stimulus difference of the tactile in the modulated vibratory stimuli. In this study, the stimulus difference has been estimated by detecting the P300 wave in the averaged waveform of the ERP. However, it is indispensable to evaluate the stimulus difference efficiently to make the differential limen in the frequency region to express the modulated vibration clear. In this paper, the single trial analysis on the ERP that the significant difference of the EEG before and after the stimulus presentation is evaluated by the test technique is proposed.

View full abstract

This paper describes the hardware design of core functions of the Internet protocol IP over IEEE1394 interface (IP over 1394) and its implementation on an FPGA. The design was evaluated by counting the number of FPGA logic elements required for the implementation. Using a system clock of 49.152MHz, we verified that packets sent from an application on top of the protocol stack were correctly received by the other protocol stack via the IEEE1394 port at a transfer rate of 400 Mbps. We also verified the communication behaviors of the design with an isochronous resource manager to reserve a channel prior to data transmissions. The hardware cost of the core IP layer was less than that of the link layer. The evaluation results will help the IP-over-1394 designers explore quantitatively various spectrum of the software/hardware design alternatives.

View full abstract

There are rear projection large 3D displays that allow you to view 3D TV motion pictures without special 3D glasses. The screen employs lenticular plates whereby both the projection side and viewing side have the same double symmetrical shape. A double symmetrical design calls for two lenticular plates having approximately the same shape fabricated to enable a stable 3-dimensional view. When designing the lenticular shape, however, it is difficult to satisfy light absorption characteristics on both the projection and viewing sides while maintaining the geometric relationship of both the projection and viewing sides.
This paper describes a design method of symmetrical double lenticular screen for rear projection large 3D displays, and also describes a description of the result of applying the method to a 70-inch HDTV 3D display using a symmetrical double lenticular screen. As a result, we were able to obtain a quality 3D effect that requires no special 3D glasses, and by using a lenticular screen, we are able to view a crisp, high definition 3D image.

View full abstract

In this paper, we present a method of face recognition using 3D images. We first compensate for the poses of 3D original facial images using feature points and geometrical measurement. Then, the contour maps which are invariant under different illumination conditions are extracted for recognition. In the second step, a method is adopted for face recognition based on fuzzy clustering and parallel neural networks (NN's). Experimental results for 35 persons with different poses and illumination conditions demonstrate the efficiency of our algorithm

View full abstract

This paper addresses the fuzzy control problem for a class of nonlinear systems using the Lyapunov synthesis approach. In order to deal with cases where the system state is unavailable, a state observer is proposed. Consequently, the whole system behavior can be attributed to a kind of the standard singularly perturbed form. At the same time, to deal with the gap, if any, between the real state and its estimated value from the state observer, we view it as a part of system disturbance, and propose a unique way to deal with the disturbance, i.e., adopt a switching function with an alterable coefficient, which is tuned by adaptive law based on the tracking error between the output of the considered system and the desired value. Finally, it is shown that the fuzzy controller proposed guarantees the tracking error will shrink to zero, while maintaining all signals involved in the system stable.

View full abstract

In this paper, a new system identification scheme is proposed based on a memory-based modeling (MBM) method. According to the MBM method, some local models are automatically generated using input/output data pairs of the controlled object stored in the data-base. Especially, it is well known that the MBM method works suitably on nonlinear systems. Therefore, even if the nonlinearities are contained in the controlled object, accuracy identification can be performed by the proposed method. Moreover, since the parameter estimates are easily applied to many existing controllers, the good control result can be obtained for nonlinear systems. In this paper, the generalized predictive control (GPC) is used as the one of existing controllers, because the GPC is designed based on multi-step prediction, and is effective for systems with large, ambiguous and/or time-variant time-delays. Finally, the effectiveness of the newly proposed control scheme is numerically evaluated on some simulation examples.

View full abstract

As the atmospheric electric field always fluctuates its observed waveforms are very complicated. It is important to detect the characteristic variation from the observed signals and find out its cause. The natural observation method is adequate to analyze such non-periodic signals as the atmospheric electric field signals. In this method the aspectum is proposed as the quantity to estimate the instantaneous power. In this study signal processing by the aspectum was applied to signals of atmospheric electric field variation. It was clarified that there exists the strong correlation between the instantaneous fluctuation of atmospheric electric field and the wind speed in the convenient meteorological environment.

View full abstract

In recent years, the various controller design schemes have been examining for nonlinear systems with uncertainties. As one of such mechanical systems, an acrobot attract attention in last decade, which belongs to nonholonomic systems. Several controller design schemes have been proposed for the acrobot.
In this paper, a new approach to control the acrobot is considered, in which a suitable set of swing up pattern is created using the real-coded genetic algorithm(GA). The fitness function includes some terms, i.e., the time, the extra motion and the height in the inverted state. The effectiveness of the newly proposed scheme is numerically verifyed on some simulation results.

View full abstract

In this paper, the robust tracking control problem for rigid robot with discontinuous uncertainty is considered. The controller design approach is based on Filippov's framework and generalized Lyapunov stability theory. It is shown that no Lipschitz continuous controller to stabilize the error system asymptotically exists because of the existence of nontrivial equilibrium set including the origin. A robust controller which makes the error system asymptotically stable is derived in Filippov's framework.

View full abstract

A neural algorithm for the two-layer planarization problem using a gradient ascent learning of the Hopfield network is presented. This algorithm which is designed to embed a two-layer graph on a plane, uses the Hopfield network to get a near-maximal two-layer planar subgraph, and increases the energy by modifying the weights and the thresholds in gradient ascent direction to help the network escape from the state of near-maximal two-layer planar subgraph to the state of the maximal two-layer planar subgraph. The experimental results show that the proposed algorithm generates much better solutions than traditional Hopfield network and simulated annealing.

View full abstract

This paper presents an immune system derived approach for planning transportation of materials between manufacturing processes in the factory. Transportation operations are modeled by Petri Net, and divided into submodels. Transportation orders are derived from the firing sequences of those submodels through convergence calculation by the immune system derived excitation and suppression operations. Basic evaluation of this approach is conducted by simulation-based investigation.

View full abstract

Diagnosis for discrete event systems has been investigated. In this paper, authors examine a state estimation problem of a system modeled by a finite state automaton in which each state has its corresponding logical formulas. We formalize a diagnosis problem of truth values of atomic propositions which constitute the logical formulas. Our approach to the problem is based on the discrete event system theory by use of linear-time temporal logic.

View full abstract

A cutting stock problem against flawed and connectable resource is a kind of a constrained two-dimensional orthogonal guillotine cutting stock problem where orders are assigned to resources. A resource is cut by “sets" which is combinations of fixed cutting blade for length direction, and cut end to end for width direction. Moreover, there are features that flaws that cannot be assigned to orders exist on resources, and that connection which enables to cut several parts of resources by the same set is permitted. This research proposes a method by two phases. In first phase, the rectangles called “available area" which is generated by avoiding flaws and connecting, are created by deciding positions of set change according to an expected extracting rate and a number of set changes. Second, assignment for each available area is done from an order with a larger width by using a branch and bound method in order to achieve a higher yield ratio. At this time, an effective degree of a set pattern is calculated for each useful area. The effective degree means a rate of improvement of an extraction rate per length of an assigned target order. It is possible to decide assignment rapidly by searching a set pattern which has a large effective degree earlier. The proposed method has been applied to a real cutting stock problem.

View full abstract

This paper describes application system architecture using cellular phones as user interface devices, which enables users to interact with the system by graphic symbols on a client screen. Our approach has the following features: (i) divided interaction logics running on a server and a Java phone client; both interaction logics cooperate to accomplish a user's operation using a simplified script, (ii) local interaction which enables users to handle figures on a client screen without connecting to a server, and (iii) device-independent script which hides the differences of API sets among various cellular phones. By using this architecture, complicated figures including lots of graphic symbols can be displayed in spite of program-size limitation on a client device, and application programs including same interaction logics are just described once for various cellular phones. Our experiments show the advantage of the local interaction. A client program can respond immediately when handling complicated figures. The ratio of requests to the server is reduced to 23%. It takes less than 9 seconds to display typical contents, which is good enough for practical use. This method also reduces development costs at the second development or later.

View full abstract

A navigation system was installed in many cars, and a large amount of information came to be offered to drivers in recent years. However, characters and symbols on the display are not easy for the aged to see because of a reduction in visual functions. We have proposed a method of deciding an appropriate character size not only for the youth but also for the elderly. This method uses the relation between the static visual acuity and the viewing distance by age. In addition, this method is based on the relation between the accommodation time and the age. In this method, a minimum character size is decided in which drivers are able to read the characters and symbols in a short gaze time while driving. An evaluation experiment has showed that this method is very effective to improve the display visibility including the elderly.

View full abstract

The decay of the recorded magnetization due to the thermal fluctuation has become one of the most important problems to be overcome in the technology of high density magnetic recording. In this paper, a self-consistent 2D read/write simulation based on the Poisson equation by a finite element method(FEM) is performed on a system of an AFC medium and a thin film head by using an extended Stoner-Wohlfarth(SW) model for the recording layers, where the switching rate of each SW particle at finite temperatures is taken into account by using the Neel-Arrhenius law and the master equation.

View full abstract

Aiming an IC implementation of a DC-DC converter which can provide a 4.5˜5 V stepped-up voltage for diode-lamps, a switched-capacitor (SC) DC-DC converter is proposed in this paper. Different from a conventional approach employing doubler circuits, the proposed circuit provides the output voltage by achieving a 3/2 step-up conversion. Therefore, decline in power efficiency for the proposed circuit is gentle. The process of DC-DC conversion is analyzed theoretically. To confirm the validity of the circuit design, SPICE simulations are performed. For the input voltage 3.2˜4.5 V, the power efficiency is 73˜92 % in the output current about 150 mA.

View full abstract

This letter presents an analyzing method on the series-connected Fabry-Perot resonator composed of a lossless nonlinear media. Simulation results on the bistability are demonstrated, and differences in the bistable characteristic from the conventional single Fabry-Perot resonator are also described.

View full abstract

Raman amplification has many attractive features. In case of WDM transmission through standard SMFs a dispersion compensation fiber (DCF) is effectively united with a Raman amplifier by launching pump light into the DCF. We propose a novel Raman amplifier with a reflective structure that has advantages such as high efficiency of pumping, easy combination with a tunable dispersion compensator, and easy protection against pump power leakage into the transmission line. Optical power budget of 6dB is obtained comparing to the case without amplification in an experiment where optical signals of 40-Gbit/s 8-channls WDM with a 0.8-nm spacing are transmitted over 60km.

View full abstract

One-by-one transmission and the IP routing through one-fiber ring have been proposed. The network was constructed, the routing functions were implemented, and the throughputs were measured. No additional degradations were found between the throughputs by the proposed transmission and the direct uploading.

View full abstract

The paper describes a control technique for an active stereo vision robot. For the purpose of achieving both rigid tracking and accurate estimating, the proposed approach unifies the image-based high gain PD control law for tracking and the mechanical impedance control law for stable motion of the base axis.

View full abstract

The Particle Swarm Optimization (PSO) method is one of the most powerful methods available for solving both unconstrained and constrained global optimization problems. However, knowledge about how the PSO method works or finds a globally optimal solution of a global optimization problem is as yet limited. This paper deals with a physical interpretation of the PSO method in order to obtain strategies about how to tune its parameters. The results of this study will contribute to a reduction of labor for tuning the parameters of PSO.

View full abstract