Recent communication networks has astonishingly advanced and they are widely used in the world. For the next generation networks, further researches supporting network technologies have been devoted in many areas. This paper provides an overview of the existing key technologies and future trends on communication systems such as cipher, error control coding, OCDMA (Optical Code Division Multiple Access) and UWB (Ultra Wide Band ).

View full abstract

This paper proposes a method for estimating noise-power spectrum and reducing stationary noise components in noisy speech signal. Noise reduction generally suppresses the stationary noise by applying a multiplicative gain calculated from the estimated noise-power spectrum. However, the accuracy of the noise-power estimation is degraded by the presence of superimposed speech. The proposed method aims to maintain the accuracy of the noise-power estimation in a period of speech. The method first estimates the power ratio of noise to input signal for each frequency bin by assuming the noise amplitude spectrum to be constant in a short time period which cannot be applied to the amplitude spectrum of speech. The method then improves the estimation accuracy by compensating for the errors caused by time variations in the stationary noise. Simulation results demonstrate the accuracy improvement of the noise-power estimation for the stationary noise and the better performance in terms of noise reduction.

View full abstract

We construct a speaker authentication system, where 3-layerd neural networks with ensemble learning algorithm are used, and investigate the effect of ensemble learning on the performance of the system. From the results, we found that the authentication rates of the system for a person became to 100% by using ensemble learning. Therefore, the new ensemble leaning used in this study is thought to be useful for the speaker authentication system with layered neural networks. In addition, a new multi-step authentication system for many persons by extending the system for a person was suggested. In the system, the ensemble learning was also useful for the speaker authentication system of neural networks for many persons.

View full abstract

Bonohöeffer-van der Pol (BVP) oscillator is a planar nonlinear model exhibiting a rich variety of bifurcation phenomena for both equilibria and limit cycles. Since it has two ports regarding its state variables, many topologically different coupled systems are obtained even when two identical BVP oscillators are resistively coupled. Although individual coupled systems have been already studied, it is not able to compare them because models of nonlinearity and normalizations are different. In this paper, we firstly unify them, and investigate their bifurcations of equilibria and synchronization modes of limit cycles. As results, common bifurcations of equilibria for all coupled structures are found, and some properties on parameter ranges, synchronization modes and their stabilities are clarified.

View full abstract

In this paper, a fast approximate feature pyramid construction method for deformable part models (DPMs) is proposed. DPMs can describe significant variations in the appearance of objects. Therefore, they are widely used for object detection due to their high accuracy and efficiency. However, their high computational cost is a major issue to be tackled for time-critical tasks, such as the real-time image recognition of video. For such tasks, a fast detection process is desired. In the proposed method, we focus on the optimization of the feature pyramid constructor and replace a conventional constructor with a resampling-based method. The conventional feature pyramid constructor requires multiple recursive rescaling processes, which are very time-consuming. By replacing these rescaling processes with an image-resampling process, we can improve the speed of the object detector. Approximate sampling is effective because the original scaling method requires a moderate image-size reduction. To avoid performance loss, we also propose a feature pyramid constructor framework that combines approximate scaling and conventional scaling. The evaluation results show that the proposed resampling-based feature pyramid method is 7% faster than the conventional method. In addition, this method can be combined with other optimization methods for searching the feature space.

View full abstract

We propose a method of detecting asbestos in building materials by integrating various kinds of classifiers. Recently, asbestos-related illnesses have become a nation wide problem in Japan. Now, human inspectors check whether asbestos is contained in building materials. An asbestos detection method using a support vector machine (SVM) with a weight summation kernel of color and shape has been proposed. It was effective but it did not work well for asbestos with very thin and low contrast because only a single detector with 40×40 pixels was used. Since the color, shape and size of asbestos vary in microscope images, it is difficult to detect them with a single classifier. Therefore, we train many classifiers with various region sizes and feature types, and integrate them along the orientation of asbestos. We collect the asbestos detection with high accuracy and a small number of false positives by considering the asbestos orientation in classifier integration.

View full abstract

This paper is concerned with single nonlinear feedback control for nonlinear systems with noisy measurement. A given nonlinear system is linearized piecewise to design the linear optimal controllers, which are then smoothly united into a single nonlinear feedback controller by an automatic choosing function. The state estimation is carried out using a nonlinear filter on the basis of the following formal linearization. This filter is synthesized by applying the extended Kalman filter technique to the augmented system that adds some polynomials of nonlinear measurement terms to the given nonlinear system. This is called an augmented automatic choosing control of a formal linearization filter type (AACCFLF). Simulation results are shown to demonstrate the effectiveness of the proposed method.

View full abstract

For an asymmetric lossless and reciprocal circuit we can obtain image parameters and iterative parameters. The characters between the image circuit and the iterative circuit are different. For example, we can make a symmetric circuit by using image parameters, and the pass band of image circuit is narrower than that of iterative circuit.

View full abstract