Takahiro Murakami

Controller design approach based on linear programming.

This study explains and demonstrates the design method for a control system with a load disturbance observer. Observer gains are determined by linear programming (LP) in terms of the Routh-Hurwitz stability criterion and the final-value theorem. In addition, the control model has a feedback structure, and feedback gains are determined to be the linear quadratic regulator. The simulation results confirmed that compared with the conventional method, the output estimated by our proposed method converges to a reference input faster when a load disturbance is added to a control system. In addition, we also confirmed the effectiveness of the proposed method by performing an experiment with a DC motor.

Speech enhancement based on a combined higher frequency regeneration technique and RBF networks

In this paper, we propose a new method for noise reduction in speech signals. In the method, the nonlinear spectral subtraction (NSS) method is applied to the down-sampled noisy speech signal. Then, the proposed method restores the missing higher frequency components from the narrow-bandwidth enhanced speech signal in order to improve the subjective speech quality. For recovering the broad-bandwidth speech signal, the proposed method exploits the radial basis function (RBF) network, which is one of the wellknown models of artificial neural networks.

Improved internal model control based on optimal control for servo system with dead time

This paper describes an improved internal model control based on PID control for servo system with dead time. The internal model control works for more stability of disturbance. The proposed system uses the feedback based on internal model control and PID control when the disturbance occurs. The proposed method can be applied to not only servo stable plant with dead time but also unstable plant with dead time. It can be applicable to both simulation and actual DC motor control. The proposed method has good robustness compared with general PID design method.

Design of internal model control based on an optimal control for a servo system

This paper describes a design of internal model control based on an optimal control for a servo system. The control system has the feedback based on the proposed disturbance compensator in the disturbance response. The compensator is designed to become the denominator of the transfer function without a dead time in the disturbance responses. The disturbance response of the proposed method is faster than that of the previous method.

A design of a robust discrete-time controller.

In this paper, we proposed a robust discrete-time controller. This control system, which is derived from the idea of the normalized plant, does not include plant parameters. Thus, we obtain a control system independent of plant parameters and that has the same structure as a conventional optimal servo control system. Simulation and experimental results show that the proposed method is fairly robust to plant parameter variations and external disturbances.

Speech Enhancement by Spectral Subtraction Based on Subspace Decomposition

This paper presents a novel algorithm for spectral subtraction (SS). The method is derived from a relation between the spectrum obtained by the discrete Fourier transform (DFT) and that by a subspace decomposition method. By using the relation, it is shown that a noise reduction algorithm based on subspace decomposition is led to an SS method in which noise components in an observed signal are eliminated by subtracting variance of noise process in the frequency domain. Moreover, it is shown that the method can significantly reduce computational complexity in comparison with the method based on the standard subspace decomposition. In a similar manner to the conventional SS methods, our method also exploits the variance of noise process estimated from a preceding segment where speech is absent, whereas the noise is present. In order to more reliably detect such non-speech segments, a novel robust voice activity detector (VAD) is then proposed. The VAD utilizes the spread of eigenvalues of an autocorrelation matrix corresponding to the observed signal. Simulation results show that the proposed method yields an improved enhancement quality in comparison with the conventional SS based schemes.

FIR system identification based on a nonparametric Bayesian model using the Indian buffet process

In this paper, we propose a nonparametric Bayesian model combined with the Indian buffet process (IBP) for a finite impulse response (FIR) system. We develop an FIR system identification method that can simultaneously estimate the number of FIR taps and coefficients. In the proposed model, each FIR tap consists of a coefficient and a gain, and the gain is a binary value. An infinite-dimensional binary vector is composed of binary values, and we assume that this binary vector is generated by the IBP. To identify the FIR system, we specify the likelihood function and prior distributions of the parameters and derive their posterior distributions. We can simultaneously estimate the number of FIR taps and coefficients by sampling from posterior distributions using the Gibbs sampler. Our simulations demonstrate that although the number of FIR taps is unknown, the identification performance of the proposed method in a high signal-to-noise ratio environment is similar to or better than that of the conventional least square solution.

An approach to model-following controller design based on a stabilized digital inverse system

This article explains our approach to designing a model-following controller based on a stabilized digital inverse system. We use an inter-sample auxiliary observation output from the model as an input to the inverse system. We then design the inverse system for minimum and non-minimum phase plants. In order to reposition the unstable poles of the inverse system on a stable region, we determine the optimal gain vector using a linear quadratic regulator. We confirm the effectiveness of our proposed method by performing simulations for several plants and conducting an experiment with a direct current motor.

Blind speech extraction using subband independent component analysis with scale adjustment function and neural memory

This paper presents a new method for speech extraction by a combined subband independent component analysis (subband ICA) and neural memory. In the method, subband ICA separates the signals obtained from microphones into the signal components of interest and the rest at each subband. Subband ICA approach has two fundamental problems. The neural memory represented by probabilistic neural networks (PNNs) is then used for identifying the signal components of interest among the separated components and thereby solved the permutation problem. Then we adjust the scale of each output signal to that of the corresponding input signal. Simulation results for both the instantaneous and delayed mixture cases show that the proposed subband ICA approach consistently yields the performance improvement in comparison with the conventional fullband/subband approaches.

Time-scale and pitch-scale modification by the phase vocoder without occurring the phase unwrapping problem

A phase vocoder is one of the well-known methods for time-scale and pitch-scale modification. This method performs time-scale and pitch-scale modification by increasing or decreasing an estimated instantaneous angular frequency. However, due to the phase unwrapping problem, which is a well-known problem in the phase vocoder, the phase vocoder distorts the waveform of modified signal. The phase unwrapping problem occurs when the phase vocoder estimates the instantaneous angular frequency. In this paper, we propose a phase vocoder without occurring the phase unwrapping problem. The proposed method uses the difference of phases between adjacent analysis frames directly for the phase modification. Because this method does not use the estimated instantaneous angular frequency, the phase unwrapping problem can be avoided. We verified the effectiveness of the proposed method by simulations of time-scale and pitch-scale modification.