• MALSORI
• /
• no.57
• /
• pp.153-164
• /
• 2006

The MMSE-STSA based speech enhancement algorithm is widely used as a preprocessing for noise robust speech recognition. It weighs the gain of each spectral bin of the noisy speech using the estimate of noise and signal power spectrum. In this paper, we investigate the influence of parameters used to estimate the speech signal and noise power in MMSE-STSA upon the recognition performance of noisy speech. For experiments, we use the Aurora2 DB which contains noisy speech with subway, babble, car, and exhibition noises. The HTK-based continuous HMM system is constructed for recognition experiments. Experimental results are presented and discussed with our findings.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.14 no.2
• /
• pp.26-30
• /
• 2000

In this paper, we obtain a mathematical model of a gas turbine control system from experimental data. The gas turbine in Gunsan power plant is selected as controlled system. The recursive least square algorithm is used to model the plant. For parameter estimation, plant is assumed as second order system and forgetting factor is 0.98 and the period of input and output signal period is 1sec. As a result, input and output characteristics of real system and modeling are identified.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06e
• /
• pp.165-168
• /
• 1998

본 논문에서는 충격성 잡음에 강인하기 위한 시변 주파수 추정 기법을 제안하였다 충격성 잡음에 강인하기 위해서는 충격성 잡음에 의한 추정 변수의 동요를 제한하고 추정된 오차가 향후 추정시 영향을 미치는 오차의 전파현상을 제한하여야 한다. 충격성 잡음에 의한 추정오차의 전파를 제한하기 위해서는 망각인자의 도입이 필요함을 증명하였고 보다 효과적으로 사용하기 위해서 가변 망각인자를 도입하였다. 가변 망각인자의 도입으로 충격성 잡음에 의한 오차의 전파를 선택적으로 제한할 수 있으며 충격성 잡음에 의한 추정계수의 변동은 영향함수 측면에서 Huber함수를 이용하여 제한하였다. 제안된 알고리듬은 Huber함수와 가변망각인자의 도입으로 충격성 잡음에 의해 생기는 오차의 크기와 오차의 영향이 전파되는 것을 적응적으로 제한하기 때문에 모의실험을 통해 기존의 칼만 알고리듬보다 나은 성능을 보임을 알 수 있었다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.04a
• /
• pp.495-498
• /
• 2004

In this paper, passive telemetry capacitive humidity sensor system using a RLSE(Recursive Least Square Estimation) technique Is proposed. To overcome the problem like power limits and complications that general passive telemetry sensor system including IC chip has, the principle of inductive coupling is applied to model the sensor system. Specially, by applying the forgetting factor, we show that the accuracy of its estimation can be improved even in the case of time varying parameter and also the convergence time can be reduced.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.98.6-98
• /
• 2001

In the double-talk situation where both the near-end and far-end signal present, the performance of echo cancellation using the conventional LMS algorithm is easily degraded because it freezes the adaptation in this situation. Recently CLMS and ECLMS algorithms were proposed to solve this problem. These algorithms could be used to adapt the filter´s parameters continuously even in the double-talk situation. In this paper, we propose new recursion formulas to calculate the ECLMS algorithm. And we compare and analyze the performances of double-talk echo canceller according to changing the value of channel tracking factors ${\alpha}$, ${\beta}$ and forgetting factor λ. The computer simulation was performed and the results showed that, ...

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.30B no.5
• /
• pp.52-61
• /
• 1993

This paper deals with adaptive control method of a robot manipulator with one-flexible link. ARMA model is used as a prediction and estimation model, and adaptive control scheme consists of parameter estimation part and adaptive controller. Parameter estimation part estimates ARMA model's coefficients by using recursive least-squares(RLS) algorithm and generates the predicted output. Variable forgetting factor (VFF) is introduced to achieve an efficient estimation, and adaptive controller consists of reference model, error dynamics model and minimum prediction error controller. An optimal input is obtained by minimizing input torque, it's successive input change and the error between the predicted output and the reference output.

• Proceedings of the Acoustical Society of Korea Conference
• /
• autumn
• /
• pp.361-364
• /
• 2001

지금까지 수중음향 시스템에서 jammer 신호를 제거하는 방법에 관한 많은 연구가 진행되어 왔다. 그러나, 기존의 빔형성 기법은 간섭 신호원(interference source)이 움직일 경우 그 성능이 현저히 떨어지는 문제점을 갖고 있다. 이러한 현상은 수중 음향 시스템이 간섭 신호원의 움직임에 대하여 즉각적으로 null의 위치를 변화시키지 못하기 때문에 발생하게 된다. 이를 해결하기 위해서는 시간에 따라 위치가 변하는 jammer 환경에 대하여 대응할 수 있는 새로운 알고리즘이 필요하게 된다. 이러한 단점을 보완하기 위해 본 논문에서는 가변 망각인자를 갖는 적응 빔형성 기법을 제안하고, 컴퓨터 모의실험을 통하여 제안된 알고리즘이 기존의 적응 빔형성 기법에 비하여 출력 SINR(signal to interference plus noise ratio)의 측면에서 성능 향상을 가짐을 보였다.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.60 no.1
• /
• pp.150-155
• /
• 2011

In an active noise control systems, an IIR filter may cause a problem in stability beacause of its poles. For IIR filter, its poles goes sometimes out of a unit circle in a z-plane in the transition state, where the adaptive algorithm converges to the optimum value, which causes the system to diverge. Fuzzy LMS algorithm has a better convergence property than conventional LMS algorithms, but is not applicable to IIR filter because of the reasons. Stabilized adaptive algorithm could be improves stability by moving the pole of IIR filer toward the origin forcibly in the transient state, and by introducing forgetting factor to maintain the optimum convergence when it reaches to the steady state. In this paper, We proposed stabilized adaptive fuzzy LMS algorithms with IIR filter structures, for single channel active noise control with ill conditioned signal case. Computer simulations were performed to show the effectiveness of a proposed algorithm.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.20 no.6
• /
• pp.534-539
• /
• 2015

The DC overhead line voltage of an electric railway substation swings depending on the accelerating and regenerative-braking energy of trains, and it deteriorates the energy quality of the electric facility in the DC railway substation and restricts the powering and braking performance of subway trains. Recently, an energy storage system or a regenerative inverter has been introduced into railway traction substations to diminish both the variance of the overhead line voltage and the peak power consumption. In this study, the variance of the overhead line voltage in a DC railway substation is modelled by RC parallel circuits in each feeder, and the RC parameters are estimated using the recursive least mean square (RLMS) scheme. The forgetting factor values for the RLMS are selected using simulated annealing optimization, and the modelling scheme of the overhead line voltage variation is evaluated through raw data measured in a downtown railway substation.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.237-240
• /
• 2002

지면에 가까이 있는 비행체나 수면에 가까이 있는 수중 운동체에서 도래하는 신호는 직접파와 반사파로 구성되어 있다. 그런 환경에서는 반사파가 직접파와 coherent한 관계를 갖는다. 이런 현상으로 해서 DOA추정을 위해 추정한 correlation 행렬이 singular한 성질을 갖게 된다. 이렇게 singular한 행렬은 DOA추정에 큰 오류를 내게 된다. 또 움직이는 대상의 경우 추정 correlation 행렬의 성질이 시간에 따라 변하게 된다. 본 논문에서는 위 두 상황을 함께 해결하기 위해서 PASTd알고리즘을 변형하여 Forward/Backward Variable forgetting factor를 도입한 PASTd알고리즘을 제안한다.