• 한국전자통신학회논문지
• /
• 제18권1호
• /
• pp.71-76
• /
• 2023

본 논문은 음향잡음감쇠기에서 신경망 구조의 Wiener 필터를 이용하여 비선형 잡음을 감쇠시키는 방법에 대하여 연구하였다. 이 시스템은 기존의 적응필터를 이용하는 대신 신경망 위너필터를 이용한 심층학습 알고리즘으로 비선형 잡음감쇠 성능을 개선한다. 128-neuron, 8-neuron 은닉층과 오차 역전파(back propagation) 알고리즘을 이용하여 비선형 잡음이 포함된 단일입력 음성신호로부터 음성을 추정한다. 본 연구에서 비선형 잡음에 대한 감쇠 성능을 검증하기 위하여 Keras 라이브러리를 사용한 시뮬레이션 프로그램을 작성하고 모의실험을 수행하였다. 모의실험 결과, 본 시스템은 비선형 잡음이 포함되어 있는 경우에도 위너필터 대신 FNN 필터를 사용하면 잡음감쇠 성능이 상당히 개선되는 것을 볼 수 있다. 이는 FNN 필터의 복잡한 구조가 어떤 형태의 비선형 특성도 잘 표현하기 때문이다.

• 제어로봇시스템학회논문지
• /
• 제21권1호
• /
• pp.34-39
• /
• 2015

In this paper, a finite impulse response (FIR) filter is proposed for discrete-time nonlinear systems. The proposed filter is designed by combining the estimate of the perturbation state and nominal state. The perturbation state is estimated by adapting the optimal time-varying FIR filter for the linearized perturbation model and the nominal state is directly obtained from the nonlinear nominal trajectory model. Since the FIR structured estimators use the finite horizon information on the most recent time interval, the proposed extended FIR filter satisfies the bounded input/bounded output (BIBO) stability, which can't be obtained from infinite impulse response (IIR) estimators. Thus, it can be expected that the proposed extended FIR filter is more robust than IIR structured estimators such as an extended Kalman filter for the round-of errors and the uncertainties from unknown initial states and uncertain system model parameters. The simulation results show that the proposed filter has better performance than the extended Kalman filter (EKF) in both robustness and fast convergency.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.1669-1674
• /
• 2004

In this paper, a nonlinear information filter (IF) for curvilinear motions in an interacting multiple model (IMM) algorithm to track a maneuvering vehicle on a road is investigated. Driving patterns of vehicles on a road are modeled as stochastic hybrid systems. In order to track the maneuvering vehicles, two kinematic models are derived: A constant velocity model for linear motions and a constant-speed turn model for curvilinear motions. For the constant-speed turn model, a nonlinear IF is used in place of the extended Kalman filter in nonlinear systems. The suggested algorithm reduces the root mean squares error for linear motions and rapidly detects possible turning motions.

• Journal of the Optical Society of Korea
• /
• 제20권2호
• /
• pp.228-233
• /
• 2016

In this paper, we propose a nonlinear maximum average correlation height (MACH) filter for information authentication of photon counting double random phase encryption (DRPE). To enhance the security of DRPE, photon counting imaging can be applied because of its sparseness. However, under severely photon-starved conditions, information authentication of DRPE may not be implemented successfully. To visualize the photon counting DRPE, a three-dimensional imaging technique such as integral imaging can be used. In addition, a nonlinear MACH filter can be utilized for helping the information authentication. Therefore, in this paper, we use integral imaging and nonlinear MACH filter to implement the information authentication of photon counting DRPE. To verify our method, we implement optical experiments and computer simulation.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2003년도 춘계 학술대회 학술발표 논문집
• /
• pp.303-306
• /
• 2003

In radar tracking, since the sensor measures range, azimuth and elevation angle of a target, the measurement equation is nonlinear and the extended Kalman filter (EKF) is applied to nonlinear estimation. The conventional EKF has been widely used as a nonlinear filter for radar tracking, but the considerably large measurement error due to the linearization of nonlinear function in highly nonlinear situations may deteriorate the performance of the EKF To solve this problem, a fuzzy-model-based Kalman filter (FMBKF) is proposed for radar tracking. The FMBKF uses a local model approximation based on a TS fuzzy model instead of a Jacobian matrix to linearize nonlinear measurement equation. The hybrid GA and RLS method is used to identify the premise and the consequent parameters and the rule numbers of this TS fuzzy model. In two-dimensional radar tracking problem, the proposed method is compared with the conventional EKF.

• 대한전기학회논문지:전기기기및에너지변환시스템부문B
• /
• 제50권7호
• /
• pp.347-355
• /
• 2001

The conventional output filter design methods of a voltage source inverter includes two main problems: the performance is not sure to satisfy the specification under the nonlinear load; the designed filter must be occasionally modified when the controller is designed. In this paper, we analyze the relation between both linear and nonlinear loads, the output voltage magnitude, LC lowpass filter parameters, and the control response time. Upon the basis, both filter and controller are simultaneously designed for the performance to satisfy the specification under the nonlinear load as well as linear load. The proposed method sharply enhances the reliability of the performance. P.U is used for the method to be applied to all the quantities of the system. The simulation and the experiment of the proposed method carried out respectively to verify the validity.

• 전기전자학회논문지
• /
• 제17권2호
• /
• pp.111-118
• /
• 2013

본 논문에서는 부대역 신호를 사용한 병렬 적응 비선형 Volterra 필터를 소개하고, 이 시스템의 성능을 분석하는 것을 주요 내용으로 한다. 연구 결과 제시한 부대역 신호를 사용한 적응 Volterra 필터는 수렴성이 우수함을 입력신호의 상관함수의 eigenvalue 분포를 사용하여 해석적으로 도출되었으며, 이러한 응용이 최근에 보고된 적응 반향 제거기에서 유용하게 사용될 수 있음을 이론적으로 밝혔다. 또한 각 부대역 에서의 최적필터를 이론적으로 유도하였고 컴퓨터 모의실험으로 이를 검증하였다.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2003년도 ISIS 2003
• /
• pp.311-314
• /
• 2003

In radar tracking, since the sensor measures range, azimuth and elevation angle of a target, the measurement equation is nonlinear and the extended Kalman filter (EKF) is applied to nonlinear estimation. The conventional EKF has been widely used as a nonlinear filter for radar tracking, but the considerably large measurement error due to the linearization of nonlinear function in highly nonlinear situations may deteriorate the performance of the EKF. To solve this problem, a fuzzy-model-based Kalman filter (FMBKF) is proposed for radar tracking. The FMBKP uses a local model approximation based on a TS fuzzy model instead of a Jacobian matrix to linearize nonlinear measurement equation. The hybrid GA and RLS method is used to identify the premise and the consequent parameters and the rule numbers of this TS fuzzy model. In two-dimensional radar tracking problem, the proposed method is compared with the conventional EKF.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.864-866
• /
• 1996

The attitude error is expressed using four kinds of quaternion errors. And the explicit relation equations between them are derived four kinds of nonlinear error models of SDINS using the their explicit relation are also proposed for a nonlinear filter which may be available for a system in the presence of a large attitude error the concept of the proposed nonlinear error model is applied to the velocity aided SDINS using a linear Kalman filter and an extended Kalman filter the simulation results reveal a improvement of performance using the nonlinear error model.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1995년도 하계학술대회 논문집 B
• /
• pp.937-940
• /
• 1995

The purpose of this paper is to present a nonlinear system identification method, where an adaptive nonlinear recursive state-spare(ANRSS) filter is employed as its filter structure, and a variable step (VS) algorithm is applied as its adaptation law. To demonstrate the validity of the proposed method, some simulation results are included.