• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06c
• /
• pp.209-212
• /
• 1998

소나나 레이다분야에서 목표물로부터 오는 신호의 산란점을 추정하는 것은 추정한 산란점 특성을 분석하여 목표물을 식별하기 위해서 연구되어 오고 있다. 지금까지 연구되어온 모델링에 의한 산란점 추정을 보면 많은 경우 Prony 모델에 근거하여 추정하고 있다. Prony 모델을 레이다나 소나에 적용할 때에 몇 가지 단점을 갖고 있다. 그 첫째는 Prony모델이 잡음에 약하다는 점이고 둘째는 Prony모델이 모델차수에 대한 사전정보를 요구한다는 점이다. 본 논문에서는 위에 든 단점중에서 잡음에 취약한 점을 보완하기 위해서 입력 신호 성분만을 자동 추적하는 창함수(Automatic Tracking Window)를 전처리기로 사용한 Prony 산란점 추정 방법을 제안한다. 또 그 성능을 기존 Prony방법만을 사용한 산란점 추정방법과 비교하여 잡음에 대한 성능 향상을 보였다.

• Journal of Biomedical Engineering Research
• /
• v.22 no.2
• /
• pp.187-195
• /
• 2001

시각자극 유발전위 추론시에 사용되는 평균법은 많은 회수의 단일-시도 유발전위(single-trial evoked potential)의 기록이 필요하다. 본 논문에서는 이러한 평균법의 후처리기로서 Prony 방법을 적용하여 향상된 신호-대-잡음비를 제공하는 방법을 제안하였다. Prony 방법은 유발전위와 같은 잡음이 섞여 있는 신호에서 지수함수를 추론하는데 많이 적용되어져 왔다. 여러 가지 잡음 모델을 이용한 시뮬레이션과 실제 시각자극 유발전위 데이터를 이용한 실험을 통해 Prony 방법이 평균법의 후처리기(post-processor) 기능으로서 사용될 때, 신호의 질적인 향상 측면 뿐 아니라 평균법 사용시 단일-시도 유발전위의 기록 횟수의 단축이라는 결과도 가져옴을 보였다.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.54 no.6
• /
• pp.293-305
• /
• 2005

This paper describes an improved Prony method in its speed, accuracy and reliability by efficiently determining the optimal sampling interval with use of DFT (discrete Fourier transformation). In the Prony method the computation time is dominated by the size of the linear prediction matrix, which is given by the number of data times the modeling order The size of the matrix in a general Prony method becomes large because of large number of data and so does the computation time. It is found that the Prony method produces satisfactory results when SNR is greater than three. The maximum sampling interval resulting minimum computation time is determined using the fact that the spectrum in DFT is inversely proportional to sampling interval. Also the process of computing the modes is made efficient by applying Hessenberg method to the companion matrix with complex shift and computing selectively only the dominant modes of interest. The proposed method is tested against the 2003 KEPCO system and found to be efficient and reliable. The proposed method may play a key role in monitoring in real time low frequency oscillations of power systems .

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.20 no.9
• /
• pp.863-875
• /
• 2010

The vibration of a cylindrical shell is generated due to elastic waves propagating on the shell. Those elastic waves include propagating waves such as flexural, longitudinal and shear waves. Those also include non-propagating decaying waves, i.e. evanescent waves. In order to separate contributions of each type of waves to the data for the vibration of the cylindrical shell, spatial signal processing techniques for wavenumber analysis are investigated in this paper. Those techniques include Fast Fourier transform(FFT) algorithm, Extended Prony method and Overdetermined Modified Extended Prony method(OMEP). Those techniques have been applied to identify the waves from simulated vibration signals with various signal-to-noise ratios. Futhermore, the experimental data for in-plane vibration of the cylindrical shell has been processed with those techniques to identify propagating waves(longitudinal, shear and flexural waves) and evanescent waves.

• Journal of Biomedical Engineering Research
• /
• v.10 no.2
• /
• pp.131-138
• /
• 1989

In this paper the feature of background EEG is extracted by LS Prony Method for the analysis of background EEG in the time domain. Autocorrelation leg estimates are not required with the LS Prony method. The Prony method is required any the solution of two serfs of simultaneous linear equation and a polynominal rooting. That the optimal order of this model is the 6-th order is determined by using Akaike' s Information Criterial test. From the experimential results the alpha band amplitude is the largest among alpha band beta band theta band delta band and beta band amplitude is larger than that of the delta band and theta band. The sustained time for the alph a band, the beta band, the delta band and the theta band is 2, 3461 (sec), 0.6490(sec), 0.3120(sec), 0.7046(sec) respectively. Consequenty the alpha band is maintained in the whole subjects, the beta band, the delta band, the theta band are existed intermittently in each subjects.

• The Journal of the Acoustical Society of Korea
• /
• v.15 no.1E
• /
• pp.65-69
• /
• 1996

This paper presents an algorithm for multipath channel parameter estimation by an improved Prony method. This algorithm applies a modified regularized spectral estimation to the conventional SVD Prony method. This method requires no a priori information on the number of multipath. The performance of the proposed algorithm is almost the same as that of the SVD based multipath channel parameter estimation algorithm.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.43 no.4
• /
• pp.687-690
• /
• 1994

This paper presents a method of obtaining transmission network equivalents from the network's response to the pulse excitation signal. Proposed method is base on Prony signal analysis and jtransfer function identification technique. As a result Thevenin-type of discrete-time filter model can be generated. It can reproduce the driving point impedance characteristic of the network.

• Journal of Biomedical Engineering Research
• /
• v.11 no.2
• /
• pp.289-294
• /
• 1990

In this paper, for acoustic source reconstruction using angular frequency propagation method, the extended Prony method is propsed which is useful due to not having the inherent property of DFT. The simulation is carried out and its improved results are shown explicitly by comparing with DFT case.

• Journal of Electrical Engineering and Technology
• /
• v.6 no.2
• /
• pp.154-160
• /
• 2011

A new algorithm for estimating the fundamental frequency of power system signals is presented. The proposed algorithm consists of two stages: orthogonal decomposition and a complex Prony analysis. First, the input signal is decomposed into two orthogonal components using cosine and sine filters, and a variable window is adapted to enhance the performance of eliminating harmonics. Then a complex Prony analysis that is proposed in this paper is used to estimate the fundamental frequency by approximating the cosine-filtered and sine-filtered signals simultaneously. To evaluate the performance of the algorithm, amplitude modulation and harmonic tests were performed using simulated test signals. The performance of the algorithm was also assessed for dynamic conditions on a single-machine power system. The Electromagnetic Transients Program was used to generate voltage signals for a load increase and single phase-to-ground faults. The performance evaluation showed that the proposed algorithm accurately estimated the fundamental frequency of power system signals in the presence of amplitude modulation and harmonics.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.54 no.3
• /
• pp.122-132
• /
• 2005

This paper presents a novel approach based on Prony method to analysis of small signal stability in power system. Prony method is a valuable tool in identifying transfer function and estimating the modal parameter of power system oscillation from measured or computed discrete time signal. This paper define the relative residue of time signal and propose the condition to select low frequency oscillation in each generator. This paper describes the application results of proposed algorithm with respect to KEPCO systems. Simulation results show that the proposed algorithm can be used as another tools of power systems analysis.