• 응용통계연구
• /
• 제34권3호
• /
• pp.507-521
• /
• 2021

고차원 자료에서 이상치를 탐지하기 위해서는 변수를 선별해야 할 필요성이 있다. 이상치 탐지에 적합한 정보가 종종 일부 변수에만 포함되어 있기 때문이다. 많은 수의 부적합한 변수가 자료에 포함될 경우 모든 관측치의 거리가 비슷해지는 집중효과가 발생하고 이로 인해 모든 관측치의 이상정도가 비슷해지는 문제가 발생하게 된다. 부분공간 이상치 탐지기법은 전체 변수 중 이상치 탐지에 적합한 변수들의 집합을 선별하여 관측치의 이상정도를 측정함으로써 이러한 문제를 극복한다. 본 논문은 대표적인 부분공간 이상치 탐지기법을 부분공간 선정 방식에 따라 세가지 유형으로 분류하고 각 유형에 속한 방법론을 부분공간 선정 기준과 이상 정도 측정 방식에 따라 요약한다. 더하여, 부분공간 이상치 탐지기법들을 적용할 수 있는 컴퓨팅 프로그램을 소개하고 집중효과에 대한 간단한 가상 실험과 자료 분석 결과를 제시한다.

• ETRI Journal
• /
• 제37권2호
• /
• pp.317-327
• /
• 2015

A reliable analysis of consumer preference from a large amount of purchase data acquired in real time and an accurate customer characterization technique are essential for successful direct marketing campaigns. In this study, an optimal segmentation of post office customers in Korea is performed using a subspace projection-based clustering method to generate an accurate customer characterization from a high-dimensional census dataset. Moreover, a traditional temporal mining method is extended to an algorithm using the MapReduce framework for a consumer preference analysis. The experimental results show that it is possible to use parallel mining through a MapReduce-based algorithm and that the execution time of the algorithm is faster than that of a traditional method.

• Smart Structures and Systems
• /
• 제7권2호
• /
• pp.153-167
• /
• 2011

The Subspace based System Identification Techniques (SSIT) have been very popular within the research circles in the last decade due to their proven superiority over the other existing system identification techniques. For operational (output only) modal analysis, the stochastic SSIT and for operational modal analysis in the presence of exogenous inputs, the combined deterministic stochastic SSIT have been used in the literature. This study compares the application of the two alternative techniques on a typical school building in Istanbul using 100 Monte Carlo simulations. The study clearly shows that the combined deterministic stochastic SSIT performs superior to the stochastic SSIT when the techniques are applied on noisy data from low to mid rise stiff structures.

• Structural Engineering and Mechanics
• /
• 제38권3호
• /
• pp.315-332
• /
• 2011

The Combined Deterministic Stochastic Subspace based System Identification Technique (CDSSSIT) is a powerful input-output system identification technique which is known to be always convergent and numerically stable. The technique determines a Kalman state sequence from the projection of the output-input data. The state space matrices are determied subsequently from this Kalman state sequence using least squares. The objective of this paper is to examine the efficiency of the CDSSSIT in identifying the modal parameters (frequencies and mode shapes) of a stiff structure. The results show that the CDSSSIT predicts the modal parameters of stiff buildings quite accurately but is very sensitive to the location of sensors.

• 한국공간구조학회논문집
• /
• 제3권4호
• /
• pp.53-58
• /
• 2003

In order to control seismic responses of building structures effectively and stably, it is very important to estimate the dynamic characteristics of target structure exactly based on input-output signal data. In this paper, it is shown that Subspace Identification Method is able to be applied effectively to system identification of building structures. To verify the efficiency of Subspace Identification Method, the vibration experiments were conducted on a specimen structure which is a 5-storied building structure model consisted of H-shaped steel beam, and the simulated seismic responses of the identified structure model were compared with the observed ones under the same excitation. It was observed that the experimental results coincided with the analyzed ones proposed in this paper.

• 산업기술연구
• /
• 제31권B호
• /
• pp.3-7
• /
• 2011

This paper addresses a computationally simple adaptive beamforming method to cancel interferences arriving onto a sensor array. In the proposed method, an estimate of the interference subspace is extracted from a submatrix of the sample covariance matrix and an orthonormal basis for the estimated subspace is efficiently found, one basis vector being updated every sample. Its computational burden is just $O(M{\eta})$ in an M-sensor array when ${\eta}$ directional signals are present. The new method does not make any premises of the geometrical structure of arrays, and can be applied to arbitrary arrays.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2009년도 정보 및 제어 심포지움 논문집
• /
• pp.408-410
• /
• 2009

We present an efficient and robust measurement model for visual tracking. This approach builds on and extends work on subspace representations of measurement model. Subspace-based tracking algorithms have been introduced to visual tracking literature for a decade and show considerable tracking performance due to its robustness in matching. However the measures used in their measurement models are often restricted to few approaches. We propose a novel measure of object matching using Angle In Feature Space, which aims to improve the discriminability of matching in subspace. Therefore, our tracking algorithm can distinguish target from similar background clutters which often cause erroneous drift by conventional Distance From Feature Space measure. Experiments demonstrate the effectiveness of the proposed tracking algorithm under severe cluttered background.

• 한국전자파학회논문지
• /
• 제15권9호
• /
• pp.863-871
• /
• 2004

본 논문에서는 정합 필터 출력을 이용하여 연산 복잡도가 감소된 DS/CDMA DMB 하향 링크 시스템에서의 부분 공간 채널 추정 기법을 제안한다. 연산 복잡도 감소 효과는 채널 벡터의 길이가 짧고 시스템의 부하가 적절할 때 매우 크게 나타난다. 이전에 제안된 부분 공간 채널 추정 알고리즘은 화산 이득이 큰 시스템에서 막대한 연산 복잡도 문제를 겪게 된다. 제안된 알고리즘에 약간의 성능 손실이 발생하지만 데이터 행렬의 길이가 길어지게 되면 그 영향은 미미해진다. 시뮬레이션과 분석적인 MSE 성능의 유도를 통해 성능을 평가한다.

• Structural Engineering and Mechanics
• /
• 제59권1호
• /
• pp.187-207
• /
• 2016

Practical ambient excitations of engineering structures usually do not comply with the stationary-white-noise assumption in traditional operational modal analysis methods due to heavy traffic, wind guests, and other disturbances. In order to eliminate spurious modes induced by non-white noise inputs, the improved stochastic subspace identification based on a delay index is proposed in this paper for a representative kind of stationary non-white noise ambient excitations, which have nonzero autocorrelation values near the vertical axis. It relaxes the stationary-white-noise assumption of inputs by avoiding corresponding unqualified elements in the Hankel matrix. Details of the improved stochastic subspace identification algorithms and determination of the delay index are discussed. Numerical simulations on a four-story frame and laboratory vibration experiments on a simply supported beam have demonstrated the accuracy and reliability of the proposed method in eliminating spurious modes under non-white noise ambient excitations.

• 대한수학회지
• /
• 제53권4호
• /
• pp.847-868
• /
• 2016

Let M be an invariant subspace of $H^2$ over the bidisk. Associated with M, we have the fringe operator $F^M_z$ on $M{\ominus}{\omega}M$. It is studied the Fredholmness of $F^M_z$ for (generalized) zero based invariant subspaces M. Also ker $F^M_z$ and ker $(F^M_z)^*$ are described.