• 대한전자공학회논문지TC
• /
• 제40권1호
• /
• pp.15-23
• /
• 2003

본 논문에서는 입사 신호의 근사 공분산 행렬을 이용하여 신호의 입사각을 빠르게 추정하는 입사각 추정 알고리듬을 제안한다. MUSIC(MUltiple Signal Classification) 알고리듬과 같은 기존의 부분공간 입사각 추정 알고리듬은 입력 공분산 행렬을 구하기 위해서 다수의 표본 신호를 필요로 하며, 입력 공분산 행렬을 획득하기 위한 표본 신호의 수신시간 동안 입사각 추정이 수행될 수 없으므로 빠른 신호처리가 불가능하다. 또한 코히어런트 신호가 입사하는 경우에 코히어런트 신호간의 간섭으로 신호의 입사각을 정확하게 추정할 수 없다. 제안한 입사각 추정 알고리듬은 빔 형성기를 이용하여 매 표본 신호의 공간적인 빔 형성을 먼저 수행하여 신호간의 간섭을 제거한 후에 센서의 출력 값을 이용하여 방위각 응답(bearing response)과 방향 스펙트럼(directional spectrum)을 구한다. 방위각 응답으로 대략적인 신호의 입사각을 추정한 후에 방향 스펙트럼을 이용하여 정착하게 신호의 입사각을 추정한다. 제안 입사각 추정 알고리듬은 공분산 행렬을 구하기 위하여 그 순간의 각 어레이 소자에 입사되는 표본 신호만을 사용하고 방위각 응답을 구하기 위해서 몇 순간 동안의 표본 신호만 필요로 하므로 기존 입사각 추정 알고리듬에 비하여 크게 향상된 입사각 추정 속도를 갖는다.

• 한국식품영양학회지
• /
• 제23권3호
• /
• pp.419-427
• /
• 2010

The focus of the current study was to investigate the physicochemical properties of a corn starch-sponge matrix prepared at a low concentration below gel forming by freeze-drying. The effect of variables(starch concentration, heating temperature, and heating hold time) on the physicochemical properties of the samples was analyzed by response-surface methodology. Regression models on the properties of samples such as hardness, springiness, and water solubility index(WSI) showed high correlation coefficients(r>0.95) and significant F values, but regression models for the other properties(swelling power, apparent viscosity, reducing sugar content, and digestibility) showed them to have relatively low significance. Sample hardness of sample showed the highest value at condition of $90^{\circ}C$ and 5%, whereas springiness was at a maximum at $130^{\circ}C$ and 5%. Also, at 1% of starch concentration, mechanical properties were greatly decreased as the relative humidity increased, compared with the 3% and 5%, especially in the hardness of samples. The WSI showed an increasing trend with heating temperature regardless of starch concentration. Overall, the physicochemical properties of freeze-dried corn starch-sponge matrix were influenced much more by starch concentration and heating temperature than by heating hold time. The results of this study show that the basic properties of freeze-dried corn starch-sponge matrix can be used for the specific food applications or as a functional material for its stability.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권9호
• /
• pp.2991-3007
• /
• 2022

Two dimensional locality preserving projections (2D-LPP) is an improved algorithm of 2D image to solve the small sample size (SSS) problems which locality preserving projections (LPP) meets. It's able to find the low dimension manifold mapping that not only preserves local information but also detects manifold embedded in original data spaces. However, 2D-LPP is simple and elegant. So, inspired by the comparison experiments between two dimensional linear discriminant analysis (2D-LDA) and linear discriminant analysis (LDA) which indicated that matrix based methods don't always perform better even when training samples are limited, we surmise 2D-LPP may meet the same limitation as 2D-LDA and propose a novel matrix exponential method to enhance the performance of 2D-LPP. 2D-MELPP is equivalent to employing distance diffusion mapping to transform original images into a new space, and margins between labels are broadened, which is beneficial for solving classification problems. Nonetheless, the computational time complexity of 2D-MELPP is extremely high. In this paper, we replace some of matrix multiplications with multiple multiplications to save the memory cost and provide an efficient way for solving 2D-MELPP. We test it on public databases: random 3D data set, ORL, AR face database and Polyu Palmprint database and compare it with other 2D methods like 2D-LDA, 2D-LPP and 1D methods like LPP and exponential locality preserving projections (ELPP), finding it outperforms than others in recognition accuracy. We also compare different dimensions of projection vector and record the cost time on the ORL, AR face database and Polyu Palmprint database. The experiment results above proves that our advanced algorithm has a better performance on 3 independent public databases.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2010년도 하계학술대회
• /
• pp.15-17
• /
• 2010

적응 어레이(adaptive array) 기법은 이동통신 시스템에서 발생하는 동일채널 간섭신호를 제거하는 데에 효과적인 기술이다. 적응 어레이 기법 중 가장 널리 이용되는 SMI 기법은 높은 성능을 보이는 데 반해 상대적으로 많은 계산량을 필요로 한다. 본 논문에서는 역행렬의 크기를 줄임으로써 SMI 기법의 복잡도를 줄이는 방법을 제안한다. 제안하는 기법은 안테나 어레이 중 받은 신호의 파워가 가장 작은 안테나를 사용하지 않음으로써 안테나 어레이의 개수를 하나 줄이는 방식을 사용한다. 시뮬레이션 결과는 제안하는 방법의 복잡도가 낮음에도 불구하고 우수한 성능을 갖는다는 것을 입증한다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.109.3-109
• /
• 2001

The so-called curse of dimensionality arises when Gaussian mixture is used on high-dimensional small-sample-size data, since the number of free elements that needs to be specied in each covariance matrix of Gaussian mixture increases exponentially with the number of dimension d. In this paper, by constraining the covariance matrix in its decomposed orthonormal form we get a local PCA model so as to reduce the number of free elements needed to be specified. Moreover, to cope with the small sample size problem, we adopt BYY data smoothing learning which is a regularization over maximum likelihood learning obtained from BYY harmony learning to implement this local PCA model.

• 통합자연과학논문집
• /
• 제10권1호
• /
• pp.33-39
• /
• 2017

Consider the problem of estimating a $p{\times}1$ mean vector ${\theta}(p-q{\geq}3)$, $q=rank(P_V)$ with a projection matrix $P_v$ under the quadratic loss, based on a sample $X_1$, $X_2$, ${\cdots}$, $X_n$. We find a James-Stein type decision rule which shrinks towards projection vector when the underlying distribution is that of a variance mixture of normals and when the norm ${\parallel}{\theta}-P_V{\theta}{\parallel}$ is restricted to a known interval, where $P_V$ is an idempotent and projection matrix and rank $(P_V)=q$. In this case, we characterize a minimal complete class within the class of James-Stein type decision rules. We also characterize the subclass of James-Stein type decision rules that dominate the sample mean.

• 대한화학회지
• /
• 제19권3호
• /
• pp.163-168
• /
• 1975

쌀, 생선 및 표준 기준 배추시료 속에 함유된 수은의 함량을 두 개의 독립적인 분석 방법 즉, 과거 본 연구실에서 개발한 방법과 Sjostrand 방법에 의하여 분석하였다. 분석 결과는 시료의 종류에 따라서 서로 다른 매트릭스 효과를 보였다. 예를 들면 배추과 생선 시료는 같은 매티릭스 효과를 나타내었으나 쌀 시료는 이 두 종의 시료과 비교하여 다른 효과를 보였다.

• 전기학회논문지
• /
• 제62권3호
• /
• pp.380-386
• /
• 2013

This paper represents an adaptive beamforming technique to suppress interference or jamming signals in wideband. In order to maintain low side lobe level(SLL) at an antenna element level, Taylor-weighting was used. Also, to make a nulling beam pattern toward jammer's directions in wideband, we used the modified Loaded Sample Matrix Inversion(LSMI) algorithm and Tapped Delay Line(TDL). To verify the proposed algorithm, we applied it to a rectangular array antenna. Finally, the results show beam pattern with low SLL and jammers suppression.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권11호
• /
• pp.5382-5403
• /
• 2018

In this paper, for a spectrum sensing purpose, we heuristically established a test statistic (TS) from a sample covariance matrix (SCM) for multiple antennas based cognitive radio. The TS is formulated as a scaled-energy which is calculated as a sum of scaled diagonal entries of a SCM; each of the diagonal entries of a SCM scaled by corresponding row's Euclidean norm. On the top of that, by combining theoretical results together with simulation observations, we have approximated a decision threshold of the TS which does not need prior knowledge of noise power and primary user signal. Furthermore, simulation results - which are obtained in a fading environment and in a spatially correlating channel model - show that the proposed method stands effect of noise power mismatch (non-uniform noise power) and has significant performance improvement compared with state-of-the-art test statistics.

• 정보보호학회논문지
• /
• 제10권3호
• /
• pp.37-48
• /
• 2000

서울Visual cryptography scheme is a simple method in which can be directly decoded the secret information in human visual system without performing any cryptographic computations. This scheme is a kind of secret sharing scheme in which if a secret of image type is scattered to n random images(slides) and any threshold (or more) of them are stacked together the original image will become visible. In this paper we consider (2, n) visual cryptography scheme and propose a new construction method in which the number of expanded pixels can be reduced by using the sample matrix. The proposed scheme can futhermore distribute the multiple secret image to each group according to the difference of relative contrast.