• 한국통신학회논문지
• /
• 제24권9B호
• /
• pp.1785-1794
• /
• 1999

Affine projection 알고리즘은 RLS보다 적은 계산량으로 LMS보다 우수한 수렴성능을 나타낸다. 그러나 affine projection 알고리즘은 역행렬 연산을 필요로 하기 때문에 여전히 LMS에 비해 과중한 계산을 필요로 한다. 본 논문에서는 affine projection 알고리즘을 분석하여 이 알고리즘이 Gram-Scheme 구조로 해석될 수 있음을 보이고 이를 이용하여 NLMS와 비슷한 계산량으로 affine projection 알고리즘을 근사적으로 구현할 수 있는 새로운 알고리즘을 제안하였다. 제안한 방법은 NLMS와 비슷한 계산량을 가지면서 기존의 affine projection 알고리즘과 비슷한 수렴성능을 나타내었다.

• Journal of applied mathematics & informatics
• /
• 제30권1_2호
• /
• pp.93-100
• /
• 2012

In this paper, we present a strongly convergent parallel projection algorithm by introducing some parameter sequences for convex feasibility problem. To prove the strong convergence in a simple way, we transmit the parallel algorithm in the original space to an alternating one in a newly constructed product space. Thus, the strong convergence of the parallel projection algorithm is derived with the help of the alternating one under some parametric controlling conditions.

• 한국음향학회지
• /
• 제33권6호
• /
• pp.420-426
• /
• 2014

본 논문에서는 충격성 잡음 환경 하에서 인접 투사 알고리즘(Affine Projection Algorithm, APA)의 안정성을 높이기 위해 제안된 인접 투사 부호 알고리즘(Affine Projection Sign Algorithm, APSA)의 수렴 속도를 향상시키기 위한 새로운 APSA을 제안하였다. APSA은 충격성 잡음에 안정적으로 동작하고 역행렬 연산을 요구하지 않는다는 장점을 가지고 있다. 제안된 알고리즘은 기존 알고리즘의 장점을 가지면서도 기존의 알고리즘보다 더 빠른 수렴속도를 갖는다. 기존의 알고리즘은 i번째 과거 입력 신호를 필터 갱신을 위해 사용되는 모든 입력 신호의 $l_2$-norm으로 정규화 하지만 제안된 알고리즘은 i번째 과거 입력 신호를 i번째 과거 입력 신호의 $l_2$-norm 만으로 정규화한다. 시스템 식별 환경에서 두 알고리즘의 성능을 비교하는 컴퓨터 모의 실험을 수행하여 제안된 알고리즘의 수렴속도가 기존의 알고리즘보다 개선되었음을 보였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.2389-2392
• /
• 2005

We present fast stereo matching algorithm using edge projection. Traditional stereo matching algorithm uses 2D template for the search of corresponding point thus it requires huge the computational cost. In this paper, we reduce the 2D search problem into 1D using edge projection along vertical and horizontal direction inside the region of interest. Also, by accumulation of edge projection along vertical and horizontal direction, the edge projection within the region of interest could simply be obtained by just subtracting two values. Experimental results show that matching algorithm using edge projection also gives comparable discriminating power compared to that of using intensity.

• 대한수학회보
• /
• 제50권3호
• /
• pp.823-832
• /
• 2013

We present a new algorithm for solving a system of nonlinear equations with convex constraints which combines proximal point and projection methodologies. Compared with the existing projection methods for solving the problem, we use a different system of linear equations to obtain the proximal point; and moreover, at the step of getting next iterate, our projection way and projection region are also different. Based on the Armijo-type line search procedure, a new hyperplane is introduced. Using the separate property of hyperplane, the new algorithm is proved to be globally convergent under much weaker assumptions than monotone or more generally pseudomonotone. We study the convergence rate of the iterative sequence under very mild error bound conditions.

• 대한전자공학회논문지SP
• /
• 제40권5호
• /
• pp.410-416
• /
• 2003

인접투사 알고리즘(Affine Projection Algorithm: APA)에서 투사 차원은 적응필터 계수갱신에 사용하는 입력 벡터 요소 수이며, 이는 계수갱신을 위해 사용되는 투사 기저의 개수이다. 그리고 투사차원은 적응 필터의 수렴속도와 밀접한 관계가 있으며 알고리즘에서 필요한 계산량을 결정한다. 적응필터가 수렴함에 따라 수렴속도가 감소하게 되므로, 수렴속도를 결정하는 투사차원을 줄일 수 있다. 본 논문에서는 적응필터 계수갱신 시 계수벡터의 변화량을 수렴속도와 연관시켜 인접투사 알고리즘의 투사차원을 조정하는 가변 차원 인접투사 알고리즘(Variable Dimension APA: VDAPA)을 제안한다. 제안한 방법은 적응필터의 수렴정도에 따라 투사 기저 수를 조정함으로써 APA의 계산량을 줄인다. 컴퓨터 모의 실험을 통해 기존의 인접투사 알고리즘 및 NLMS 알고리즘과의 수렴속도, 계산량 면에서 성능을 비교하여 제안한 방법의 우수성을 확인한다.

• Journal of information and communication convergence engineering
• /
• 제8권1호
• /
• pp.23-28
• /
• 2010

According to increasing projection order, the AP algorithm bas noise amplification problem in large background noise. This phenomenon degrades the performances of the AP algorithm. In this paper, we analyze convergence characteristic of the AP algorithm and then suggest a noise robust modified AP algorithm for reducing this problem. The proposed algorithm normalizes the update equation to reduce noise amplification of AP algorithm, by adding the multiplication of error power and projection order to auto-covariance matrix of input signal. By computer simulation, we show the improved performance than conventional AP algorithm.

• ETRI Journal
• /
• 제37권6호
• /
• pp.1251-1258
• /
• 2015

We investigate an image recovery method for sparse-view computed tomography (CT) using an iterative shrinkage algorithm based on a second-order approach. The two-step iterative shrinkage-thresholding (TwIST) algorithm including a total variation regularization technique is elucidated to be more robust than other first-order methods; it enables a perfect restoration of an original image even if given only a few projection views of a parallel-beam geometry. We find that the incoherency of a projection system matrix in CT geometry sufficiently satisfies the exact reconstruction principle even when the matrix itself has a large condition number. Image reconstruction from fan-beam CT can be well carried out, but the retrieval performance is very low when compared to a parallel-beam geometry. This is considered to be due to the matrix complexity of the projection geometry. We also evaluate the image retrieval performance of the TwIST algorithm -sing measured projection data.

• 디지털산업정보학회논문지
• /
• 제8권2호
• /
• pp.49-58
• /
• 2012

This paper is on an improved orthogonal projection method which can cancel the acoustic feedback signals in the digital hearing aids. Comparing with the NLMS algorithm which is widely used for simplicity and stability, it shows that this method has the improvement of the convergence performances, and has small computational quantities, for signals with the large auto-correlation as speech signals. This uses the improved orthogonal projection algorithm which reduces the correlation of signals. To verify the convergence characteristics of the proposed algorithm, we simulated about various input signals. The acoustic feedback canceller has a 12-bit resolution with 64-tap adaptive FIR filter. And we compared the results of simulation for this algorithm with the ones for the NLMS algorithm. By these works, it is proved that the feedback canceller adopting the proposed algorithm shows about 3.5dB more high SNR than the NLMS algorithm in the colored input signals.

• 한국통신학회논문지
• /
• 제22권10호
• /
• pp.2339-2347
• /
• 1997

The performance of the adaptive noise canceller using LMS algorithm is degraded by the gradient noise due to target speech signals. An adaptive noise canceller with speech detector was proposed to reduce this performande degradation. The speech detector utilized the adaptive prediction-error filter adapted by the NLMS algorithm. This paper discusses to enhance the performance of the adaptive noise canceller forthecorlored noise. The affine projection algorithm, which is known as faster than NLMS algorithm for correlated signals, is used to adapt the adaptive filter and the adaptive prediction error filter. When the voice signals are detected by the speech detector, coefficients of adaptive filter are adapted by the sign-error afine projection algorithm which is modified to reduce the miaslignment of adaptive filter coefficients. Otherwirse, they are adapted by affine projection algorithm. To obtain better performance, the proper step size of sign-error affine projection algorithm is discussed. As resutls of computer simulation, it is shown that the performance of the proposed ANC is better than that of conventional one.