• ETRI Journal
• /
• 제35권5호
• /
• pp.931-934
• /
• 2013

The weight calculation in an iterative algorithm is the most computationally costly task in computed tomography image reconstruction. In this letter, a fast algorithm to speed up the weight calculation is proposed. The classic square pixel rotation approximate calculation method for computing the weights in the iterative algorithm is first analyzed and then improved by replacing the square pixel model with a circular pixel model and the square rotation approximation with a segmentation method of a circular area. Software simulation and hardware implementation results show that our proposed scheme can not only improve the definition of the reconstructed image but also accelerate the reconstruction.

• 대한전자공학회논문지TC
• /
• 제45권12호
• /
• pp.105-113
• /
• 2008

유비쿼터스 사회에서는 사용자의 요구를 충족시키기 위하여 사용자가 갖고 있는 기기에 대한 정밀한 위치측정을 필요로 한다. 위치 측정은 송수신기간에 신호의 전송을 기반으로 한 거리측정을 통해 이뤄지기 때문에 위치측정의 오차는 거리측정의 오차로부터 발생한다. 신호가 전송되는 기기 간에 장애물이 존재하게 되면 LoS(Line of Sight)신호 성분이 줄어들게 되어 NLoS(Non-Line of Sight) 채널이 발생하게 되고 정확한 시점에서 신호를 검출할 수 없게 되어 거리오차가 발생하게 된다. 일반적인 위치측정 알고리즘은 참조기기(Reference Device)의 거리측정 성능에 관계없이 참조기기와 목표기기(Target Device)간의 거리측정 값을 위치 계산에 그대로 사용하기 때문에 거리측정 값으로부터 발생되는 오차가 위치 계산에 더해지게 된다. 따라서 본 논문에서는 각 참조기기가 속해 있는 채널특성을 판별하고 NLoS채널로부터 계산된 거리와 LoS채널로부터 계산된 거리를 다른 비율로 적용하여 위치측정의 오차를 줄이는 Iterative Calculation 기법을 제안한다. 참조기기는 수신된 신호의 Kurtosis, Mean, Excess Delay, RMS Delay spread를 통해 NLoS와 LoS 채널을 구분한다. 이를 통해 구분된 채널마다 각기 다른 비율로 랜덤 거리를 계산된 거리에 더하여 위치를 계산하는 것을 반복적으로 수행한 뒤 평균값을 계산하여 확률적으로 존재할 가능성이 높은 목표기기의 위치를 찾아감으로써 NLoS채널로부터 계산된 거리오차가 위치측정에 미치는 영향을 줄이는 방법을 제안하고 시뮬레이션을 통해 기존의 방식과 비교했을 때 성능향상을 확인하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권5호
• /
• pp.1649-1665
• /
• 2021

In view of the low accuracy of the traditional FunkSVD algorithm, and in order to improve the computational efficiency of the algorithm, this paper proposes a parallel algorithm of improved FunkSVD based on Spark (SP-FD). Using RMSProp algorithm to improve the traditional FunkSVD algorithm. The improved FunkSVD algorithm can not only solve the problem of decreased accuracy caused by iterative oscillations but also alleviate the impact of data sparseness on the accuracy of the algorithm, thereby achieving the effect of improving the accuracy of the algorithm. And using the Spark big data computing framework to realize the parallelization of the improved algorithm, to use RDD for iterative calculation, and to store calculation data in the iterative process in distributed memory to speed up the iteration. The Cartesian product operation in the improved FunkSVD algorithm is divided into blocks to realize parallel calculation, thereby improving the calculation speed of the algorithm. Experiments on three standard data sets in terms of accuracy, execution time, and speedup show that the SP-FD algorithm not only improves the recommendation accuracy, shortens the calculation interval compared to the traditional FunkSVD and several other algorithms but also shows good parallel performance in a cluster environment with multiple nodes. The analysis of experimental results shows that the SP-FD algorithm improves the accuracy and parallel computing capability of the algorithm, which is better than the traditional FunkSVD algorithm.

• 대한전기학회논문지:전력기술부문A
• /
• 제51권3호
• /
• pp.127-133
• /
• 2002

This paper presents an efficient improvement of the iterative eigenvalue calculation method of the AESOPS algorithm. The intuitively and heuristically approximated iterative eigenvalue calculation method of the AESOPS algorithm is transformed to the Second Order Newton Raphson Method which is generally used in numerical analysis. The equations of second order partial differentiation of external torque, terminal and internal voltages are derived from the original AESOPS algorithm. Therefore only a few calculation steps are added to transform the intuitively and heuristically approximated AESOPS algorithm to the Second Order Newton Raphson Method, while the merits of original algorithm are still preserved.

• 대한조선학회지
• /
• 제14권3호
• /
• pp.1-4
• /
• 1977

The computer aided ship design and construction has become very popular one in a ship yard recently. For one of such a purpose a program is developed with Aitken's iterative interpolation method. From the sample calculation we can conclude that the program has a reliable acquracy for the calculation of hydrostatic data or loading manual. And also the program can be applicable to a ship construction by careful selecting of input data.

• 대한전기학회논문지:전력기술부문A
• /
• 제48권11호
• /
• pp.1394-1400
• /
• 1999

This paper presents and efficient improvement of the iterative eigenvalue calculation method and the selection of initial values in AESOPS algorithm. To determine the initial eigenvalues of the system, system state matrix is constructed with the two-axis generator model. From the submatrices including synchronous and damping coefficients, the initial eigenvalues are calculated by the QR method. Participation factors are also calculated from the above submatrices in order to determine the generators which have a important effect to the specific oscillation mode. Also, the heuristically approximated eigenvalue calculation method in the AESOPS algorithm is transformed to the Newton Raphson Method which is largely used in the nonlinear numerical analysis. The new methods are developed from the AESOPS algorithm and thus only a few calculation steps are added to practice the proposed algorithm.

• 대한기계학회논문집A
• /
• 제25권8호
• /
• pp.1206-1212
• /
• 2001

The kinematic analysis of Stewart platform manipulator(SPM) is carried out in order to reduce the calculation time for its forward kinematic solution when the iterative numerical method is employed. The kinematic equations for three substructures of the 6-3 SPM are newly derived by introducing Denavit-Hartenberg link parameters and using kinematic constraints associated with the SPM and substructure kinematics. It is shown that the forward kinematics can be easily solved from three nonlinear equations with three unknown variables only, leading to a great reduction in calculation time.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 하계학술대회 논문집 C
• /
• pp.1506-1509
• /
• 1999

This paper presents a new economic dispatch algorithm to improve the unit commitment solution while guaranteeing the near optimal solution without reducing calculation speed. The conventional economic dispatch algorithms have the problem that it is not applicable to the unit commitment formulation due to the frequent on/off state changes of units during the unit commitment calculation. Therefore, piecewise linear iterative method have generally been used for economic dispatch algorithm for unit commitment. In that method, the approximation of the generator cost function makes it hard to obtain the optimal economic dispatch solution. In this case, the solution can be improved by introducing a inverse of the incremental cost function. The proposed method is tested with sample system. The results are compared with the conventional piecewise linear iterative method. It is shown that the proposed algorithm yields more accurate and economical solution without calculation speed reduction.

• 한국전자파학회논문지
• /
• 제26권11호
• /
• pp.1012-1019
• /
• 2015

본 논문은 공동구조의 RCS(Radar Cross Section)을 계산하는 반복적 물리 광학법(Iterative Physical Optics: IPO)의 연산속도를 가속하는 기법들을 효과적으로 적용하는 방법을 제시한다. IPO는 기존에 공동 구조 내부에서 발생하는 다중 반사 효과 계산 시 기하 광학법(Geometric Optics: GO)를 사용하는 SBR(Shooting and Bouncing Rays)과는 달리 근거리 필드 식을 활용하기 때문에 정확도가 향상된 산란 계산이 가능하다. 하지만 PO(Physical Optics)에 비해 크게 느리며, 실질적인 사용을 위해서는 계산속도의 향상을 위한 기법이 필요하다. 이를 해결하기 위해 IPO에서 특징적으로 사용되는 반복적 부분을 GPU(Graphic Processing Unit)으로 계산하고, AIPO-CR(Adaptive Iterative Physical Optics-Change Rate)으로 반복횟수를 최적화하여 효과적으로 연산속도를 향상시킨다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2003년도 하계학술대회 논문집 A
• /
• pp.125-127
• /
• 2003

The Conventional time-domain simulation of transient stability requires iterative calculation procedures to consider the saliency of generator. Recently, a non-iterative algorithm has successfully developed to take into account the generator saliency exactly with the use of $E_q'$-model. This study proposes an extended non-iterative algorithm by adopting the two-axis generator model. Given internal voltages and rotor angles of the generators, network voltages and generator currents can be directly calculated by solving a linear algebraic equation, which enables us to reduce the computation time remarkably.