• 한국통신학회논문지
• /
• 제25권5B호
• /
• pp.830-841
• /
• 2000

We extend the sue of the method of least square to develop a recursive algorithm for the design of adaptive transversal filters such that, given the least-square estimate of this vector of the filter at iteration n-1, we may compute the updated estimate of this vector at iteration a upon the arrival of new data. We begin the development of the RLS algorithm by reviewing some basic relations that pertain to the method of least squares. Then, by exploiting a relation in matrix algebra known as the matrix inversion lemma, we develop the RLS algorithm. An important feature of the RLS algorithm is that it utilizes information contained in the input data, extending back to the instant of time when the algorithm is initiated. In this paper, we propose new tap weight updated RLS algorithm in adaptive transversal filter with data-recycling buffer structure. We prove that convergence speed of learning curve of RLS algorithm with data-recycling buffer is faster than it of exiting RL algorithm to mean square error versus iteration number. Also the resulting rate of convergence is typically an order of magnitude faster than the simple LMS algorithm. We show that the number of desired sample is portion to increase to converge the specified value from the three dimension simulation result of mean square error according to the degree of channel amplitude distortion and data-recycle buffer number. This improvement of convergence character in performance, is achieved at the (B+1)times of convergence speed of mean square error increase in data recycle buffer number with new proposed RLS algorithm.

• 제어로봇시스템학회논문지
• /
• 제17권2호
• /
• pp.125-130
• /
• 2011

This paper shows a method to design complementary filter using least square. The complementary filter is one of useful filters estimating angle. The basic concept of this filter is to enhance advantages of each sensor that angle detecting using a gyroscope has good accuracy at a high frequency and an accelerometer at a low frequency. When designing complementary filter, the most commonly used method is using cut-off frequency. However, it may be not easy to obtain a cut-off frequency. This paper presents a systematic method to determine the coefficients of the complementary filter using well-known linear least squares minimizing error between estimating angle and true angle.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1994년도 제11회 음성통신 및 신호처리 워크샵 논문집 (SCAS 11권 1호)
• /
• pp.107-110
• /
• 1994

In this paper, we describe the features of the fundamental frequency contour of Korean read speech, and propose a new stylization method to characterize the Fø pattern of segments. Our algorithm consists of three stylization processes : the segment level, the syllable level, and the sord level. For stylization of Fø contour in the segment level , we applied least square error minimization method to determine Fø values at initial, medial, and final position in a segment. In the syllable level, we determine the stylized Fø pattern of a syllable using the mean Fø value of each word and style information for each word, syllable and segment, we reconstruct Fø contour of sentences. The simulation results show that the error is less than 10% of the actual Fø contour for each sentence. In perception test, there is little difference between the synthesized speech with the original difference between the synthesized speech with the original Fø contour and the synthesized speech with the stylized Fø contour.

• 디지털산업정보학회논문지
• /
• 제10권3호
• /
• pp.197-205
• /
• 2014

In applications of adaptive noise control or active noise control, the presence of a transfer function in the secondary path following the adaptive controller and the error path, been shown to generally degrade the performance of the Least Mean Square (LMS) algorithm. Thus, the convergence rate is lowered, the residual power is increased, and the algorithm can become unstable. In general, in order to solve these problems, the filtered-x LMS (FX-LMS) type algorithms can be used. But these algorithms have slow convergence speed and weakness in the environment that the secondary path and error path are varied. Therefore, I present the new algorithm called the "Bi-directional Filtered-x (BFX) LMS" algorithm with nearly equal computation complexity. Through experimental study, the proposed BFX-LMS algorithm has better convergence speed and better performance than the conventional FX-LMS algorithm, especially when the secondary path or error path is varied and the impulsive disturbance is flow in.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2008년도 하계학술대회 논문집 Vol.9
• /
• pp.525-526
• /
• 2008

For measuring nonlinear Vehicle Modeling based on 7-Depth Sensor, the neural networks are proposed m adaptive and in realtime. The structure of it is similar to recurrent neural networks; a delayed output as the input and a delayed error between the output of plant and neural networks as a bias input. In addition, we compute the desired value of hidden layer by an optimal method instead of transfering desired values by backpropagation and each weights are updated by RLS(Recursive Least Square). Consequently, this neural networks are not sensitive to initial weights and a learning rate, and have a faster convergence rate than conventional neural networks. This new neural networks is Error Estimated Neural Networks. We can estimate nonlinear models in realtime by the proposed networks and control nonlinear models.

• 약학회지
• /
• 제47권5호
• /
• pp.261-265
• /
• 2003

Near-infrared (NIR) system was used to determine rapidly and simply indomethacin in buffer solution for a dissolution test of tablets and capsules. Indomethacin standards were prepared ranging from 10 to 50 ppm using the mixture of phosphate buffer (pH 7.2) and water (1 : 4). The near-infrared (NIR) transmittance spectra of indomethacin standard solutions were collected by using a quartz cell in 1 mm and 2 mm pathlength. Partial least square regression (PLSR) was explored to develop calibration models over the spectral range 1100∼1700 nm. The model using 1 mm quartz cell was better than that using 2 mm quartz cell. The PLSR models developed gave standard error of prediction (SEP) of 0.858 ppm. In order to validate the developed calibration model, routine analysis was performed using another standard solutions. The NIR routine analysis showed good correlation with actual values. Standard error of prediction (SEP) is 1.414 ppm for 7 indomethacin samples in routine analysis and its error was permeable in the regulation of Korean Pharmacopoeia (VII). These results show the potential use of the real time monitoring for indomethacin during a dissolution test.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 2006년도 International Symposium on GPS/GNSS Vol.1
• /
• pp.485-489
• /
• 2006

In last decades, several linearization methods for the AOA measurements have been proposed, for example, Gauss-Newton method and closed-form solution. Gauss-Newton method can achieve high accuracy, but the convergence of the iterative process is not always ensured if the initial guess is not accurate enough. Closed-form solution provides a non-iterative solution and it is less computational. It does not suffer from convergence problem, but estimation error is somewhat larger. This paper proposes a self-tuning weighted least square AOA algorithm that is a modified version of the conventional closed-form solution. In order to estimate the error covariance matrix as a weight, two-step estimation technique is used. Simulation results show that the proposed method has smaller positioning error compared to the existing methods.

• 소음진동
• /
• 제4권1호
• /
• pp.59-69
• /
• 1994

A new LMS algorithm titled constrained LMS' is proposed for problems with constrained structure. The conventional LMS algorithm can not be used because it destroys the constrained structures of the weights or parameters. Proposed method uses error-back propagation, which is popular in training neural networks, for error minimization. The illustrative examplesare shown to demonstrate the applicability of the proposed algorithm.

• 한국통신학회논문지
• /
• 제32권3C호
• /
• pp.338-347
• /
• 2007

본 논문에서는 순환 보호 구긴(cyclic-prefix)을 사용하는 시공간 블록 부호 (STBC: Space-Time Block-Coding) 단일 반송파 시스템에서 향상된 채널 성능을 위한 가중된 블록 적응형 주파수 영역 채널 추정기를 제안한다. 제안된 채널 추정기 구조는 필터 입력 신호에 대해 STBC로 구성된 블록을 형성하며, 이후 형성된 입력 블록에 대해 사후 오차 (a posteriori error)를 이용하는 가중된 LS (least-square) 규준을 적용하여 알고리즘을 유도한다. 또한 정적 채널에서 steady-state EMSE (excess mean-square error) 분석을 통해 블록 길이가 늘어남에 따라 EMSE를 분석한다. 전산 모의실험에서는 시변 TU (typical urban) 채널에서 블록 길이를 증가시킬수록 제안한 채널 추정기는 기존 NLMS와 RLS 채널 추정기들 보다 우수한 성능을 나타냄을 확인 할 수 있다.

• 한국전자통신학회논문지
• /
• 제9권5호
• /
• pp.549-554
• /
• 2014

본 논문에서 ICS(Interference Cancellation System) 중계기를 위한 Signed LMS(Least Mean Square) 알고리즘을 제안한다. 제안된 Signed LMS 알고리즘은 스텝 사이즈를 조절함에 따라 성능이 개선된다. 제안된 Signed LMS 알고리즘에서 스텝사이즈가 0.067인 경우 수렴횟수 1000 회 일 때 평균 자승 에러는 기존 CMA 알고리즘보다 약 3 ~ 18 dB정도 더 낮다. 그리고, 평균 자승 에러 -25 dB 일 때 LMS(Least Mean Square)와 CMA보다 수렴횟수가 500 ~ 4000 회 정도 줄어든다.