• The Journal of the Acoustical Society of Korea
• /
• v.28 no.8
• /
• pp.815-821
• /
• 2009

Information processing in variable and noisy environments is usually accomplished by means of adaptive filters. Among various adaptive algorithms, Least Mean Square (LMS) has become the most popular for its robustness, good tracking capabilities and simplicity, both in terms of computational load and easiness of implementation. In practical application of the LMS algorithm, the most important key parameter is the Step Size. As is well known, if the Step Size is large, the convergence rate of the algorithm will be rapid, but the steady state mean square error (MSE) will increase. On the other hand, if the Step Size is small, the steady state MSE will be small, but the convergence rate will be slow. Many researches have been proposed to alleviate this drawback by using a variable Step Size. In this paper, a new variable Step Size LMS(VSSLMS) called Categorized VSSLMS (CVSSLMS) is proposed. CVSSLMS updates the Step Size by categorizing the current status of the gradient, hence significantly improves the convergence rate. The performance of the proposed algorithm was verified from the view point of convergence rate, Excessive Mean Square Error(EMSE), and complexity through experiments.

• Journal of IKEEE
• /
• v.15 no.3
• /
• pp.235-240
• /
• 2011

The scattered pilots employed in DVB-T take advantage of the merits of both the block type and comb type pilot arrangement to increase the transmission efficiency. To estimate the channel transfer functions for data subcarriers, it is required to conduct time-frequency domain 2D estimation using the pilots. Though 2D Wiener estimator is optimal in sense of MSE (mean square error), it is too complex to implement in hardware. In this paper a new channel estimation method for the scattered pilot based OFDM system by measuring the power of AWGN and removing the noise in the LS (least square) estimate of the channel is proposed. And the simulation results reveal the proposed method outperforms the 2D linear interpolation in the fading channel.

• Journal of Broadcast Engineering
• /
• v.15 no.1
• /
• pp.63-75
• /
• 2010

In this paper, the equalizer techniques that is able to adopt the digital on-channel repeater for 8VSB-based DTV system has been analyzed and we propose an effective equalizer structure which can reduce the error propagation phenomenon by the feedback signal and improve the receiver performance at the same time. In order to confirm the effective cancellation of the feedback signal, the multi-level Correlation LMS scheme is proposed through the analysis of conventional basic LMS based DFE and Correlation LMS algorithm and as compared with the conventional method, we can confirm the reduction of error propagation. When performing the computer simulation, as the Brazil channel model which is very popular for DTV broadcasting system is adopted, the result is drawn by comparing and analysing the equalizer algorithm. We have examine the symbol error rate which is in the range of 15~25dB of operation receipt SNR and MSE(Mean Square Error) in the DTV broadcasting system. As a result of comparing with the existing method, the signal-noise ratio which is necessary for maintain the bit error correction ability that the means of proposal is same is reduced by about 2~5dB, and in the rate of convergence through the MSE, we found the reduction of needed time.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.3
• /
• pp.355-359
• /
• 1993

This paper presents a derivation of the optimum convergence constant to yield the most rapid convergence under a desired mean-square error (MSE) for echo canceller using the LMS algorithm. For white input data, the optimum convergence constant is a simple closed-form function of the number of filter taps, the input signal variance, the initial MSE, and the desired MSE. This characteristic makes it easily designed in many practical applications. Computer simulations are also employed to show the correctness and effectiveness of the derived results.

• Speech Sciences
• /
• v.10 no.2
• /
• pp.77-84
• /
• 2003

In this research, we seek the beginning of the speech and detect the stationary speech region using lip information. Performing running average of the estimated speech signal in the stationary region, we reduce the effect of musical noise which is inherent to the conventional MlMSE (Minimum Mean Square Error) speech enhancement algorithm. In addition to it, SFM (Spectral Flatness Measure) is incorporated to reduce the speech signal estimation error due to speaking habit and some lacking lip information. The proposed algorithm with Wiener filtering shows the superior performance to the conventional methods according to MOS (Mean Opinion Score) test.

• Journal of information and communication convergence engineering
• /
• v.8 no.2
• /
• pp.155-158
• /
• 2010

In this article, we proposed a new interference canceller using the adaptive algorithm. We designed constant modulus algorithm-correlation normailized least mean square (CMA-CNLMS) for wireless system. This structure is normalized LMS algorithm using correlation between the desired and input signal for cancelling the interference signals in the wideband code division multiple access (WCDMA) system. We showed that the proposed algorithm could improve the Mean Square Error (MSE) performance of LMS algorithm. MATLAB (Matrix Laboratory) is employed to analyze the proposed algorithm and to compare it with the experimental results. The MSE value of the LMS with mu=0.0001 was measured as - 12.5 dB, and that of the proposed algorithm was -19.5 dB which showed an improvement of 7dB.

• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.1475-1478
• /
• 2002

This paper presents a performance analysis of an improved adaptive algorithm proposed by the authors recently. It is based on the normalized least mean square (NLMS) algorithm, which Is one of the major techniques to adapt the cofficients of a transversal filter. Generally, the performance of an adaptive algorithm is often discussed by investigating the mis-adjustment. In this paper, unlike these approaches, a novel analytical method is considered. letting the parameters so that the residual mean square error (MSE) after the convergence of the algorithm is equal to that of the NLMS algorithm, the MSE level is compared. It is shown that the theoretical analysis is agreed with the simulation results.

• Journal of Information Display
• /
• v.5 no.1
• /
• pp.41-46
• /
• 2004

In this paper, a new stereo 3D moving-target tracking system using the MSE (mean square error) and BPEJTC (binary phase extraction joint transform correlator) algorithms is proposed. A moving target is extracted from the sequential input stereo image by applying a region-based MSE algorithm following which, the location coordinates of a moving target in each frame are obtained through correlation between the extracted target image and the input stereo image by using the BPEJTC algorithm. Through several experiments performed with 20 frames of the stereo image pair with $640{\times}480$ pixels, we confirmed that the proposed system is capable of tracking a moving target at a relatively low error ratio of 1.29 % on average at real time.

• Journal of the Korean Statistical Society
• /
• v.31 no.4
• /
• pp.433-445
• /
• 2002

Using some available information about the unknown variance $\sigma$$^2$ of a normal distribution with mean $\mu$, a sequential approach is used to estimate $\sigma$$^2$. Two cases have been considered regarding the mean $\mu$ being known or unknown. The mean square error (MSE) of the new estimators are compared to that of the usual estimator of $\sigma$$^2$, namely, the sample variance based on a sample of size equal to the expected sample size. Simulation results indicates that, the new estimator is more efficient than the usual estimator of $\sigma$$^2$whenever the actual value of $\sigma$$^2$ is not too far from the prior information.

• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.120-126
• /
• 2019

Recently, accurate prediction of power consumption based on machine learning techniques in Internet of Energy (IoE) has been actively studied using the large amount of electricity data acquired from advanced metering infrastructure (AMI). In this paper, we propose a deep learning model based on Gated Recurrent Unit (GRU) as an artificial intelligence (AI) network that can effectively perform pattern recognition of time series data such as the power consumption, and analyze performance of the prediction based on real household power usage data. In the performance analysis, performance comparison between the proposed GRU-based learning model and the conventional learning model of Long Short Term Memory (LSTM) is described. In the simulation results, mean squared error (MSE), mean absolute error (MAE), forecast skill score, normalized root mean square error (RMSE), and normalized mean bias error (NMBE) are used as performance evaluation indexes, and we confirm that the performance of the prediction of the proposed GRU-based learning model is greatly improved.