• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.8
• /
• pp.2722-2742
• /
• 2014

We propose methods of linear transceiver design for two different power constraints, sum relay power constraint and per relay power constraint, which determine signal processing matrices such as base station (BS) transmitter, relay precoders and user receivers to minimize sum mean square error (SMSE) for multi-relay multi-user (MRMU) networks. However, since the formulated problem is non-convex one which is hard to be solved, we suboptimally solve the problems by defining convex subproblems with some fixed variables. We adopt iterative sequential designs of which each iteration stage corresponds to each subproblem. Karush-Kuhn-Tucker (KKT) theorem and SMSE duality are employed as specific methods to solve subproblems. The numerical results verify that the proposed methods provide comparable performance to that of a full relay cooperation bound (FRCB) method while outperforming the simple amplify-and-forward (SAF) and minimum mean square error (MMSE) relaying in terms of not only SMSE, but also the sum rate.

• Journal of electromagnetic engineering and science
• /
• v.12 no.4
• /
• pp.290-292
• /
• 2012

This paper proposes a setup for a best multisine design method that uses a time-domain optimization. The method is based on minimization of the time-domain error, so its resulting multisine has a very accurate ACLR estimation. This is because its probability distribution and sample-to-sample correlation are close to those of the original signal, which are crucial for the testing of nonlinear power amplifiers. In addition, a hyperbolic-tangent function is introduced to control the ripple of tone magnitudes within signal bandwidth. For the verification, multisines were generated and compared for many aspects such as normalized error, in-band ripple, and ACLR estimation. Test results with different numbers of tones provide supporting evidence that the suggested multisine design has better ripple suppression, by up to 7 dB, and better accuracy, by up to 0.2 dB, when compared to the conventional method. The accuracy of the ACLR was improved by about 5 dB when the number of tones was 4. The suggested method improves the ACLR estimation performance of multisine testing due to its closer resemblance to the target modulation signal.

• Proceedings of the IEEK Conference
• /
• 2005.11a
• /
• pp.1241-1246
• /
• 2005

This paper described an 8bit, 100Msample/s CMOS D/A converter using a glich-time minimization technique for the high-speed sampling rate of 100MHz level. The proposed DAC was implemented in 0,35um Hynix CMOS technology and adopts a current mode architecture to optimize sampling rate, resolution, chip area. The DAC linear characteristics was similar to the proposed specification the prototype error between DNL and INL is less than ${\pm}0.09LSB$ respectively. Also, fab-out chip was tested, analysed the cause of error operation, and proposed the field considerations for chip test.

• Journal of Communications and Networks
• /
• v.17 no.1
• /
• pp.1-11
• /
• 2015

In this paper, the transmission of an improper-complex second-order stationary data sequence is considered over a strictly band-limited frequency-selective channel. It is assumed that the transmitter employs linear modulation and that the channel output is corrupted by additive proper-complex cyclostationary noise. Under the average transmit power constraint, the problem of minimizing the mean-squared error at the output of a widely linear receiver is formulated in the time domain to find the optimal transmit and receive waveforms. The optimization problem is converted into a frequency-domain problem by using the vectorized Fourier transform technique and put into the form of a double minimization. First, the widely linear receiver is optimized that requires, unlike the linear receiver design with only one waveform, the design of two receive waveforms. Then, the optimal transmit waveform for the linear modulator is derived by introducing the notion of the impropriety frequency function of a discrete-time random process and by performing a line search combined with an iterative algorithm. The optimal solution shows that both the periodic spectral correlation due to the cyclostationarity and the symmetric spectral correlation about the origin due to the impropriety are well exploited.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.22 no.3
• /
• pp.688-694
• /
• 1998

The ability to locate the defects in materials is one of the major attrations of the acoustic emission(AE) technique. The most conventional method for planar AE source localization is to place three or more AE sensors on the plate and to determine the source position by measuring the differences in the arrival times of the AE wave at the sensors, which is called as triangulation method. But this method can not be applied in the material of which elastic wave propagtion velocity is not known. In this paper, we propose two methods, vector method and error minimization method, for AE source location on the material with unknown AE wave velocity. In this method, it is not needed to know the propagation velocity previously, that is, we can apply this method to arbitrary material of which properties are not known exactly. Also, in this paper, the robustness to the error in the measurement of time differences are discussed for both methods. Finally, in order to evaluate the actual performances, experiments using a pencil lead break as the AE source were carried out on the aluminum plate.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.1438-1441
• /
• 2005

To establish of standard technique of nano-length measurement in 2D plane, new AFM system has been designed. In this system, measurement uncertainty is dominantly affected by the Abbe error of XY scanning stage. No linear stage is perfectly straight; in other words, every scanning stage is subject to tilting, pitch and yaw motion. In this paper, an AFM system with minimum offset of XY sensing is designed. And XY scanning stage is designed to minimize rotation angle because Abbe errors occur through the multiply of offset and rotation angle. To minimize the rotation angle optimal design has performed by maximizing the stiffness ratio of motion direction to the parasitic motion direction of each stage. This paper describes the design scheme of full AFM system, especially about XY stage. Full range of fabricated XY scanner is $100um\times{100um}$. And tilting, pitch and yaw motion are measured by autocollimator to evaluate the performance of XY stage. Using this AFM system, 3um pitch specimen was measured. As a result, the uncertainty of total system has been evaluated.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.4 no.3
• /
• pp.585-594
• /
• 2000

The exact path following of an autonomous mobile robot in a factory and an unreliable environment has many disadvantages in case of a classical control algorithm. In this paper, a neural network control approach based on an error back propagation algorithm is proposed for controlling a mobile robot to follow a line installed on the road. Since not only the three recognized informations from three sensors attached on a mobile robot but also the ten detailed informations in non recognition area are learned with input patterns, a mobile robot moves smoothly an installed line in spite of non perception space. The mobile robot has an effect of error minimization with a short time till a destination. To test an effectiveness of the proposed controller, the two motor velocity changes which is affected from a moving angle change of a mobile robot are simulated with computer.

• Journal of Intelligence and Information Systems
• /
• v.5 no.1
• /
• pp.95-101
• /
• 1999

In this paper, a comparison between different methods to combine predictions from neural networks will be given. These methods are bagging, bumping, and balancing. Those are based on the analysis of the ensemble generalization error into an ambiguity term and a term incorporating generalization performances of individual networks. Neural Networks and AI machine learning models are prone to overfitting. A strategy to prevent a neural network from overfitting, is to stop training in early stage of the learning process. The complete data set is spilt up into a training set and a validation set. Training is stopped when the error on the validation set starts increasing. The stability of the networks is highly dependent on the division in training and validation set, and also on the random initial weights and the chosen minimization procedure. This causes early stopped networks to be rather unstable: a small change in the data or different initial conditions can produce large changes in the prediction. Therefore, it is advisable to apply the same procedure several times starting from different initial weights. This technique is often referred to as training ensembles of neural networks. In this paper, we presented a comparison of three statistical methods to prevent overfitting of neural network.

• Journal of Power Electronics
• /
• v.18 no.5
• /
• pp.1398-1408
• /
• 2018

In order to further improve the steady-state control performance of model predictive torque control (MPTC), a double-vector-based model predictive torque control without a weighting factor is proposed in this paper. The extended voltage vectors synthesized by two basic voltage vectors are used to increase the number of feasible voltage vectors. Therefore, the control precision of the torque and the stator flux along with the steady-state performance can be improved. To avoid testing all of the feasible voltage vectors, the solution of deadbeat torque control is calculated to predict the reference voltage vector. Thus, the candidate voltage vectors, which need to be evaluated by a cost function, can be reduced based on the sector position of the predicted reference voltage vector. Furthermore, a cost function, which only includes a reference voltage tracking error, is designed to eliminate the weighting factor. Moreover, two voltage vectors are applied during one control period, and their durations are calculated based on the principle of reference voltage tracking error minimization. Finally, the proposed method is tested by simulations and experiments.

• Journal of the Korean Society for Precision Engineering
• /
• v.26 no.7
• /
• pp.142-149
• /
• 2009

The aim of this study was to identify both the mechanical and reflex properties associated with spasticity in hemiplegic patients. Ten hemiplegic patients were included in this study. Multiple pendulum tests were executed for each subject, and knee joint angle and EMG of Rectus Femoris muscle were measured. The neuromusculoskeletal system model was developed from generally accepted mechanism and identified through minimization of the error in the model-predicted pendulum trajectories. The identification was successful in terms of small error in simulated kinematics and high sensitivity and precision of simulated torque against EMG activity. The reflex threshold showed significant difference between different clinical scores (p<0.01) and significant negative correlation (r=-0.93) with the EMG duration. It is expected that the suggested method may help in understanding mechanisms underlying spasticity.