• 제어로봇시스템학회:학술대회논문집
• /
• 1987.10b
• /
• pp.470-472
• /
• 1987

This paper deals with the continuous-discrete estimation problem using FIR filters and performs modeling error analyses of the FIR filters, compared to Kalman filter and the limited memory filters, via computer simulations. It is shown that, the less driving noise the system has, the better performance the FIR filter exhibits and that this characteristic appears rare distinctly in nonlinear system than in linear systems.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2001.05a
• /
• pp.77-83
• /
• 2001

In recent years, the AES in North America and the NESSIE project in Europe have been in progress. Six proposals have been submitted to the NESSIE project including the LILI-128 by Simpson in Australia in the synchronous stream cipher category. These proposals tend towards a design with parallelism of the algorithms in order to facilitate speed-up. In this paper, we consider the PS-LFSR and propose the effective implementation of various nonlinear combiners: memoryless-nonlinear combiner, memory-nonlinear combiner, nonlinear filter function, and clock-controlled function. Finally, we propose m-parallel SUM-BSG and LILI-l28's parallel implementation as examples, and we determine their securities and performances.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.6
• /
• pp.177-186
• /
• 2005

This paper introduces nonlinear estimators with reduced complexity, and proposes the Bussgang blind equalization algorithm employing the nonlinear estimators. The proposed algorithm utilized the facts that the Bayesian estimator is well approximated to the sigmoid estimator in initial stage of equalization with closed eye and is well approximated to the threshold estimator under open eye condition. The proposed method adopts selectively one of the two nonlinear estimators, i.e., the sigmoid estimator and the threshold estimator, according to channel distortion level at each iteration. As a result, by using the sigmoid estimator with reduced constellation, the proposed scheme, as it is applied to blind equalization of high-order QAM signals, simplifies the computational complexity extremely, and enhances the blind convergence capability and steady-state performance.

• Journal of the Institute of Convergence Signal Processing
• /
• v.7 no.2
• /
• pp.53-59
• /
• 2006

Dynamic neural networks have been applied to diverse fields requiring temporal signal processing such as system identification and signal prediction. This paper proposes the gamma neural network(GAM), which uses gamma memory kernel in the hidden layer of feedforward multilayered network, to improve dynamics of networks and then describes nonlinear adaptive prediction using the proposed network as an adaptive filter. The proposed network is evaluated in nonlinear signal prediction and compared with feedforword(FNN) and recurrent neural networks(RNN) for the relative comparison of prediction performance. Simulation results show that the GAM network performs better with respect to the convergence speed and prediction accuracy, indicating that it can be a more effective prediction model than conventional multilayered networks in nonlinear prediction for nonstationary signals.

• Journal of Ocean Engineering and Technology
• /
• v.19 no.4 s.65
• /
• pp.1-8
• /
• 2005

A nonlinear time-domain strip theory for vertical wave loads and ship responses is to be investigated. The hydrodynamic memory effect is approximated by a higher order differential equation without convolution. The ship is modeled as a non-uniform Timoshenko beam. Numerical calculations are presented for the S175 Containership translating with the forward speed in regular waves. The approach described in this paper can be used in evaluating ship motions and wave loads in extreme wave conditions and validating nonlinear phenomena in ship design.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.51 no.1
• /
• pp.39-42
• /
• 2002

For the purpose of modeling EEG signal which has nonstationary and nonlinear dynamic characteristics, this paper propose a state feedback real time recurrent neural network model. The state feedback real time recurrent neural network is structured to have memory structure in the state of hidden layers so that it has arbitrary dynamics and ability to deal with time-varying input through its own temporal operation. For the model test, Mackey-Glass time series is used as a nonlinear dynamic system and the model is applied to the prediction of three types of EEG, alpha wave, beta wave and epileptic EEG. Experimental results show that the performance of the proposed model is better than that of other neural network models which are compared in this paper in some view points of the converging speed in learning stage and normalized mean square error for the test data set.

• Proceedings of the KIEE Conference
• /
• 1995.07b
• /
• pp.804-806
• /
• 1995

This paper discusses Self-organized Distributed Networks(SODN) as identifier of nonlinear dynamical systems. The structure of system identification employs series-parallel model. The identification procedure is based on a discrete-time formulation. The learning with the proposed SODN is fast and precise. Such properties arc caused from the local learning mechanism. Each local networks learns only data in a subregion. Large number of memory requirements and low generalization capability for the untrained region, which are drawbacks of conventional local network learning, are overcomed in the SODN. Through extensive simulation, SODN is shown to be effective for identification of nonlinear dynamical systems.

• Journal of electromagnetic engineering and science
• /
• v.8 no.1
• /
• pp.17-22
• /
• 2008

It is presented that sampling rates for behavioral modeling of quasi-memory less nonlinear devices can be far less than the Nyquist rate of the input signal. Although it has been believed that the sampling rate of nonlinear device modeling should be at least the Nyquist rate of the output signal, this paper suggests that far less than the Nyquist rate of the input signal can be applied to the modeling of quasi-memoryless nonlinear devices, such as frequency multipliers. To verify, a QPSK signal at 820 MHz were applied to a frequency tripler, whereby the device can be utilized as an up-converting mixer into 2.46 GHz with the aid of digital predistortion. AM-AM, AM-PM and PM-PM can be successfully measured regardless of sampling rates.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2011.10a
• /
• pp.349-354
• /
• 2011

Tuned mass damper is widely used in many applications of industry. The main advantage of tuned mass damper is that it can increase the damping ratio of system and reduce the vibration amplitude. Meanwhile, the natural frequency of system will be divided by two peaks, and the peak speeds are closely related to the mass and the stiffness of auxiliary mass system added. In addition, the damping ratio will also affect the peak frequency of the dynamic response. In the present research, the nonlinear mechanical characteristics of rubber is investigated and put into use, since it is usually manufactured as the spring element of tuned mass damper. By the sense of the nonlinear stiffness as well as the damping ratio which can be changed by preload applied on, the shape memory alloy is proposed to control the auxiliary mass system by self-optimizing. Supported by the experiment data of rubber, the 1 DOF theoretical model and finite element model based on computer simulation are implemented to perform the feasibility of the proposed semi-active tuned mass damper working on the drive shaft.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.10
• /
• pp.1518-1526
• /
• 1993

It has been known that the amplifiers for high power signal in satellite communication channels suffer from nonlinear distortions, which reduce the performance of the communication channel significantly. In order to compensate the nonlinear distortion, a new data predistortion method with the LMS algorithm is proposed in this paper, Whereas the previous approach handles this problem by assigning corresponding predistorter to each symbol for the case of 16-QAM, the proposed approach uses the same memory for the symbols, which have identical amplitudes, and predistors the input of high-power amplifiers by the amplitude and phase differences, resulting in better adaptive data predistorter with small number of digital memory (3 predistorters) and fast convergence rate. Superiority of the proposed approach in the paper is demonstrated by comparing it with the previous approach.