• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.40 no.8
• /
• pp.301-306
• /
• 2003

This paper proposes a novel detection scheme using a radial basis function (RBF) network in a multi-input multi-output (MIMO) environment. In order to evaluate the performance of the proposed MIMO-RBF receiver, simulations are performed over the rich-scattering fading channel. Simulation results confirm that the proposed scheme shows the similar bit-error rate (BER) performance of a maximum likelihood detection (MLD) and outperforms Vertical-Bell Laboratories Layered Space-Time using minimum-mean-square-error nulling (VBLAST-MMSE) as well as VBLAST using zero-forcing nulling (VBLAST-ZF). Moreover, we investigate the effect on the performance of the number of RBF center with two modulation formats (BPSK and QPSK) and different number of transmit and receive antennas. The performance of the proposed detector is verified with respect to an initialization-rate of RBF centers.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.11 no.5
• /
• pp.45-52
• /
• 2002

In an end milling process, the undeformed chip section area and cutting forces vary periodically with the phase change of the tool. However, the real undeformed chip section area deviates from the geometrically ideal one owing to the cutter runout and tool shape error. In the current study, a method of estimating the real undeformed chip section area which reflects the cutter runout and tool shape error is presented during up-end milling processes of Inconel 718. The specific cutting forces, $K_r$ and $K_t$ are defined as the radial and tangential cutting forces divided by the modified chip section area, respectively. Both of the $K_{r}$ and $K_t$ values become smaller as the helix angle increases from $30^{\circ}$ to $40^{\circ}$. Whereas they become larger as the helix angle increases from $40^{\circ}$ to $50^{\circ}$. The $K_r$ and $K_t$ values show a tendency to decrease with increase of the modified chip section area.a.

• Korean Journal of Applied Biomechanics
• /
• v.23 no.4
• /
• pp.327-333
• /
• 2013

The purpose of this study was to investigate the effects of different types of attentional focus(internal focus vs. external focus) on the dart throwing mechanics. Seven expert dart throwing athletes were assigned to an internal focus group and other seven athletes were assigned to an external focus group. Each group was asked to throw dart either under verbal instruction or without instruction. During dart throwing, accuracy(radial error), consistency(bivariate variable error), dart velocity, acceleration, elbow joint ROM, elbow joint angular velocity, EMD(electromechanical delay), iEMG of biceps brachii and triceps brachii, and CI(coactivation index) were collected and analyzed. Nither instruction type nor instruction itself affected in accuracy and consistency. However, in dart velocity and acceleration, there was an interaction between instruction and attentional focus types. Velocity and acceleration increased in the internal condition, where as they decreased in the external condition. The ROM of elbow joint did not affected by instruction and attention type. However, similar to dart velocity and acceleration, angular velocity increased in internal focus group, while it decreased in external focus group. EMG showed no difference with any condition. In conclusion, internal focus is better than external focus for dart throwing.

• Korean Journal of Applied Biomechanics
• /
• v.24 no.3
• /
• pp.309-315
• /
• 2014

The aim of the present study was to investigate the effects of 6-week wrist resistance training on wrist torque control. Nineteen subjects were randomly assigned to either the wrist training group (n=9) or the control group (n=10). The training group performed wrist exercises for six directions (flexion, extension, pronation, supination, radial deviation, and ulnar deviation) while the control group did not. Testing for the isometric torque control error, one-repetition maximum (1-RM) strength, and isokinetic maximum torque (angular velocity of $60^{\circ}/s$ wrist movements) were conducted before and after six weeks of resistance training and after every two-week interval of training. The wrist training group showed significant decreases in isometric torque control error in all six directions after the 2-week resistance training, while the control group did not show significant increase or decrease. The training group showed significant increases in the maximum strength in all six directions assessed by 1-RM strength and isokinetic strength tests after the 4-week resistance training, while the control group did not show any statistically significant changes. This study shows that motor control ability significantly improves within the first two weeks of resistance training, while the wrist strength significantly improves within the first four weeks of resistance training in wrist training group compared to the control.

• Computers and Concrete
• /
• v.16 no.5
• /
• pp.741-757
• /
• 2015

In the last decade, several modeling approaches have been proposed and applied to estimate the high-performance concrete (HPC) slump flow. While HPC is a highly complex material, modeling its behavior is a very difficult issue. Thus, the selection and application of proper modeling methods remain therefore a crucial task. Like many other applications, HPC slump flow prediction suffers from noise which negatively affects the prediction accuracy and increases the variance. In the recent years, ensemble learning methods have introduced to optimize the prediction accuracy and reduce the prediction error. This study investigates the potential usage of bagging (Bag), which is among the most popular ensemble learning methods, in building ensemble models. Four well-known artificial intelligence models (i.e., classification and regression trees CART, support vector machines SVM, multilayer perceptron MLP and radial basis function neural networks RBF) are deployed as base learner. As a result of this study, bagging ensemble models (i.e., Bag-SVM, Bag-RT, Bag-MLP and Bag-RBF) are found superior to their base learners (i.e., SVM, CART, MLP and RBF) and bagging could noticeable optimize prediction accuracy and reduce the prediction error of proposed predictive models.

• Journal of the Society of Naval Architects of Korea
• /
• v.44 no.3 s.153
• /
• pp.219-227
• /
• 2007

A stereo-PIV (particle image velocimetry) technique is used to investigate the vortical structure of the wake behind a rotating propeller in the present study. A four bladed propeller is tested in a cavitaion tunnel without any wake screen. Hundreds of instantaneous velocity fields are phase-averaged to reveal the three dimensional spatial evolution of the flow behind the propeller. The results of conventional 2-D PIV are also compared with those of the stereo-PIV to understand the vortical structure of propeller wake deeply. The variations of radial and axial velocities in the 2-D PIV results seem to be affected by the out-of-plane motion. generating a little perspective error in the in-plane velocity components of the slipstream. The strong out-of-plane motion around the hub vortex also causes the perspective error to vary the axial velocity component a little at the near wake region. The out-of-plane velocity component had the maximum value of about 0.3U0 in the tip vortices and continued its magnitude in the wake region.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.12C
• /
• pp.1262-1267
• /
• 2005

Elliptical basis function (EBF) neural network works stably under high-level background noise environment and makes the nonlinear processing possible. It can be adapted real time VAD with simple design. This paper introduces VAD implementation using EBF and the experimental results show that EBF VAD outperforms G729 Annex B and RBF neural networks. The best error rates achieved by the EBF networks were improved more than $70\%$ in speech and $50\%$ in silence while that achieved by G.729 Annex B and RBF networks respectively.

• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.2949-2952
• /
• 2000

A new neural networks and learning algorithm are proposed in order to measure nonlinear heights of complexed road environments in realtime without pre-information. This new neural networks is Error Self Recurrent Neural Networks(ESRN), 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 back-propagation 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. We can estimate nonlinear models in realtime by ESRN and learning algorithm and control nonlinear models. To show the performance of this one. we control 7 degree of freedom full car model with several control method. From this simulation. this estimation and controller were proved to be effective to the measurements of nonlinear road environment systems.

• Journal of Intelligence and Information Systems
• /
• v.4 no.2
• /
• pp.23-34
• /
• 1998

Travel time forecasting, especially public bus travel time forecasting in urban areas, is a difficult and complex problem which requires a prohibitively large computation time and years of experience. As the network of target area grows with addition of streets and lanes, computational burden of the forecasting systems exponentially increases. Even though the travel time between two neighboring intersections is known a priori, it is still difficult, if not impossible, to compute the travel time between every two intersections. For the reason, previous approaches frequently have oversimplified the transportation network to show feasibilities of the problem solving algorithms. In this paper, forecasting of the travel time between every two intersections is attempted based on travel time data between two neighboring intersections. The time stamps data of public buses which recorded arrival time at predetermined bus stops was extensively collected and forecast. At first, the time stamp data was categorized to eliminate white noise, uncontrollable in forecasting, based on wavelet conversion. Then, the radial basis neural networks was applied to remaining data, which showed relatively accurate results. The success of the attempt was confirmed by the drastically reduced relative error when the nodes between the target intersections increases. In general, as the number of the nodes between target intersections increases, the relative error shows the tendency of sharp increase. The experimental results of the novel approaches, based on wavelet conversion and neural network teaming mechanism, showed the forecasting methodology is very promising.

• Journal of Positioning, Navigation, and Timing
• /
• v.9 no.4
• /
• pp.347-356
• /
• 2020

In this study, Inclined Geosynchronous Orbit (IGSO) and Geostationary Orbit (GEO) of BeiDou System (BDS) and Quasi Zenith Satellite System (QZSS) satellites positions and clock errors calculated by broadcast ephemeris and compared with Multi-GNSS Experiment (MGEX) products provided by five Analysis Centers (ACs). Root Mean Square Errors (RMSE) calculated for satellite position error. The IGSO results showed that 1.82 m, 0.91 m, 1.28 m in BDS and 1.34 m 0.36 m 0.49 m in QZSS and the GEO results showed that 2.85 m, 6.34 m, 6.42 m in BDS and 0.47 m, 4.79 m, 5.82 m in QZSS in the direction of radial, along-track and cross-track respectively. RMS calculated for satellite clock error. The IGSO result showed that 2.08 ns and 1.24 ns and the GEO result showed that 1.28 ns and 1.12 ns in BDS and QZSS respectively.