• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.11
• /
• pp.151-162
• /
• 1994

A new neural network structure called Self-organized Distributed Networks (SODN) is proposed for developing the neural network-based multidimensional system models. The learning with the proposed networks is fast and precise. Such properties are caused from the local learning mechanism. The structure of the networks is combination of dual networks such as self-organized networks and multilayered local networks. Each local networks learns only data in a sub-region. 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 proposed networks. The simulation results of the proposed networks show better performance than the standard multilayer neural networks and the Radial Basis function(RBF) networks.

• KSBB Journal
• /
• v.24 no.6
• /
• pp.515-520
• /
• 2009

This study presented facile method of cell patterning using fabricated PDMS patterns on polyelectrolyte coated surface. This basic principle is the fabrication of functional surface presenting two orthogonal surfaces such as cell adhesive and repellent properties. Cell adhesive surface was firstly fabricated with simple coating of polyelectrolyte multilayer. And then, the desired patterns of PDMS for the prevention of nonspecific binding of cells were transferred onto the previously formed thin film of polyelectrolyte multilayer. Thus, we could prepare novel functional surface simultaneously containing PDMS and polyelectrolyte region. As expected, the PDMS regions showed effective prevention of nonspecific binding of cell and the other region, exposed polyelectrolyte area, provided cell adhesive environment. The height of formed PDMS structure was about 100 nm. Based on this method, cell patterning can be successfully obtained with various pattern shapes and sizes. Therefore, we expect that this simple method will be useful platform technology for the development of cell chip, cell based assay system, and biochip.

• KIEE International Transactions on Electrophysics and Applications
• /
• v.3C no.5
• /
• pp.182-188
• /
• 2003

We have demonstrated a fluidic technique for self-assembly of microfabricated parts onto substrate using patterned shapes of magnetic force self-assembled monolayers (SAMs). The metal particles and the array were fabricated using the micromachining technique. The metal particles were in a multilayer structure (Au, Ti, and Ni). Sidewalls of patterned Ni dots on the array were covered by thick negative photoresist (SU-8), and the array was magnetized. The array and the particles were mixed in buffer solution, and were arranged by magnetic force interaction. Binding direction of the metal particle onto Ni dots was controlled by multilayer structure and direction of magnetization. A quarter of total Ni dots were covered by the particles. The binding direction of the particles was controllable, and condition of particles was almost even with the Au surface on top. The particles were successfully arranged on the array.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.549-550
• /
• 2006

It has become topic continuously at MEMS or semiconductor process to form three-dimensional multilayer structure. In this paper, we devised the new polymer pattern method that has multilayer structure. This is method that uses different kind of polymeric material. Specially, polymers used in this study that we propose became all pattern by photolithography, prevented that process increases. Here, polymer that we use used polymer of epoxy order called "SU-8" and polymer of novolak resin called "AZ-1518". The result, "SU-8" was formed pattern to 3.5um thickness, and "AZ-1518" about pattern 3um thickness. Also, It was been 6um thickness at same pattern area.

• Journal of Power Electronics
• /
• v.16 no.3
• /
• pp.1200-1208
• /
• 2016

Planar electromagnetic interference (EMI) filter has significant engineering significance to power electronic system integration and miniaturization. However, the value of differential mode capacitance cannot meet the demand of noise suppression because of the size limit of ceramics. In this case, the EMI filter of novel multilayers is recommended to address this issue. A novel integrated structure of EMI filter based on multilayer ceramic is proposed in this study. The inductance and capacitance of the new structure can be designed separately, which is an advantage in manufacturing. Insertion loss is measured more closely to the actual situation in this study, which is different from the condition where source and load impedances are both 50 Ω. In the process of designing a novel EMI filter, noise impedance is considered. Moreover, the prototype is created and applied to a small switching power supply, which verifies the effectiveness of the developed EMI filter.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.12
• /
• pp.2418-2425
• /
• 1994

Although the error backpropagation(EBP) algorithm based on the gradient descent method is a widely-used learning algorithm of neural networks, learning sometimes takes a long time to acquire accuracy. This paper develops a novel learning method to alleviate the problems of EBP algorithm such as local minima, slow speed, and size of structure and thus to improve performance by adopting other new networks. Gaussian Potential Function networks(GPFN), in parallel with multilayer neural networks. Empirical simulations show the efficacy of the proposed algorithm in function approximation, which enables us to train networks faster with the better generalization capabilities.

• Smart Structures and Systems
• /
• v.25 no.6
• /
• pp.765-778
• /
• 2020

The present study intends to find a proper solution for the estimation of the physical behaviors of enlarged piles through a combination of small-scale laboratory tests and a hybrid computational predictive intelligence process. In the first step, experimental program is completed considering various critical influential factors. The results of the best multilayer perceptron (MLP)-based predictive network was implemented through three mathematical-based solutions of dragonfly algorithm (DA), whale optimization algorithm (WOA), and ant lion optimization (ALO). Three proposed models, after convergence analysis, suggested excellent performance. These analyses varied based on neurons number (e.g., in the basis MLP hidden layer) and of course, the level of its complexity. The training R² results of the best hybrid structure of DA-MLP, WOA-MLP, and ALO-MLP were 0.996, 0.996, and 0.998 where the testing R² was 0.995, 0.985, and 0.998, respectively. Similarly, the training RMSE of 0.046, 0.051, and 0.034 were obtained for the training and testing datasets of DA-MLP, WOA-MLP, and ALO-MLP techniques, while the testing RMSE of 0.088, 0.053, and 0.053, respectively. This obtained result demonstrates the excellent prediction from the optimized structure of the proposed models if only population sensitivity analysis performs. Indeed, the ALO-MLP was slightly better than WOA-MLP and DA-MLP methods.

• Journal of Magnetics
• /
• v.11 no.1
• /
• pp.1-4
• /
• 2006

We report on the growth, structural and transport properties of zinc-blende CrAs/GaAs/MnAs/GaAs multilayers on GaAs(001) substrates. Structural analyses using in-situ reflection high-energy electron diffraction and exsitu cross-sectional transmission electron microscopy confirmed the realization of a zinc-blende crystal structure. Room temperature Hall measurements reveal that the as-grown multilayer exhibits p-type conductivity with a very low resistivity of $0.052\;\omega{cm}$, a high carrier concentration of $6.2X10^{19}\;cm^{-3}$ and a Hall mobility of $1.8\;cm^2/Vs$. We also observed a clear decrease of the resistivity in samples after low temperature annealing.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.8 no.6
• /
• pp.153-157
• /
• 2008

Recently, electronic devices is produced by multilayer structure, therefore analysis for hidden layers is important nondestructive inspection. This paper presents visibility enhancement methods for the ultrasonic multiple echoes reflected from adhesive layer in the multilayers using digital signal processing. The reflected signals from the multilayers come out interval of the peaks in the power cepstrum. In the experiment, the adhesive layers of settled thickness using epoxy were formed. The reflected signals from the multilayer is detected by pulse-echo method and power cepstrum is processed for enhancement of visibility.

• Proceedings of the KIEE Conference
• /
• 2006.07d
• /
• pp.1781-1782
• /
• 2006

In this paper, a methodology has been proposed according to which, after carrying out a set of soil's resistivity measurements using Wenner Four-probe data, one can compute the parameters of the multilayer earth structure using a genetic algorithm(GA). The results provided by the GA constitute the indispensable data that can be used in circuital or field simulations of grounding systems. The methodology allows to proceed toward a very efficient simulation of the grounding system and an accurate calculation of potential on the ground's surface.