• Journal of information and communication convergence engineering
• /
• v.9 no.6
• /
• pp.637-640
• /
• 2011

The hybrid neural networks have characteristics such as fast learning times, generality, and simplicity, and are mainly used to classify learning data and to model non-linear systems. The middle layer of a hybrid neural network clusters the learning vectors by grouping homogenous vectors in the same cluster. In the clustering procedure, the homogeneity between learning vectors is represented as the distance between the vectors. Therefore, if the distances between a learning vector and all vectors in a cluster are smaller than a given constant radius, the learning vector is added to the cluster. However, the usage of a constant radius in clustering is the primary source of errors and therefore decreases the recognition success rate. To improve the recognition success rate, we proposed the enhanced hybrid network that organizes the middle layer effectively by using the enhanced ART1 network adjusting the vigilance parameter dynamically according to the similarity between patterns. The results of experiments on a large number of calling card images showed that the proposed algorithm greatly improves the character extraction and recognition compared with conventional recognition algorithms.

• Journal of Digital Convergence
• /
• v.10 no.2
• /
• pp.287-292
• /
• 2012

In this paper, user personalized services through the emotion perception required to support location-based sensing data preprocessing techniques and emotion data preprocessing techniques is studied for user's emotion data building and preprocessing in V-A emotion model. For this purpose the granular context tree and string matching based emotion pattern matching techniques are used. In addition, context-aware and personalized recommendation services technique using probabilistic reasoning is studied for personalized services based on context awareness.

• Journal of information and communication convergence engineering
• /
• v.14 no.3
• /
• pp.137-146
• /
• 2016

In this paper, we introduce a quality of experience (QoE)-provisioning mobility management scheme. The emphasis is on a mobility-aware QoE solution enabling network components to recognize the mobility pattern of an end-user and to prepare a handover in advance. We further focus on an energy-adaptive QoE solution based on the energy profile providing the preferred pattern of energy consumption and an energy preference check engine determining whether the provision of the service that the end-user requested is suitable to QoE or not. Lastly, we concentrate on a network-based intelligent mobility management scheme adopting the calm service and the balance. Consequently, we conclude that the proposed schemes improve the handover latency, QoE metrics, and energy efficiency simultaneously.

• Journal of information and communication convergence engineering
• /
• v.10 no.2
• /
• pp.123-128
• /
• 2012

In communication systems, there are various types of microstrip antenna that can be used for many applications. This paper mainly focuses on the simple design of an inset rectangular microstrip patch antenna to operate at a frequency of 2.45 GHz for rectenna design. The study involves using an high frequency structure simulator to design the antenna dimensions and to determine its performance. This antenna is based on a thickness of 1.6 mm flame retardant 4 (FR-4) substrate having a dielectric constant of approximately 4.7, an inset feed, and a ground plane. After simulation, the antenna performance characteristics such as its return loss, voltage standing wave ratio, gain, and radiation pattern were obtained and compared with the fabricated measured antenna.

• Journal of information and communication convergence engineering
• /
• v.12 no.2
• /
• pp.83-88
• /
• 2014

The simultaneous suppression of sidelobes and the sharpening of the central peak in the process of diffraction pattern detection based on asymmetric apodization have been investigated. Asymmetric apodization is applied to a semicircular array of two-dimensional (2D) aperture functions, which is a series of 'coded-phase arrays of semicircular rings randomly distributed over the central circular region of a pupil function' and is similar to that used in the field of diffractive optics. The point spread function (PSF) of an imaging system with asymmetric apodization of the discrete type has been found to possess a good side with suppressed sidelobes, whereas its bad side contains enhanced sidelobes. Further, the diffracted field characteristics are obtained in the presence of these aperture functions. Asymmetric apodization is helpful in improving the performance of the optical gratings or 2D arrays used in real-time imaging techniques.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2003.04a
• /
• pp.357-365
• /
• 2003

The major problems with the conventional neural network, especially Back Propagation Neural Network, arise from the necessity of many training data for neural network learning and ambiguity in the relation of neural network structure to the convergence of solution. In this paper, the PNN is used as a pattern classifier to detect the damage of structure to avoid those drawbacks of the conventional neural network. In the PNN-based pattern classification problems, the probability density function for patterns is usually assumed by Gaussian distribution. But, in this paper, several probability density functions are investigated in order to select the most approriate one for structural damage assessment.

• International journal of advanced smart convergence
• /
• v.2 no.1
• /
• pp.21-23
• /
• 2013

The antenna pattern in this case has a main beam pointed in the desired signal direction, and has a null in the direction of the interference.The conventional antenna pattern concepts of beam width, side lobes, and main beams are not used, as the antenna weights are designed to achieve a set performance criterion such as maximization of the output SNR.A new direction of arrival estimation method using effects of model errors and sensitivity analysis is proposed. Two subspaces are used to form a signal space whose phase shift between the reference signal and its effects of model error signal. Through simulation, the performance showed that the proposed method leads to increased resolution and improved accuracy of DOA estimation relative to those achieved with existing method. Since a desired signal is obtained after interference rejection through correction effects of model error, the effect of channel interference on the estimation is significantly reduced.

• KIEE International Transactions on Electrophysics and Applications
• /
• v.3C no.5
• /
• pp.194-198
• /
• 2003

A wide-band E-plane notch phased array antenna having bandwidths of 3:1 and a scan volume of $\pm$ 45 is designed considering the active element pattern (AEP) with analysis of the full structure of E-plane notch phased array antenna. Using the numerical E-plane waveguide simulator as an infinite linear array in the broadside angle, the active reflection coefficient (ARC) of the unit element is optimized in the design frequency range. To evaluate the convergence of the AEP, the simulation of full array as changing the number array is investigated, and the minimum numbers of array that have characteristics similar to the AEP of an infinite array are determined.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.2
• /
• pp.219-224
• /
• 2020

In this paper, we studied the compact MIMO(: Multi-Input Multi-Output) antenna for multi-beam pattern generation and control in narrow space. The characteristics and performance of the basic Dipole SPA and Monopole SPA structures are shown. We implemented a monopole SPA antenna and measured its performance in an 802.11g system. When the average transmission rate was measured using the SPA antenna, the SPA antenna improved the best performance by 8.7 times compared to commercial antenna.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.62 no.3
• /
• pp.380-386
• /
• 2013

This paper represents an adaptive beamforming technique to suppress interference or jamming signals in wideband. In order to maintain low side lobe level(SLL) at an antenna element level, Taylor-weighting was used. Also, to make a nulling beam pattern toward jammer's directions in wideband, we used the modified Loaded Sample Matrix Inversion(LSMI) algorithm and Tapped Delay Line(TDL). To verify the proposed algorithm, we applied it to a rectangular array antenna. Finally, the results show beam pattern with low SLL and jammers suppression.