• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.63 no.4
• /
• pp.534-540
• /
• 2014

In this paper, we propose a fuzzy combined Polynomial Neural Network(PNN) for pattern classification. The fuzzy combined PNN comes from the generic TSK fuzzy model with several linear polynomial as the consequent part and is the expanded version of the fuzzy model. The proposed pattern classifier has the polynomial neural networks as the consequent part, instead of the general linear polynomial. PNNs are implemented by stacking the simple polynomials dynamically. To implement one layer of PNNs, the various types of simple polynomials are used so that PNNs have flexibility and versatility. Although the structural complexity of the implemented PNNs is high, the PNNs become a high order-multi input polynomial finally. To estimate the coefficients of a polynomial neuron, The weighted linear discriminant analysis. The output of fuzzy rule system with PNNs as the consequent part is the linear combination of the output of several PNNs. To evaluate the classification ability of the proposed pattern classifier, we make some experiments with several machine learning data sets.

• Journal of Internet Computing and Services
• /
• v.10 no.3
• /
• pp.17-25
• /
• 2009

This paper presents a methodology for a fall detection using the feature extraction method based on the neural network with weighted fuzzy membership functions (NEWFM). The distributed non-overlap area measurement method selects the minimized number of input features by removing the worst input features one by one. Nineteen number of wavelet transformed coefficients captured by a triaxial accelerometer are selected as minimized features using the non-overlap area distribution measurement method. The proposed methodology shows that sensitivity, specificity, and accuracy are 95%, 97.25%, and 96.125%, respectively.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.8
• /
• pp.731-738
• /
• 2011

In this paper, we introduce an advanced architecture of K-Means clustering-based polynomial Radial Basis Function Neural Networks (p-RBFNNs) designed with the aid of SSOA (Space Search Optimization Algorithm) and develop a comprehensive design methodology supporting their construction. In order to design the optimized p-RBFNNs, a center value of each receptive field is determined by running the K-Means clustering algorithm and then the center value and the width of the corresponding receptive field are optimized through SSOA. The connections (weights) of the proposed p-RBFNNs are of functional character and are realized by considering three types of polynomials. In addition, a WLSE (Weighted Least Square Estimation) is used to estimate the coefficients of polynomials (serving as functional connections of the network) of each node from output node. Therefore, a local learning capability and an interpretability of the proposed model are improved. The proposed model is illustrated with the use of nonlinear function, NOx called Machine Learning dataset. A comparative analysis reveals that the proposed model exhibits higher accuracy and superb predictive capability in comparison to some previous models available in the literature.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.3
• /
• pp.1336-1356
• /
• 2017

The spectrum scarcity crisis has resulted in a shortage of resources for many emerging wireless services, and research on dynamic spectrum management has been used to solve this problem. Game theory can allocate resources to users in an economic way through market competition. In this paper, we propose a bidding game-based spectrum allocation mechanism in cognitive radio network. In our framework, primary networks provide heterogeneous wireless service and different numbers of channels, while secondary users have diverse bandwidth demands for transmission. Considering the features of traffic and QoS demands, we design a weighted interference graph-based grouping algorithm to divide users into several groups and construct the non-interference user-set in the first step. In the second step, we propose the dynamic bidding game-based spectrum allocation strategy; we analyze both buyer's and seller's revenue and determine the best allocation strategy. We also prove that our mechanism can achieve balanced pricing schema in competition. Theoretical and simulation results show that our strategy provides a feasible solution to improve spectrum utilization, can maximize overall utility and guarantee users' individual rationality.

• Journal of Internet Computing and Services
• /
• v.10 no.5
• /
• pp.127-133
• /
• 2009

This paper presents an approach to classify normal and epilepsy from electroencephalogram(EEG) using a neural network with weighted fuzzy membership functions(NEWFM). To extract input features used in NEWFM, wavelet transform is used in the first step. In the second step, the frequency distribution of signal and the amount of changes in frequency distribution are used for extracting twenty-four numbers of input features from coefficients and approximations produced by wavelet transform in the previous step. NEWFM classifies normal and epilepsy using twenty four numbers of input features, and then the accuracy rate is 98%.

• Journal of Internet Computing and Services
• /
• v.9 no.1
• /
• pp.129-135
• /
• 2008

This paper presents a methodology to forecast KOSPI index by extracting fuzzy rules based on the neural network with weighted fuzzy membership functions (NEWFM) and the minimized number of input features using the distributed non-overlap area measurement method. NEWFM classifies upward and downward cases of KOSPI using the recent 32 days of CPPn,m (Current Price Position of day n for n-1 to n-m days) of KOSPI. The five most important input features among CPPn,m and 38 wavelet transformed coefficients produced by the recent 32 days of CPPn,m are selected by the non-overlap area distribution measurement method. For the data sets, from 1991 to 1998, the proposed method shows that the average of forecast rate is 67.62%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.12
• /
• pp.5422-5441
• /
• 2016

Despite the abundant literature in the field, there is still the need to find a time-efficient, highly accurate, easy to deploy and robust localization algorithm for real use. The algorithm only involves minimal human intervention. We propose an enhanced Received Signal Strength Indicator (RSSI) based positioning algorithm for Wi-Fi capable devices, called the Dynamic Weighted Evolution for Location Tracking (DWELT). Due to the multiple phenomena affecting the propagation of radio signals, RSSI measurements show fluctuations that hinder the utilization of straightforward positioning mechanisms from widely known propagation loss models. Instead, DWELT uses data processing of raw RSSI values and applies a weighted posterior-probabilistic evolution for quick convergence of localization and tracking. In this paper, we present the first implementation of DWELT, intended for 1D location (applicable to tunnels or corridors), and the first step towards a more generic implementation. Simulations and experiments show an accuracy of 1m in more than 81% of the cases, and less than 2m in the 95%.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.44 no.7
• /
• pp.25-35
• /
• 2002

In this study, the effect of DEM (Digital Elevation Model) resolution (15m, 30m, 50m, 70m, 100m, 200m, 300m) on the hydrological simulation was examined using the BASINS (Better Assessment Science Integrating point and Nonpoint Source) for the Heukcheon watershed (303.3 ㎢) data from 1998 to 1999. Generally, as the cell size of DEM increased, topographical changes were observed as the original range of elevation decreased. The processing time of watershed delineation and river network needed more time and effort on smaller cell size of DEM. The larger DEM demonstrated had some errors in the junction of river network which might affect on the simulation of water quantity and quality. The area weighted average watershed slope became milder but the length weighted average channel slope became steeper as the DEM size increased. DEM resolution affected substantially on the topographical parameter but less on the hydrological simulation. Considering processing time and accuracy on hydrological simulation, DEM grid size of 100m is recommended for this range of watershed size.

• Journal of Electrical Engineering and Technology
• /
• v.12 no.3
• /
• pp.996-1006
• /
• 2017

Phasor Measurement Unit (PMU) placement is a crucial problem for State Estimation (SE) of the power system, which can ensure that the power network is fully observed. Further, the observation reliability problem of the system has been concerned in the operation conditions. In this paper, based on modified weighted adjacent matrix ($A_w$), an optimal placement method is proposed to solve simultaneously two problems involving the optimal PMU placement problem and the observation reliability enhancement problem of the system. The purpose of the proposed method is to achieve both the minimum total cost and the maximum observation reliability, with a focus on increasing the security of observability, strengthening the observation reliability of buses as well as enhancing the effectiveness of redundancy. Simulations on IEEE 14, 24, 30 and 57 bus test systems are presented to justify the methodology. The results of this study show that the proposed method is not only ensuring the power network having the observability effectively but also enhancing significantly the observation reliability. Therefore, it can be a useful tool for SE of the power system.

• Journal of Internet Computing and Services
• /
• v.10 no.2
• /
• pp.179-186
• /
• 2009

This paper presents an approach to classify normal and Ventricular Tachycardia/Ventricular Fibrillation(VT/VF) from the Creighton University Ventricular Tachyarrhythmia DataBase(CUDB) using the neural network with weighted fuzzy membership functions(NEWFM). In the first step, wavelet transform is used for producing input values which are used in the next step. In the second step, two numbers of input features are extracted by phase space reconstruction method and peak extraction method using coefficients produced by wavelet transform in the previous step. NEWFM classifies normal and VT/VF beats using two numbers of input features, and then the accuracy rate is 90.13%.