• Journal of Electrical Engineering and Technology
• /
• v.10 no.5
• /
• pp.2062-2069
• /
• 2015

In the series hybrid active power filter (SHAPF) with magnetic flux compensation (MFC), the system current oscillate in the experimental results when adding the same phase harmonic current command in current control block. This condition endangers the security of the SHAPF. Taking the digit period average arithmetic as example, this paper explains the inter-harmonics current oscillation in the experiment. The conclusion is that the SHAPF is unstable to the inter-harmonics current in theory. Limited by the capacity of the inverter, the system current and the inverter output current do not increase to infinite. At last, some methods are proposed to solve this problem. From the practical viewpoint, the voltage feed-forward control is easy to achieve. It can suppress the current oscillation problems, and also improve the filtering effect. The feasibility of the methods is validated by both the emulation and experiment results.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.6A
• /
• pp.782-793
• /
• 2000

This paper presents a simple dynamical prediction scheme of the signal level which is attenuated and varied due to rain fading in satellite communication links using above 10GHz frequency bands. The proposed prediction scheme has four functional blocks for discrete-time low-pass filtering, slope-based prediction, mean-error correction and hybrid fixed/variable prediction margin allocation. Through simulations using Ka-band attenuation data obtained from the data measured over Ku-band by frequency-scaling, it is shown that the slope-based prediction with the mean-error correction has as small standard deviation of prediction error as below 1 dB, and that the error is about 1.5 to 2.5 times as small as that without the mean-error correction. The hybrid prediction margin allocation requires smaller average margin than those of both fixed and variable methods.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.6
• /
• pp.175-186
• /
• 2004

We present a hybrid approach of texture-based method and connected component (CC)-based method for text extraction in complex images. Two primary methods, which are mainly utilized in this area, are sequentially merged for compensating for their weak points. An automatically constructed MLP-based texture classifier can increase recall rates for complex images with small amount of user intervention and without explicit feature extraction. CC-based filtering based on the shape information using NMF enhances the precision rate without affecting overall performance. As a result, a combination of texture and CC-based methods leads to not only robust but also efficient text extraction. We also enhance the processing speed by adopting appropriate region marking methods for each input image category.

• ETRI Journal
• /
• v.38 no.3
• /
• pp.579-588
• /
• 2016

To improve the ability to determine a vehicle's movement information even in a challenging environment, a hybrid approach called non-Gaussian square rootunscented particle filtering (nGSR-UPF) is presented. This approach combines a square root-unscented Kalman filter (SR-UKF) and a particle filter (PF) to determinate the vehicle state where measurement noises are taken as a finite Gaussian kernel mixture and are approximated using a sparse Gaussian kernel density estimation method. During an outage-free GPS period, the updated mean and covariance, computed using SR-UKF, are estimated based on a GPS observation update. During a complete GPS outage, nGSR-UPF operates in prediction mode. Indeed, because the inertial sensors used suffer from a large drift in this case, SR-UKF-based importance density is then responsible for shifting the weighted particles toward the high-likelihood regions to improve the accuracy of the vehicle state. The proposed method is compared with some existing estimation methods and the experiment results prove that nGSR-UPF is the most accurate during both outage-free and complete-outage GPS periods.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.01a
• /
• pp.71-73
• /
• 2014

본 논문에서는 지하철 내의 인구밀집도 파악을 통한 승객의 이용 편의성을 극대화하고, 군중밀도가 높은 지하철 구간의 경우 고른 인구분포 측정을 통한 승객의 편의성을 극대화하는 플랫폼 및 모바일 앱을 제안한다. 제안하는 시스템으로 기존의 지하철 역내에 설치된 CCTV에 모션벡터 영상처리와 RFID 기술을 결합한 Hybrid CDE로 구성되며, Size-Filtering을 통해서 재검출 과정을 거친다. 이러한 결과 값은 전동차 각 구간의 인구밀집도 정보를 정확히 측정 할 수 있다. 또한 결과 값을 바탕으로 효율적인 인구 유동을 유도할 수 있으며 정보 소외 계층 및 사회적 약자 등, 승객을 안전하게 보호할 수 있는 환경을 조성한다. 시스템 관리자는 학습기능 알고리즘을 통해서 오차 범위를 최소화한 플랫폼 설계를 통해 실시간 모니터링 함으로써 정보 습득 및 제공면에서도 새로운 시스템 설계 제안이 될 것이다.

• Proceedings of the KIEE Conference
• /
• 2003.07b
• /
• pp.1096-1098
• /
• 2003

In this paper, the control algorithm and control methods for a combined system of shunt passive filter and series active filter in 3-phase 4-wire system are discussed. Moreover, the 3-phase 4-wire system is widely employed in distributing electric energy to several office building and manufacturing plants. In such systems, the third harmonic and odd multiples of $3^{rd}$($9^{th}$, $15^{th}$, etc.) are termed as triple and zero sequence components that do not cancel each other in the system neutral. As a result, the triple harmonics add together creating a primary source of excessive neutral current. Regarding this concern, this paper presents a new control scheme for a series hybrid active system. This series active power filter acts not only as a harmonic compensator but also as a harmonic isolator. Hence the required rating of the series active filter is much smaller than that of a conventional shunt active filter. However, the performance of the combined system is greatly influenced by the filtering algorithm employed in the active power filter. This paper proposes a series active power filter scheme based on performance function. Some experiments was executed and experimental results from a prototype active power filter confirm the suitability of the proposed approach.

• The Korean Journal of Applied Statistics
• /
• v.33 no.6
• /
• pp.683-696
• /
• 2020

The aim of this research is to find an optimal gene set that provides highly accurate multi-class classification with a minimum number of genes. A two-stage procedure is proposed: Based on minimum redundancy and maximum relevance (mRMR) framework, several statistics to rank differential expression genes and K-means clustering to reduce redundancy between genes are used for data filtering procedure. And a particle swarm optimization is modified to select a small subset of informative genes. Two well known multi-class microarray data sets, ALL and SRBCT, are analyzed to indicate the effectiveness of this hybrid method.

• International Journal of Computer Science & Network Security
• /
• v.22 no.10
• /
• pp.237-245
• /
• 2022

The prominence of IoTs (Internet of Things) and exponential advancement of computer networks has resulted in massive essential applications. Recognizing various cyber-attacks or anomalies in networks and establishing effective intrusion recognition systems are becoming increasingly vital to current security. MLTs (Machine Learning Techniques) can be developed for such data-driven intelligent recognition systems. Researchers have employed a TFDNNs (Tensor Flow Deep Neural Networks) and DCNNs (Deep Convolution Neural Networks) to recognize pirated software and malwares efficiently. However, tuning the amount of neurons in multiple layers with activation functions leads to learning error rates, degrading classifier's reliability. HTFDNNs ( Hybrid tensor flow DNNs) and MRNs (Modified Residual Networks) or Resnet CNNs were presented to recognize software piracy and malwares. This study proposes HTFDNNs to identify stolen software starting with plagiarized source codes. This work uses Tokens and weights for filtering noises while focusing on token's for identifying source code thefts. DLTs (Deep learning techniques) are then used to detect plagiarized sources. Data from Google Code Jam is used for finding software piracy. MRNs visualize colour images for identifying harms in networks using IoTs. Malware samples of Maling dataset is used for tests in this work.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.11
• /
• pp.3030-3049
• /
• 2023

With the development and wide application of hybrid network computing modes like cloud computing, edge computing and fog computing, the customer service requests and the collaborative optimization of various computing resources face huge challenges. Considering the characteristics of network environment resources, the optimized deployment of service resources is a feasible solution. So, in this paper, the optimal goals for deploying service resources are customer experience and service cost. The focus is on the system impact of deploying services on load, fault tolerance, service cost, and quality of service (QoS). Therefore, the alternate node filtering algorithm (ANF) and the adjustment factor of cost matrix are proposed in this paper to enhance the system service performance without changing the minimum total service cost, and corresponding theoretical proof has been provided. In addition, for improving the fault tolerance of system, the alternate node preference factor and algorithm (ANP) are presented, which can effectively reduce the probability of data copy loss, based on which an improved cost-efficient replica deployment strategy named ICERD is given. Finally, by simulating the random occurrence of cloud node failures in the experiments and comparing the ICERD strategy with representative strategies, it has been validated that the ICERD strategy proposed in this paper not only effectively reduces customer access latency, meets customers' QoS requests, and improves system service quality, but also maintains the load balancing of the entire system, reduces service cost, enhances system fault tolerance, which further confirm the effectiveness and reliability of the ICERD strategy.

• Asia pacific journal of information systems
• /
• v.29 no.4
• /
• pp.752-770
• /
• 2019

The purpose of this study is to incorporate telemarketing processes to improve telemarketing performance. For this application, we have attempted to mix the model of machine learning to extract potential customers with personalisation techniques to derive recommended products from actual contact. Most of traditional recommendation systems were mainly in ways such as collaborative filtering, which predicts items with a high likelihood of future purchase, based on existing purchase transactions or preferences for products. But, under these systems, new users or items added to the system do not have sufficient information, and generally cause problems such as a cold start that can not obtain satisfactory recommendation items. Also, indiscriminate telemarketing attempts can backfire as they increase the dissatisfaction and fatigue of customers who do not want to be contacted. To this purpose, this study presented a multi-purpose hybrid recommendation algorithm to achieve two goals: to select customers with high possibility of contact, and to recommend products to selected customers. In addition, we used subscription data from telemarketing agency that handles insurance products to derive realistic applicability of the proposed recommendation system. Our proposed recommendation system would certainly solve the cold start and scarcity problem of existing recommendation algorithm by using contents information such as customer master information and telemarketing history. Also. the model could show excellent performance not only in terms of overall performance but also in terms of the recommendation success rate of the unpopular product.