• The Journal of the Korea institute of electronic communication sciences
• /
• v.5 no.2
• /
• pp.145-151
• /
• 2010

This paper proposes a reverse filtering system of sound field obtaining a state of sound field transmitted from two sounds, using an adaptive filter and neural network. The proposed system uses the reverse filtering method with calculating and renewing a coefficient of a filter, using least mean square. Based on training the neural network, experiments confirm that the proposed system is effective for a simple waveform with non-linear distortion, by using neural network and adaptive filtering method.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.695-698
• /
• 2009

본 논문에서는 음원으로부터 나온 음과 동일한 음을 들을 수 있는 시스템을 구축하는 것을 목적으로 하여 이 두 개의 음으로부터 전달되어온 음장의 상태를 구하여 이 역 필터를 구성하는 방법을 연구한다. 본 논문에서는 최소 2승 평균법(Least Mean Square, LMS)을 사용하여 FIR 필터(Finite Impulse Response)의 계수를 계산하여 이를 갱신함으로써 역 필터법을 구축하는 방법을 사용한다. 또한 이 방법과는 별도로 LMS법의 부분을 신경회로망에 대처하는 알고리즘을 제안하였다. 시뮬레이션 실험으로부터 상당히 간단한 파형에 비선형인 왜곡이 있는 것을 본 논문에서 제안한 신경회로망에 의한 학습 가능한 것을 확인하였다.

• 한국정보통신설비학회:학술대회논문집
• /
• 2005.08a
• /
• pp.313-317
• /
• 2005

The network ingress filtering is a simple and efficient method for preventing IP source spoofing of fixed nodes. Since mobile hosts cannot communicate with its correspondent nodes if the network ingress filtering is configured in mobile IPv4 network, reverse tunneling was considered as a method for avoiding network ingress filtering. But, unfortunately this method does not solve IP source spoofing of mobile nodes. In this paper, we propose a simple and efficient method for preventing IP source spoofing of mobile nodes assuming that only the mobile hosts connected to foreign agents and the network that foreign agent manages is small.

• The Korean Journal of Applied Statistics
• /
• v.17 no.3
• /
• pp.535-545
• /
• 2004

The clustering method suggested in this paper produces clusters based on the 'rules of variables' by merging the 'training' and the identically structured reference data and then by filtering it to obtain the clusters of the 'training data' through the use of the 'tree classification model'. The reference dataset is generated by spatially contrasting it to the 'training data' through the 'reverse arcing' algorithm to effectively identify the clusters. The strength of this method is that it can be applied even to the mixture of continuous and discrete types of 'training data' and the performance of this algorithm is illustrated by applying it to the simulated data as well as to the actual data.

• Journal of Advanced Marine Engineering and Technology
• /
• v.28 no.8
• /
• pp.1313-1323
• /
• 2004

In accordance with adoption of new convention for the control of ship's ballast water at the diplomatic conference held in London Feb, 2004, every country has to regulate the ballast water and deposit matters. When this Resolution comes into effect in 2009, all vessels engaged in international voyage must have ballast water control program, ballast water records, equipments which are suitable to the standard of exchange and performance for the ballast water. This study estimates objectively their performances, merits and demerits of the ballast water treatment technique and exchanging techniques for safe operation of ships. It is desirable to design an equipment to control the ballast water using the brush-type vacuum suction nonstop reverse cleaning system to overcome the clogging phenomenon and the direct disc filtering to maximize filtering area for the optimum process considering biological availabilities. It will be expected to protect against marine pollution and to maintain clean sea if it is secured to develop new ballast water treatment techniques. And it will also be expected to cope with the Resolution and each regulation of the developed countries from the ballast water.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.42 no.4 s.304
• /
• pp.25-34
• /
• 2005

We had research to conduct interrelationships between sensors using postprocessing analysis with the Fuzzy-Kalman Filtering Auto-Correlation about Real Time Simulator data of the NaroSC LCS in case of a fully blind situation scenario. The conducted interrelations are same harmony with relations in scenario. We had analyzed signals of sensors reverse-using a optimization character of Fuzzy-Kalman Filter. As our research conclusion, We had recognized possibilities of signal processing about the KSLV-1, on-board payloads, general equipments of ground support which apply to multi sensor systems.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.37 no.4
• /
• pp.1-8
• /
• 2000

A defect of decrease in TCP throughput can be investigated in asymmetric environment of different uplink and downlink bandwidths. Under two-way TCP traffic, the total link utilization is decreased by the successive injection of data packets in buffer. To solve these problems, terminal ACK filtering and packet scheduling mechanisms are introduced in this paper. ACK filtering eliminates the buffered ACK packets and transmits recent ACK packets in the uplink with limited bandwidth. Packet scheduling is the method of preventing 'clustering' and 'ack compression' states which are generated in the two-way TCP traffic. The guarantee of the data traffic in reverse TCP connection and the high throughput in forward TCP connection are investigated by simulation.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10c
• /
• pp.664-666
• /
• 1999

Mobile IP 기술에 의해 시간 및 장소의 제한을 받지 않고 사용자가 원격지 또는 이동 중에도 정보를 처리할 수 있는 환경이 제공된다. 패킷 필터링 기술 같은 경우에는 패킷의 출발지, 목적지 주소뿐만 아니라 패킷의 이동 방향도 고려하므로 Mobile IP의 동작이 방해를 받게 될 수가 있다. 이러한 문제점을 해결하기 위해 나온 것이 역 터널링 기법인데 이 경우는 효율성이 떨어지므로 상황에 따라 역 터널링을 수행하는 선택적 역터널링 기법에 나오게 되었다. 효율성을 증진시킨 선택적 역 터널링 기법은 외부 네트워크의 보안 정책에 위배되는 패킷 전송이 이루어질 수도 있으므로 새로운 보안 정책이 절실히 요구된다. 본 논문에서는 선택적 역 터널링시 외부 에이전트가 패킷 필터링 기능을 갖는 기법에 대해 제안한다. 이 기법을 적용함으로써 이동노드가 선택적 역 터널링시 외부 네트워크의 보안 정책에 부합되게 행동할 수 있도록 하였다.

• Biomolecules & Therapeutics
• /
• v.15 no.2
• /
• pp.123-126
• /
• 2007

A sensitive, simple and highly selective liquid chromatography method of determination for extraction of pseudoephedrine hydrochloride from Allegra D tablet was developed. The chief benefit of the present method is the minimal sample preparation, as the procedure is only filtering through pore syringe filter. Two drugs (pseudoephedrine hydrochloride, fexofenadine) were separated on a C$_{18}$ column and analyzed by high performance liquid chromatography (HPLC). The method had a chromatographic run time of 8.0 min. 1 ml of pseudoephedrine hydrochloride solution (1 mg/ml) was filtered through 0.22 um pore syringe filter. 50 ul of filtering solution was injected to HPLC pump and we knew the retention time (1.85 min) of separating of pseudoephedrine hydrochloride using UV detector at 280 nm. We used C$_{18}$ column (4.6 mm${\times}$250 mm), mobile phase solution (<0.05 mol/L NaH$_2$PO$_4$, 2 ml/L H$_3$PO$_4$>/CH$_3$CN / sodium dodesyl sulfate = 60 ml / 40 ml / 1 g). We separated psedoephedrine hydrochloride at run time of 1.85 min from Allegra D tablet solution (1 mg/ml) filtered through 0.22 um pore syringe filter using UV detector at 280 nm. Flow rate was set at 1.0 ml/min and the column temperature was set at 40$^{\circ}C$. Psedoephedrine hydrochloride solution (1 mg/ml) separated from Allegra D tablet was filtered through 0.22 um pore syringe filter and injected 50 ul. We confirmed the peak of psedoephedrine hydrochloride at same retention time and the separating solution was freeze-dried. In conclusion, A simple isocratic reverse-phase HPLC method has been developed that provides excellent separation of pseudoephedrine from Allegra D tablet.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.3
• /
• pp.129-134
• /
• 2015

Since automatic social engineering based spam attacks induce for users to click or receive the short message service (SMS), e-mail, site address and make a relationship with an unknown friend, it is very easy for them to active in online social networks. The previous spam detection schemes only apply manual filtering of the system managers or labeling classifications regardless of the features of social networks. In this paper, we propose the spam detection metric after reflecting on a couple of features of social networks followed by analysis of real social network data set, Twitter spam. In addition, we provide the online social networks based unsupervised scheme for automated social engineering spam with self organizing map (SOM). Through the performance evaluation, we show the detection accuracy up to 90% and the possibility of real time training for the spam detection without the manager.