• 한국위성정보통신학회논문지
• /
• 제10권2호
• /
• pp.102-108
• /
• 2015

일반적으로 TDMA(Time Division Multiple Access)를 기반으로 하는 양방향 위성 통신 시스템에서는 중심국과 멀리 떨어진 단말국간의 네트워크 동기를 필요로 한다. 본 논문은 양방향 위성시스템을 위한 국제규격인 DVB-S2/RCS2에서 제시한 NCR(Network Clock Reference) 복원 구조를 기반으로 네트워크 클럭 동기에 대한 개념을 설명한다. 또한 DVB-S2에서 CCM(Constant Coding & Modulation) 운용 모드와 부호율, 변조방식을 바꿔 throughput을 최적화 하는 ACM(Adaptive Coding & Modulation) 운용 모드를 지원하는 새로운 NCR 삽입 구조를 제안하였으며 이에 대한 시뮬레이션을 통해 타당성을 검증하였다.

• Journal of Communications and Networks
• /
• 제13권2호
• /
• pp.160-166
• /
• 2011

In u-healthcare services based on wireless body sensor networks, reliable connection is very important as many types of information, including vital signals, are transmitted through the networks. The transmit power requirements are very stringent in the case of in-body networks for implant communication. Furthermore, the wireless link in an in-body environment has a high degree of path loss (e.g., the path loss exponent is around 6.2 for deep tissue). Because of such inherently bad settings of the communication nodes, a multi-hop network topology is preferred in order to meet the transmit power requirements and to increase the battery lifetime of sensor nodes. This will ensure that the live body of a patient receiving the healthcare service has a reduced level of specific absorption ratio (SAR) when exposed to long-lasting radiation. We propose an efficientmethod for delivering delay-intolerant data packets over multiple hops. We consider forward error correction (FEC) in an erasure correction mode and develop a mathematical formulation for packet-level scheduling of delay-intolerant FEC packets over multiple hops. The proposed method can be used as a simple guideline for applications to setting up a topology for a medical body sensor network of each individual patient, which is connected to a remote server for u-healthcare service applications.

• 동력기계공학회지
• /
• 제14권6호
• /
• pp.67-74
• /
• 2010

Many tanks are installed in ship and marine structures. They are often in contact with inner or outer fluid, like ballast, fuel and cargo tanks. Fatigue damages are sometimes observed in these tanks which seem to be caused by resonance with exciting force of engine and propeller. Vibration characteristics of these thin walled tanks in contact with fluid near engine and propeller are strongly affected by added mass of containing fluid. Therefore it is essentially important to estimate the added mass effect to predict vibration of the tanks. Many authors have studied vibration of cylindrical and rectangular tanks containing fluid. Few research on dynamic interaction among tank walls through fluid are reported in the vibration of rectangular tanks recently. In case of rectangular tanks, structural coupling between adjacent panels and effect of vibration modes of multiple panels on added mass have to be considered. In the previous report, A numerical tool of vibration analysis of a 3-dimensional tank is developed by using finite element method for plates and boundary element method for fluid region. In this paper, the coupling effect between panels of a tank on added mass of containing fluid, the effect of structural constraint between panels on each vibration mode for fluid region and mode characteristics in accordance with changing breadth of the plates are investigated numerically and discussed.

• Journal of Information Processing Systems
• /
• 제17권3호
• /
• pp.556-570
• /
• 2021

The existing video expression recognition methods mainly focus on the spatial feature extraction of video expression images, but tend to ignore the dynamic features of video sequences. To solve this problem, a multi-mode convolution neural network method is proposed to effectively improve the performance of facial expression recognition in video. Firstly, OpenFace 2.0 is used to detect face images in video, and two deep convolution neural networks are used to extract spatiotemporal expression features. Furthermore, spatial convolution neural network is used to extract the spatial information features of each static expression image, and the dynamic information feature is extracted from the optical flow information of multiple expression images based on temporal convolution neural network. Then, the spatiotemporal features learned by the two deep convolution neural networks are fused by multiplication. Finally, the fused features are input into support vector machine to realize the facial expression classification. Experimental results show that the recognition accuracy of the proposed method can reach 64.57% and 60.89%, respectively on RML and Baum-ls datasets. It is better than that of other contrast methods.

• 소음진동
• /
• 제11권1호
• /
• pp.118-124
• /
• 2001

The effect of a concentrated mass on the regions of dynamic instability of an axially oscillating cantilever beam is investigated in this paper. The equations of motion are derived using Kane's method and the assumed mode method. It is found that the bending stiffness is harmonically varied by axial inertia forces due to oscillating motion. Under the certain conditions between oscillating frequency and the natural frequencies, dynamic instability may occur and the magnitude of the bending vibration increase without bound. By using the multiple time scales method, the regions of dynamic instability are obtained. The regions of dynamic instability are found to be depend on the magnitude of a concentrated mass or its location.

• 한국통신학회논문지
• /
• 제31권1A호
• /
• pp.65-71
• /
• 2006

IEEE 802.11b 무선 네트워크에서 스테이션들이 저전력 모드로 동작할 때, 액세스 포인트(AP)가 주기적으로 비콘(beacon)을 전송하여 스테이션들을 관리한다. 비콘을 수신한 후 스테이션들이 전송매체에 대해 경쟁하고 전송매체가 이용 가능할 때까지 전송매체의 상태를 계속해서 검사한다. 따라서 각 스테이션에서 에너지 소모는 무선 네트워크의 부하에 따라 증가한다. 이 논문에서 다수의 스테이션들이 있는 무선 네트워크에서 가상비트맵을 제어하여 스테이션에서 에너지 소모를 줄이는 방법을 제안한다. IEEE 802.11b 무선 인터페이스의 저전력 모드의 문제점이 설명되었고 에너지 소모를 줄이기 위한 새로운 방법이 제안되었다. 네트워크 시뮬레이터인 ns2를 이용하여 제안된 방법에 대해 모의실험이 수행되었고 IEEE 802.11b의 저전력 모드와 비교했다. 스테이션이 깨어있는 시간으로 측정된 결과는 무선 네트워크에서 다수의 스테이션들이 AP를 경유하여 통신이 이루어질 때 에너지 소모가 개선되는 것을 보여준다.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2003년도 봄 학술발표회 논문집
• /
• pp.382-387
• /
• 2003

In this paper, a neural networks-based damage detection method using the modal properties is presented, which can effectively reduce the effect of the modeling errors in the baseline finite element model from which the training patterns for the networks are to be generated. The differences or the ratios of the mode shape components between before and after damage are used as the input to the neural networks in this method, since they are found to be less sensitive to the modeling errors than the mode shapes themselves. Results of laboratory test on a simply supported bridge model and field test on a bridge with multiple girders confirm the applicability of the present method.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2002년도 하계학술대회 논문집 D
• /
• pp.2337-2340
• /
• 2002

This paper presents the method of AGV(Automatic guided vehicle)'s moving environment(plane, corner, edge) recognition using SONAR sensor configuration. As for the SONAR sensor, the Crosstalk effect has been generally considered as an inevitable noisy phenomenon in the indoor environment. However, this effect can be used as a clue for classifying and localizing targets in the indoor environment if those can be controlled and used well. EERUF(error eliminate rapid ultrasonic firing) is a method for firing multiple ultrasonic sensors in mobile robot application and multi-echo mode of POLARIOD Device can reduce the Crosstalk effect. Here, Crosstalk effect was reduced using EERUF and applied to the AGV with a simple wall-following algorithm in the indoor environment. This method was tesed by a typical AGV with multi SONAR sensors in the laboratory environment.

• 한국정보디스플레이학회:학술대회논문집
• /
• 한국정보디스플레이학회 2009년도 9th International Meeting on Information Display
• /
• pp.91-94
• /
• 2009

We propose advanced liquid crystal display (LCD) modes through surface modification using UV curable reactive mesogen (RM) mixed with alignment layers. The LC directors on the modified alignment layer are controlled and memorized by the polymerized RMs under an applied voltage. Using the method proposed here, we can improve the response time and viewing angle characteristics through surface controlled patterned vertical alignment (SC-PVA) mode and 8-domains PVA mode, respectively. Also, we found that the method can be applicable to fabricate multi-domain structures using multiple UV exposure as same as conventional UV alignment method.

• Structural Engineering and Mechanics
• /
• 제33권6호
• /
• pp.747-763
• /
• 2009

In this work, the linear vibration characteristics of $90^{\circ}$ pipe bends and their cylindrical and toroidal shell components are studied. The finite element method, based on shear-deformation shell elements, is used to carry out a vibration analysis of metallic multiple $90^{\circ}$ mitred pipe bends. Single, double, and triple mitred bends are considered, as well as a smooth bend. Sample natural frequencies and mode shapes are given. To validate the procedure, comparison of the natural frequencies is made with existing results for cylindrical and toroidal shells. The influence of the multiplicity of the bend, the boundary conditions, and the various geometric parameters on the natural frequency is described. The differential quadrature method, based on classical shell theory, is used to study the vibration of components of these bends. Regression formulas are derived for cylindrical shells (straight pipes) with one or two oblique edges, and for sectorial toroidal shells (curved pipes, pipe elbows). Two types of support are considered for each case. The results given provide information about the vibration characteristics of pipe bends over a wide range of the geometric parameters.