• 한국전산유체공학회:학술대회논문집
• /
• 1998.05a
• /
• pp.35-42
• /
• 1998

Time marching algorithms applied to compressible Navier-Stokes equation have a convergence problem at low Mach number. It is mainly due to the eigenvalue stiffness and pressure singularity as Mach number approaches to zero. Among the several methods to overcome the shortcomings of time marching scheme, time derivative preconditioning method have been used successfully. In this numerical analysis, we adopted a preconditioner of K.H. Chen and developed a two-dimensional, axisymmetric Navier-Stokes program. The steady state driven cavity flow and backward facing step flow problems were computed to confirm the accuracy and the robustness of preconditioned algorithm for low Mach number flows. And the transonic and supersonic flows insice the JPL axisymmetric nozzle internal flow is exampled to investigate the effects of preconditioning at high Mach number flow regime. Test results showed excellent agreement with the experimental data.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.46 no.9
• /
• pp.24-30
• /
• 2009

This paper presents a simple resource allocation scheme for throughput enhancement in relay based orthogonal frequency division multiple access (OFDMA) cellular systems. The resource allocation schemes, which are based on the optimization problem, have high computational complexity. That is why a searching process is required on the overall allocable resources. Since these schemes should be performed in real time, we propose a simple resource allocation scheme which has very low computational complexity. Firstly, we formulate the optimization problem and draw observations for throughput maximization. Based on observations, we propose a three step allocation scheme that separates the allocable resources into three (i.e. relay, frequency and time). By doing so, the computational complexity can be reduced. Simulation results show that the proposed scheme has near-optimum performance in spite of its low computational complexity.

• Journal of applied mathematics & informatics
• /
• v.22 no.1_2
• /
• pp.453-460
• /
• 2006

A discrete-time priority queueing system with place reservation discipline is studied, in which two different types of packets arrive according to batch geometric streams. It is assumed that there is a reserved place in the queue. Whenever a high-priority packet enters the queue, it will seize the reserved place and make a new reservation at the end of the queue. Low-priority arrivals take place at the end of the queue in the usual way. Using the probability generating function method, the joint distribution of system state and the delay distribution for each type are obtained.

• Advances in biomechanics and applications
• /
• v.1 no.2
• /
• pp.127-141
• /
• 2014

A preliminary study of a new pulsatile pump that will work to a frequency greater than 1 Hz, is presented. The fluid-structure interaction between a Newtonian blood flow and a piston drive that moves with periodic speed is simulated. The mechanism is of double effect and has four valves, two at the input flow and two at the output flow; the valves are simulated with specified velocity of closing and reopening. The simulation is made with finite elements software named COMSOL Multiphysics 3.3 to resolve the flow in a preliminary planar configuration. The geometry is 2D to determine areas of high speeds and high shear stresses that can cause hemolysis and platelet aggregation. The opening and closing valves are modelled by solid structure interacting with flow, the rhythmic opening and closing are synchronized with the piston harmonic movement. The boundary conditions at the input and output areas are only normal traction with reference pressure. On the other hand, the fluid structure interactions are manifested due to the non-slip boundary conditions over the piston moving surfaces, moving valve contours and fix pump walls. The non-physiologic frequency pulsatile pump, from the viewpoint of fluid flow analysis, is predicted feasible and with characteristic of low hemolysis and low thrombogenesis, because the stress tension and resident time are smaller than the limit and the vortices are destroyed for the periodic flow.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.11
• /
• pp.62-71
• /
• 1996

In this paper we propose a method that improves PSNR at low bit rate and reduces computational complexity in fractal image coding based on discrete wavelet transform. The proposed method, which uses significant coefficient tree, improves PSNR of the reconstructed image and reduces computational comlexity of mapping domain block onto range block by matching only the significant coefficients of range block to coefficients of domain block. Also, the proposed method reduces error propagation form lower resolution subbands to higher resolution subbands by correcting error of lower resolution subbands. Some experimental results confirm that the proposed method reduces encoding and decoding time significantly and has fine reconstructed images having no blocking effect and clear edges at low bit rate.

• International Journal of Fluid Machinery and Systems
• /
• v.4 no.3
• /
• pp.341-348
• /
• 2011

The study of bacterial flagellar swimming motion remains an interesting and challenging research subject in the fields of hydrodynamics and bio-locomotion. This swimming motion is characterized by very low Reynolds numbers, which is unique and time reversible. In particular, the effect of rotation of helical flagella of bacterium on swimming motion requires detailed multi-disciplinary analysis. Clear understanding of such swimming motion will not only be beneficial for biologists but also to engineers interested in developing nanorobots mimicking bacterial swimming. In this paper, computational fluid dynamics (CFD) simulation of a three dimensional single flagellated bacteria has been developed and the fluid flow around the flagellum is investigated. CFD-based modeling studies were conducted to find the variables that affect the forward thrust experienced by the swimming bacterium. It is found that the propulsive force increases with increase in rotational velocity of flagellum and viscosity of surrounding fluid. It is also deduced from the study that the forward force depends on the geometry of helical flagella (directly proportional to square of the helical radius and inversely proportional to pitch).

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.12
• /
• pp.2791-2797
• /
• 2010

In recent years, the goal of providing high speed wireless data services has generated a great amount of interest among the research community. Several researchers have shown that the capacity of the system, in the presence of flat Rayleigh fading, improves significantly with the use of combined MIMO and LDPC. To feed the soft values to LDPC decoder, the soft values must be calculated from multiple transmitter and receiver antennas in Rayleigh fading channel. It requires high computational complexity to get the soft symbols by increasing number of transmitter and receiver antennas. Therefore, this thesis proposed on effective algorithm for calculation of soft values from multiple antennas based on LLR. As result, This thesis shows that maximum 61% of computational complexity is reduced with a little loss of performance.

• 유체기계공업학회:학술대회논문집
• /
• 2002.12a
• /
• pp.337-345
• /
• 2002

The present study addresses a computational result of unsteady gas flow through a critical nozzle. The axisymmetric, unsteady, compressible, Wavier-Stokes equations are solved using a finite volume method that makes use of the second order upwind scheme for spatial derivatives and the multi-stage Runge-Kutta integral scheme for time derivatives. The steady solutions of the governing equation system are validated with the previous experimental data to ensure that the present computational method is valid to predict the critical nozzle flows. In order to simulate the effects of back pressure fluctuations on the critical nozzle flows, an excited pressure oscillation with an amplitude and frequency is assumed downstream of the exit of the critical nozzle. The results obtained show that for low Reynolds numbers, the unsteady effects of the pressure fluctuations can propagate upstream of the throat of critical nozzle, and thus giving rise to the applicable fluctuations in mass flow rate through the critical nozzle, while for high Reynolds numbers, the pressure signals occurring at the exit of the critical nozzle do not propagate upstream beyond the nozzle throat. For very low Reynolds number, it is found that the sonic line near the throat of the critical nozzle remarkably fluctuateswith time, providing an important mechanism for pressure signals to propagate upstream of the nozzle throat, even in choked flow conditions. The present study is the first investigation to clarify the unsteady effects on the critical nozzle flows.

• Journal of Computing Science and Engineering
• /
• v.10 no.2
• /
• pp.68-73
• /
• 2016

An efficient texture compression method is proposed based on a block matching process between the current block and the previously encoded blocks. Texture mapping is widely used to improve the quality of rendering results in real-time applications. For fast texture mapping, it is important to find an optimal trade-off between compression efficiency and computational complexity. Low-complexity methods (e.g., ETC1 and DXT1) have often been adopted in real-time rendering applications because conventional compression methods (e.g., JPEG) achieve a high compression ratio at the cost of high complexity. We propose a block matching-based compression method that can achieve a higher compression ratio than ETC1 and DXT1 while maintaining computational complexity lower than that of JPEG. Through a comparison between the proposed method and existing compression methods, we confirm our expectations on the performance of the proposed method.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.5
• /
• pp.187-193
• /
• 2021

Our final goal is to implement nursing robots that can recognize patient's faces and their medicine on prescription. They can help patients to take medicine on time and prevent its abuse for recovering their health soon. As the first step, we proposed a medicine classifier with a low computational network that is able to run on embedded PCs without GPU in order to be applied to universal nursing robots. We confirm that our proposed model called MedicineNet achieves an 99.99% accuracy performance for classifying 15 kinds of medicines and background images. Moreover, we realize that the calculation time of our MedicineNet is about 8 times faster than EfficientNet-B0 which is well known as ImageNet classification with the high performance and the best computational efficiency.