• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38C no.2
• /
• pp.196-201
• /
• 2013

In this paper, we proposed high speed LDPC encoder architecture for DVB-S2 standard. In conventional algorithm, the processes of parity calculations are serial fashion. Therefore conventional algorithm need clocks of number of parity. The proposed LDPC encoding architecture is based on a parallel 360 bits-wise operations. The key issues for realizing high speed are using the two kinds of index addresses and make use of memories efficiently. We implemented a half rate LDPC encoder on an FPGA, and confirmed its maximum throughput is up to 10 Gbps on 100MHz clock.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.25 no.2
• /
• pp.97-104
• /
• 2016

In this study, the qualitative effects of the positional error of carrier pin holes on the planet load sharing characteristics of the three-point suspension gearbox of wind turbines were investigated experimentally. A 35-kW gearbox comprising two planetary gear stages and a parallel gear stage and size one-fourth of that of a 2-MW three-point suspension gearbox was used as the test gearbox. The strain gauges attached to the ring gear teeth of the input planetary gear stage were used for the purpose of this study. The applied loading conditions were 50%, 75%, and 100% of the rated torque, and the mesh load factor was used as the load sharing index. The experimental results indicated that both the magnitude and direction of the positional error of pin holes had a significant effect on the planet load sharing characteristics of the three-point suspension gearbox. In addition, an increase in the applied torque results in uniform load sharing.

• Journal of IKEEE
• /
• v.17 no.1
• /
• pp.29-35
• /
• 2013

In this paper, we suggest an improved Convex Hull algorithm considering sort in plane point set. This algorithm has low computational complexity since processing data are reduced by characteristic of extreme points. Also it obtains a complete convex set with just one processing using an convex vertex discrimination criterion. Initially it requires sorting of point set. However we can't quickly sort because of its heavy operations. This problem was solved by replacing value and index. We measure the execution time of algorithms by generating a random set of points. The results of the experiment show that it is about 2 times faster than the existing algorithm.

• Journal of IKEEE
• /
• v.21 no.1
• /
• pp.92-95
• /
• 2017

In this paper, we proposed a way to improve function of small scale GP-GPU. Instead of using superscalar which increase scheduling-complexity, we suggested the application of simple core to maximize GP-GPU performance. Our studies also demonstrated that simplified Stream Processor is one of the way to achieve functional improvement in GP-GPU. In addition, we found that developing of optimal thread-assigning method in Warp Scheduler for specific application improves functional performance of GP-GPU. For examination of GP-GPU functional performance, we suggested the thread-assigning way which coordinated with Deep-Learning system; a part of Neural Network. As a result, we found that functional index in algorithm of Neural Network was increased to 90%, 98% compared with Intel CPU and ARM cortex-A15 4 core respectively.

• Structural Engineering and Mechanics
• /
• v.55 no.5
• /
• pp.979-999
• /
• 2015

This paper proposes a novel reliability analysis method which computes reliability index, most probable point and probability of failure of uncertain systems more efficiently and accurately with compared to Monte Carlo, first-order reliability and response surface methods. It consists of Initial and Simulation steps. In Initial step, a number of space-filling designs are selected throughout the variables space, and then in Simulation step, performances of most of samples are estimated via interpolation using the space-filling designs, and only for a small number of the samples actual performance function is used for evaluation. In better words, doing so, we use a simple interpolation function called "reduced" function instead of the actual expensive-to-evaluate performance function of the system to evaluate most of samples. By using such a reduced function, total number of evaluations of actual performance is significantly reduced; hence, the method can be called Reduced Function Evaluations method. Reliabilities of six examples including series and parallel systems with multiple failure modes with truncated and/or non-truncated random variables are analyzed to demonstrate efficiency, accuracy and robustness of proposed method. In addition, a reliability-based design optimization algorithm is proposed and an example is solved to show its good performance.

• The KSFM Journal of Fluid Machinery
• /
• v.18 no.2
• /
• pp.67-72
• /
• 2015

The purpose of this study is to achieve energy saving effect of inverter driving multistage centrifugal pump. For determining the operation points in the pump system, the system curves should be obtained experimentally. To get the system curves, three pumps combined in parallel and one pump operated with different rotational speeds. But for variable speed pump system, energy saving rates can not be evaluated from operation efficiencies. That is why operation efficiencies, system curves, duty cycles, and input powers of the pump system were measured by the constructed experimental apparatus. The duty cycle segmented into different flow rates and weighting the average value for each segment by the interval time. The system was operated with two different periods. The mean duty cycles were collected from apartment and found that the system operated at 40% and at 50% or below capacity. Measured energy saving rate was 58.16%. Estimating method of energy saving rate could be more effective operation index than that of operation efficiency.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.11
• /
• pp.569-584
• /
• 2016

In this paper, we present a spatial knowledge extractor implemented in Hadoop MapReduce parallel, distributed computing environment. From a large spatial dataset, this knowledge extractor automatically derives a qualitative spatial knowledge base, which consists of both topological and directional relations on pairs of two spatial objects. By using R-tree index and range queries over a distributed spatial data file on HDFS, the MapReduce-enabled spatial knowledge extractor, MRSPAKE, can produce a web-scale spatial knowledge base in highly efficient way. In experiments with the well-known open spatial dataset, Open Street Map (OSM), the proposed web-scale spatial knowledge extractor, MRSPAKE, showed high performance and scalability.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.29 no.4 s.235
• /
• pp.425-434
• /
• 2005

In order to analyze pulsating flows in elastic blood vessels, a method based on the ALE concept and finite volume method was reformed and modulated to include wall motion of elastic vessels and impedance phase angle(phase difference between wall motion and blood flow). Our study indicated wall shear rates(WSR) were significantly influenced by the wall motion and the impedance phase angle. For larger wall motion more than $5{\%},$ the computed WSR started to deviate from the results of the perturbation theory that assumed smaller wall motion. The study showed that oscillatory shear index increased as the impedance phase angle became more negative like $-70{\circ}\;or\;-80{\circ}$ due to reduced mean WSR and increased amplitude of WSR. This result indicated that hypertensive patients are more vulnerable to atherosclerosis than normal persons because of the role of more negative impedance phase angles usually observed in these patients.

• Korean Journal of Optics and Photonics
• /
• v.9 no.1
• /
• pp.25-30
• /
• 1998

In this paper, a simple scheme of optical interconnections between M input and N output terminals using mode speckle patterns generated in the mulitmode optical fiber is described. It is shown that fully parallel nonblocking interconnections are accomplished by using properly generated mode speckle patterns as destination codes and dectecting the generated mode speckle patterns separately using M multiplexed Vander Lugt matched filters at each output terminal. The orthogonality condition of mode speckle patterns is investigated by either changing the launching angles or rotating the polarization direction of the beam. Preliminary experiments by use of a 4$\times$4 graded refractive-index multimode fiber coupler whose core diameter was 50 ${\mu}{\textrm}{m}$ were performed to show the feasibility of our scheme.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.4
• /
• pp.1922-1940
• /
• 2019

IP address lookup is a function to determine nexthop for a given destination IP address. It takes an important role in modern routers because of its computation time and increasing Internet traffic. TCAM-based IP lookup approaches can exploit the capability of parallel searching but have a limitation of its size due to latency, power consumption, updatability, and cost. On the other hand, multibit trie-based approaches use SRAM which has relatively low power consumption and cost. They reduce the number of memory accesses required for each lookup, but it still needs several accesses. Moreover, the memory efficiency and updatability are proportional to the number of memory accesses. In this paper, we propose a novel architecture using an Indexed Multibit Trie (IMT) which is based on combined TCAM and SRAM. In the proposed architecture, each lookup takes at most two memory accesses. We present how the IMT is constructed so as to be memory-efficient and fast updatable. Experiment results with real-world forwarding tables show that our scheme achieves good memory efficiency as well as fast updatability.