• Journal of Broadcast Engineering
• /
• v.24 no.2
• /
• pp.251-263
• /
• 2019

Recently the huge demand for Internet media streaming has dramatically increased the cost of the CDN (Content Delivery Network) and the need for a means to reduce it is increasing day by day. In this situation, a P2P-assisted DASH technology has recently emerged, which uses P2P (Peer-to-Peer) communications based on WebRTC (Web Real-Time Communication) standards to reduce the CDN cost. This paper proposes an algorithm that can significantly improve CDN cost savings in this technology by selecting peers based on the transport history. Also we implemented this algorithm in an experimental system and, after setting experimental conditions that emulate the actual mobile network environment, we measured the performance of the experimental system. As a result, we demonstrated that the proposed algorithm can achieve higher CDN cost savings compared to the conventional algorithm where peers are selected at random.

• Journal of Broadcast Engineering
• /
• v.24 no.2
• /
• pp.281-291
• /
• 2019

Depth from defocus estimates the 3D depth by using a phenomenon in which the object in the focal plane of the camera forms a clear image but the object away from the focal plane produces a blurred image. In this paper, algorithms are studied to estimate 3D depth by analyzing the degree of blur of the image taken with a single camera. The optimized object range was obtained by 3D depth estimation derived from depth from defocus using one image of a single camera or two images of different focus of a single camera. For depth estimation using one image, the best performance was achieved using a focal length of 250 mm for both smartphone and DSLR cameras. The depth estimation using two images showed the best 3D depth estimation range when the focal length was set to 150 mm and 250 mm for smartphone camera images and 200 mm and 300 mm for DSLR camera images.

• Journal of the Institute of Convergence Signal Processing
• /
• v.21 no.2
• /
• pp.73-78
• /
• 2020

Musculoskeletal disease is often caused by sitting down for long period's time or by bad posture habits. In order to prevent musculoskeletal disease in daily life, it is the most important to correct the bad sitting posture to the right one through real-time monitoring. In this study, to detect the sitting information of user's without any constraints, we propose posture measurement system based on multi-channel pressure sensor and CNN model for classifying sitting posture types. The proposed CNN model can analyze 5 types of sitting postures based on sitting posture information. For the performance assessment of posture classification CNN model through field test, the accuracy, recall, precision, and F1 of the classification results were checked with 10 subjects. As the experiment results, 99.84% of accuracy, 99.6% of recall, 99.6% of precision, and 99.6% of F1 were verified.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.5
• /
• pp.157-163
• /
• 2022

As the online game market grows, the use of game bots is causing the most serious problem for game services. We propose a harvest coordinate analysis model to detect harvesting bots among game bots of the Massively Multiplayer Online Role-Playing Games(MMORPGs) genre. The proposed model analyzes the player's harvesting behavior using the coordinate data. Game bots can obtain in-game goods and items more easily than normal players and are not affected by realistic restrictions such as sleep time and character manipulation fatigue. As a result, there is a difference in harvesting coordinates between normal players and game bots. We divided the coordinate zones and used these coordinate zone differences to distinguish between game bot players and normal players. We created a dataset with NCSoft's AION log and applied it to a random forest model to detect game bots, and as a result, we derived performance with a recall of 0.72 and a precision of 0.92.

• Composites Research
• /
• v.34 no.6
• /
• pp.380-386
• /
• 2021

This paper aims to improve the critical speed of power-take-off (PTO) shafts by using carbon fiber reinforced plastics (CFRPs). The PTO shaft was designed with titanium-CFRPs hybrid structure in order to compensate the low shear strength of CFRPs. Based on the requirements for PTO shafts, the dimensions of PTO shafts were determined through a parametric study. To evaluate the performance of the PTO shaft, a vibration test, a static torsion test, and a torsion durability test were performed. In the vibration test, the critical speed of PTO shafts was 20570 rpm, which was 7.5% higher than that of titanium shafts. Additionally, it was confirmed that the maximum allowable torque of the PTO shaft was 2300 N·m. Finally, under repeated load in the range of 11.3 to 113 N·m, the fatigue failure in the PTO shaft did not occur up to 10⁶ cycles.

• Composites Research
• /
• v.22 no.1
• /
• pp.22-31
• /
• 2009

Recently the wind energy has been alternatively used as a renewable energy resource instead of the mostly used fossil fuel due to its lack and environmental issues. This work is to propose a structural design and analysis procedure for development of the 500W class small wind turbine system which will be applicable to relatively low speed region like Korea and for the domestic use. The wind turbine blade was performed structural analysis including stress, deformation, buckling, vibration and fatigue. In addition, the blade should be safe from the impact damage due to FOD(Foreign Object Damage) including the bird strike. MSC.Dytran was used in order to analyze the bird strike penomena on the blade, and the applied method Arbitrary Lagrangian-Eulerian was evaluated by comparison with the previous study results. Finally, the structural test was carried out and its test results were compared with the estimated results for evaluation of the designed structure.

• Membrane Journal
• /
• v.33 no.6
• /
• pp.416-426
• /
• 2023

In this study, hindered amine-grafted graphene oxide (HA-GO) with antioxidant properties was prepared and incorporated into Nafion-based composite membranes as an effective filler material for polymer electrolyte membrane fuel cell applications. HA-GO was synthesized via a ring-opening reaction between amine groups in 4-amino-2, 2, 6, 6-tetramethyl piperidine and epoxy groups on the surface of GO. Nafion-based composite membranes containing different weight contents of HA-GO were fabricated to compare the polymer electrolyte membrane properties with those of the pure Nafion membrane. The composite membranes with HA-GO were found to have better mechanical properties, chemical stability, and proton conductivity than the pure Nafion membrane. In particular, the conductivity retention behavior confirmed by the decrease in proton conductivity after Fenton's test of the composite membranes was better than that of the pure Nafion membrane due to the incorporation of HA-GO with effective antioxidant properties.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.05a
• /
• pp.826-829
• /
• 2008

모바일 환경에서는 모바일 기기의 제한된 에너지를 효율적으로 사용하는 것이 요구된다. 특히 모바일 Peer-to-Peer(P2P) 시스템에서 모바일 기기가 피어로서 저장된 에너지를 모두 소모하면 더 이상파일 탐색 등과 같은 작업에 참여할 수 없게 된다. 또한 에너지가 없는 피어가 증가하게 되면 파일탐색 실패율이 증가하게 되고, 결국 파일탐색이 불가능하게 된다. 슈퍼피어(super peer)를 기반으로 한 계층적 P2P 시스템은 기존 단일 계층 P2P 시스템보다 더 적은 양의 메시지로 파일탐색을 가능하게 해준다. 하지만 하나의 슈퍼피어가 관리하는 서브피어(sub-peer)가 많을 경우, 서브피어의 관리 비용과 파일 탐색 요청의 증가로 슈퍼피어의 에너지 소모 비율이 크게 증가하게 되며, 이로 인해 네트워크 참여 시간이 줄어들게 된다. 이는 전체 네트워크 유지시간 감소에 영향을 미친다. 본 논문은 무선 통신 기반의 모바일 환경에서 슈퍼피어를 기반으로 한 계층적 P2P 시스템을 기반으로, 슈퍼피어가 관리하는 서브피어들의 수를 제한하고, 만약 일정한 수를 초과하였을 경우, 초과한 슈퍼피어의 서브피어들 중에서 새로운 슈퍼피어를 선정하고, 기존 슈퍼피어로부터 일부 서브피어들을 분배하여 관리하게 하는 시스템을 제안하였다. 실험을 통해 성능을 평가 하였으며, 실험 결과 파일 탐색 요청을 위한 네트워크 유지시간이 더 오래 지속됨을 확인하였다.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.6
• /
• pp.875-882
• /
• 2023

In this paper, we propose a virtual tuning technique for a temperature controller using the PID algorithm. Virtual simulation on a computer was used using the mathematical expression of the control object. A technique for accurately calculating the gain of the PID algorithm was introduced through detailed computer data analysis, and superior performance compared to conventional experimental tuning results was verified. In addition, it has the advantage of replacing tuning experiments conducted on actual control subjects, so there are no temporal or spatial limitations. Tuning experiments that actually operate the control object do not show detailed data that appears during the process. The accuracy of the experiment could not be guaranteed, and the results could not be confirmed immediately. Through the proposed technique, the entire tuning process can be accurately checked on a computer and the cause of problems that occur can also be analyzed.

• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.2
• /
• pp.90-96
• /
• 2023

Securing large amounts of training data in deep learning neural networks, including convolutional neural networks, is of importance for avoiding overfitting phenomenon or for the excellent performance. However, securing labeled training data in deep learning neural networks is very limited in reality. To overcome this, several augmentation methods have been proposed in the literature to generate an additional large amount of training data through transformation or manipulation of the already acquired traing data. However, unlike training data such as images and texts, it is barely to find an augmentation method in the literature that additionally generates bio-signal training data for convolutional neural network based human activity recognition. Thus, this study proposes a simple but effective augmentation method of bio-signal training data for convolutional neural network based human activity recognition. The usefulness of the proposed augmentation method is validated by showing that human activity is recognized with high accuracy by convolutional neural network trained with its augmented bio-signal training data.