• 방송공학회논문지
• /
• 제23권5호
• /
• pp.614-621
• /
• 2018

본 논문에서는 generative adversarial network (GAN)을 이용한 비감독 학습을 통해 깊이 카메라로 깊이 영상을 취득할 때 발생한 손실된 부분을 복원하는 기법을 제안한다. 제안하는 기법은 3D morphable model convolutional neural network (3DMM CNN)와 large-scale CelebFaces Attribute (CelebA) 데이터 셋 그리고 FaceWarehouse 데이터 셋을 이용하여 학습용 얼굴 깊이 영상을 생성하고 deep convolutional GAN (DCGAN)의 생성자(generator)와 Wasserstein distance를 손실함수로 적용한 구별자(discriminator)를 미니맥스 게임기법을 통해 학습시킨다. 이후 학습된 생성자와 손실 부분을 복원해주기 위한 새로운 손실함수를 이용하여 또 다른 학습을 통해 최종적으로 깊이 카메라로 취득된 얼굴 깊이 영상의 손실 부분을 복원한다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 하계종합학술대회 논문집(3)
• /
• pp.145-148
• /
• 2000

ATM network is considered as best suitable technology for multimedia service in various aspects such as bandwidth, capability of traffic expandability and so on. In this paper, we suggested a buffer management algorithm in ATM network to improve overall network performance with threshold and auxiliary buffer whose input consists of superpositiI on of voice and multimedia data traffic. To evaluate the proposed buffer management algorithm simulations are executed with four priorities, that is delay and loss priorities and the results are proved that network throughput is improved better than the existing partial buffer method.

• Nuclear Engineering and Technology
• /
• 제53권7호
• /
• pp.2229-2236
• /
• 2021

In nuclear-related facilities, such as nuclear power plants, research reactors, accelerators, and nuclear waste storage sites, radiation detection, and mapping are required to prevent radiation overexposure. Sensor network systems consisting of radiation sensor interfaces and wxireless communication units have become promising tools that can be used for data collection of radiation detection that can in turn be used to draw a radiation map. During data collection, malfunctions in some of the sensors can occasionally occur due to radiation effects, physical damage, network defects, sensor loss, or other reasons. This paper proposes a reproduction strategy for radiation maps using a U-net model to compensate for the loss of radiation detection data. To perform machine learning and verification, 1,561 simulations and 417 measured data of a sensor network were performed. The reproduction results show an accuracy of over 90%. The proposed strategy can offer an effective method that can be used to resolve the data loss problem for conventional sensor network systems and will specifically contribute to making initial responses with preserved data and without the high cost of radiation leak accidents at nuclear facilities.

• 정보처리학회논문지B
• /
• 제9B권2호
• /
• pp.173-180
• /
• 2002

기존의 VOB(Virtual Output Buffer) 모델에서 신경회로망의 학습 패턴 처리를 위해 가상 셀 손실율이 도입되었다. VOB모델은 신경망이 실제 셀 손실율 없이도 연결 수락 경계을 잘 찾을 수 있음을 보여주었다. 그러나 VOB 모델은 셀 손실율을 과다 평가하는 경향이 있어 결과적으로 망 자원의 이용률이 낮은 단점이 있다. 본 논문에서는 이러한 단점을 보완하는 방법으로 연결 수락 경계에서 셀 손실율의 평균에 대한 정보를 충분히 포함하는 셀 손실율 참조 곡선의 개념을 제안하였다. 그리고 제안된 셀 손실율 참조 곡선을 이용하여 가상 셀 손실율을 처리하는 방법을 제안하였다. 제안된 한습 패턴 처리 방법은 ATM 트래픽 중에 가장 대표적인 두 가지 호원에 대하여 실험하였다. 실험에 사용된 호원은 LAN 데이터의 그래픽 특성을 가시는 On-Off 트래픽과 비디오 화상 통신의 특성을 가지는 Auto-Regressive 트래픽이다.

• International journal of advanced smart convergence
• /
• 제12권2호
• /
• pp.173-181
• /
• 2023

Data collection is an essential element in the construction and operation of a smart factory. The quality of data collection is greatly influenced by network conditions, and existing wireless network systems for IoT inevitably lose data due to wireless signal strength. This data loss has contributed to increased system instability due to misinformation based on incorrect data. In this study, I designed a distributed MQTT IoT smart sensor and gateway structure that supports wireless multicasting for smooth sensor data collection. Through this, it was possible to derive significant results in the service latency and data loss rate of packets even in a wireless environment, unlike the MQTT QoS-based system. Therefore, through this study, it will be possible to implement a data collection management system optimized for the domestic smart factory manufacturing environment that can prevent data loss and delay due to abnormal data generation and minimize the input of management personnel.

• 한국항행학회논문지
• /
• 제14권2호
• /
• pp.253-259
• /
• 2010

무선 메쉬 네트워크는 Base Station 기반과 더불어 애드혹 네트워크나 블루투스와 같이 유연성을 가진 네트워크이다. 하지만 무선 메쉬 네트워크는 높은 패킷 손실률을 보이고 TCP(Transport Control Protocol) 알고리즘은 무선 메쉬 네트워크의 패킷손실 원인을 네트워크 내의 혼잡으로 인식하기 때문에 TCP 혼잡제어 알고리즘(Congestion Control Algorithm)을 실행하게 된다. 이러한 TCP 동작은 혼잡 손실이 아닌 패킷 손실로 발생 할 경우, 상당한 성능 저하를 초래하게 된다. 본 논문에서는 무선 메쉬 네트워크에서 TCP 혼잡제어 알고리즘의 성능을 개선시키기 위해 혼잡 윈도우를 무선망에서도 적응력 있게 조절하는 연구를 제안하였다.

• 한국통신학회논문지
• /
• 제34권11B호
• /
• pp.1225-1233
• /
• 2009

기업이나 대학 등 대규모 기관들은 고속 네트워크를 구축하여 운영하고 있다. 이와 같은 고속 엔터프라이즈 네트워크 내에서 네트워크 이용률(utilization)이 저조함에도 불구하고 종단간 성능은 네트워크 속도에 비하여 상당히 저조하다. 이러한 원인은 특정 장비에서 아주 작은 시간에 트래픽이 폭주하는 미세폭주(micro-congestion)가 발생하기 때문이다. 미세폭주는 패킷손실(Packet Loss)나 지연(Delay), 패킷역전(Packet Reordering)을 발생시키고 이것이 성능 저하의 원인이 된다. 본 논문에서는 미세폭주를 발생하는 시점을 검출하는 방법과 미세폭주를 검출한 시점에 트래픽을 수집하여 패킷손실, 지연, 패킷역전과 트래픽의 특성과의 상관관계 분석하여 성능저하를 발생시키는 미세폭주의 특성을 규명한다.

• 한국정보통신설비학회:학술대회논문집
• /
• 한국정보통신설비학회 2008년도 정보통신설비 학술대회
• /
• pp.45-50
• /
• 2008

Many applications of sensor network require connection to the Internet. The transmission protocol of traditional sensor network was designed within the sensor network itself. However, based on 6LoWPAN which can be accessed using IPv6, direct connection is possible between the sensor network and the TCP/IP network outside. Transmission of data in applications of sensor network falls into two main categories. One is a small packet that is periodically produced such as packet related to temperature and humidity. The other is a relatively large packet that brings about network overheads such as images. We investigated the conformance test and pros and cons of application data over the transmission protocol of Zigbee and 6LoWPAN. As a result, both Zigbee and 6LoWPAN have shown low rate of loss for periodic data and have in creased reliability of data transfer. When transmitting streaming image data, both ACK, non ACK mode of Zigbee and UDP of 6LoWPAN minimized transmission time but suffered the consequences of high packet loss. Even though TCP of 6LoWPAN required a long transmission time, we were able to confirm that no loss has occurred.

• Nuclear Engineering and Technology
• /
• 제31권2호
• /
• pp.226-235
• /
• 1999

Applying digital network technology for advanced nuclear plant requires deterministic communication for tight safety requirements, timely and reliable data delivery for operation-critical and mission-critical characteristics of nuclear plant. Communication protocols, such as IEEE 802/4 Token Bus, IEEE 802/5 Token Ring, FDDI, and ARCnet, which have deterministic communication capability are partially applied to several nuclear power plants. Although digital communication technologies have many advantages, it is necessary to consider the noise immunity from electromagnetic interference (EMI), electrical interference, impulse noise, and heat noise before selecting specific digital network technology for nuclear plant. In this paper, we consider the token frame loss and data frame loss rate due to the link error event, frame size, and link data rate in different protocols, and evaluate the possibility of failure to meet the hard real-time requirement in nuclear plant.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2006년도 하계종합학술대회
• /
• pp.59-60
• /
• 2006

TCP applied to the wireless-wired integrated network is the one that was applied to the existing wired network. In the wireless-wired integrated network, both wireless and congestion loss can occur. When wireless packet losses occur, the congestion control of TCP causes performance degradation by reducing its transmission rate. In this paper, we propose the algorithm to distinguish the wireless packet loss from congestion packet loss using MIB of the 802.11 MAC which has been generally used recently in wireless links.