• Journal of the Korea Computer Graphics Society
• /
• v.27 no.3
• /
• pp.65-74
• /
• 2021

3D reconstruction means that reconstructing the 3D shape of the object in an image and a video. We proposed a progressive occupancy network architecture that can recover not only the overall shape of the object but also the local details. Unlike the original occupancy network, which uses a feature vector embedding information of the whole image, we extract and utilize the different levels of image features depending on the receptive field size. We also propose a novel network architecture that applies the image features sequentially to the decoder blocks in the decoder and improves the quality of the reconstructed 3D shape progressively. In addition, we design a novel decoder block structure that combines the different levels of image features properly and uses them for updating the input point feature. We trained our progressive occupancy network with ShapeNet. We compare its representation power with two prior methods, including prior occupancy network(ONet) and the recent work(DISN) that used different levels of image features like ours. From the perspective of evaluation metrics, our network shows better performance than ONet for all the metrics, and it achieved a little better or a compatible score with DISN. For visualization results, we found that our method successfully reconstructs the local details that ONet misses. Also, compare with DISN that fails to reconstruct the thin parts or occluded parts of the object, our progressive occupancy network successfully catches the parts. These results validate the usefulness of the proposed network architecture.

• Journal of KIISE
• /
• v.42 no.12
• /
• pp.1600-1610
• /
• 2015

While today's networks have been shown to exhibit HBDP (High Bandwidth Delay Product) characteristics, the legacy TCP increases the size of the congestion window slowly and decreases the size of the congestion window drastically such that it is not suitable for HBDP Networks. In order to solve this problem with the legacy TCP, many congestion control TCP mechanisms have been proposed. C-TCP (Compound-TCP) is a hybrid TCP which is a synergy of delay-based and loss-based approaches. C-TCP adapts the decreasing rate of the delay window without considering the congestion level, leading to degradation of performance. In this paper, we propose a new scheme to improve the performance of C-TCP. By controlling the increasing and decreasing rates according to the congestion level of the network, our proposed scheme can improve the bandwidth occupancy and fairness of C-TCP. Through performance evaluation, we show that our proposed scheme offers better performance in HBDP networks as compared to the legacy C-TCP.

• Information and Communications Magazine
• /
• v.21 no.3
• /
• pp.96-105
• /
• 2004

SK 텔레콤은 국내 제일의 이동 통신 사업자로서, 이동전화 및 무선인터넷 분야에서 국내 최고 수준의 시장점유율과 기술력을 확보하고 있으며 국내 IT 환경의 급속한 디지털 (Digital)화 및 유비쿼터스(Ubiquitous) 환경 도래에 대비하여 홈 네트워크 사업 활성화를 추진 중이다 그 일환으로 정보통신부에서 추진하는 홈 네트워크 시범사업의 주관 사업자로서 성공적인 사업 제공을 위한 컨소시엄 구성 및 시범 사업 준비를 완료하고 관련 기술 개발을 진행 중이다.(중략)

• The Journal of the Korea institute of electronic communication sciences
• /
• v.11 no.6
• /
• pp.581-590
• /
• 2016

This paper proposes an efficient collision avoidance algorithm that a group of RFID readers manage each of their affiliated active RFID tags in a space by establishing wireless network and avoids communication collision between RFID reader to reader and RFID tag to tag, and reader to tag. RF readers operate as wireless network nodes, create synchronous links each other, and can exchange messages. Active tags also are operated communicating each reader devices synchronously, and competing slot position between tags algorithm using unique tag identification number is implemented. Each reader node operates their own reader function different time slot network communication period to prevent collision between readers communication.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.11
• /
• pp.2437-2445
• /
• 2012

This paper presents the efficient measurement method of radio resource by analyzing various medium occupied elements. The medium occupied time consists of 802.11 frames, wireless interference, and protocol waiting time from a wireless node on a current channel. And it is used to performance metric. Existing research is only measured partial occupied elements, and is lack of validation of measurement unit and scalability on various IEEE 802.11 radio. This paper presents the measurement method of classified occupied elements. To achieve this, we modified 802.11n based OpenHAL device driver to collect the register information of wireless chipset, and to analyze receiving frames in an virtual monitor mode. We conclude accurate medium occupied time measurement system from various validation methods.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.10B
• /
• pp.621-629
• /
• 2007

In cognitive radio (CR) network, the channels are generally classified into either the unavailable channels that are occupied by incumbent users or the available channels that are not occupied. The conventional channel classification scheme may result in poor utilization of spectrum holes since it does not take the spatial relationship between CR node and incumbent users into consideration. In this paper, we propose an efficient channel management scheme for the centralized CR network to maximize the spectrum holes by overcoming the shortcomings of conventional scheme. In addition, we mathematically analyze the effectiveness of proposed scheme. Based on the proposed channel management scheme, we also propose the rendezvous algorithm, which can establish the control channels between base station and CR node under the dynamically changing spectrum environment.

• Journal of Advanced Navigation Technology
• /
• v.18 no.1
• /
• pp.67-73
• /
• 2014

Due to the development of the communication technology and the high-performance mobile device, the number of video streaming service user has been increasing. Video streaming services in the restricted network environments need error control to support users for the Ultra High Definition(UHD) video delivery. Packet-level FEC method was used to control the error for delay-sensitive video service, but it has a defect in huge occupation of network resources by redundant packets. In this paper, we proposed a packet-level FEC algorithm which maintains the video quality by previous FEC methods and also reduces the occupation of network resource in the restricted network environments. The proposed algorithm controls the redundant rate with using the delivery characteristic of the video coding technology, and improves the efficiency of network resources by reducing the traffic by around 33% in comparison with the previous FEC methods.

• Journal of KIISE
• /
• v.42 no.9
• /
• pp.1162-1174
• /
• 2015

These days, the networks have exhibited HBDP (High Bandwidth Delay Product) characteristics. The legacy TCP slowly increases the size of the congestion window and drastically decreases the size of a congestion window. The legacy TCP has been found to be unsuitable for HBDP networks. TCP mechanisms for solving the problems of the legacy TCP can be categorized into the loss-based TCP and the delay-based TCP. Most of the TCP mechanisms use the standard slow start phase, which leads to a heavy packet loss event caused by the overshoot. Also, in the case of congestion avoidance, the loss-based TCP has shown problems of wastage in terms of the bandwidth and RTT (Round Trip Time) fairness. The delay-based TCP has shown a slow increase in speed and low occupancy of the bandwidth. In this paper, we propose a new scheme for improving the over shoot, increasing the speed of the bandwidth and overcoming the bandwidth occupancy and RTT fairness issues. By monitoring the buffer condition in the bottleneck link, the proposed scheme does congestion control and solves problems of slow start and congestion avoidance. By evaluating performance, we prove that our proposed scheme offers better performance in HBDP networks compared to the previous TCP mechanisms.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.05a
• /
• pp.1097-1100
• /
• 2006

최근 네트워크 기술이 발전하면서 사용자들의 요구가 다양화되고 대용량의 멀티미디어 데이터에 대한 필요성이 증가하고 있다. 실시간성과 동시성이 중요시 되는 멀티미디어 데이터의 전송에는 UDP(User Datagram Protocol)가 사용되고 있는데, UDP는 현재의 인터넷 기반인 TCP (Transmission Control Protocol )와 경쟁관계에 있어 네트워크의 혼잡을 초래하는 경우가 많았다. 또한 TCP 트랙픽은 네트워크의 혼잡제어를 수행하지만 수신자의 관점을 고려하지 않은 점이 있어서 스트리밍 전송에는 적합하지 않은 단점을 가지고 있다. 본 논문에서는 UDP(User Datagram Protocol) 트래픽의 네트워크 형평성 문제를 해결하고 수신자의 관점을 고려하기 위해 네트워크에 우선순위를 적용하여 RTP/RTCP를 기반으로 수신자의 버퍼를 고려한 혼잡제어 알고리즘을 제안하였다. 시뮬레이션 결과를 통해 우선 순위가 있는 네트워크가 우선 순위가 없는 네트워크에 비해 패킷 손실률과 버퍼 점유도 면에서 스트리밍 전송에 적합한 환경을 제공하고 있음을 확인할 수 있었다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.273-276
• /
• 2020

기존의 물리 센서를 활용한 주차 감지는 주차장 규모가 클수록 큰 비용이 필요하고 이미지 기반의 분석은 개별 주차장에 대한 데이터 라벨링과 학습의 노력이 필요했다. 본 논문은 LoRa(Long Range) 네트워크와 마이크로프로세서를 활용한 IoT기반의 시스템으로 영상데이터를 서버로 전송하고 COCO(Common Object in context) 데이터셋으로 학습된 Mask R-CNN 기반의 모델을 활용한 주차장 내 차량점유 감지 알고리즘을 통해 개별 주차장에 대한 학습 또는 라벨링 없이 주차장 내 주차상태를 식별하고 사용자에게 인터페이스를 통해 실시간으로 주차정보를 제공하는 시스템을 구현한다.