• Proceedings of the KIEE Conference
• /
• 2002.11b
• /
• pp.111-113
• /
• 2002

본 논문에서는 경쟁적 전력시장에서 송전 계약에 관한 비용을 모선에서 한계비용 만감도를 이용해서 분할하는 방법을 제시하였다. 구조개편된 전력사장에서는 단순하면서도 분명한 전력 요금 시스템이 요구된다. Zonal pricing은 이런 면에서 좋은 pricing 시스템이 요구된다. 하지만, zone을 효과적으로 나누는 것은 상당히 어렵다. 기존의 방법에서는 오직 부하만을 고려하여 분류를 하였지만, 에너지 시장에서의 전력의 매매행위를 고려할 때, 시스템 내부의 모든 사업자에 대한 고려가 필요하다. 그래서, 본 논문에서는 한계비용 민감도를 이용해서 모든 에너지 시장 참여자를 고려하여, 네트워크를 분할하였다. 모든 노드에서의 민감도의 패턴이 분석되었고 동일한 패턴을 가진 노드들을 같은 zone으로 분류하였다. 6-모선 전력시스템을 이용해서 제안한 방법을 설명하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.07a
• /
• pp.395-396
• /
• 2014

최근 급증하는 교통 혼잡으로 인해 시간적/물리적 손실이 크게 발생하고 있다. 이러한 교통난 해소는 시설투자만으로는 근본적인 해결책이 될 수 없다는 판단 하에 지난 수년간 보다 정확한 교통량을 예측하기 위해 다양한 교통량 예측 모델들이 개발되어왔다. 그러나 기존 모델들은 회기분석을 통해 과거 교통량을 분석하고 과거의 교통패턴이 미래에 지속적으로 연장된다는 가정 하에 연구되었기 때문에 실시간으로 급변하는 불규칙한 교통 패턴에 대한 예측의 신뢰성을 떨어트린다. 이를 위해 본 논문에서는 큐잉 네트워크 모델 기반의 교통량 예측 모델을 설계 하고 이를 바탕으로 안드로이드 기반의 애플리케이션을 구현하였다.

• Journal of the Korea Society for Simulation
• /
• v.32 no.3
• /
• pp.43-53
• /
• 2023

In this detailed and comprehensive study, our primary focus has been placed on accurately gauging the number of visitors and their real-time locations in commercial spaces. Particularly, in a real cafe, using security cameras, we have developed a system that can offer live updates on available seating and predict future congestion levels. By employing YOLO, a real-time object detection and tracking algorithm, the number of visitors and their respective locations in real-time are also monitored. This information is then used to update a cafe's indoor map, thereby enabling users to easily identify available seating. Moreover, we developed a model that predicts the congestion of a cafe in real time. The sophisticated model, designed to learn visitor count and movement patterns over diverse time intervals, is based on Long Short Term Memory (LSTM) to address the vanishing gradient problem and Sequence-to-Sequence (Seq2Seq) for processing data with temporal relationships. This innovative system has the potential to significantly improve cafe management efficiency and customer satisfaction by delivering reliable predictions of cafe congestion to all users. Our groundbreaking research not only demonstrates the effectiveness and utility of indoor location tracking technology implemented through security cameras but also proposes potential applications in other commercial spaces.

• Journal of Internet Computing and Services
• /
• v.8 no.5
• /
• pp.125-132
• /
• 2007

Transmission mechanisms that include an available bandwidth estimation algorithm and a packet loss differentiation scheme, in general, exhibit higher TCP performance in wireless networks. TCP New Jersey, known as the best existing scheme in terms of goodput, improves wireless TCP performance using the available bandwidth estimation at the sender and the congestion warning at intermediate routers. Although TCP New Jersey achieves 17% and 85% improvements in goodput over TCP Westwood and TCP Reno, respectively, we further improve TCP New Jersey by exploring improved available bandwidth estimation, retransmission timeout, and recovery mechanisms. Hence, we propose TCP New Jersey PLUS (shortly TCP NJ+), showing that under 1% packet loss rate, it outperforms 3% by TCP New Jersey and 5% by TCP Wes1wood. In 5% packet loss rate, a characteristic of high bit-error-rate wireless network, it outperforms other TCP variants by 19% to 104% in terms of goodput even when the network is in bi-directional congestion.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.12C
• /
• pp.1239-1247
• /
• 2005

It is important to improve TCP performance in Self-similar TCP network where signalling between the same end nodes through bidirectional traffic routes. In wireless link, the traffic limitation pattern occurred in two or more TCP connections is applied into MPEC video control as multi time-interval congestion control. For TCP update variable, we extend TCP and perform as function call, and we study a method of relating TCP with LTS module controlling with the information type that is overcoming the limit of feedback loop determined by RTT. For comparison, we measure the TCP throughput without LTS and verify the fairness by means of meta control. The improved TCP performance is shown by that the number of connections of traffic congestion control increases when RTT increases.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.5
• /
• pp.786-792
• /
• 2022

In the cloud computing environment, there has been a lot of research into the Fog/Edge Computing (FEC) paradigm for securing user proximity of application services and computation offloading to alleviate service delay difficulties. The method of predicting dynamic location change patterns of edge devices (moving objects) requesting application services is critical in this FEC environment for efficient computing resource distribution and deployment. This paper proposes an optimal moving pattern extraction algorithm in which variable weights (distance, time, congestion) are applied to selected paths in addition to a support factor threshold for frequency patterns (moving objects) of edge devices. The proposed algorithm is compared to the OPE_freq [8] algorithm, which just applies frequency, as well as the A^* and Dijkstra algorithms, and it can be shown that the execution time and number of nodes accessed are reduced, and a more accurate path is extracted through experiments.

• Journal of Korean Society of Transportation
• /
• v.28 no.5
• /
• pp.55-64
• /
• 2010

In order to prepare an efficient solution for PM10 reduction in underground stations, the authors measured PM10 concentration levels every 30 minutes in the concourse, platform, and tunnel of Gaehwasan Station of Seoul's subway line 5. Through a correlation analysis of each changing pattern of PM10 concentration, the direction and velocity of diffusion in underground stations were estimated. The PM10 concentration levels were highest in the tunnel, followed by the platform and concourse. PM10 concentrations in the tunnel, platform, and concourse showed a pattern of increasing in the rush hours and decreasing in the non-rush hours. According to the statistical analysis of PM10 concentrations and changing patterns in each location, the higher PM10 concentration in the tunnel expanded to the platform, and some from the platform expanded to the concourse. Therefore, to efficiently reduce PM10 concentrations, it is essential to detect the centralized generation, diffusion factor, expanding route, expanding measure, and other variables and to remove or reduce the diffusion factor and level. Through operating the ventilation system in the right time frame while the PM10 concentration level increases, the power consumption and peak power consumption can be reduced.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2016.05a
• /
• pp.287-288
• /
• 2016

선박과 선박간의 사고 위험도를 예측하는 교통정보 생성 기술을 해상교통관제센터에 적용하기에는 위험도 정보의 정확성에 한계가 있다. 또한 대상 해역에 대한 교통 패턴을 파악하는 밀집도 및 혼잡도와 같은 교통정보 생성 기술은 위험 우선순위 선박을 도출하는 것이 불가능하다. 복잡한 교통 패턴을 보이는 해상교통관제 해역에서 위험 선박을 인지하여 관제사의 관제 업무를 지원하기 위해서는 새로운 접근이 필요하다. 본 연구에서는 관제대상해역의 교통 상황을 총체적으로 파악하고 위험 선박을 사전에 인지할 수 있는 교통정보 생성을 위해서 기계학습 기법을 검토하였으며, 기존의 인공지능 한계를 극복하기 위한 딥러닝 프레임워크 도입을 검토하였다. 해상교통관제센터의 이미지, 메시지, 음성 등 다양한 형태의 연속적 자료들을 통합하고 이를 토대로 총체적인 분석을 통해 관제 업무를 지원할 수 있는 교통 상황 인지 정보를 생성할 수 있을 것으로 파악되었다. 빅데이터 기반의 기계학습은 보다 의미 있는 상황 인지 정보를 생성할 수 있기 때문에 이를 위한 관제 센터의 각종 데이터 통합이 필요하다.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.5
• /
• pp.396-401
• /
• 2017

In this paper, we propose a solution for interference with Bluetooth when connecting Bluetooth and other devices. The random frequency hopping technique is a technique of generating a hopping pattern using the entire Bluetooth channel without considering the interference of the wireless LAN. The proposed adaptive frequency hopping technique is a technique for generating a hopping pattern of Bluetooth channel considering periodic carrier sensing of Bluetooth and considering WLAN interference. Simulation results show that the use of adaptive frequency hopping reduces the packet error rate as the Bluetooth carrier sensing interval decreases even in the congestion of WLAN interference. Especially, the frequency hopping technique improves the average packet error rate by about 13% compared to the adaptive frequency hopping technique.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2014.06a
• /
• pp.323-325
• /
• 2014

전체 충돌사고 중 어선과 관련된 충돌사고가 약 80퍼센트를 차지하고 있다. 어선사고의 잠재적 위험요소를 찾기 위해 본 연구에서는 완도 VTS 관제구역 내 통항하는 어선의 항해특성 및 패턴을 분석함으로써 어선 비어선간 발생할 수 있는 충돌사고의 위험해역 및 Near Collision을 분석하고자 한다. 이 결과를 바탕으로 관제사의 의사결정을 지원함으로써 해상교통관제서비스 향상과 해양 사고 예방에 기여하고자 한다.