• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.292-292
• /
• 2021

최근 데이터 과학의 획기적인 발전으로 딥러닝(Deep Learning) 알고리즘이 개발되어 다양한 분야에 널리 적용되고 있다. 본 연구에서는 인공신경망 중 하나인 LSTM(Long-Short Term Memory) 네트워크를 기반으로 정확한 댐유입량 예측을 수행하는 SWLSTM 모델을 제안하였다. SWLSM은 모델의 정확도를 개선하기 위해 세 가지 주요 아이디어를 채택하였다. (1) 통계적 속성 (PACF) 및 교차 상관 함수(CCF)를 사용하여 적절한 입력 변수와 시퀀스 길이를 결정하였다. (2) 선택된 입력 예측 변수 시계열을 웨이블릿 변환(WT)을 사용하여 하위 시계열로 분해한다. (3) k-folds cross validation 및 random search 기법을 사용하여 LSTM의 하이퍼 매개변수들을 효율적으로 최적화하고 검증한다. 제안된 SWLSTM의 효과는 한강 유역 5개 댐의 시단위/일단위/월단위 유입량을 예측하고 과거 자료와 비교함으로써 검증하였다. 모델의 정확도는 다양한 평가 메트릭(R2, NSE, MAE, PE)이 사용하였으며, SWLSTM은 모든 경우에서 LSTM 모델을 능가하였다. (평가 지표는 약 30 ~ 80 % 더 나은 성능을 보여줌). 본 연구의 결과로부터, 올바른 입력 변수와 시퀀스 길이의 선택이 모델 학습의 효율성을 높이고 노이즈를 줄이는 데 효과적임을 확인하였다. WT는 홍수 첨두와 같은 극단적인 값을 예측하는 데 도움이 된다. k-folds cross validation 및 random search 기법을 사용하면 모델의 하이퍼 매개변수를 효율적으로 설정할 수 있다. 본 연구로부터 댐 유입량을 정확하게 예측한다면 정책 입안자와 운영자가 저수지 운영, 계획 및 관리에 도움이 될 것이다.

• Journal of the Korea Society for Simulation
• /
• v.19 no.4
• /
• pp.11-19
• /
• 2010

ZigBee is a standard for wireless personal area networks(WPANs) based on the IEEE 802.15.4 standard. It has been developed for low cost and low power consumption. There are two alternative routing schemes that have been proposed for the ZigBee standard: Ad-hoc On-Demand Distance Vector(AODV) and tree routing. The tree routing forwards packets from sensors to a sink node based on the parent-child relationships established by the IEEE 802.15.4 MAC topology formation procedure. In order to join the network, a sensor node chooses an existing node with the strongest RSSI(Received signal strength indicator) signal as a parent node. Therefore, some nodes carry a large amount of traffic load and exhaust their energy rapidly. To overcome this problem, we introduce a new metric based on link quality and traffic load for load balancing. Instead of the strength of RSSI, the proposed scheme uses the new metric to choose a parent node during the topology formation procedure. Extensive simulation results using TOSSIM(TinyOS mote SIMulator) show that the CFR scheme outperforms well in comparison to the conventional tree routing scheme.

• The KIPS Transactions:PartC
• /
• v.8C no.5
• /
• pp.647-656
• /
• 2001

Video delivery from a server to a client across a network system is an important part of many multimedia applications. Usually, the network system has constraint in both the amount of network bandwidth and the buffer size in the client. While delivering a video stream across such a constrained network system, loss of frames may be unavoidable. The system resources consumed by the dropped frames are wasted, and the losses of frames would result in discontinuous display at the client. In this paper, for delivering hierarchically encoded video stream, we introduce the notion of selective layer discard algorithm at the server which not only preemptively discards data at the server but also drops less important part of a frame instead of the entire frame. By the simulation, we compare the proposed selective layer discard algorithm and the existing selective frame discard algorithm. The simulation results show that the proposed algorithm may improve the quality of decoded video, and decrease the replay discontinuity at the client.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10d
• /
• pp.561-565
• /
• 2006

무선 모바일 애드 혹 네트워크에서 에너지 효율을 위한 많은 라우팅 알고리즘이 연구되어 왔다. 이러한 제안된 다른 알고리즘들을 살펴보면, 에너지 효율 뿐 만 아니라 성능의 향상을 위한 노력도 기울여 왔다. 일반적으로 목적지 노드가 전송받은 RREQ의 메트릭을 통하여 에너지 효율적인 경로를 선택 하게 되는데, 이것은 결국 데이터 전송 지연이 발생하게 된다. 그리고 경로가 설정된 이후에 경로를 유지하는데 있어서, 노드의 이동이나 배터리의 소진으로 링크가 끊기게 되면 경로를 다시 찾는 과정에서 제어 패킷을 다시 플러딩 하기 때문에 여러 노드의 에너지 소비를 초래하게 된다. 본 논문에서는 이러한 문제들을 해결하기 위해 기존의 AODV(On-Demand Distance Vector Routing) 프로토콜을 기반으로 다음과 같은 방법을 제안한다. 먼저 RREQ를 받은 노드가 남은 배터리의 양을 고려하여 전송의 여부를 결정하는 방법을 제안하고, 경로를 유지하는데 있어서 전송의 세기를 통해 노드의 이동성을 인지하고 링크가 끊기기 전에 다른 노드로 경로를 대체함으로써 경로 재설정을 통한 에너지 소비를 줄이는 방안을 제안한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.11a
• /
• pp.196-198
• /
• 2014

Wireless Sensor Networks(WSN) 기반의 Border Surveillance Networks(BSN)는 다양한 제한조건을 동반하며 침입 혹은 환경 등을 감시함을 목적으로 하고 있기 때문에 이벤트 발생시 신뢰적인 데이터 전송은 매우 중요한 부분을 차치하게 된다. 본 논문에서 적용한 대표적인 요구기반 라우팅 방법인 Ad hoc On-demand Distance Vector Routing(AODV)는 BSN 에 적용하였을 경우 홉 수(Hop Count)를 메트릭으로 최적의 경로를 계산하고 선택하여 Uni-Path 로 데이터를 전송하게 된다. 이때 AODV 는 Uni-path 로 데이터 전송을 진행하게 되며 링크의 단절이 빈번히 발생하는 BSN 의 환경적 요인으로 데이터 전송의 신뢰성이 감소하게 될 것이며 이는 비효과적인 데이터 전송을 야기하게 된다. 따라서 본 논문에서는 Multi-path 개념을 도입하여 데이터 전송 시 링크단절에 의해 발생하는 시간의 지연을 줄이고 더욱 신뢰성 있게 데이터를 전송할 수 있는 방법을 제안한다. 제안하는 방법은 송, 수신하는 두 노드의 전송범위에 공통으로 속한 노드들이 전송되는 데이터를 중복 전송하게 된다. 그리하여 링크의 단절이 일어날 경우 발생하는 경로 재 탐색 및 경로복구의 발생 빈도를 줄이고 동시에 데이터 전송률을 증가시켜 신뢰적인 데이터 전송을 하게 된다.

• Journal of Broadcast Engineering
• /
• v.14 no.2
• /
• pp.134-143
• /
• 2009

In this paper, we propose a new relay selection algorithm for a two-way multiple relay channel. In the two-way multiple relay channel, two users exchange information with each other via multiple relays. The relays use a decode-and-forward or amplify-and-forward protocol, and exploit the combining process of the received packets to reduce the required channel resources. In the multiple relay network, diversity gain is achieved as the number of relays increases, and various schemes are proposed. In this paper, we propose a single best relay selection scheme based on instantaneous channel conditions. First of all, relays obtain the instantaneous channel state information in the handshaking process, and a single best relay is selected in a distributed methods prior to data transmissions. The relay selection metric is proposed so that the end-to-end channel condition is evaluated based on the intantaneous channel state informations. Simulation results show that the proposed relay selection algorithm achieve the increased throughput and diversity order when the number of potential relays is increased.

• Journal of Korea Society of Industrial Information Systems
• /
• v.22 no.1
• /
• pp.1-9
• /
• 2017

With the Development of Wireless Network Technology and Wireless Link Technology, Wireless Mesh Network (WMN) is Attracting Attention as a Key Technology to Construct the Wireless Transit Network. The WMN has been Studied for a Long Time in Various Fields, however there are still many Problems that have not been solved yet. One of them is the Routing Problem to find an Optimal path in a Multi-hop Network Composed of Wireless Links. In the Hybrid-WMN, which is one of the Three Types of WMN, Optimal Path Selection Requires Research on Path Search Protocols that Effectively use the Infrastructure Mesh as a Transit Network, Together with Research for a Routing Metric with Excellent Performance. Therefore, this Paper Proposes a Dual Mode-AODV(Ad hoc On-demand Distance Vector) for Hybrid-WMN. Simulation result shows that the Path Selection Delay was Reduced by 52% than AODV when the Proposed Dual Mode-AODV was applied.

• The KIPS Transactions:PartD
• /
• v.11D no.3
• /
• pp.577-590
• /
• 2004

Today, the growth of software industry leads to e quantitative expansion of software package products. Due to this rapid increase of software package products, qualify certification has been required fur software products which users select. Unlike the quality certification of industrial products, the history of software product certification has not been so long. For this reason, software quality evaluation and certification methods have not matured yet. When certifying software products, one of most important factors is the systematic generation of evaluation sets. The evaluation sets include checklists with metrics, and criteria for the software quality certification according to the classification of software product type. This paper presents a metamodel for the systematic generation and maintenance of the evaluation sets. Then, we construct prototype level evaluation sets to show the validity of the metamodel.

• The KIPS Transactions:PartC
• /
• v.15C no.1
• /
• pp.45-50
• /
• 2008

Mobile ad hoc networks (MANETs) allow mobile nodes to communicate among themselves via wireless multiple hops without the help of the wired infrastructure. Therefore, in the MANET, a route setup mechanism that makes nodes not within each other's transmission range communicate is required and, for this, the Ad-hoc On-demand Distance Vector (AODV) was proposed as one of the reactive routing protocols well suited for the characteristics of the MANET. AODV uses the hop count as the routing metric and, as a result, a node selects the farthest neighbor node as its next hop on a route, which results in a problem of deteriorating the overall network throughput because of selecting a relatively low data rate route. In this paper, we propose an efficient reactive routing protocol based on the multi-rate aware MAC. Through the simulations, we analyze the performance of our proposed mechanism and, from the simulation results, we show that our proposed mechanism outperforms the existing mechanism.

• The KIPS Transactions:PartA
• /
• v.12A no.5 s.95
• /
• pp.451-460
• /
• 2005

We present a methodology to design energy-efficient data paths using FPGAs. Our methodology integrates domain specific modeling, coarse-grained performance evaluation, design space exploration, and low-level simulation to understand the tradeoffs between energy, latency, and area. The domain specific modeling technique defines a high-level model by identifying various components and parameters specific to a domain that affect the system-wide energy dissipation. A domain is a family of architectures and corresponding algorithms for a given application kernel. The high-level model also consists of functions for estimating energy, latency, and area that facilitate tradeoff analysis. Design space exploration(DSE) analyzes the design space defined by the domain and selects a set of designs. Low-level simulations are used for accurate performance estimation for the designs selected by the DSE and also for final design selection We illustrate our methodology using a family of architectures and algorithms for matrix multiplication. The designs identified by our methodology demonstrate tradeoffs among energy, latency, and area. We compare our designs with a vendor specified matrix multiplication kernel to demonstrate the effectiveness of our methodology. To illustrate the effectiveness of our methodology, we used average power density(E/AT), energy/(area x latency), as themetric for comparison. For various problem sizes, designs obtained using our methodology are on average $25\%$ superior with respect to the E/AT performance metric, compared with the state-of-the-art designs by Xilinx. We also discuss the implementation of our methodology using the MILAN framework.