• 한국BIM학회 논문집
• /
• 제8권3호
• /
• pp.12-19
• /
• 2018

The research aims to find implications of machine learning and urban big data as a way to construct the flexible transportation network system of smart city by responding the urban context changes. This research deals with a problem that existing a bus headway model is difficult to respond urban situations in real-time. Therefore, utilizing the urban big data and machine learning prototyping tool in weathers, traffics, and bus statues, this research presents a flexible headway model to predict bus delay and analyze the result. The prototyping model is composed by real-time data of buses. The data is gathered through public data portals and real time Application Program Interface (API) by the government. These data are fundamental resources to organize interval pattern models of bus operations as traffic environment factors (road speeds, station conditions, weathers, and bus information of operating in real-time). The prototyping model is implemented by the machine learning tool (RapidMiner Studio) and conducted several tests for bus delays prediction according to specific circumstances. As a result, possibilities of transportation system are discussed for promoting the urban efficiency and the citizens' convenience by responding to urban conditions.

• 한국정보통신학회논문지
• /
• 제13권5호
• /
• pp.988-995
• /
• 2009

텔레매틱스 서비스 중 가장 보편적으로 사용되는 것이 출발지에서 목적지까지의 최단 경로 안내 서비스이다. 본 논문에서는 미래 시간에 대한 교통흐름 예측 결과를 바탕으로 한 동적 최단 경로 탐색 시스템을 개발하고 실시간교통정보를 이용한 다양한 실험을 수행하여 성능을 분석하였다. 교통흐름 예측은 베이지안 네트워크 (Bayesian network)를 이용한 예측 시스템을 사용하였다. 동일한 출발지와 목적지에 대해 동적 최단 경로와 정적 및 누적 최단 경로를 탐색하고 각 경로에 대한 통행 시간을 계산하여 실제 최단 경로의 통행시간과 비교하였다. 실험 결과 75% 이상의 비율로 동적 최단 경로의 통행시간이 정적이나 누적 최단 경로의 통행시간보다 실제 최단경로의 통행시간에 가깝게 나타났다. 따라서 중간 경유지에 도착 예정인 시간대의 교통 흐름을 예측하여 동적 최단 경로를 구하는 것이 출발시간의 교통흐름을 모든 구간에 적용하여 최단 경로를 구하는 정적 최단 경로에 비해 더 정확한 교통정보를 제공하여 텔레매틱스 서비스의 품질을 향상시킬 수 있음을 보여 주었다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.105-107
• /
• 2006

As internet streaming data increase, transport protocol such as TCP, TGP-Friendly is important to study control transmission rate and share of Internet bandwidth. In this paper, we propose a TCP-Friendly protocol using Neural Network for media delivery over wired Internet which has various traffic size(PTFRC). PTFRC can effectively send streaming data when occur congestion and predict one-step ahead round trip time and packet loss rate. A multi-layer perceptron structure is used as the prediction model, and the Levenberg-Marquardt algorithm is used as a traning algorithm. The performance of the PTFRC was evaluated by the share of Bandwidth and packet loss rate with various protocols.

• 한국통신학회논문지
• /
• 제33권5B호
• /
• pp.304-309
• /
• 2008

본 논문에서는 시계열 예측 모델을 이용하여 웡 또는 바이러스 등과 같은 공격 트래픽에 의해 네트워크상에 발생할 수 있는 트래픽 이상 징후를 탐지할 수 있는 예측 모델 기반 트래픽 이상 징후 탐지 기법을 제안한다. 제안 기법은 비교적 정확한 예측모델로 알려져 있는 ARIMA 모델을 이용하였고 이상 징후 여부를 확률값으로 변화하여 확률 임계값에 따라 이상 징후를 탐지하도록 하여 그 성능을 극대화할 수 있도록 하였다. 이를 위해 제안 기법을 네트워크상에 발생시킨 웜과 같은 비정상 공격 트래픽을 포함한 전체 트래픽과 웹 트래픽에 적용하여 트래픽의 이상 징후를 신뢰성 있는 수준에서 탐지함을 보여주었다. 이 기법을 네트워크 기반의 침입탐지시스템에 적용할 강제 큰 효과 가져올 수 있을 것이다.

• 전자공학회논문지C
• /
• 제36C권7호
• /
• pp.36-45
• /
• 1999

ATM망에서 트래픽 흐름을 제어하고 망 자원 사용을 효율적으로 사용하기 위해서는 폭주(Congestion)발생에 의한 망 성능 저하를 막고 폭주현상에 대처할 수 있는 적응적인 제어가 필요하다. 본 논문에서는 모든 트래픽에 대해 고정된 형태의 제어를 하는 Buffered Leaky Bucket과 적응성과 예측 기능을 갖는 신경회로망(Neural Network)을 이용하여 버퍼의 효율성을 높이고 망의 서비스 품질(QoS)로 구별되는 셀 손실율과 버퍼 지연을 테스트 및 성능 비교를 하였다. 또한 입력 트래픽의 다중화를 위해 사용되는 DWRR과 DWEDF의 셀 스케쥴링 알고리즘이 균등 지연을 만족할 수 있도록 개선하였다. 셀 스케쥴러로부터 망의 폭주 정보는 신경회로망을 이용한 Leaky Bucket에서 예측된 트래픽 손실율을 제어하고 손실율 정도에 따라 토큰 발생율과 버퍼 한계값은 제어된다. 이러한 트래픽 손실율 예측은 다음 입력 트래픽에 대한 손실과 버퍼지연을 줄일 수 있도록 제어의 효율성을 높일 수 있으며 다른 제어방식에도 응용될 수 있다. ATM 트래픽에 대한 신경회로망 학습과 예측 테스트를 위해 확률 랜덤 변수에 의해 발생된 셀 발생과 예측을 모의 실험하였으며, 이때 다양한 트래픽의 QoS가 향상되었음을 알 수 있었다.

• 한국도로학회논문집
• /
• 제20권2호
• /
• pp.57-65
• /
• 2018

PURPOSES : The study aims to predict the service life of national highway asphalt pavements through deep learning methods by using maintenance history data of the National Highway Pavement Management System. METHODS : For the configuration of a deep learning network, this study used Tensorflow 1.5, an open source program which has excellent usability among deep learning frameworks. For the analysis, nine variables of cumulative annual average daily traffic, cumulative equivalent single axle loads, maintenance layer, surface, base, subbase, anti-frost layer, structural number of pavement, and region were selected as input data, while service life was chosen to construct the input layer and output layers as output data. Additionally, for scenario analysis, in this study, a model was formed with four different numbers of 1, 2, 4, and 8 hidden layers and a simulation analysis was performed according to the applicability of the over fitting resolution algorithm. RESULTS : The results of the analysis have shown that regardless of the number of hidden layers, when an over fitting resolution algorithm, such as dropout, is applied, the prediction capability is improved as the coefficient of determination ($R^2$) of the test data increases. Furthermore, the result of the sensitivity analysis of the applicability of region variables demonstrates that estimating service life requires sufficient consideration of regional characteristics as $R^2$ had a maximum of between 0.73 and 0.84, when regional variables where taken into consideration. CONCLUSIONS : As a result, this study proposes that it is possible to precisely predict the service life of national highway pavement sections with the consideration of traffic, pavement thickness, and regional factors and concludes that the use of the prediction of service life is fundamental data in decision making within pavement management systems.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 하계종합학술대회논문집
• /
• pp.126-129
• /
• 1998

In order to control the flow of traffics in ATM networks and optimize the usage of network resources, an efficient control mechanism is necessary to cope with congestion and prevent the degradation of network performance caused by congestion. This paper proposes a new UPC(Usage Parameter Control) mechanism that varies the token generation rate and the buffer threshold of leaky bucket by using a Neural Network controller observing input buffers and token pools, thus achieving the improvement of performance. Simulation results show that the proposed adaptive algorithm uses of network resources efficiently and satisfies QoS for the various kinds of traffics.

• 한국도로학회논문집
• /
• 제14권6호
• /
• pp.103-110
• /
• 2012

PURPOSES : In Korea, over 70 percent of the land consists of mountainous and rolling area. Thus, tunnels continue its upward trend as road network are extended. In these circumstances, the importance of tunnel has been increased nowadays and then its safety investigation and research should be performed. This study is focus on confirming and improving the safety of tunnel. On tunnel hood, sunglare effect can irritate driver's behavior instantly and this can result in incident. METHODS : The study of this phenomenon is rarely conducted in domestic and foreign papers, so there is no proper measure for this. This study analyzes the driving environment of the effect of sunglare effect on tunnel hood. RESULTS : Traffic accidents stem from complex set of factors. This study build the Traffic Accident Prediction Models to find out the effect of sunglare effect on tunnel's hood. The independent variables are traffic volume, geometric design of road, length of tunnel and road side environment. Using these variables, this model estimates accident frequency on tunnel hood by Poisson regression model and Negative binomial regression model. Although Poisson regression model have more proper goodness of fit than Negative binomial regression model, Poisson regression model has overdipersion problem. So the Negative binomial regression model is used in this analysis. CONCLUSIONS : Consequently, the model shows that sunglare effect can play a role in driving safety on tunnel hood. As a result, the information of sunglare effect should be noticed ahead of tunnel hood so this can prevent drivers from being in hazard situation.

• 전기학회논문지
• /
• 제60권9호
• /
• pp.1767-1769
• /
• 2011

In a CR (cognitive radio) network, channel selection is one of the important issues for the efficient channel utilization. When the CR user exploits the spectrum of primary network, the interference to the primary network should be minimized. In this paper, we propose a spectrum hole prediction based channel selection scheme to minimize the interference to the primary network. To predict spectrum hole, statistic properties of primary user's traffic is used. By using the predicted spectrum hole, channel is selected and it can reduce the possibility of interference to the primary user and increase the efficiency of spectrum utilization. The performance of proposed channel selection scheme is evaluated by the computer simulation.

• International Journal of Precision Engineering and Manufacturing
• /
• 제6권3호
• /
• pp.29-36
• /
• 2005

With the rapid growth of the Internet, the Internet-based robot has been realized by connecting off-line robot to the Internet. However, because the Internet is often irregular and unreliable, the varying time delay in data transmission is a significant problem for the construction of the Internet-based robot system. Thus, this paper is concerned with the development of an Internet-based robot system, which is insensitive to the Internet time delay. For this purpose, the PPS (Position Prediction Simulator) is suggested and implemented on the system. The PPS consists of two parts : the robot position prediction part and the projective virtual scene part. In the robot position prediction part, the robot position is predicted for more accurate operation of the mobile robot, based on the time at which the user's command reaches the robot system. The projective virtual scene part shows the 3D visual information of a remote site, which is obtained through image processing and position prediction. For the verification of this proposed PPS, the robot was moved to follow the planned path under the various network traffic conditions. The simulation and experimental results showed that the path error of the robot motion could be reduced using the developed PPS.