• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 하계학술대회 논문집 D
• /
• pp.2956-2958
• /
• 2000

This paper describes the design of a real time traffic data acquisition system using loop detector and piezzo sensor. Loop detector is the cheapest method to measure the speed and piezzo is used to detect the vehicle axle information. A ISA slot based I/O board is designed for data acquisition and PC process the raw traffic data and transfer the data to the host system. Simulation kit is designed with toy car kits. simulated loop detector and piezzo sensor. The data acquisition system collects up to 10 lane highway traffic data such as vehicle count. speed. length axle count. distance between the axles. The data is processed to generate traffic count, vehicle classification, which are to be used for ITS. The system architecture and simulation data is included and the system will be tested for field operation.

• ETRI Journal
• /
• 제41권5호
• /
• pp.560-573
• /
• 2019

Two supervised learning algorithms, a basic neural network and a long short-term memory recurrent neural network, are applied to traffic including DDoS attacks. The joint effects of preprocessing methods and hyperparameters for machine learning on performance are investigated. Values representing attack characteristics are extracted from datasets and preprocessed by two methods. Binary classification and two optimizers are used. Some hyperparameters are obtained exhaustively for fast and accurate detection, while others are fixed with constants to account for performance and data characteristics. An experiment is performed via TensorFlow on three traffic datasets. Three scenarios are considered to investigate the effects of learning former traffic on sequential traffic analysis and the effects of learning one dataset on application to another dataset, and determine whether the algorithms can be used for recent attack traffic. Experimental results show that the used preprocessing methods, neural network architectures and hyperparameters, and the optimizers are appropriate for DDoS attack detection. The obtained results provide a criterion for the detection accuracy of attacks.

• 한국ITS학회 논문지
• /
• 제19권6호
• /
• pp.96-106
• /
• 2020

본 연구에서는 루프 센서를 통한 교통량 수집방식의 오류를 해결하기 위해 1종(승용차)과 3종(일반 트럭)의 구분이 어려운 부분 및 영상 이미지의 단점을 보완하기 위해 도로변에 열화상 카메라를 설치하여 영상 이미지를 수집하였다. 수집된 영상 이미지를 레이블링 단계를 거쳐 1종(승용차)과 3종(일반 트럭)의 학습데이터를 구성하였다. 정지영상을 대상으로 labeling을 진행하였으며, 총 17,536대의 차량 이미지(640x480 pixel)에 대해 시행하였다. 열화상 영상 기반의 차종 분류를 달성하기 위해 CNN(Convolutional Neural Network)을 이용하였으며, 제한적인 데이터량과 품질에도 불구하고 97.7%의 분류정확도를 나타내었다. 이는 AI 영상인식 기반의 도로 교통량 데이터 수집 가능성을 보여주는 것이라 판단되며, 향후 더욱더 많은 학습데이터를 축적한다면 12종 차종 분류가 가능할 것이다. 또한, AI 기반 영상인식으로 도로 교통량의 12종 차종뿐만 아니라 다양한(친환경 차량, 도로 법규 위반차량, 이륜자동차 등) 차종 분류를 할 수 있을 것이며, 이는 국가정책, 연구, 산업 등의 통계 데이터로 활용도가 높을 것으로 판단된다.

• 대한토목학회논문집
• /
• 제29권3D호
• /
• pp.347-356
• /
• 2009

본 연구는 기존의 도로기능분류 정의와 방법론을 벗어나 교통특성에 따른 도로분류 방법론인 도로특성분류를 기초로 분석을 수행하였다. 도로특성분류에 대한 일련의 과정 중에서 다양한 교통특성을 반영하는 설명변수를 기초로 요인점수를 산출하고, 동질한 도로구간을 그룹핑하는 군집화 분석과정과 적정 군집수 도출에 따른 군집결과비교에 본 연구는 초점을 맞추었다. 도로분류를 위해 병합적 계층 군집분석인 Ward법, 비계층적 군집분석인 K-means법, 자율신경 회로망을 이용한 K-SOM을 사용하여 비교분석하였다. 각 군집기법에 대한 결과를 토대로 비교분석한 결과, 군집 수 5 이하에서는 K-means법, 군집 수 14 이상에서는 Kohonen selforganizing maps가 가장 우수한 것으로 나타났으며, 군집수 5~9사이에서는 Ward법과 Kmeans법의 군집 성능이 불규칙한 패턴을 보임에 따라 세밀한 결과분석을 통해 우수성을 결정하는 것이 바람직할 것으로 분석되었다. 본 연구결과는 다양한 교통특성을 고려한 도로구간의 군집 속성을 분석하고 예측하는 분류화 작업에 중요한 기초적인 자료로 사용될 것으로 기대된다.

• 한국도로학회논문집
• /
• 제18권3호
• /
• pp.109-114
• /
• 2016

PURPOSES : The demand for extending national highways is increasing, but traffic monitoring is hindered because of resource limitations. Hence, this study classified highway segments into 5 types to improve the efficiency of short-term traffic count planning. METHODS : The traffic volume trends of 880 highway segments were classified through R-squared and linear regression analyses; the steadiness of traffic volume trends was evaluated through coefficient of variance (COV), and the normality of the data were determined through the Shapiro-Wilk W-test. RESULTS : Of the 880 segments, 574 segments had relatively low COV and were classified as type 1 segments, and 123 and 64 segments with increasing and decreasing traffic volume trends were classified as type 2 and type 3 segments, respectively; 80 segments that failed the normality test were classified as type 4, and the remaining 39 were classified as type 5 segments. CONCLUSIONS : A theoretical basis for biennial count planning was established. Biennial count is recommended for types 1~4 because their mean absolute percentage errors (MAPEs) are approximately 10%. For type 5 (MAPE =19.26%), the conventional annual count can be continued. The results of this analysis can reduce the traffic monitoring budget.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권10호
• /
• pp.3771-3792
• /
• 2021

With the rapid development of science and technology, several high-performance networks have emerged with various new applications. Consequently, financially or socially motivated attacks on specific networks have also steadily become more complicated and sophisticated. To reduce the damage caused by such attacks, administration of network traffic flow in real-time and precise analysis of past attack traffic have become imperative. Although various traffic analysis methods have been studied recently, they continue to suffer from performance limitations and are generally too complicated to apply in existing systems. To address this problem, we propose a method to calculate the correlation between the malicious and normal flows and classify attack traffics based on the corresponding correlation values. In order to evaluate the performance of the proposed method, we conducted several experiments using examples of real malicious traffic and normal traffic. The evaluation was performed with respect to three metrics: recall, precision, and f-measure. The experimental results verified high performance of the proposed method with respect to first two metrics.

• 대한약침학회지
• /
• 제10권2호통권23호
• /
• pp.133-137
• /
• 2007

Objectives : This study was conducted to classify the TA(traffic accident) patients who admissed in Oriental Medical Hospital. Methods : We divided TA patients who admissed in Oriental Medical Hospital by age, sex, visited period after onset, impact region, disease, and chief complaint according to a patient classification table, and analyzed their characteristics. Results and Conclusion : The patients who visited within 1 week after TA numbered most(65.4%) in classification by period. Most impact region was neck(80%). Most of TA patients were suffered and treated by cervical(80%) and lumbar(56.9%) sprain.

• 한국정보시스템학회지:정보시스템연구
• /
• 제14권3호
• /
• pp.97-104
• /
• 2005

The paper proposes an implementation architecture of DiffServ-over-MPLS traffic engineering (TE) on Intel IXP2400 network processor using Intel IXA SDK 4.0 Framework. Program architecture and functions are described. Also fast and scalable range-match classification scheme is proposed for DiffServ-over-MPLS TE that has been integrated with functional blocks from Intel Microblocks library. Performance test shows that application can process packets at approximate data rate of 3.5 Gbps. The proposed implementation architecture of DiffServ-over-MPLS TE on Network processor can provide guaranteed QoS on high-speed next generation Internet, while being flexible and easily modifiable.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2005년도 추계학술대회논문집
• /
• pp.666-669
• /
• 2005

Social noise survey and auditory experiment on residential noises such as floor impact, air-borne, bathroom, drainage and traffic noises were conducted to classily a noise insulation Performance in apartment building. The survey results showed that annoyance among subjective responses to residential noises was most greatly affecting to satisfaction with noises. In the survey, boundary limit between satisfaction and dissatisfaction was also determined. Auditory experiments was also undertaken to determine noise insulation performance according to the percent of satisfaction for individual noise source. Result of auditory experiment showed that the noise insulation performance for floor impact, airborne, drainage and traffic noise corresponding to 40 % satisfaction is 49 dB (L$_{i,Fmax,AW}$), 48 dB (R'w), N-41, and N-40, respectively. Finally, classes of noise insulation performance in apartment building were proposed according to satisfaction with noises

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2019년도 제59차 동계학술대회논문집 27권1호
• /
• pp.147-148
• /
• 2019

네트워크 응용 서비스들은 점점 더 복잡해지고 있으며, 네트워크 통신 기술의 발전과 함께 네트워크의 특성, 네트워크 관리 및 혼잡 제어에 대한 높은 요구 사항을 제시하므로 네트워크 트래픽 분류가 점점 더 중요해지고 있다. 트래픽 분류는 다양한 특성에 따라 네트워크 트래픽을 여러 클래스로 분류하여 처리하는 작업이다. 본 논문에서는 현재 네트워크 분야에서 적용된 여러 트래픽 분류 기법을 조사한다. 이를 통해 SDN(Software Defined Networking) 환경에서 효율적인 트래픽 분류가 가능한 기법 선택을 위해 비교하며 향후 연구를 위해 트래픽 분류 기법들을 소개한다.