http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Analysis of Correlation between Regional Traffic Congestion Index and Population Distribution
Navin Ranjan(란잔나빈),Sovit Bhandari(반다리소빗),Hongping Zhao(조홍평),Hoon Kim(김훈) 대한전자공학회 2019 대한전자공학회 학술대회 Vol.2019 No.11
Traffic congestion has become a serious problem faced by large and growing cities as it hinders the growth of the city, lengthen the commuting time, and increase the frequency of the accidents. Many approaches to quantify the traffic congestion index (TCI) of the city and its correlation factor has been proposed in order to effectively ameliorate the congestion of city. However, they all refer to city level TCI, the normalized value over large region, which fail to emphasize on the impact of TCI on small region of the city on total congestion. This paper suggest that TCI for multiple small region is more informative than city level TCI value, in addition analyze the correlation of TCI with population density.
Neural Network Learning-based Traffic Jam Prediction Technique
Navin Ranjan(반다리소빗),Sovit Bhandari(란잔나빈),Hongping Zhao(조홍평),Hoon Kim(김훈) 대한전자공학회 2019 대한전자공학회 학술대회 Vol.2019 No.11
The accurate city-wide traffic congestion prediction can plays a vital role in management of the Transportation Network, it can assists both traffic administrators to take measures for maintaining smooth traffic flow and commuters to plan the optimal route in advance. Few algorithm has been proposed for congestion prediction based on Image data, but cannot perform well for large road network. This paper proposed an efficient congestion prediction algorithm based on convolutional neural network (CNN), long short-term memory (LSTM) and de -convolutional neural network (deCNN). The effectiveness of the proposed model is evaluated on Traffic Image data capture from Seoul Transportation Operation and Information Service (TOPIS), an online web service.
Artificial Intelligence Enabled Fog Radio Access Networks : A Case Study
Sovit Bhandari(반다리소빗),Navin Ranjan(란잔나빈),Hongping Zhao(조홍평),Hoon Kim(김훈) 대한전자공학회 2019 대한전자공학회 학술대회 Vol.2019 No.11
Fog computing is a newly evolved network architecture that brings cloud computing benefit to the edge of the network. This structure is designed to solve the drawbacks of conventional cellular structure. Meanwhile, Artificial Intelligence (AI) applications are booming with the breakthroughs in deep learning (DL) models. Many iterative algorithm has been proposed to optimize the performance of fog radio access networks (FRANs), but cannot perform well in the live networks. To improve this, AI based optimization solution should be implemented. This paper discusses about;1) the prospective areas of fog computing where AI can be used; 2) existing challenges and future trends.