Summary: | 碩士 === 長庚大學 === 資訊工程學系 === 105 === Traffic congestion is a problem in big cities, this problem arises because the
number of vehicles is not balanced by the growth of road, besides that due to several
unforeseen incidents, such as an accident. Thereby causing congestion. However, at
present the development and use of positioning technology and sensors can help to
collect huge amount of Vehicular Ad Hoc Network (VANET) traffic data in a real
time basis. Large volume of real-time data at a tremendous speed can be gathered
to help of roadside units using wireless sensors and on-board units of the vehicles.
In our works, we propose an algorithm to predict the possible of traffic density by
analyzing the VANET Big Data, by using classication and data mining models.
We have designed several designs on graph theory, for classication, of the existing
traffic by measurement of the affinity vector of each existing intersection and road
segment. The results of our experiments show that our proposed method has better
performance in terms of the accuracy.
|