碩士 / 國立成功大學 / 電腦與通信工程研究所 / 100 / Nowadays, traffic congestion arises to be a very serious problem especially in metropolitan cities. It makes drivers to spend more time to their destinations. In this work, by leveraging the techniques of Vehicular Ad-hoc Networks (VANET) a dynamic navigation protocol, called STN, is proposed for individual vehicles to find time-efficient paths toward their given destinations. The trajectory information of vehicles is maintained in server to assist the planning of navigation path. In STN, a source vehicle sending a request message toward nearest access point to acquire the driving path. By comparing the trajectory and time information in the system, the future traffic load can be predicted. The traffic load information enables the server to estimate driving speed within different paths toward the destination and then further get a time-efficient path. In addition, adjustment, update and re-plan mechanisms are proposed to reduce the deviation of prediction. To evaluate its performance, the real road map of Shalu, including 20 road segments, is used. The simulation results demonstrate that STN, on average, could save around 14% driving time, compared to traveling along the vehicle-assisted shortest-time paths under density 40.
| Identifer | oai:union.ndltd.org:TW/100NCKU5652048 |
| Date | January 2012 |
| Creators | Pei-JinHe, 何佩瑾 |
| Contributors | Kuo-Feng Ssu, 斯國峰 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | en_US |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 39 |
Page generated in 0.0019 seconds