面向智能交通系统的时空索引方法研究-计算机应用技术专业毕业论文.docxVIP

河海大学硕士研究生沦文 河海大学硕士研究生沦文 摘要 ABSTRACT With the continued adVances in wircless communica“ons，geo—posjtionin吕and ∞nsumer electronicS，an infrast兀lcture is emerging which enables 10cation_based seⅣices that rely on the tracking ofthe contjnuously changing positions of entire populations of service users，tenned moVing objects．This scenario is characterized by h唱e VoIumes of updates，for whjch rcason locatjon updat。tecbn0109ies become important．The main issue considered is how to represent the location of a mOVing object in a database so that tracking can be done with as few updates as possible． PreVjous works fbcus their research on update approaches，with better approaches could reduce update and improVe system pe—b肌an∞．But although these researches， they didn’t百ve proper index st兀lcture to sustain their update approaches．On the other hand，the researches On the spatio—temporal access methods fo册ally base their precondjtion on t王le free moVement(such as R—tree aJld TPR-trec)，not considcr road ne“vod(s． This paper first introduces LBS，then describes three existing tracking approaches， of：fIers empirical eValuations and comparisons arnong them，and presents pfoper access method for each one． Specially， it presents a new access stnIcture for the segment-based tracl【ing approaches and its optimization which indexes moVing objects not only by their absolute position but also by the roads they driVe． The paper gives the strIlcture description，jnsertion and search algorithms，then compares this stmcture with existing spatjO—temporal access method TPR—tree in sIructure size，insenion and query efficiency，which proVes it h培h—power for indexing moving objects in road networks． Key Words：Location Based Service(LBS)，tracMng，spatio·temporal access method，R-tree，TPR—tree 学位论文独创性声明： 学位论文独创性声明： 本人所呈交的学位论文是我个人在导师指导下进行的研究工作 及取得的研究成果。尽我所知，除了文中特别加以标注和致谢的地方 外，论文中不