Server Side Data Selection Framework For Mobile GIS

• Main Authors: Chin-Sung Yang, 楊錦松
• Other Authors: Chung-Ming Chiu
• Format: Others
• Language: zh-TW
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/10141924213057324279

碩士 === 國立成功大學 === 測量工程學系碩博士班 === 92 === 　　The most significant feature of Mobile GIS is that its services may vary dependent on the different situation, demand, and location of individual users. Rather than the success of data transition, the current bottleneck lies on how to provide appropriate information and service based on the understanding of the users’ location and demand with rather limited bandwidth of transition. Challenges and problems include (1) users may lose willingness of using such a system with frequent and repeated communications due to the constraint of hardware/software and low bandwidth of transition, (2)ready-made maps do not possess the flexibility to change its content according to users requirements and position changes. (3) current systems usually lack the ability of Location-aware service. How to have a better understanding about users’ current situation and corresponding environment factors, so as to analyze possible data request from users or change provided services become a major challenge in the future. 　　By introducing location-aware concept, we develop a server-side data selection framework capable of dynamically selecting corresponding data that meets users’ demands based on their demands and locations. XML is chosen as the data transition format, as well as the way for recording request and response, between the server and clients. The designed framework also consists of the design of personal database, business data, and metadata of relating maps data and rule selection knowledge, such that the data selection framework can effectively generate the most appropriate data constraint to meet users’ demands. The spatial and thematic constraints then retrieve the corresponding data layers from data archives and send the results back to the clients with only the information needed by the users. This data selection framework, if necessary, can expand its data selection conditions to fit with different system requirements.