文摘
Uncertain data has already widely existed in many practical applications, such as sensor networks, RFID networks, location-based services and mobile object management, etc. The skyline queries over uncertain data as an important aspect of uncertain data management, has received extensive attention from the database research community currently, due to its importance in many application including multi-criteria decision making, preference answering, market analysis, etc. However, in most uncertainty applications, the uncertain data are usually collected from vast number of independent data sources among geographically scattered sites, which makes the central assembly of data at one location for storage and query is infeasible and inefficient. Taking account of the network delay and limited bandwidth associated with sharing and communicating large amounts of distributed data over an internet, an important and challenging problem in the scenario is to retrieve all the global skyline tuples from all the distributed local sites with minimum communication cost. In this paper, we propose GFS, which is an efficient scheme for probabilistic skyline over distributed uncertain data. GFS firstly prunes the unqualified tuples with the global grid information and further iteratively prune the unqualified tuples with an improved feedback mechanism. Extensive experiments confirm that the effectiveness and the efficiency of the GFS scheme. Keywords Uncertain data Probabilistic skyline Distributed skyline Grid filtration