Computing and Information Systems - Theses
Permanent URI for this collection
Search Results
1 - 1 of 1
-
ItemPrivacy-aware spatial surveys( 2010)Surveying spatial knowledge may invade the privacy of individuals. The privacy concerns may significantly affect the quality of the collected data. We address the privacy issues for two classes of monitoring, implicit monitoring and explicit monitoring. In implicit monitoring, spatial data is collected without notifying the individuals being monitored. Implicit monitoring is a common means to surveying moving objects in large areas. Different to implicit monitoring, explicit monitoring needs the consent of individuals to collect private data. Such monitoring is normally conducted through public surveys. Both classes of monitoring can cause significant privacy issues. To address the privacy issues, we propose three approaches in this research.For implicit monitoring, we develop two data structures. The first data structure, Distributed Euler Histograms (DEHs), is designed for monitoring objects that can move freely in a space. It guarantees a high level of privacy protection by avoiding the collection of any identification information. The second data structure, Euler Histograms on Short ID (EHSID), is suitable for monitoring objects with constrained movements such as vehicles in road networks. A monitoring system built upon EHSID not only solves a broad range of aggregate queries on the spatial data, but also guarantees a high level of privacy protection as real identification data is not collected. For explicit monitoring, we focus on an innovative type of public surveys, negative surveys, which collect data that is complementary to the truth. We develop Gaussian Negative Surveys that significantly improve the accuracy level from the existing negative surveys.