Processing range-monitoring queries in mobile computing environment

Until now, range queries over mobile objects are typically supported by indexing linear trajectories of mobile objects. This strategy provides only static query results. As the objects continue to move, the query results could become obsolete quickly. Thus, a more realistic solution is to provide not only instant query results, but also continuous and real-time update until the queries are terminated explicitly. We call this new type of queries range-monitoring queries. Obviously, the traditional computing model cannot be used to support such queries. To keep query result updated, the same query has to be issued and processed repeatedly at a high frequency. While this could exhaust the server resource, it cannot provide real-time monitoring update.; In this thesis, we address the challenge of processing real-time range-monitoring queries by proposing efficient Monitoring Query Management (MQM) techniques. Since range-monitoring queries are continuous queries and many can be active simultaneously, existing database management systems need to be extended with real-time query management capability in order to support range-monitoring queries. Our research could be viewed as a step toward enhancing databases with such functionality in order to support mobile applications. In our environment, each mobile object is made aware of its nearby monitoring queries. When an object moves, it monitors its spatial relationship with these queries and updates server with its current location if its movement affects any query result. The benefit of this distributive approach is two fold. First, it relieves the server from overwhelming query evaluation since the server does not need to track the continuous movement of mobile objects. Because the same server can now be used to support a much larger number of mobile objects and monitoring queries, this strategy is highly scalable. Second, it requires only minimum location updates from mobile objects. Our approach is able to achieve accurate and real-time monitoring results without requiring constant location update from mobile objects. Thus, it is highly energy-efficient. At the server side, monitoring queries are managed and indexed using our new spatial access methods for efficient query distribution and real-time processing. Our strategy exploits the computing capability of mobile units to conserve their power in a fair way. The amount saved at each mobile unit is proportional to the amount of computing it is willing to contribute. Our simulation results indicate that, by exploring the heterogeneous mobile computing capability, our techniques offer significant performance improvement in terms of power consumption as well as service scalability.