A partial index for distributed broadcasting in wireless mobile networks

To increase the scalability of location-based services (LBSs), researchers have begun to propose methods for processing spatial queries in wireless broadcast environments. In this paper, we examine the issue of spatial query processing within wireless broadcast systems and propose a distributed broadcast system (D-Bcast) model that can be applied to spatial query processing in such environments. Instead of receiving data directly from the server, the client in our proposed system model receives, via a subserver (Ssub), only the portion of the spatial data that has been requested by the querier, which reduces the broadcasting cycle and thus the access time. This distributed data broadcast model represents an efficient method of transmitting data to an arbitrary number of clients, including those that are moving across different main server (Smain) ranges. To implement this model, a hierarchical bitmap-based spatial index (HBI) employing a tree-based indexing scheme is proposed that divides the broadcast space into grids according to the distributions of objects. The grids are constructed such that an index tree for each grid can easily be configured. A full index of the broadcast program is constructed by applying the HBI to each Ssub, with the indexes of the Ssubs corresponding to the respective portions of the broadcast program used to broadcast data to the clients. In this manner, the total broadcasting cycle required to be read by the client is reduced and the query processing is decreased. Our proposed method is more effective than existing wireless broadcasting methods in reducing the location error that occurs when an update is induced by object movement. A cost model and experimental results are used to verify the efficacy of the proposed method.