A study of localized on-demand data collection in sensor networks

In sensor networks, users can collect the sensing data in an on-demand manner. Upon receiving a user's query message, each sensor node that is corresponding to the query delivers the sensing data to the sink node. However, directly forwarding query message throughout the network will cause a large traffic overhead and high energy consumption. In addition, directly forwarding all sensing data toward the sink node via many independent paths might be inefficient with regards to energy consumption and energy distribution. In this paper, we propose a localized query forwarding and data collection scheme based on the processing of the attribute content of the query's name. In contrast to the conventional distributions of data query, a query's name is resolved into the IDs and the locations of corresponding group of sensor nodes before being distributed to the network. Therefore, the query is efficiently forwarded into the network within a localized area. Furthermore, the sensing data are collected and aggregated at intermediate nodes that reside among the corresponding sensor nodes, before being forwarded to the sink node. We also attempt to extensively study the performance on an 802.15.4 integrated ns-2 simulator. The simulation results reveal that the proposed approaches highly reduce the energy consumption of data query and data collection in the network and the energy distribution is more efficiently distributed among sensor nodes than conventional approaches.