Adaptive data dissemination in wireless sensor networks

Recent years have witnessed growing research interest in wireless sensor networks. In the literature, three data storage strategies, i.e., local, data-centric, and index-centric, have been proposed to answer one-shot queries originated from inside the network. Through careful analysis, we show in this paper that each of the three strategies respectively achieves the best performance in a certain range of query-rate to update-rate (Q2U) ratios. We therefore propose several adaptive schemes which periodically adjust the data storage and dissemination strategy in accordance with the Q2U ratio observed in the network. Experimental results demonstrate that in dynamic environments the proposed adaptive schemes substantially outperform the three basic strategies.