Data aggregation scheduling on wireless mobile sensor networks

As urban traffic increases rapidly, the research in Intelligent Transportation Systems gets more and more attentions from the industry and academia. Real time navigations and traffic monitoring applications in ITS require a large number of real-time speed and location information. Such wireless mobile sensor networks composed by vehicles in ITS take such frequent change of wireless links and severe radio interference. Therefore, data collection algorithm in wireless mobile sensor networks should effectively use dynamic wireless links. This paper studies non-conflict data aggregation scheduling problem in wireless mobile sensor networks. The problem is formulated and proved as a NPC problem. Through analyzing inter-subtree conflict and inner-subtree conflict, part interference graph is constructed to filter candidate transmission time sets in order to remove inter-subtree conflicts. Then an algorithm based on dynamic programming is proposed to schedule transmission times on mobile aggregation routing tree. We conduct an extensive experimental study based on real taxi trajectory data sets. Experimental results show that our algorithm increases about 1/4 on the data collection rate compared with the existing VANET data aggregation algorithms and makes average delay less.