Adaptive mobility-assisted data collection in wireless sensor networks

Mobility-assisted data collection in wireless sensor networks offers us a new approach to reducing and balancing the energy consumption of sensor nodes, however, the resultant data collection latency may be large due to the limited travel speed. Much research effort has been made on reducing the data collection latency with the scenario of a single mobile device. A potential problem with this is its scalability, and a straightforward solution is to employ multiple mobile devices to collect data collaboratively. In this paper, the data collection with multiple heterogeneous mobile devices, in terms of the capability to conduct data collection, is investigated. Our previous work, which models the network where a mobile sink visits sensor nodes and collects data as an M/G/1 queue, is extended, and new analytical results on the response time distribution, or equivalently the distribution of the data collection latency in the network, are presented. In addition, an adaptive data collection scheme is proposed, which schedules the data collection based on the service demand to reduce the operation cost. The accuracy of our modeling and analysis, along with the performance evaluation of the proposed data collection scheme, is verified through extensive simulations.