Reconfiguration in heterogeneous mobile wireless sensor networks

We investigate reconfiguration in large-scale mobile wireless sensor networks. Consider a field of interest (FoI) divided into sub-regions, each of which has a distinct set of sensing requirements. Consider a mobile network consisting of nodes with heterogeneous sensing capabilities distributed within the FoI. We address the problem of reconfiguring the network at minimal cost so as to satisfy all sensing requirements while maintaining connectivity. We show that under certain conditions the problem may be decomposed into spatial and assignment sub-problems, and proceed to state a geometrical solution to the former and a mathematical programming approach to the latter. We state the resulting linear and convex integer programs, then develop an optimal decentralized solution algorithm to the convex program. Finally we proceed to evaluate average-case performance of our algorithm through simulation, concluding with a discussion of effective network design, solution approximation, and future directions for the problem.