Graph matching for crowdsourced data in mobile sensor networks

Abstract

We investigate the problem of graph matching to translate topological indoor localization to geographical localization, by modeling the building map and the semantic maps as graphs. A graph matching algorithm is proposed along with a node similarity measure based on finding the minimum distance between all sets of permutations of two vectors. We provide an efficient technique to calculate the similarity measurement, and prove its correctness via a theorem. The matching algorithm is shown to find all pairs of corresponding nodes correctly on real data.