This report explains a how some partially informed attacker can estimate aggregate statistics of an unknown set of objects embedded in the Euclidean plane using available partial data, a nearest-neighbor oracle. The algorithm presented employs Voronoi Partitioning of the set space, where each object contained in the set is associated with some geo-location parameters. Next, a real world usage of the algorithm EstimateSUM is illustrated along with corresponding performance metric results. Some alternatives have been suggested to make the work of the attacker difficult.
Find complete report here: http://docdro.id/5HXe2wV