Kinetic Euclidean minimum spanning tree

From HandWiki

A kinetic Euclidean minimum spanning tree is a kinetic data structure that maintains the Euclidean minimum spanning tree (EMST) of a set P of n points that are moving continuously. For the set of points P in 2-dimensional space, there are two kinetic algorithms for maintenance of the EMST.

Rahmati and Zarei[1] build a kinetic data structure based on the kinetic Delaunay triangulation to handle updates to the EMST in polylog time per event. Their kinetic data structure handles [math]\displaystyle{ O(n*m) }[/math] events, where m is the number of all changes to the Delaunay triangulation of the moving points. Their kinetic approach can work well for maintenance of the minimum spanning tree (MST) of a planar graph whose edge weights are changing as a continuous function of time.

Abam, Rahmati, and Zarei[2] provide a significant improvement on exact kinetic maintenance on the Euclidean minimum spanning tree. Their kinetic data structure handles a nearly cubic number of events.

References

  1. Rahmati, Zahed; Zarei, Alireza (2012). "Kinetic Euclidean minimum spanning tree in the plane". Journal of Discrete Algorithms 16: 2–11. doi:10.1016/j.jda.2012.04.009. 
  2. Ali Abam, Mohammad; Rahmati, Zahed; Zarei, Alireza (2012). "Kinetic Pie Delaunay Graph and Its Applications". Algorithm Theory – SWAT 2012. Lecture Notes in Computer Science. 2012. pp. 48–58. doi:10.1007/978-3-642-31155-0_5. ISBN 978-3-642-31154-3.