Spanners, Weak Spanners, and Power Spanners for Wireless Networks
For c ∈ \(\mathbb R\), a c-spanner is a subgraph of a complete Euclidean graph satisfying that between any two vertices there exists a path of weighted length at most c times their geometric distance. Based on this property to approximate a complete weighted graph, sparse spanners have found many applications, e.g., in FPTAS, geometric searching, and radio networks. In a weakc-spanner, this path may be arbitrary long but must remain within a disk of radius c-times the Euclidean distance between the vertices. Finally in a c-power spanner, the total energy consumed on such a path, where the energy is given by the sum of the squares of the edge lengths on this path, must be at most c-times the square of the geometric distance of the direct link.
While it is known that any c-spanner is also both a weak C 1-spanner and a C 2-power spanner (for appropriate C 1,C 2 depending only on c but not on the graph under consideration), we show that the converse fails: There exists a family of c 1-power spanners that are no weak C-spanners and also a family of weak c 2-spanners that are no C-spanners for any fixed C (and thus no uniform spanners, either). However the deepest result of the present work reveals that any weak spanner is also a uniform power spanner. We further generalize the latter notion by considering (c,δ)-power spanners where the sum of the δ-th powers of the lengths has to be bounded; so (·,2)-power spanners coincide with the usual power spanners and (·,1)-power spanners are classical spanners. Interestingly, these (·,δ)-power spanners form a strict hierarchy where the above results still hold for any δ ≥ 2; some even hold for δ > 1 while counterexamples exist for δ < 2. We show that every self-similar curve of fractal dimension d > δ is no (C,δ)-power spanner for any fixed C, in general.
Unable to display preview. Download preview PDF.
- 2.Eppstein, D.: The Geometry Junkyard: Fractals, http://www.ics.uci.edu/~eppstein/junkyard/fractal.html
- 3.Eppstein, D.: Beta-skeletons have unbounded dilation. Technical Report ICS-TR-96-15 (1996)Google Scholar
- 4.Eppstein, D.: Spanning trees and spanners. In: Handbook of Computational Geometry, pp. 425–461 (2000)Google Scholar
- 6.Fischer, M., Meyer auf der Heide, F., Strothmann, W.-B.: Dynamic data structures for realtime management of large geometric scenes. In: 5th Annual European Symposium on Algorithms (ESA 1997), pp. 157–170 (1997)Google Scholar
- 8.Jia, L., Rajaraman, R., Scheideler, C.: On local algorithms for topology control and routing in ad hoc networks. In: Proc. 15th ACM Symposium on Parallel Algorithms and Architectures (SPAA 2003), pp. 220–229 (2003)Google Scholar
- 9.Li, X.-Y., Wan, P.-J., Wang, Y.: Power efficient and sparse spanner for wireless ad hoc networks. In: IEEE International Conference on Computer Communications and Networks (ICCCN 2001), pp. 564–567 (2001)Google Scholar
- 10.Meyer auf der Heide, F., Schindelhauer, C., Volbert, K., Grünewald, M.: Energy, congestion and dilation in radio networks. In: Proc. 14th Symposium on Parallel Algorithms and Architectures (SPAA 2002), pp. 230–237 (2002)Google Scholar
- 13.Rao, S.B., Smith, W.D.: Approximating geometrical graphs via spanners and banyans. In: Proceedings of the 30th annual ACM symposium on Theory of computing, pp. 540–550 (1998)Google Scholar
- 15.Volbert, K.: Experimental Analysis of Adjustable Sectorized Topologies for Static Ad Hoc Networks. Accepted for DIALM-POMC (2004)Google Scholar
- 16.Wang, Y., Li, X.-Y.: Distributed Spanner with Bounded Degree for Wireless Ad Hoc Networks. In: Parallel and Distributed Computing Issues in Wireless networks and Mobile Computing, p. 120 (2002)Google Scholar
- 17.Wang, Y., Li, X.-Y., Wan, P.-J., Frieder, O.: Sparse power efficient topology for wireless networks. In: Proc. ACM Hawaii International Conference on System Sciences (HICSS 2002), p. 296 (2002)Google Scholar