Abstract: We analyze the behavior of two line arrangement algorithms,
a sweepline algorithm and an incremental algorithm, in approximate
arithmetic. The algorithms have running times
O(n
2
log n) and O(n
2
) respectively. We show that each of
these algorithms can be implemented to have O(nffl) relative
error. This means that each algorithm produces an
arrangement realized by a set of pseudolines so that each
pseudoline differs from the corresponding line relatively by
at most O(nffl). We also show... (Update)
...point arithmetic is a di#cult task even if robustness only means that the program should always run to completion. The papers [FM91, Mil88, SOI90] illustrate the di#culty of designing robust algorithms. Second, the repair step is non trivial if the floating point algorithm...
...types of error in numeric computations. Such errors has been studied, for example, in the context of computational geometry [8, 9, 13, 14, 21, 22, 23]. We discuss this further in Section 6. Approximate sorting. Bern, Karloff, Raghavan, and Schieber [3] introduced approz...
FORTUNE, S., AND MILENKOVIC, V. Numerical stability of algorithms for line arrangements. In Proc. 7th Annu. ACM Sympos. Comput. Geom. (1991), pp. 334--341. http://citeseer.nj.nec.com/fortune91numerical.html More
@inproceedings{ fortune91numerical,
author = "Steven Fortune and Victor Milenkovic",
title = "Numerical Stability of Algorithms for Line Arrangements",
booktitle = "Symposium on Computational Geometry",
pages = "334-341",
year = "1991",
url = "citeseer.nj.nec.com/fortune91numerical.html" }