Near-optimal algorithms for shortest paths in weighted unit-disk graphs

From MaRDI portal



Abstract: We revisit a classical graph-theoretic problem, the extit{single-source shortest-path} (SSSP) problem, in weighted unit-disk graphs. We first propose an exact (and deterministic) algorithm which solves the problem in O(nlog2n) time using linear space, where n is the number of the vertices of the graph. This significantly improves the previous deterministic algorithm by Cabello and Jejv{c}iv{c} [CGTA'15] which uses O(n1+delta) time and O(n1+delta) space (for any small constant delta>0) and the previous randomized algorithm by Kaplan et al. [SODA'17] which uses O(nlog12+o(1)n) expected time and O(nlog3n) space. More specifically, we show that if the 2D offline insertion-only (additively-)weighted nearest-neighbor problem with k operations (i.e., insertions and queries) can be solved in f(k) time, then the SSSP problem in weighted unit-disk graphs can be solved in O(nlogn+f(n)) time. Using the same framework with some new ideas, we also obtain a (1+varepsilon)-approximate algorithm for the problem, using O(nlogn+nlog2(1/varepsilon)) time and linear space. This improves the previous (1+varepsilon)-approximate algorithm by Chan and Skrepetos [SoCG'18] which uses O((1/varepsilon)2nlogn) time and O((1/varepsilon)2n) space. More specifically, we show that if the 2D offline insertion-only weighted nearest-neighbor problem with k1 operations in which at most k2 operations are insertions can be solved in f(k1,k2) time, then the (1+varepsilon)-approximate SSSP problem in weighted unit-disk graphs can be solved in O(nlogn+f(n,O(varepsilon−2))) time. Because of the Omega(nlogn)-time lower bound of the problem (even when approximation is allowed), both of our algorithms are almost optimal.





Cited in
(27)








This page was built for publication: Near-optimal algorithms for shortest paths in weighted unit-disk graphs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2223616)