A Note on Reachability and Distance Oracles for Transmission Graphs

10/11/2022
by   Mark de Berg, et al.
0

Let P be a set of n points in the plane, where each point p∈ P has a transmission radius r(p)>0. The transmission graph defined by P and the given radii, denoted by 𝒢_tr(P), is the directed graph whose nodes are the points in P and that contains the arcs (p,q) such that |pq|≤ r(p). An and Oh [Algorithmica 2022] presented a reachability oracle for transmission graphs. Their oracle uses O(n^5/3) storage and, given two query points s,t∈ P, can decide in O(n^2/3) time if there is a path from s to t in 𝒢_tr(P). We show that the clique-based separators introduced by De Berg et al. [SICOMP 2020] can be used to improve the storage of the oracle to O(n√(n)) and the query time to O(√(n)). Our oracle can be extended to approximate distance queries: we can construct, for a given parameter ε>0, an oracle that uses O((n/ε)√(n)log n) storage and that can report in O((√(n)/ε)log n) time a value d_hop^*(s,t) satisfying d_hop(s,t) ≤ d_hop^*(s,t) < (1+ε)· d_hop(s,t) + 1, where d_hop(s,t) is the hop-distance from s to t. We also show how to extend the oracle to so-called continuous queries, where the target point t can be any point in the plane. To obtain an efficient preprocessing algorithm, we show that a clique-based separator of a set F of convex fat objects in R^d can be constructed in O(nlog n) time.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset
Success!
Error Icon An error occurred

Sign in with Google

×

Use your Google Account to sign in to DeepAI

×

Consider DeepAI Pro