Frequency-Competitive Query Strategies to Maintain Low Congestion Potential Among Moving Entities

TL;DR

This paper develops strategies for location queries that maintain low congestion among moving entities by balancing query frequency and congestion bounds, extending previous work with new algorithms and tight bounds.

Contribution

It introduces novel algorithms to guarantee fixed congestion bounds while minimizing query frequency, providing tight competitive bounds in a dynamic setting.

Findings

Established tight bounds on query frequency for congestion control.

Designed algorithms that guarantee fixed congestion with minimal queries.

Extended previous models to dynamic, unpredictable entity movement.

Abstract

We consider the problem of using location queries to monitor the congestion potential among a collection of entities moving, with bounded speed but otherwise unpredictably, in $d$-dimensional Euclidean space. Uncertainty in entity locations due to potential motion between queries gives rise to a space of possible entity configurations at each moment in time, with possibly very different congestion properties. We define different measures of what we call the congestion potential of such spaces, in terms of the (dynamic) intersection graph of the uncertainty regions associated with entities, to describe the congestion that might actually occur. Previous work [SoCG'13, EuroCG'14, SICOMP'16, SODA'19], in the same uncertainty model, addressed the problem of minimizing congestion potential using location queries of some bounded frequency. It was shown that it is possible to design a query scheme that is $O(1)$-competitive, in terms of worst-case congestion potential, with other, even clairvoyant query schemes (that know the trajectories of all entities), subject to the same bound on query frequency. In this paper we address the dual problem: how to guarantee a fixed bound on congestion potential while minimizing the query frequency, measured in terms of total number of queries or the minimum spacing between queries (granularity), over any fixed time interval. This complementary objective necessitates quite different algorithms and analyses. Nevertheless, our results parallel those of the earlier papers, specifically tight competitive bounds on required query frequency, with a few surprising differences.