Dynamic Bi-Colored Graph Partitioning

Conference Paper (2022)
Author(s)

Yanbin He (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Mario Coutino (TNO)

Elvin Isufi (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Geert Leus (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Research Group
Signal Processing Systems
More Info
expand_more
Publication Year
2022
Language
English
Research Group
Signal Processing Systems
Pages (from-to)
692-696
ISBN (electronic)
978-908279709-1
Event
30th European Signal Processing Conference, EUSIPCO 2022 (2022-08-29 - 2022-09-02), Belgrade, Serbia
Downloads counter
250
Collections
Institutional Repository

Abstract

In this work, we focus on partitioning dynamic graphs with two types of nodes (bi-colored), though not necessarily bipartite graphs. They commonly appear in communication network applications, e.g., one color being base stations, the other users, and the dynamic process being the varying connection status between base stations and moving users. We introduce a partition cost function that incorporates the coloring of the graph and propose solutions based on the generalized eigenvalue problem (GEVP) for the static two-way partition problem. The static multi-way partition problem is then handled by a heuristic based on the two-way partition problem. Regarding the adaptive partition, an eigenvector update-based method is proposed. Numerical experiments demonstrate the performance of the devised approaches.

No files available

Metadata only record. There are no files for this record.