Year: 2023
Type: article
Abstract: Networks provide a powerful description of the complex systems that surround us. These structures support dynamical processes such as diffusion, synchronisation or percolation. The study of networks has remained limited to the pairwise formalism, where interactions take place between two participants. Simplicial complexes capture interactions between any number of elements. They are naturally equipped with a topological description, and support generalised dynamical variables called topological signals. Such signals can be found in many systems, yet have received little interest so far. Where studied, signals supported by structures of different dimension have been typically treated independently. This thesis explores the use of the Dirac operator as a tool rooted in topology to model the coupled dynamics of topological signals, and capture the interplay between signals of different orders. We apply this framework to describe the synchronisation dynamics of coupled phase os- cillators supported by nodes and edges in networks. We show through analytical and numerical investigations that the system undergoes explosive transitions between syn- chronised and incoherent states, while the synchronised state is characterised by coherent emergent oscillations. In the context of reaction-diffusion systems, we demonstrate how the Dirac operator is key to the formation of spatial heterogeneous patterns in topological signals supported by nodes and edges of a network. As a final illustration, we investi- gate two methods to filter and process real topological signals supported by nodes, edges and triangles in a simplicial complex. With the Dirac operator, we capture the interplay between signals of different orders and successfully jointly filter topological signals from artificial noise. This thesis illustrates how the Dirac operator provides a convenient formalism rooted in topology to consider the coupled dynamics of topological signals. This opens many further research avenues at the intersections of applied topology, dynamical systems, simplicial signal processing and higher-order network science.
Author: Calmon, L
Language: English
Subfield: Statistical and Nonlinear Physics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7011631404
