Interlayer synchronization in time-varying Kuramoto-Sakaguchi networks

TL;DR

This paper studies how time-varying interlayer connections in Kuramoto-Sakaguchi networks cause abrupt synchronization transitions, revealing that topology changes can significantly lower the number of links needed for high synchrony.

Contribution

It introduces a novel analysis of abrupt interlayer synchronization transitions driven by periodic topology changes in multiplex networks.

Findings

Abrupt transitions occur at lower interlayer link counts with time-varying topology.

Periodic topology changes facilitate synchronization by shrinking the basin of attraction.

The interlayer order parameter effectively measures synchronization states.

Abstract

We investigate the abrupt transition from low interlayer synchrony to high interlayer synchrony in a system of two identical layers of non-locally coupled Kuramoto-Sakaguchi oscillators using time-switching of the interlayer topology, while keeping the intralayer topology constant. As a measure of interlayer synchronization we introduce the interlayer order parameter, and determine the interlayer order parameter first in the static case as a function of the number of interlayer links observing a monotonically smooth transition to the fully synchronized state in the case that all nodes in layer 1 are connected to their corresponding nodes in layer 2. In the second step, we analyze the synchronization behavior of the multiplex network for a time-varying interlayer topology, observing abrupt transitions from the state of low synchrony to the state of high synchrony for a significantly lower number of interlayer connections. These abrupt transitions, even if not to full synchrony, appear only because of the periodic change of the interlayer topology, fostering the transition to occur for a significantly lower number of interlayer connections. We interpret this phenomenon as the shrinking, and ultimately, the nearly disappearance of the basin of attraction of a desynchronized network state. A comprehensive statistical analysis of this phenomenon is needed to interpret the results properly.