Local phase synchronization and clustering for the delayed phase-coupled oscillators with plastic coupling (Q739502)

The authors start with a network of \(n\) Kuramoto phase oscillators with dynamic coupling strengths. There are up to \(n^2\) connections, giving a total of \(n+n^2\) variables. They then incorporate distributed delays, so that sums are no longer over the current values of other oscillators' phases, but over the integrals of their values in the past. They then linearise this system about the perfectly synchronous (in terms of phases) state, but for some reason treat the connection strengths as constant, giving an \(n\)-dimensional linear system, instead of the expected \(n+n^2\)-dimensional one. This aside, the linear system will generically have solutions in the form of decaying or growing exponentials, giving the stability or otherwise of the synchronous state. However, the authors seem to concentrate on non-generic cases, where solutions neither grow nor decay, but are determined by the initial condition. I am doubtful about the relevance of the results obtained in this case to the original nonlinear system. The English is poor. E.g. ``Noting \(D_i\) is \(M\)-matrix and \(t_i\) is the minimum number''. ``locally frequency synchronization'' is mentioned but not defined. ``phase state set is spitted'' probably means ``phase state set is split'' etc.