A growth estimate for the monodromy matrix of a canonical system

TL;DR

This paper establishes a new upper bound on the growth of the monodromy matrix in canonical systems, linking it to the spectral distribution, and demonstrates the bound's sharpness through examples.

Contribution

It provides a generic, flexible upper estimate for the monodromy matrix's growth, improving upon earlier results and applicable to continuous Hamiltonians.

Findings

Derived a sharp growth estimate for the monodromy matrix.

Connected spectral sparsity with monodromy matrix growth.

Constructed examples showing the bound's optimality.

Abstract

We investigate the spectrum of 2-dimensional canonical systems in the limit circle case. It is discrete and, by the Krein-de Branges formula, cannot be more dense than the integers. But in many cases it will be more sparse. The spectrum of a particular selfadjoint realisation coincides with the zeroes of one entry of the monodromy matrix of the system. Classical function theory thus establishes an immediate connection between the growth of the monodromy matrix and the distribution of the spectrum. We prove a generic and flexibel upper estimate for the monodromy matrix, use it to prove a bound for the case of a continuous Hamiltonian, and construct examples which show that this bound is sharp. The first two results run along the lines of earlier work of R.Romanov, but significantly improve upon these results. This is seen even on the rough scale of exponential order.