Fast simulation of Gaussian-mode scattering for precision interferometry. (5th January 2016)

• Record Type: Journal Article
• Authors: Brown, D; Smith, R J E; Freise, A
• Is Part Of: Journal of optics. Volume 18:Number 2(2016:Feb.)
• Publication Date: 2016-01-05
• Subjects: simulation -- scattering -- gaussian beams -- reduced order quadrature; Optics -- Periodicals; 535.05
• DOI: 10.1088/2040-8978/18/2/025604
• Languages: English
• ISSNs: 2040-8978

Abstract

Abstract: Understanding how laser light scatters from realistic mirror surfaces is crucial for the design, commissioning and operation of precision interferometers, such as the current and next generation of gravitational-wave detectors. Numerical simulations are indispensable tools for this task but their utility can in practice be limited by the computational cost of describing the scattering process. In this paper we present an efficient method to significantly reduce the computational cost of optical simulations that incorporate scattering. This is accomplished by constructing a near optimal representation of the complex, multi-parameter 2D overlap integrals that describe the scattering process ( referred to as a reduced order quadrature ). We demonstrate our technique by simulating a near-unstable Fabry–Perot cavity and its control signals using similar optics to those installed in one of the LIGO gravitational-wave detectors. We show that using reduced order quadrature reduces the computational time of the numerical simulation from days to minutes (a speed-up of ) while incurring negligible errors. This significantly increases the feasibility of modelling interferometers with realistic imperfections to overcome current limits in state-of-the-art optical systems. While we focus on the Hermite–Gaussian basis for describing the scattering of the optical fields, our method is generic and could be applied with any suitable basis. An implementation of this reduced order quadrature method is provided in the open source interferometer simulation software Finesse .