First stage of LISA data processing II: Alternative filtering dynamic models for LISA

TL;DR

This paper investigates alternative dynamic models and state vectors for filtering in LISA data processing to improve efficiency and extendability of gravitational wave detection algorithms.

Contribution

It introduces new dynamic models and reduced state vectors to enhance filtering speed and flexibility in LISA data analysis.

Findings

Reduced redundancy in state vectors

Faster filtering algorithms

Enhanced extendability to complex scenarios

Abstract

Space-borne gravitational wave detectors, such as (e)LISA, are designed to operate in the low-frequency band (mHz to Hz), where there is a variety of gravitational wave sources of great scientific value. To achieve the extraordinary sensitivity of these detector, the precise synchronization of the clocks on the separate spacecraft and the accurate determination of the interspacecraft distances are important ingredients. In our previous paper (Phys. Rev. D 90, 064016 [2014]), we have described a hybrid-extend Kalman filter with a full state vector to do this job. In this paper, we explore several different state vectors and their corresponding (phenomenological) dynamic models, to reduce the redundancy in the full state vector, to accelerate the algorithm, and to make the algorithm easily extendable to more complicated scenarios.