Imaging a moving point source from multi-frequency data measured at one and sparse observation directions (part I): far-field case

TL;DR

This paper introduces a multi-frequency algorithm for imaging the trajectory of a moving point source using sparse far-field data at observable directions, effectively recovering the source's path and domain in 2D and 3D.

Contribution

It develops a novel approach to image moving sources from limited far-field data by identifying observable directions and leveraging monotonicity properties.

Findings

Successfully images source trajectories in 2D and 3D.

Reveals the importance of observable directions for accurate imaging.

Demonstrates effectiveness with numerical examples.

Abstract

We propose a multi-frequency algorithm for imaging the trajectory of a moving point source from one and sparse far-field observation directions in the frequency domain. The starting and terminal time points of the moving source are both supposed to be known. We introduce the concept of observable directions (angles) in the far-field region and derive all observable directions (angles) for straight and circular motions. At an observable direction, it is verified that the smallest trip containing the trajectory and perpendicular to the direction can be imaged, provided the orbit function possesses a certain monotonical property. Without the monotonicity one can only expect to recover a thinner strip. The far-field data measured at sparse observable directions can be used to recover the $\Theta$-convex domain of the trajectory. Both two- and three-dimensional numerical examples are implemented to show effectiveness and feasibility of the approach.