On the relation between time-reversed acoustics and Green’s function retrieval in space-variant and in time-variant materials
Kees Wapenaar a , Johannes Aichele b , Dirk-Jan van Manen b · 2026
Read On the relation between time-reversed acoustics and Green’s function retrieval in space-variant and in time-variant materials by Kees Wapenaar a , Johannes Aichele b , Dirk-Jan van Manen b (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-18382.txt.
Introduction
Kees Wapenaar, Johannes Aichele, and Dirk-Jan van Manen study two sound tricks and how they link. One trick plays a sound field back in time to focus it at its source. The next trick pulls a key response curve out of noise, with no need for a real source at all. For still stuff that does not change with time, the link between the tricks is well known.
Fresh work has pushed both tricks to stuff whose traits shift as time goes by. The team asks how the two tricks link in that new case. They also ask how that link maps to the old case. Sound tools of this kind serve scans of the earth, tests of parts, and care of health. A clear link helps staff pick the right trick for each job.
Catalog record
| CREATOR | Kees Wapenaar a , Johannes Aichele b , Dirk-Jan van Manen b |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 84 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-18382
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On the relation between time-reversed acoustics and Green’s function retrieval in space-variant and in time-variant materials Kees Wapenaar a , Johannes Aichele b , Dirk-Jan van Manen b Abstract Abstract The methods of time-reversed acoustics and Green’s function retrieval are traditionally deployed for classical inhomogeneous, time-invariant materials. The mutual relation between these methods is well-established. Recently, similar methods have been proposed for homogeneous, time-variant materials. Here we investigate their mutual relation and their relation with the corresponding methods in classical materials. For this analysis we make use of the fact that the wave equations for both types of material are similar, with the roles of time and space interchanged. However, the principle of causality holds for both types of material, hence, here the roles of time and space are not interchanged. We find that • whereas classical time-reversed acoustics involves emission of a time-reversed single-component wave field from a (ideally closed) boundary into the inhomogeneous material, its idealized counterpart involves emission of a sign-reversed two-component wave field, recorded in a time-reversed material, from a single time instant into the actual time-variant material; • whereas classical Green’s function retrieval involves temporal crosscorrelation of wave fields at two space locations in response to single-component sources on a (ideally closed) boundary, its counterpart involves spatial crosscorrelation of wave fields at two time instants in response to two-component sources at a single time instant. I Introduction Waves propagating through time-variant materials are scattered by time boundaries, just as waves propagating through space-variant materials are scattered by spatial boundaries. For electromagnetic waves, scattering by time boundaries has been studied for decades Morgenthaler58IRE ; Jiang75IEEE ; Mendonca2002PS . The research in this area expanded significantly since the introduction of dynamic metamaterials Koutserimpas2018IEEE ; Caloz2020IEEE1 ; Ramaccia2020OL ; Apffel2022PRL ; Moussa2023NP ; Pacheco2025PRB ; Stefanini2025PRL . For acoustic waves, the phenomenon of time scattering has been introduced by Fink and coworkers Bacot2016NP ; Fink2017EPJ ; Bal2019SIAM ; Hidalgo2023PRL ; Delory2025PRL . In particular, they demonstrated theoretically and experimentally that a time boundary sends waves back in space to focus at their origin, similar to time-reversed waves that are sent back from a spatial boundary Fink92IEEE . Whether we consider acoustic or electromagnetic waves, there exists much analogy between wave phenomena in homogeneous, time-variant materials, and those in “classical” inhomogeneous, time-invariant materials Mendonca2002PS ; Xiao2014OL ; Hoop2014WM ; Salem2015arXiv ; Torrent2018PRB ; Caloz2020IEEE2 ; Manen2024arXiv ; Wapenaar2024WM . In particular, the roles of time and space are interchanged between the wave equations for both types of material. However, the analogy also has its limitations. For example, waves reflected by a time boundary propagate forward in time and backward in space, similar as waves reflected by a spatial boundary Xiao2014OL ; Salem2015arXiv ; Caloz2020IEEE2 . Hence, here the roles of time and space are not interchanged. More generally, the principle of causality is a time-related property, whether we consider time-variant or space-variant materials Wapenaar2024WM . In other words, whereas the roles of time and space are interchanged between the wave equations for both types of material, they are in general not interchanged in the initial and boundary conditions, and hence the relation between the solutions of the wave equations (i.e., the wave fields in both types of material) is in general not straightforward. Another aspect that complicates the analogy is the fact that the number of time dimensions (one) is different from the number of space dimensions (two or three), except of course for 1D situations. Despite the aforementioned limitations, we investigate the relation between time-reversed acoustics and Green’s function retrieval in space-variant and in time-variant materials. For classical inhomogeneous, time-invariant materials, the research fields of time-reversed acoustics Fink92IEEE ; Draeger99JASA ; DeRosny2002PRL and Green’s function retrieval by temporal crosscorrelation Weaver2001PRL ; Campillo2003Science ; Wapenaar2003GEO ; Snieder2004PRE ; Schuster2004GJI ; VanManen2005PRL originated independently. However, it was soon discovered that these fields are closely related Derode2003JASA ; Wapenaar2005JASA . It is reasonable to assume that a similar close relationship exists between time-reversed acoustics Bacot2016NP ; Fink2017EPJ and Green’s function retrieval by spatial crosscorrelation Wapenaar2024WM ; Wapenaar2025RS in homogeneous, time-variant materials. In a companion paper Aichele2025arXiv we investigate the relations between temporal and spatial crosscorrelation, underlying time-reversed acoustics and Green’s function retrieval. In that analysis we keep the material space- and time-invariant, except for an impulsive temporal disruption, representing a time boundary. In this paper we extend the analysis for materials that are either arbitrarily space-variant or arbitrarily time-variant. After a brief discussion of the general equations for waves in space- and time-variant materials (section II ), in section III we review classical time-reversed acoustics, Green’s function retrieval and their mutual relation for arbitrarily inhomogeneous, time-invariant materials. This serves as an introduction to section IV , where we discuss time-reversed acoustics, Green’s function retrieval and their mutual relation for homogeneous, arbitrarily time-variant materials. In that section we also discuss the relation with the classical methods reviewed in section III . We present conclusions in section V . II General equations for waves in space- and time-variant materials Although in this paper we deal primarily with 2D and 3D acoustic waves, we use a unified notation that also covers 2D horizontally-polarized shear (SH) waves, 2D transverse electric (TE) and 2D transverse magnetic (TM) waves (the latter two are relevant because much of the theory for waves in time-variant materials was initially developed for electromagnetic waves). We denote position in space by the Cartesian coordinate vector {\bf x}=(x,y,z) (for the 3D situation) or {\bf x}=(x,z) (for the 2D situation), and time by t . With reference to Table 1 , we consider space- and time-variant wave field quantities U({\bf x},t) , {\bf V}({\bf x},t) , P({\bf x},t) and {\bf Q}({\bf x},t) , space- and time-variant material parameters \alpha({\bf x},t) and \beta({\bf x},t) , and space- and time-variant source quantities a({\bf x},t) , {\bf b}({\bf x},t) . The boldface quantities {\bf V} , {\bf Q} and {\bf b} denote vectors with three components, i.e. {\bf V}=(V_{x},V_{y},V_{z}) etc. (for the 3D situation), or with two components, i.e. {\bf V}=(V_{x},V_{z}) etc. (for the 2D situation). The wave field quantities obey the following unified equations \displaystyle\partial_{t}U+{\bf\nabla}\cdot{\bf Q} \displaystyle= \displaystyle a, (1) \displaystyle\partial_{t}{\bf V}+{\bf\nabla}P \displaystyle= \displaystyle{\bf b}, (2) where \partial_{t} stands for the partial differential operator \frac{\partial}{\partial t} and {\bf\nabla} is the nabla operator, defined as {\bf\nabla}=(\partial_{x},\partial_{y},\partial_{z}) (for the 3D situation) or {\bf\nabla}=(\partial_{x},\partial_{z}) (for the 2D situation). For acoustic waves, using the quantities defined in row 1 of Table 1 , equation ( 1 ) is the acoustic deformation equation, -\partial_{t}\Theta+{\bf\nabla}\cdot{\bf v}=q , and equation ( 2 ) stands for equilibrium of mechanical momentum, \partial_{t}{\bf m}+{\bf\nabla}p={\bf f} Hoop95Book ; Torrent2018PRB . Assuming the material is isotropic and instantaneously and locally reacting, the unified wave field quantities in equations ( 1 ) and ( 2 ) are related via the constitutive relations \displaystyle U \displaystyle= \displaystyle\alpha P, (3) \displaystyle{\bf V} \displaystyle= \displaystyle\beta{\bf Q}. (4) For acoustic waves these relations are -\Theta=\kappa p and {\bf m}=\rho{\bf v} , respectively. Equations ( 1 ) – ( 4 ) hold in materials of which the parameters \alpha({\bf x},t) and \beta({\bf x},t) vary continuously with space and time. They need to be supplemented with boundary conditions at those positions and times where these