Time reversal super resolution in randomly layered media

Jean‐Pierre Fouque, Josselin Garnier, Knut SølnaWave 2006 · DOI

TL;DR

This paper explores how time reversal techniques can achieve super-resolution imaging in randomly layered media, overcoming the diffraction limit. By exploiting multipathing of waves scattered by the medium's layers, they demonstrate that a larger time-reversal mirror can create a focal spot significantly smaller than what's possible in homogeneous media.

What Problem Does This Paper Address?

The ability to focus waves with a resolution finer than the classical diffraction limit (Rayleigh criterion) is a long-standing goal in various fields like acoustics and optics. While time reversal has shown promise for super-resolution, its effectiveness in complex, heterogeneous environments like randomly layered media has been a challenge. Previous work often focused on small mirrors comparable to the wavelength, or specific regimes. This paper tackles the problem of achieving super-resolution with a larger time-reversal mirror in a randomly layered medium, a scenario where multipathing becomes a dominant factor. They also extend this to focusing on a buried passive scatterer.

Methods

The authors employ a theoretical framework based on linear acoustics in three spatial dimensions. They model the randomly layered medium using a slab with fluctuating bulk modulus dependent only on depth. The core of their method involves:

  • Integral Representations: They derive detailed integral representations for the wavefield generated by both internal and surface sources, and for the time-reversed field. This is a standard approach in wave propagation but is meticulously applied here.
  • Time Reversal Procedure: This involves:
  • Recording the wavefield at a time-reversal mirror (a surface with spatial extent).
  • Time-reversing the recorded signal.
  • Re-emitting the time-reversed signal back into the medium.
  • Stationary Phase Analysis: This is a key mathematical tool used to analyze the asymptotic behavior of the integrals describing the wavefield, particularly to identify the conditions for refocusing and super-resolution.
  • Statistical Averaging: For the random medium case, they analyze the expectation and variance of the refocused pulse, leveraging results from previous work on the statistical properties of transmission and reflection coefficients.

Datasets

Not discussed in the paper. The paper is a theoretical study and does not present experimental data or introduce new datasets.

Key Insights and Ideas

  • Multipathing is Key to Super-Resolution: In randomly layered media, multipathing (waves taking multiple paths due to scattering) is not just noise; it's crucial. It effectively enhances the aperture of the time-reversal mirror, allowing for resolution beyond the diffraction limit.
  • Larger Mirrors Benefit from Randomness: Unlike homogeneous media where a larger mirror offers a predictable gain, in random media, a larger mirror, when combined with multipathing, can lead to super-resolution – a focal spot smaller than the diffraction limit.
  • Coherent vs. Incoherent Waves: The paper distinguishes between the refocusing of the coherent front wave and the incoherent coda waves. Both contribute to the overall refocusing, but the incoherent waves, due to their rich angular diversity from scattering, are particularly important for achieving super-resolution.
  • Focusing on Buried Scatterers: The time-reversal technique can be adapted to focus on a passive scatterer within the random medium, even if it's not directly illuminated by the mirror.

Key Results

  • Super-Resolution Achieved: The paper demonstrates theoretically that super-resolution can be achieved in randomly layered media with a time-reversal mirror larger than the wavelength. The focal spot radius can be significantly smaller than that obtained in a homogeneous medium.
  • Quantitative Focal Spot Sizes: Formulas are derived for the focal spot sizes in both homogeneous and random media. In the random medium, the spot sizes are shown to be dependent on the recording time window and the properties of the scattering.
  • Application to Passive Scatterers: The framework is extended to show that time reversal can also focus on a buried passive scatterer, with super-resolution effects also possible in this scenario.

Limitations and Future Work

  • Theoretical Framework: The results are derived theoretically. Experimental validation would be a crucial next step.
  • Simplified Medium Model: The random medium is modeled as a layered medium with z-dependent fluctuations. A truly 3D random medium might exhibit different behaviors.
  • Idealized Mirror: The mirror is assumed to have a large spatial extent and perfect recording capabilities.
  • Future Work: The authors suggest exploring the behavior in a more general 3D random medium and investigating the impact of different mirror shapes and recording strategies. They also mention that the O'Doherty-Anstey theory can quantitatively describe some differences in refocusing at a scatterer.

Summary updated

BibTeX
@inproceedings{fouque2006time,
  author = {Jean‐Pierre Fouque and Josselin Garnier and Knut Sølna},
  title = {{Time reversal super resolution in randomly layered media}},
  booktitle = {Wave},
  year = {2006},
  doi = {10.1016/j.wavemoti.2006.06.002},
}
Figure