Wave propagation in waveguides with random boundaries

Ricardo Alonso, Liliana Borcea, Josselin GarnierarXiv 2011 · DOI · arXiv

Abstract

This paper provides a detailed analysis of long-range cumulative scattering effects resulting from rough boundaries in waveguides. Assuming small random fluctuations of the boundaries, we derive a quantitative statistical description of the wave field. Our methodology utilizes coordinate transformations to straighten the boundaries, mapping the geometric complexity into the wave equation's coefficients. This approach reveals that boundary scattering is mathematically similar to propagation in waveguides with internal inhomogeneities, yet it exhibits distinct physical characteristics. We quantify the net effect of boundary scattering and demonstrate how it differs fundamentally from scattering caused by internal obstacles.

Introduction

In practical applications such as underwater acoustics or fiber optics, waveguides are rarely perfectly smooth. Natural variations or manufacturing imperfections introduce "roughness"—random fluctuations in the boundary geometry. While these fluctuations may be microscopic, their impact accumulates over long propagation distances, leading to a loss of coherence and the redistribution of energy among different wave modes. Predicting these effects is computationally intensive if one attempts to model every physical bump. Instead, this research develops a statistical framework to predict the long-range behavior of waves in such environments.

Methodology

To analyze the wave field without the geometric complexity of jagged boundaries, the study employs a three-step analytical framework:

  1. Coordinate Straightening: The authors apply a coordinate transformation that maps the fluctuating boundaries onto straight lines. This shifts the problem from a complex domain geometry to a wave equation with variable coefficients, effectively treating the boundary roughness as a perturbation in the medium's properties.
  2. Mode Decomposition: The wave field is decomposed into a superposition of discrete propagating and evanescent modes. By substituting the transformed equation into this modal framework, the authors derive a system of coupled random differential equations describing how modes interact.
  3. Diffusion Approximation: To capture the asymptotic behavior over long ranges, a diffusion approximation is applied. As the propagation distance increases, the complex amplitudes of the wave modes converge to a Markov diffusion process. This allows for the calculation of an infinitesimal generator, which describes the evolution of the wave's statistical properties, such as mean power and coherence, over distance.

Key Insights: Boundary vs. Internal Scattering

A central contribution of this work is the distinction between boundary scattering and internal scattering (e.g., scattering from internal "fog" or inhomogeneities).

  • The "Bouncy Ball" Effect: High-order modes—waves that strike the walls at steeper angles—interact with the rough boundaries more frequently. Consequently, these modes lose phase coherence much faster than low-order modes. However, they are less efficient at exchanging energy with other modes, leading to a significantly longer transport mean free path.
  • Low-Order Resilience: Low-order modes, which travel more parallel to the waveguide axis, maintain coherence over much longer distances. In boundary scattering, the primary degradation mechanism is a cumulative random phase shift. In contrast, internal scattering typically causes faster coherence loss in low-order modes because the primary mechanism is energy exchange between neighboring modes.

Quantitative Results and Scaling Laws

The study defines three critical, mode-dependent length scales to characterize the scattering process:

  • Scattering Mean Free Path: The distance at which a mode loses its initial phase coherence.
  • Transport Mean Free Path: The distance required for a wave to "forget" its initial direction of travel.
  • Equipartition Distance: The distance at which energy becomes uniformly distributed across all propagating modes.

In the high-frequency regime, where the number of modes $N$ is large, the authors establish scaling laws for attenuation. They demonstrate that the attenuation rate for high-order modes scales with $N^2$ or $N^3$ (depending on the correlation length of the boundary roughness), confirming that higher modes decay at an accelerated rate compared to lower modes. Furthermore, the research proves that while energy is exchanged between propagating and evanescent modes locally, the total energy of the propagating modes is conserved in the long-range limit.

Conclusion and Future Applications

While the current derivation assumes small fluctuations ($\epsilon \ll 1$) and focuses primarily on 2D waveguides, the statistical rules of thumb established here are foundational for understanding complex wave environments. These results have significant implications for sensor array imaging and target localization in environments like shallow water or tunnels. By understanding the statistical "fingerprint" of boundary roughness, researchers can improve the accuracy of sonar and seismic imaging systems that operate in imperfect waveguides.

Summary updated

Abstract

We give a detailed analysis of long range cumulative scattering effects from rough boundaries in waveguides. We assume small random fluctuations of the boundaries and obtain a quantitative statistical description of the wave field. The method of solution is based on coordinate changes that straighten the boundaries. The resulting problem is similar from the mathematical point of view to that of wave propagation in random waveguides with interior inhomogeneities. We quantify the net effect of scattering at the random boundaries and show how it differs from that of scattering by internal inhomogeneities.
BibTeX
@misc{alonso2011wave,
  author = {Ricardo Alonso and Liliana Borcea and Josselin Garnier},
  title = {{Wave propagation in waveguides with random boundaries}},
  year = {2011},
  eprint = {1111.1118},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1111.1118},
}

Loading related content...

Figure