Abstract
In a time reversal experiment, a signal recorded by an array of transducers and sent back time-reversed into the same medium approximately refocuses on the original source center. The refocusing resolution is improved in an inhomogeneous medium. In this work we study the effect of changes in the medium, namely, the case when back propagation takes place in a perturbed medium.
Introduction: The Robustness of Time Reversal
Time Reversal (TR) is a powerful technique where a wave signal is recorded by an array, reversed in time, and re-emitted into the medium. In complex, "random" environments—such as turbulent oceans or biological tissues—TR exhibits a remarkable phenomenon known as super-resolution. Unlike in a vacuum, where resolution is limited by the aperture of the recording array, multiple scattering in a random medium allows the array to capture a wider range of angles, leading to a tighter spatial focus.
However, real-world environments are rarely static. Between the initial recording and the subsequent re-emission, the medium may undergo small changes or perturbations. This paper investigates how such environmental shifts affect the stability and quality of the refocused signal.
Methodology
The researchers utilized a rigorous mathematical framework centered on the paraxial (parabolic) approximation. This approach assumes that waves primarily travel in one direction and that backscattering is negligible. To analyze the wave's energy density across both position and momentum, the authors employed Wigner transforms.
A key technical achievement was proving that as medium fluctuations become rapid, the complex random wave equations converge to a deterministic transport-diffusion equation (using the Kesten-Papanicolaou technique). This allowed the team to predict the behavior of the refocused signal analytically. These theoretical findings were then validated using 2D finite-difference simulations in a domain spanning thousands of wavelengths.
Key Insights: Spatial Stability vs. Temporal Smearing
The study reveals a fascinating dichotomy in how perturbations affect the refocused wave:
- Spatial Robustness: Surprisingly, small changes in the medium do not destroy the "super-resolution" effect. The wave still focuses into a tight spatial spot, often outperforming the resolution possible in a homogeneous medium. This focus is "self-averaging," meaning it depends on the statistical properties of the medium rather than a specific, lucky arrangement of scatterers.
- Temporal Smearing: While the spatial focus remains sharp, the time-domain signal is degraded. The perturbations act as a Gaussian low-pass filter. The refocused pulse becomes "blurred" or wider in time compared to the original source pulse, and its peak amplitude is attenuated.
Results and Implications
The authors derived a closed-form expression showing that the refocused signal in a perturbed medium ($u^B$) is a convolution of the unperturbed signal ($u^B_{unp}$) with a Gaussian kernel ($G$):
$$u^B(t) = (G \star u^B_{unp})(t)$$
Numerical simulations confirmed these predictions. Even when the backward medium included perturbations representing 5% of the total fluctuation, the spatial width of the focus remained narrow, while the temporal profile showed clear attenuation and broadening. This confirms that while the spatial "where" of the focus is robust, the temporal "what" of the signal carries the signature of the medium's change.
Conclusion
This work demonstrates that Time Reversal is remarkably robust to environmental changes in terms of spatial focusing. The discovery that perturbations lead to predictable temporal blurring opens new doors for environmental sensing. By measuring the degree of "smearing" in a TR experiment, researchers could potentially quantify shifts in complex media, such as monitoring changes in ocean currents or detecting subtle variations in biological tissue over time.
Summary updated