Multiple-beam Interference Spectroscopy: Instrument Analysis and Spectrum Reconstruction

Mohamad Jouni,Daniele Picone,Mauro Dalla Mura

2024-10-29

Abstract:Hyperspectral imaging systems based on multiple-beam interference (MBI), such as Fabry-Perot interferometry, are attracting interest due to their compact design, high throughput, and fine resolution. Unlike dispersive devices, which measure spectra directly, the desired spectra in interferometric systems are reconstructed from measured interferograms. Although the response of MBI devices is modeled by the Airy function, existing reconstruction techniques are often limited to Fourier-transform spectroscopy, which is tailored for two-beam interference (TBI). These methods impose limitations for MBI and are susceptible to non-idealities like irregular sampling and noise, highlighting the need for an in-depth numerical framework. To fill this gap, we propose a rigorous taxonomy of the TBI and MBI instrument description and propose a unified Bayesian formulation which both embeds the description of existing literature works and adds some of the real-world non-idealities of the acquisition process. Under this framework, we provide a comprehensive review of spectroscopy forward and inverse models. In the forward model, we propose a thorough analysis of the discretization of the continuous model and the ill-posedness of the problem. In the inverse model, we extend the range of existing solutions for spectrum reconstruction, framing them as an optimization problem. Specifically, we provide a progressive comparative analysis of reconstruction methods from more specific to more general scenarios, up to employing the proposed Bayesian framework with prior knowledge, such as sparsity constraints. Experiments on simulated and real data demonstrate the framework's flexibility and noise robustness. The code is available at <a class="link-external link-https" href="https://github.com/mhmdjouni/inverspyctrometry" rel="external noopener nofollow">this https URL</a>.

Signal Processing,Numerical Analysis,Optimization and Control

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is the limitations of current multi - beam interference (MBI) spectral reconstruction techniques. Specifically, most of the existing reconstruction techniques are based on Fourier transform spectroscopy (FTS), and these methods are mainly applicable to two - beam interference (TBI) and have many deficiencies when dealing with MBI. These problems include:

1. **Influence of non - ideal factors**: The existing FTS methods are sensitive to non - ideal factors (such as irregular sampling and noise) in MBI devices, resulting in distorted reconstruction results.
2. **Harmonic overlap problem**: In the MBI system, harmonics will overlap within the spectral range of interest, which will lead to significant distortion in the inverse discrete Fourier transform (IDFT) method during the reconstruction process.
3. **Limited resolution**: The methods based on the discrete Fourier transform (DFT) are limited by the Nyquist - Shannon sampling theorem and cannot fully utilize non - zero spectral samples caused by harmonics.
4. **Irregular sampling**: In practical applications, the collected interferograms are usually irregularly sampled, while the DFT method assumes that the interferograms are regularly sampled, so reconstruction errors will occur.

To solve these problems, the author proposes a new unified framework, which is based on Bayes' formula and combines the physical model of the MBI system. Specific contributions include:

1. **Discretization analysis**: The numerical analysis of the MBI theoretical model is extended, and the well - posedness conditions of its discretized version (i.e., the transfer matrix) are studied.
2. **Reconstruction model improvement**: A transfer matrix that contains information about the real acquisition scenario (such as irregular sampling, reflectivity changes, harmonic contributions, and measurement noise) is proposed, and the spectral reconstruction problem is transformed into a linear inverse problem in the Bayes framework. Prior knowledge (such as sparse constraints) is introduced to improve the reconstruction accuracy.

Through these improvements, this framework can not only better handle the non - ideal factors of the MBI system but also improve the accuracy and robustness of spectral reconstruction. The experimental results show that this framework exhibits good flexibility and noise - resistance performance on both simulated data and real data.

### Formula summary

- **Interferogram signal**:
\[
y(\delta)=\int_{\Omega} A(\delta, \sigma) x(\sigma) d\sigma
\]
where \(A(\delta, \sigma)\) is the response function of the device.

- **Response function of Michelson interferometer**:
\[
A(\delta, \sigma)=2 T(\sigma)(1 + \cos(2\pi\delta\sigma))
\]

- **Response function of Fabry - Perot interferometer (Airy function)**:
\[
A(\delta, \sigma)=\frac{T^{2}(\sigma)}{1 + R^{2}(\sigma)-2 R(\sigma)\cos(2\pi\delta\sigma)}
\]

- **Discrete cosine transform (DCT) representation**:
\[
y_{l}=\langle q, x\rangle+2\sum_{n = 1}^{N - 1}\text{DCT}_{l}^{(n)}(q\odot r^{n}\odot x)
\]

These formulas and analyses provide a solid theoretical basis for understanding the spectral reconstruction of the MBI system and guidance for developing more effective reconstruction algorithms.