Scattering media for reconstructive wavemeters, interrogators and sensors
Scattering media for reconstructive wavemeters, interrogators and sensors
Coherent light scattering processes can lead to speckle pattern formation by complex multi-wave diffraction. This phenomenon seems chaotic and purely stochastic, but speckle motifs are highly deterministic and unique. The speckle pattern can be reproduced for a given input optical wavelength and electric field profile, providing sufficient system stability is achieved, thus realising industrial expectations of low-cost, small footprint, stable and high-resolution optical sensing systems. In this thesis, the concept of a reconstructive speckle pattern scattering medium-based device is discussed along with system engineering approaches, performance and stability tests, measurement reconstruction techniques and characteristics of several selected media: multi-mode fibre, laser-written free space 3D scattering chips and flat fibre-inscribed scattering matrices. The experiments demonstrated the flat fibre medium, which combined long-term intrinsic stability of the 3D chip and high-resolution measurement of the multimode fibre, resulting in a Raspberry Pi-controlled device acting as a wavemeter in range 1040-1056 nm with 0.1 nm resolution limited by the laser tuning step and 15.05 dB signal-to-noise ratio. Additionally, the system performance as a strain interrogator (0-200 microstrains), polarimeter (0-20°), magnetic field sensor (0-75 mT) and displacement meter (0-180 microns) with respective resolutions up to 4 microstrains, 0.04°, 8.37 mT and 1.25 microns and temporal stability of at least 144 h was confirmed. The demonstrated fine resolutions were achieved by developing dimensionality reduction machine-learning algorithms (singular value decomposition and principal component analysis), which overcome the speckles correlation limit. The presented devices realised a proof-of-concept for low-cost, compact and stable sensing solutions with the potential to compete with existing systems, yet further research is required for full system integration and application specific environmental testing.
University of Southampton
Falak, Przemyslaw Ludwik
f414a5d1-34e8-429d-8f9a-f203b6d83324
March 2024
Falak, Przemyslaw Ludwik
f414a5d1-34e8-429d-8f9a-f203b6d83324
Brambilla, Gilberto
815d9712-62c7-47d1-8860-9451a363a6c8
Beresna, Martynas
a6dc062e-93c6-46a5-aeb3-8de332cdec7b
Lee, Timothy
beb3b88e-3e5a-4c3f-8636-bb6de8040fcc
Falak, Przemyslaw Ludwik
(2024)
Scattering media for reconstructive wavemeters, interrogators and sensors.
University of Southampton, Doctoral Thesis, 197pp.
Record type:
Thesis
(Doctoral)
Abstract
Coherent light scattering processes can lead to speckle pattern formation by complex multi-wave diffraction. This phenomenon seems chaotic and purely stochastic, but speckle motifs are highly deterministic and unique. The speckle pattern can be reproduced for a given input optical wavelength and electric field profile, providing sufficient system stability is achieved, thus realising industrial expectations of low-cost, small footprint, stable and high-resolution optical sensing systems. In this thesis, the concept of a reconstructive speckle pattern scattering medium-based device is discussed along with system engineering approaches, performance and stability tests, measurement reconstruction techniques and characteristics of several selected media: multi-mode fibre, laser-written free space 3D scattering chips and flat fibre-inscribed scattering matrices. The experiments demonstrated the flat fibre medium, which combined long-term intrinsic stability of the 3D chip and high-resolution measurement of the multimode fibre, resulting in a Raspberry Pi-controlled device acting as a wavemeter in range 1040-1056 nm with 0.1 nm resolution limited by the laser tuning step and 15.05 dB signal-to-noise ratio. Additionally, the system performance as a strain interrogator (0-200 microstrains), polarimeter (0-20°), magnetic field sensor (0-75 mT) and displacement meter (0-180 microns) with respective resolutions up to 4 microstrains, 0.04°, 8.37 mT and 1.25 microns and temporal stability of at least 144 h was confirmed. The demonstrated fine resolutions were achieved by developing dimensionality reduction machine-learning algorithms (singular value decomposition and principal component analysis), which overcome the speckles correlation limit. The presented devices realised a proof-of-concept for low-cost, compact and stable sensing solutions with the potential to compete with existing systems, yet further research is required for full system integration and application specific environmental testing.
Text
Przemyslaw_Falak_PhD_Thesis_PDFA
- Version of Record
Restricted to Repository staff only until 23 March 2025.
Text
Final-thesis-submission-Examination-Mr-Przemyslaw-Falak
Restricted to Repository staff only
More information
Published date: March 2024
Identifiers
Local EPrints ID: 488515
URI: http://eprints.soton.ac.uk/id/eprint/488515
PURE UUID: 7b8c7231-39ae-4ddf-97b2-2d8be80106d3
Catalogue record
Date deposited: 26 Mar 2024 17:37
Last modified: 16 May 2024 01:58
Export record
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics