Optical frequency comb (OFC) technology has revolutionized modern laser spec- troscopy, merging broad spectral coverage with the high accuracy and precision of atomic frequency standards. Among comb-based techniques, Dual-Comb Spectroscopy (DCS) stands out as a powerful tool for high-speed, calibration-free, and moving-part-free Fourier transform interferometry. This technique has transformative applications in a range of fields, including long-range remote sensing, human breath analysis, LiDAR (Light Detec- tion and Ranging), and hyperspectral imaging. However, the performance of DCS is fundamentally limited by the tight mutual coher- ence requirements between the two optical combs, where tiny relative phase and timing fluctuations (jitter) prevent coherent averaging, degrading the signal-to-noise ratio (SNR). Therefore, several methods were implemented to correct these fluctuations, such as hardware comb stabilisation and active feed-forward loops. Nevertheless, all these techniques require complex and bulky equipment, resulting in an expensive and not very scalable setup. This thesis presents a different approach: a Self-Corrected Dual-Comb Spectrometer that can bypass stringent hardware feedback loops via advanced numerical post-processing. By utilizing the optical interferogram itself to extract phase and frequency noise, an inverse- time-warping and Hilbert-transform-based phase correction algorithm is implemented to correct timing fluctuations a posteriori, restoring full mutual coherence and enabling long-term coherent averaging. In the final part of this work, an even lighter and more flexible approach is explored: a real-time self-correction implemented directly on dedicated hardware. To pave the way toward this kind of solution, the analytical self-correction algorithm is investigated for future hardware acceleration, evaluating its translation into a hardware-friendly Finite Impulse Response (FIR) filter framework. Preliminary steps are taken utilizing an AMD RFSoC 4x2 development board, establishing its development workflow within the Vivado design suite and testing basic test-bench programs. The board’s initial ac- quisition capabilities are characterized at a high level using synthesized waveforms from a generator, while the feasibility of the real-time pipeline is validated through Python-based simulations. This foundational exploration maps out the core architecture and identifies the primary implementation challenges, laying the groundwork for future on-chip real- time phase correction.
La tecnologia dei pettini di frequenza ottica (OFC) ha rivoluzionato la moderna spet- troscopia laser, unendo un’ampia copertura spettrale con l’elevata accuratezza e precisione degli standard di frequenza atomici. Tra le tecniche basate sui pettini, la spettroscopia a doppio pettine (DCS) si distingue come un potente strumento per l’interferometria a trasformata di Fourier ad alta velocità, senza calibrazione e priva di parti in movi- mento. Questa tecnica trova applicazioni rivoluzionarie in molteplici settori, tra cui il telerilevamento a lungo raggio, l’analisi del respiro umano, il LiDAR (Light Detection and Ranging) e l’hyperspectral imaging. Tuttavia le prestazioni della DCS sono fonda- mentalmente limitate dai rigidi requisiti di mutua coerenza tra i due pettini ottici, dove piccole fluttuazioni relative di fase e di tempo (jitter) impediscono l’operazione di media coerente, degradando il rapporto segnale-rumore (SNR). Pertanto, sono stati implemen- tati diversi metodi per correggere queste fluttuazioni, come la stabilizzazione hardware dei pettini e active feed-forward loops. Tuttavia, tutte queste tecniche richiedono apparec- chiature complesse e ingombranti, risultado in un setup costoso e poco scalabile. Questa tesi presenta un approccio differente: uno spettrometro a doppio pettine auto- corretto (self-correction) in grado di aggirare i rigidi feedback loops hardware tramite un’elaborazione numerica avanzata in post-processing. Utilizzando l’interferogramma ot- tico stesso per estrarre il rumore di fase e di frequenza, viene implementato un algoritmo di correzione di fase basato sull’inverse-time-warping e sulla trasformata di Hilbert per correggere le fluttuazioni temporali a posteriori, ripristinando la mutua coerenza e consentendo medie coerenti a lungo termine. Nell’ultima parte di questo lavoro, viene esplorato un approccio ancora più leggero e flessibile, una self-correction in tempo reale implementata direttamente su un hardware dedicato. Per spianare la strada verso questa soluzione, l’algoritmo analitico di autocor- rezione viene studiato in ottica di una futura accelerazione hardware, valutandone la con- versione nell’ambito di filtri a risposta impulsiva finita (FIR) adatti all’hardware. I passi preliminari vengono mossi utilizzando una scheda di sviluppo AMD RFSoC 4x2, definendo il flusso di lavoro all’interno della suite di progettazione Vivado e testando programmi di prova di base. Le capacità iniziali di acquisizione della scheda sono carat- terizzate ad alto livello utilizzando forme d’onda sintetizzate da un generatore, mentre la fattibilità della pipeline in tempo reale è validata attraverso simulazioni in Python. Questa esplorazione fondamentale delinea l’architettura di base e identifica le principali sfide di implementazione, gettando le basi per una futura correzione di fase in tempo reale direttamente sul chip.
Self-corrected dual-comb spectroscopy in the near-infrared by 1-GHz solid state lasers
IACONE, SERENA
2025/2026
Abstract
Optical frequency comb (OFC) technology has revolutionized modern laser spec- troscopy, merging broad spectral coverage with the high accuracy and precision of atomic frequency standards. Among comb-based techniques, Dual-Comb Spectroscopy (DCS) stands out as a powerful tool for high-speed, calibration-free, and moving-part-free Fourier transform interferometry. This technique has transformative applications in a range of fields, including long-range remote sensing, human breath analysis, LiDAR (Light Detec- tion and Ranging), and hyperspectral imaging. However, the performance of DCS is fundamentally limited by the tight mutual coher- ence requirements between the two optical combs, where tiny relative phase and timing fluctuations (jitter) prevent coherent averaging, degrading the signal-to-noise ratio (SNR). Therefore, several methods were implemented to correct these fluctuations, such as hardware comb stabilisation and active feed-forward loops. Nevertheless, all these techniques require complex and bulky equipment, resulting in an expensive and not very scalable setup. This thesis presents a different approach: a Self-Corrected Dual-Comb Spectrometer that can bypass stringent hardware feedback loops via advanced numerical post-processing. By utilizing the optical interferogram itself to extract phase and frequency noise, an inverse- time-warping and Hilbert-transform-based phase correction algorithm is implemented to correct timing fluctuations a posteriori, restoring full mutual coherence and enabling long-term coherent averaging. In the final part of this work, an even lighter and more flexible approach is explored: a real-time self-correction implemented directly on dedicated hardware. To pave the way toward this kind of solution, the analytical self-correction algorithm is investigated for future hardware acceleration, evaluating its translation into a hardware-friendly Finite Impulse Response (FIR) filter framework. Preliminary steps are taken utilizing an AMD RFSoC 4x2 development board, establishing its development workflow within the Vivado design suite and testing basic test-bench programs. The board’s initial ac- quisition capabilities are characterized at a high level using synthesized waveforms from a generator, while the feasibility of the real-time pipeline is validated through Python-based simulations. This foundational exploration maps out the core architecture and identifies the primary implementation challenges, laying the groundwork for future on-chip real- time phase correction.| File | Dimensione | Formato | |
|---|---|---|---|
|
2026_07_Iacone_Tesi.pdf
accessibile in internet per tutti
Descrizione: Tesi
Dimensione
27.66 MB
Formato
Adobe PDF
|
27.66 MB | Adobe PDF | Visualizza/Apri |
|
2026_07_Iacone_ExecutiveSummary.pdf
accessibile in internet per tutti
Descrizione: Executive Summary
Dimensione
1.32 MB
Formato
Adobe PDF
|
1.32 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/260032