Compressed Sensing of Field-resolved Molecular Fingerprints Beyond the Nyquist Frequency

TL;DR

This paper demonstrates the first experimental application of compressed sensing in field-resolved molecular spectroscopy, allowing accurate detection of molecular fingerprints beyond the Nyquist frequency with reduced data acquisition time.

Contribution

It introduces a novel experimental approach employing random scanning for compressed sensing in time-domain molecular spectroscopy, surpassing the Nyquist limit.

Findings

Successfully identified water vapor absorption peaks up to 2.5 THz

Achieved mean squared error of 12×10^-4 in peak detection

Enabled real-time field-resolved fingerprinting

Abstract

Ultrashort time-domain spectroscopy and field-resolved spectroscopy of molecular fingerprints are gold standards for detecting samples' constituents and internal dynamics. However, they are hindered by the Nyquist criterion, leading to prolonged data acquisition, processing times, and sizable data volumes. In this work, we present the first experimental demonstration of compressed sensing on field-resolved molecular fingerprinting by employing random scanning. Our measurements enable pinpointing the primary absorption peaks of atmospheric water vapor in response to terahertz light transients while sampling beyond the Nyquist limit. By drastically undersampling the electric field of the molecular response at a Nyquist frequency of 0.8 THz, we could successfully identify water absorption peaks up to 2.5 THz with a mean squared error of 12 * 10^-4. To our knowledge, this is the first experimental demonstration of time-domain compressed sensing, paving the path towards real-time field-resolved fingerprinting and acceleration of advanced spectroscopic techniques.