The sparsity of the response of a quasi-distributed fiber optic sensing system allows ‘overclocking’ its interrogation

Nadav Arbel*, Maya Yanko, Avishay Eyal

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

The response of a quasi-distributed fiber optic sensing system to a short interrogation pulse can be considered as a sparse signal in the time domain. This attribute can be exploited to enhance its performance beyond the commonly accepted limits. The work described here is analogous to some compressed sensing systems, which can be adequately sampled at sub-Nyquist rates, based on knowledge of their sparsity in the spectral domain. Equivalently, it is shown here how the sparsity of the impulse-response of a Quasi-Distributed Acoustic Sensing (Q-DAS) system can be exploited to interrogate it at rates well above the limit dictated by the total duration of its impulse response. An additional improvement in performance is achieved by using a Perfect Periodic Auto-correlation (PPA) code, rather than a pulse, for interrogation of the system. Cross-correlating the raw response of the Q-DAS system with the original code retains the sparsity of the response and provides significant gain in SNR without losing spatial resolution. Experimentally, the method was utilized to implement a 76 km long acoustic sensing system and successfully detected 250 kHz ultrasound signal. This was achieved by interrogating the system at a rate of 971 kHz which is 737 times the inherent limit dictated by the duration of the impulse-response of the system.

Original languageEnglish
Article number102373
JournalOptical Fiber Technology
Volume61
DOIs
StatePublished - Jan 2021

Keywords

  • Distributed Acoustic Sensing
  • Perfect Periodic Auto-correlation Sequence
  • Quasi-distributed sensors

Fingerprint

Dive into the research topics of 'The sparsity of the response of a quasi-distributed fiber optic sensing system allows ‘overclocking’ its interrogation'. Together they form a unique fingerprint.

Cite this