Sub-band pulse compression and SERVS-based coherent fading suppression for LFM φ-OTDR

Abstract

Coherent fading remains a critical bottleneck for phase-sensitive optical time-domain reflectometry (φ-OTDR), leading to signal "dead zones" and phase discontinuities in distributed acoustic sensing (DAS). This paper proposes and experimentally validates a robust signal processing architecture that integrates parallel sub-band pulse compression with the sub-band extraction and rotating vector sum (SERVS) algorithm. The proposed method executes parallel matched filtering directly on the raw data, introducing frequency-domain constraints to ensure sub-band decorrelation. Experimental results on a 10 km sensing fiber using 2us linear frequency modulated (LFM) pulses (70-90 MHz) demonstrate that the proposed scheme reduces fading points by 21.3% compared to traditional SERVS methods. Furthermore, the system successfully reconstructs a 30 Hz vibration signal of piezoelectric transducer (PZT). This approach provides a cost-effective solution for enhancing the engineering stability of long-range monitoring systems without hardware modifications.