High-performance optical chirp chain BOTDA by using a pattern recognition algorithm and the differential pulse-width pair technique

Optical chirp chain Brillouin optical time-domain analysis (OCC-BOTDA) has the capabilities of fast measurement, high Brillouin threshold, and freedom from the nonlocal effect; at the same time, however, it also has problems introduced by transient stimulated Brillouin scattering. The influence of the transient interaction is reflected as the broadened asymmetric Brillouin spectrum, the ghost peak, and the frequency shift of the main peak. This introduces difficulty in computing the fiber Brillouin frequency shift with good measurement accuracy. Besides, the OCC modulation causes additional noise due to the uneven amplitude response for different frequency components. In this work, we propose a high-performance OCC-BOTDA using the principal component analysis (PCA) based pattern recognition algorithm and differential pulse-width pair (DPP) technique. After building the Brillouin spectrum database (i.e., all patterns), the fiber intrinsic Brillouin frequency shift can be recognized by the PCA algorithm from a nonstandard Brillouin spectrum profile, resulting in good measurement accuracy. Meanwhile, the DPP technique, subtracting between two Brillouin signals generated by two wide-width pump pulses, is utilized to reduce the OCC modulation noise and avoid the pulse self-phase modulation effect in long-range BOTDA sensing. In the experiment, a temperature measurement with 1.3 MHz measurement precision, 4 m spatial resolution, and 5 s measurement time is achieved over a 100 km single-mode fiber.