Fiber-Optic Interferometry Using Narrowband Light Source and Electrical Spectrum Analyzer: Influence on Brillouin Measurement

Abstract

We observe an interference pattern using a simple fiber-optic interferometer consisting of an electrical spectrum analyzer and a narrowband light source, which is commonly employed for observing the Brillouin gain spectrum. This interference pattern expands well beyond the frequency range corresponding to the Brillouin frequency shift in silica fibers (ß11 GHz at 1.55 μm). Using both silica single-mode and polymer optical sensing fibers, we then experimentally prove that the distinctive noise in a selfheterodyne-based Brillouin measurement with an unoptimized polarization state originates from the interference between the reference light and the Fresnel-reflected light. This noise can be almost completely suppressed by employing a delay line that is longer than the coherence length of the light source and by artificially applying a high loss near the open end of the sensing fiber. Index Terms-Brillouinscattering, fiber-optic interferometry, nonlinear optics, polymer optical fiber. I. INTRODUCTION S UBSTANTIAL efforts have been directed toward the study of Brillouin scattering in optical fibers [1] in the last 40 years, and a number of related applications have been developed, including lasing [1], microwave signal processing [2], core alignment [3], optical memory [4], slow light generation [5], and distributed strain and temperature sensing [6]-[10]. To improve the performance of these applications, Brillouin scattering properties have been investigated not only in standard silica glass fibers [11], [12] but also in various specialty fibers, such as tellurite fibers [13], [14], bismuth-oxide fibers [14], chalcogenide fibers [15], photonic crystal fibers [16], and fibers doped with rare-earth ions (erbium, thulium, etc.) [17], [18]. Each fiber type has its own distinctive features; for instance, the Brillouin scattering power in tellurite and chalcogenide fibers is far higher than that in other fibers, whereas the Brillouin scattering power in erbium-doped fibers at 1.55 μm can be tuned by controlling the 980-nm pump power. However, all of these glass fibers are fragile and require careful handling; in sensing Manuscript

Observation of multimodal interference in millimeter-long polymer optical fibers

“…The locations of the dips were not identical to those in the SCS output. In addition, the intervals between the neighboring dips were not constant in the frequency domain, which indicates that this pattern does not originate from a fiber Fabry-Perot cavity [27,28]. Thus, we conclude that this pattern was caused by multimodal interference.…”

A Review of Hybrid Fiber-Optic Distributed Simultaneous Vibration and Temperature Sensing Technology and Its Geophysical Applications

“…DFB-Laser: Distributed Feedback-Laser, PC: Polarization Controller, OC: Optical Coupler, PD: Photodiode, and DAQ: Data Acquisition. (Adapted from [ 25 ].)…”

Measurement of the optical path length difference in an interferometer using a sinusoidally frequency-modulated light source

“…If the zero-OPLD point is located in the fiber under test (FUT; i.e., the second port of the circulator), in theory the measurement cannot be performed. Another example is a technique known as Brillouin OCDR [17][18][19] in conjunction with a distributed strain and temperature sensing system. Even when the zero-OPLD point is located outside the FUT, unless the OPLD is carefully selected, the signal-to-noise ratio (SNR) of the measurement drastically deteriorates because of the overlap of the interference pattern on the Brillouin gain spectrum [19].…”

“…Therefore, several techniques for measuring the OPLD indirectly have been developed. One such technique works by exploiting the interval between the interference fringes observed using a spectrum analyzer [19][20][21][22]. This method is simple, though it does not allow us to judge which of the two paths is longer; furthermore, the visibility of the interference pattern becomes extremely low when the OPLD is significantly longer than the coherence length of the light source (or when the frequency resolution of the spectrum analyzer is not sufficiently high).…”