Brillouin optical spectrum analyzer monitoring of subcarrier-multiplexed fiber-optic signals

Transcription

1 Brillouin optical spectrum analyzer monitoring of subcarrier-multiplexed fiber-optic signals Yonatan Stern, 1 Kun Zhong, 1,2 Thomas Schneider, 3 Yossef Ben-Ezra, 4 Ru Zhang, 2 Moshe Tur, 5 and Avi Zadok 1, * 1 Faculty of Engineering, Bar-Ilan University, Ramat-Gan 529, Israel 2 School of Science, Beijing University of Posts and Telecommunications, Beijing 1876, China 3 Institut für Hochfrequenztechnik, Hochschule für Telekommunikation, D-4277 Leipzig, Germany 4 Optiway Integrated Solutions Ltd., 11 Ha avoda St., Rosh Ha ayin, Israel 5 School of Electrical Engineering, Faculty of Engineering, Tel-Aviv University, Tel-Aviv 69978, Israel *Corresponding author: Avinoam.Zadok@biu.ac.il Received 28 June 213; accepted 27 July 213; posted 1 August 213 (Doc. ID 19357); published 23 August 213 Optical spectral analysis of closely spaced, subcarrier multiplexed fiber-optic transmission is performed, based on stimulated Brillouin scattering (SBS). The Brillouin gain window of a single, continuous-wave pump is scanned across the spectral extent of the signal under test. The polarization pulling effect associated with SBS is employed to improve the rejection ratio of the analysis by an order of magnitude. Ten tones, spaced by only 1 MHz and each carrying random-sequence on off keying data, are clearly resolved. The measurement identifies the absence of a single subcarrier, directly in the optical domain. The results are applicable to the monitoring of optical orthogonal frequency domain multiplexing and radio over fiber transmission. 213 Optical Society of America OCIS codes: (6.437) Nonlinear optics, fibers; (6.5625) Radio frequency photonics; (7.479) Spectrum analysis; (29.59) Scattering, stimulated Brillouin Introduction Optical fiber communication networks are required to increase their transmission efficiency to keep pace with growing demands for capacity. One strategy, which is rapidly being adapted from the realm of wireless communication, is the multiplexing of densely packed, orthogonal subcarrier tones, referred to as optical orthogonal frequency division multiplexing or O-OFDM [1]. The separation between neighboring data-carrying tones can be as narrow as a few megahertz [2]. Subcarrier multiplexed transmission is also prevalent in radio-over-fiber X/13/ $15./ 213 Optical Society of America (RoF) applications [3], in which fibers are used to extend the reach of wireless communication. The monitoring of transmission at the optical layer improves the quality of service, reduces downtimes and assists in the identification of faults. Optical spectrum analyzers (OSAs) are among the most fundamental and widely employed tools of signal monitoring. Most conventional OSAs rely on diffraction-grating filters, and at best provide a spectral resolution not better than 1 GHz. Such resolution is often insufficient for the monitoring of high-end analog and digital fiber-optic transmission systems. High spectral resolution may be obtained via heterodyne beating of the signal under test (SUT) with an optical local oscillator, and subsequent radio-frequency (RF) analysis of the photocurrent. However, the realization of coherent analyzers could 1 September 213 / Vol. 52, No. 25 / APPLIED OPTICS 6179

2 be challenging owing to the need for low-noise, highly stable local oscillators [4,5]. Alternatively, high-resolution incoherent OSAs have been proposed and demonstrated over the last two decades based on stimulated Brillouin scattering (SBS). In SBS, a relatively intense pump wave interacts with a counterpropagating, typically weaker signal wave, which is detuned in frequency [6]. Effective coupling, however, requires that the difference between the two optical frequencies should closely match the Brillouin frequency shift ν B 11 GHz. The power of a signal wave whose optical frequency is ν B below that of the pump is exponentially amplified by SBS, whereas a signal of frequency ν B above that of the pump is attenuated. The amplification or attenuation bandwidth achieved with continuouswave (CW) pumping is of the order of 3 MHz [6 8]. SBS interactions can be easily achieved in standard single-mode fibers at room temperature with modest pump power levels of only several milliwatts [9]. It therefore represents an attractive platform for alloptical, frequency-selective signal processing and sensing applications. Examples include optical and microwave photonic filtering [1,11], distributed fiber-optic sensing of temperature and strain variations [12,13], SBS-induced slow light [14], and more [15 18]. Several groups have demonstrated high-resolution OSAs based on SBS [19,2]. In these implementations, a pump wave, and its associated Brillouin gain line, is scanned through the spectral extent of a SUT. The power spectral density (PSD) of the SUT is reconstructed by measuring its amplified power as a function of the pump wave frequency. The spectral resolution of those previous SBS-OSAs is restricted to the order of 3 MHz by the Brillouin linewidth [6]. The resolution can be improved to a few megahertz by using a three-tone pump configuration, which introduces a single, central gain line and two symmetrically detuned SBS loss lines [21], or through saturation of the SBS gain outside a narrow spectral aperture [22]. Out-of-band spectral components of the SUT, although unamplified, are not affected by SBS and propagate to the output unattenuated. Therefore the ability to recover a relatively weak tone at the presence of a strong one using an SBS-OSA could be rather limited. This property is quantified in the out-of-band rejection ratio: the ratio of optical powers between a strong tone and a neighboring, weak spectral component that is barely detectable. The out-of-band rejection ratio has been improved through careful polarization control (as explained in the following sections) [23 25]. In combining both polarization-enhancement and three-line pump composition, Preussler et al. demonstrated an SBS-OSA with a rejection ratio of 3 db, alongside a resolution of 3 MHz [26]. However, the three-tone pump configuration was proved difficult to implement. Due to the complexity of the realization, the spectrum of the SUT in [26] had to be scanned across a fixed, composite SBS gain line, instead of the other way around. More realistic optical SUTs cannot be analyzed in this manner. To the best of our knowledge, the characterization of closely spaced, subcarrier multiplexed signals using an SBS-OSA has not yet been reported. In this work, we employ a polarization-enhanced SBS-OSA in the monitoring of fiber-optic transmission of ten subcarriers, with tone separation of 1 MHz. A rejection ratio of 35 db is obtained with a pump power of only 12 dbm. The rejection ratio and resolution of the setup are sufficient to recognize, for example, the absence of a single subcarrier, directly at the optical layer. A simple SBS-OSA configuration is used, based on a single gain line and polarization control. The results illustrate the potential of SBS analysis in the monitoring of advanced fiber-optic transmission. 2. Principle of Operation SBS gain in standard, birefringent fibers is highly polarization-dependent [25]. For a given state of polarization (SOP) of an input pump of frequency ω p, two orthogonal input signal SOPs can be identified, which provide the maximum and minimum signal output powers. The unit Jones vectors of these two SOPs are denoted ê in max and ê in min, respectively. For sufficiently long weakly and randomly birefringent fibers, and in the undepleted pump regime, the corresponding maximum and minimum complex amplitude gain values are given by [25] 1 G max ω p ω s exp 3 g ω p ω s L eff 1 G min ω p ω s exp 6 g ω p ω s L eff ; : (1) Here L eff is the effective length of the fiber, ω s is a signal frequency variable, and g ω p ω s is the exponential gain coefficient of SBS, which depends on the difference between the two optical frequencies and scales with the pump power. It is convenient to represent an arbitrarily polarized input signal in the basis of ê in max and ê in min, such that E ω s ;z E ω s α ê in max β ê in min : (2) Here E ω s is a frequency-dependent, scalar complex magnitude and jα j 2 jβ j 2 1. ê in max and ê in min will emerge at the fiber end as orthogonal SOPs to be denoted by ê out max, ê out min. The Jones vector of the output signal, corresponding to the input of Eq. (2), can be expressed as [26,27] E ω s ;z L E ω s α G max ω p ω s ê out max β G min ω p ω s ê out min : (3) In Eq. (3) L is the length of the fiber. In most practical cases G max is substantially larger than G min. 618 APPLIED OPTICS / Vol. 52, No. 25 / 1 September 213

3 Therefore the first term in Eq. (3) becomes dominant, and unless α is negligible, the arbitrarily polarized input signal will be drawn toward the SOP of ê out max. This phenomenon is referred to as SBS polarization pulling [25 27]. Further discrimination between amplified and unamplified spectral components is possible by placing a properly aligned polarizer at the output end of the signal. Let ˆp tr denote the unit Jones vector of the transmission axis of the polarizer, ˆp tr p max ê out max p min ê out min ; (4) where p max and p min are the projections of ˆp tr onto ê out max and ê out min, respectively. If ˆp tr is adjusted so that α p max β p min, out-of-band spectral components of the signal, for which there is no gain, can be blocked off entirely by the output polarizer. Subject to this constraint, the PSD of the signal wave at the polarizer output, for a given pump frequency ω p, becomes [26,27] j E out ω s j 2 je ω s j 2 jα P maxj 2 jg max ω p ω s G min ω p ω s j 2 je ω s j 2 jh ω p ω s j 2 : (5) Here, the transfer function of the signal wave through to the output of the polarizer is defined by [26] H ω p ω s α p max G max ω p ω s G min ω p ω s : (6) Given a complete out-of-band rejection, the maximum value of jα p maxj 2 is 1 4 [26,27]. At the highgain limit, the in-band power amplification of the polarizer-assisted SBS process is 6 db weaker than that of an optimized, so-called scalar process, for which jh ω p ω s j 2 jg max ω p ω s j 2. However, the rejection of the unamplified spectral components of the signal at the output of the polarizer is, in principle, infinite. In the above analysis it has been assumed that polarization mode dispersion is negligible within the spectral extent of the SUT. Figure 1 shows the simulated reconstruction of the PSD of a SUT, comprising a strong central tone and two tones, detuned by 4 MHz, which are 3 db weaker. The value of jg max ν B j 2 was chosen as 25 db. The signal power at the polarizer output was calculated by the convolution integral over Eq. (6): Z P ω p je ω s j 2 jh ω p ω s j 2 dω s : (7) Polarization Enhanced Scalar Fig. 1. Simulations of the reconstruction of the PSD of a SUT, using a SBS-based OSA. The SUT consisted of a strong, central tone and two side tones that are 3 db weaker and detuned by 4 MHz (inset). Dashed red curve, the SOP of the SUT is adjusted for maximum Brillouin amplification (scalar process). Solid black curve, polarization pulling of Brillouin amplification, together with an output polarizer, is employed to optimize the rejection of out-of-band components. in which the SUT is aligned with ê in max and the output polarizer is removed (Fig. 1, dashed red curve). The two, weak side tones are recognized more clearly by the polarization-enhanced SBS-OSA configuration. 3. Experimental Setup and Results The experimental setup is shown in Fig. 2. The output of a tunable laser source is split into two branches. Light in the upper branch (the signal branch, green) is modulated by a RF waveform, generated by an arbitrary waveform generator (AWG), and serves as an optical SUT. The SUT is then modulated again by a sine wave of a low frequency f L of 2 khz, to allow for lock-in power measurement at the output. Light in the lower branch (the pump branch, blue) is modulated by a sine wave of frequency ν, in suppressed-carrier, double-sideband format. A narrowband fiber Bragg grating is then used to select the upper modulation sideband, which is amplified to serve as CW pump. The pump and SUT counterpropagate in a 25 km long section of standard single-mode fiber, which constitutes the SBS gain medium. Normalized Power [db] Calculation is carried out by using the transfer functions of SBS polarization pulling and polarization discrimination, as described above [Eq. (6), Fig. 1, black solid curve], and that of a scalar process, Fig. 2. Experimental setup for SBS-based OSA. PC, polarization controller; MZM, Mach Zehnder modulator; AWG, arbitrary waveform generator; VOA, variable optical attenuator; FBG, fiber Bragg grating; PBS, polarization beam splitter. 1 September 213 / Vol. 52, No. 25 / APPLIED OPTICS 6181

4 In polarization-enhanced SBS-OSA measurements, the SUT at the output of the gain medium is split by using a broadband polarization beamsplitter (PBS), and the output of the fast axis is detected. The PBS is bypassed in scalar SBS measurements. The frequency offset ν is swept in 1 MHz increments (the analysis block, colored in purple), and the power of the amplified SUT at the polarizer output is measured. Lock-in measurements of the photocurrent RF power at the frequency f L block the amplified spontaneous emission associated with SBS amplification (SBS-ASE) almost entirely and allow for a more accurate reconstruction of the SUT PSD. Special attention was given to the relative alignment of the multiple polarization controllers (PCs), in accord with the considerations of the previous section. The launch SOP of the pump wave was chosen arbitrarily and was held fixed for the entire duration of the experiment. Scalar SBS-OSA measurements require that the launch SOP of the SUT be aligned with ê in max. PC 2 was therefore adjusted for maximum in-band amplification, with the PBS removed. Alignment in polarization-enhanced SBS-OSA measurements was carried out according to the following procedure. For initial calibration, the input signal was disconnected, and SBS-ASE was observed at the PBS output. The SOP of the SBS-ASE is known to be aligned with ê out max [23,25]; hence ê out max could be properly identified by setting PC 3 for maximum transmission of the SBS-ASE, through the PBS, to the photodetector (PD). Next, PC 3 was readjusted until the transmitted power of the SBS-ASE was reduced by 5%, signifying jp maxj Last, a CW SUT was reintroduced with ν detuned from ν B temporarily by several times the Brillouin linewidth. PC 2 was then adjusted until the output SUT, which is not affected by SBS, was entirely blocked by the PBS. This final adjustment guaranteed that α p max β p min, as required. The SUT power going into the Brillouin gain medium was controlled by using a variable optical attenuator (VOA). The polarization-enhanced SBS-OSA was employed in the monitoring of subcarrier multiplexed fiber-optic transmission. The SUT in these measurements consisted of 1 subcarrier tones, spaced by 1 MHz, each modulated by an on off keying pseudo-random bit sequence at 2.5 Mbit s. The overall optical power of the SUT was 23 dbm, and the SBS pump power was 13 dbm. Figure 3 shows the experimental reconstruction of the PSD of the SUT (magenta, solid), alongside the corresponding, simulated reconstruction (blue, dashed). Next, one of the subcarrier tones was intentionally removed from the SUT. Figure 4 shows the experimentally reconstructed PSD of the SUT (magenta, solid), compared with the simulated prediction (blue, dashed). The polarization-assisted SBS-OSA effectively recognized the missing subcarrier. Figures 5 and 6 show the reconstructed PSD of SUTs comprising a strong central tone and two Simulated reconstruction Experimental reconstruction Fig. 3. Experimental (magenta, solid curve) and simulated (blue, dashed curve) reconstruction of the PSD of a SUT, using a polarization-enhanced SBS-based OSA setup. The SUT consisted of ten subcarriers, separated by 1 MHz. Each subcarrier was independently modulated by an on off keying, pseudo-sequence bit sequence at 2.5 Mbit s. relatively weak tones, detuned by 4 MHz, using both scalar (red, dashed) and polarization-assisted (black, solid) SBS-OSA configurations. The input pump power level was 12 dbm, and the input power of the central tone of the SUT was 28 dbm. The maximum SBS amplification jg max ν B j 2 was 25 db. The ratio of the optical powers between the strong and the weak signal tones was 24.7 and 33.5 db for Figs. 5 and 6, respectively. The rejection ratio of the scalar SBS-OSA configuration is limited by jg max ν B j 2. Hence, the two sidelobes that were 33.5 db weaker than the central tone could not be recognized by the scalar setup (Fig. 6). The sidelobes are clearly visible in the polarization assisted SBS- OSA measurement. The rejection ratio of this configuration is limited by residual polarization pulling, observed when the pump is detuned from the frequency of maximum amplification by only 4 MHz (see also Fig. 1). The relative power of the reconstructed sidelobes is higher than that of the input, suggesting that partial pump depletion might have occurred in the amplification of the strong central tone Simulated reconstruction Experimental reconstruction Fig. 4. Experimental (magenta, solid curve) and simulated (blue, dashed curve) reconstruction of the PSD of a SUT, using a polarization-enhanced SBS-based OSA setup. The SUT was the same as that of Fig. 3 above, with subcarrier 7 intentionally removed APPLIED OPTICS / Vol. 52, No. 25 / 1 September 213

5 Polarization Enhanced Scalar Fig. 5. Experimental reconstruction of the PSD of a SUT, using scalar (red, dashed curve) and polarization-enhanced (black, solid curve) SBS-based OSA setups. The SUT consisted of a central tone and two weaker sidelobes whose optical power levels were 24.7 db lower than that of the central tone. 4. Summary In this work, the high spectral resolution monitoring of subcarrier multiplexed optical signals, directly in the optical domain, is demonstrated by using a polarization enhanced SBS-OSA setup. The setup consists of standard fiber-optic components. The resolution and rejection ratio of the SBS-OSA setup were sufficient for the monitoring of data-carrying tones, spaced by only 1 MHz. The setup successfully recognized the absence of an individual subcarrier. The reconstructed PSDs are in good agreement with the predictions of simulations. The measurement configuration includes polarization discrimination between in-band, amplified spectral components of the SUT and unamplified components. Polarization considerations improve the rejection ratio of the SBS-OSA beyond the SBS maximum gain. Hence a high rejection ratio of 35 db could be obtained by using a modest pump power of only 12 dbm. The corresponding rejection ratio in a scalar SBS-OSA measurement, for the same pump power, was an order of magnitude lower. Comparable rejection ratios in the scalar arrangement would require higher pump power levels, limited eventually by the onset of depletion Polarization Enhanced Scalar Fig. 6. Experimental reconstruction of the PSD of a SUT, using scalar (red, dashed curve) and polarization-enhanced (black, solid curve) SBS-based OSA setups. The SUT consisted of a central tone and two weaker sidelobes whose optical power levels were 33.5 db lower than that of the central tone. The spectral resolution of polarization-enhanced SBS-OSA setup is superior to that of a corresponding scalar process when the gain is relatively weak [26]. The resolutions obtained with the two methods are nearly equal for large power gain values above 2 db (see Figs. 5 and 6, for example). At this limit, the full width at half-maximum of the SBS gain line is of the order of 1 MHz. Unlike our earlier work [26], the SBS-OSA setup involved only a single CW pump, and was therefore much simpler to implement. Hence the PSD of the SUT could be acquired through the sweeping of an SBS gain window, rather than the other way around. The PSD of a SUT can be acquired over bandwidths of many nanometers by using one single tunable laser source. Broad, fast, and precise frequency sweeps, aided by calibration in an auxiliary interferometer, are widely performed, for example, in the context of optical frequency-domain reflectometry sensing [28]. Measurements over a broad optical bandwidth would require corrections for polarization mode dispersion. In conclusion, SBS-OSA measurements could be highly instrumental in the analysis and monitoring of advanced analog and digital optical transmission formats. The work of Yonatan Stern, Avi Zadok, Yossef Ben-Ezra, and Moshe Tur was supported in part by the Chief Scientist Office, the Israeli Ministry of Industry, Trade and Labor, within TERA SANTA consortium. Moshe Tur was also supported by the Israel Science Foundation (grant no. 138/12). Kun Zhong acknowledges the China Scholarship Council (CSC) for support of his research stay at Bar-Ilan University. The work of Kun Zhong and Ru Zhang was supported by the National Natural Science Foundation of China (Grant Nos and ). References 1. J. Armstrong, OFDM for optical communications, J. Lightwave Technol. 27, (29). 2. A. J. Lowery, L. B. Du, and J. Armstrong, Performance of optical OFDM in ultralong-haul WDM lightwave systems, J. Lightwave Technol. 25, (27). 3. J. Capmany and D. Novak, Microwave photonics combines two worlds, Nat. Photonics 1, (27). 4. D. M. Baney, B. Szafraniec, and A. Motamedi, Coherent optical spectrum analyzer, IEEE Photon. Technol. Lett. 14, (22). 5. F. R. Giorgetta, I. Coddington, E. Baumann, W. C. Swann, and N. R. Newbury, Fast high resolution spectroscopy of dynamic continuous-wave laser sources, Nat. Photonics 4, (21). 6. R. W. Boyd, Nonlinear Optics, 3rd ed. (Academic, 28). 7. G. P. Agrawal, Nonlinear Fiber Optics, 3rd ed. (Academic, 21). 8. A. Boh Ruffin, Stimulated Brillouin scattering: an overview of measurements, system impairments, and applications, in Technical Digest: Symposium on Optical Fiber Measurements (24), pp D. Cotter, Observation of stimulated Brillouin scattering in low-loss silica fibre at 1.3 μm, Electron. Lett. 18, (1982). 1. A. Zadok, A. Eyal, and M. Tur, GHz-wide optically reconfigurable filters using stimulated Brillouin scattering, J. Lightwave Technol. 25, (27). 1 September 213 / Vol. 52, No. 25 / APPLIED OPTICS 6183

6 11. A. Loayssa, J. Capmany, M. Sagues, and J. Mora, Demonstration of incoherent microwave photonic filters with all-optical complex coefficients, IEEE Photon. Technol. Lett. 18, (26). 12. T. Horiguchi, T. Kurashima, and M. Tateda, A technique to measure distributed strain in optical fibers, IEEE Photon. Technol. Lett. 2, pp (199). 13. A. Zadok, Y. Antman, N. Primerov, A. Denisov, J. Sancho, and L. Thevenaz, Random-access distributed fiber sensing, Laser Photon. Rev. 6, L1 L5 (212). 14. A. Zadok, A. Eyal, and M. Tur, Stimulated Brillouin scattering slow light in optical fibers, Appl. Opt. 5, E38 E49 (211). 15. S. P. Smith, F. Zarinetchi, and S. Ezekiel, Narrow-linewidth stimulated Brillouin fiber laser and applications, Opt. Lett. 16, (1991). 16. S. Loranger, V. L. Iezzi, and R. Kashyap, Demonstration of an ultra-high frequency picosecond pulse generator using an SBS frequency comb and self phase-locking, Opt. Express 2, (212). 17. A. Loayssa and F. J. Lahoz, Broad-band RF photonic phase shifter based on stimulated Brillouin scattering and singlesideband modulation, IEEE Photon. Technol. Lett. 18, (26). 18. Y. Antman, N. Levanon, and A. Zadok, Low-noise delays from dynamic Brillouin gratings based on perfect Golomb coding of pump waves, Opt. Lett. 37, (212). 19. T. Schneider, Wavelength and line width measurement of optical sources with femtometre resolution, Electron. Lett. 41, (25). 2. J. M. S. Domingo, J. Pelayo, F. Villuendas, C. D. Heras, and E. Pellejer, Very high resolution optical spectrometry by stimulated Brillouin scattering, IEEE Photon. Technol. Lett. 17, (25). 21. S. Preussler, A. Wiatrek, K. Jamshidi, and T. Schneider, Ultrahigh-resolution spectroscopy based on the bandwidth reduction of stimulated Brillouin scattering, IEEE Photon. Technol. Lett. 23, (211). 22. A. Wiatrek, S. Preußler, K. Jamshidi, and T. Schneider, Frequency domain aperture for ultra-high resolution Brillouin based spectroscopy, in Conference on Lasers and Electro-Optics 212, OSA Technical Digest (Optical Society of America, 212), paper JW4A M. O. van Deventer and J. Boot, Polarization properties of stimulated Brillouin scattering in single mode fibers, J. Lightwave Technol. 12, (1994). 24. A. Galtarossa, L. Palmieri, M. Santagiustina, L. Schenato, and L. Ursini, Polarized Brillouin amplification in randomly birefringent and unidirectionally spun fibers, IEEE Photon. Technol. Lett. 2, (28). 25. A. Zadok, E. Zilka, A. Eyal, L. Thevenaz, and M. Tur, Vector analysis of stimulated Brillouin scattering amplification in standard single-mode fibers, Opt. Express 16, (28). 26. S. Preussler, A. Zadok, A. Wiatrek, M. Tur, and T. Schneider, Enhancement of spectral resolution and optical rejection ratio of Brillouin optical spectral analysis using polarization pulling, Opt. Express 2, (212). 27. A. Wise, M. Tur, and A. Zadok, Sharp tunable optical filters based on the polarization attributes of stimulated Brillouin scattering, Opt. Express 19, (211). 28. B. Soller, D. Gifford, M. Wolfe, and M. Froggatt, High resolution optical frequency domain reflectometry for characterization of components and assemblies, Opt. Express 13, (25) APPLIED OPTICS / Vol. 52, No. 25 / 1 September 213