In-situ fiber characterization using nonlinear skirt measurement

Abstract

A system includes a processor communicatively coupled to an Amplifier Stimulated Emission (ASE) source and an optical receiver, wherein the processor is configured to cause transmission of one or more shaped ASE signals, from the ASE source, on an optical fiber, obtain received spectrum of the one or more shaped ASE signals from the optical receiver connected to the optical fiber, and characterize the optical fiber based in part on a nonlinear skirt and/or center dip depth in the received spectrum of the one or more shaped ASE signals. The one or more shaped ASE signals can be formed by the ASE source communicatively coupled to a Wavelength Selective Switch (WSS) that is configured to shape ASE from the ASE source to form the one or more shaped ASE signals with one or two or multiple peaks and with associated frequency.

Description

FIELD OF THE DISCLOSURE

The present disclosure generally relates to fiber optic systems. More particularly, the present disclosure relates to systems and methods for in-situ fiber characterization using nonlinear skirt measurement.

BACKGROUND OF THE DISCLOSURE

In optical systems, signal transmission performance is significantly impacted by fiber dispersion and nonlinearity. Chromatic dispersion, i.e., β₂, and fiber nonlinear coefficient, i.e., γ, are among the most critical fiber characteristics for photonic line system link budget, performance modeling, and optimization. The current fiber nonlinearity measurement can be with a high-power signal having a narrow linewidth or by backing out from a Stimulated Raman Scattering (SRS) measurement. The dispersion measurement, currently, is commonly implemented by measuring the phase delay difference of modulated signals on different wavelengths. For example, dispersion can also be measured by the differential time delay between two wavelengths such as an Optical Service Channel (OSC) wavelength at 1510 nm and an Optical Time Domain Reflectometer (OTDR) wavelength at 1625 nm.

Disadvantageously, Continuous Wave (CW) signals are single polarization sources by definition. This causes measurement issues in the presence of Polarization Dependent Gain (PDG), Differential Group Delay (DGD), and Polarization Mode Dispersion (PMD) which is significant in real systems. In addition, the practical implementation using CW source(s) for dispersion/nonlinear measurement requires a coherent transponder, transceiver, modem, etc. This requires equipped modules as well as end-to-end communication. This may not be available at or before turn-up. Also, the dispersion measurement using two wavelengths such as 1510 nm and 1625 nm only measures average dispersion at 1568 nm. As optical systems continue to push the limit of bandwidth over fiber, it is critical to get accurate measurements over the entire signal band.

BRIEF SUMMARY OF THE DISCLOSURE

In an embodiment, a system includes a processor communicatively coupled to an Amplifier Stimulated Emission (ASE) source and an optical receiver, wherein the processor is configured to cause transmission of one or more shaped ASE signals, from the ASE source, on an optical fiber, obtain received spectrum of the one or more shaped ASE signals from the optical receiver connected to the optical fiber, and characterize the optical fiber based in part on one or more of a nonlinear skirt and a center dip depth in the received spectrum of the one or more shaped ASE signals. The one or more shaped ASE signals can be formed by the ASE source communicatively coupled to a Wavelength Selective Switch (WSS) that is configured to shape ASE from the ASE source to form the one or more shaped ASE signals with one or more peaks and with associated frequency. The one or more shaped ASE signals can have two distinct peaks at the transmission with a significant dip at a center frequency, and the received spectrum of the one or more two-peak ASE signals has much less of a dip that the center dip depth. The processor can be further configured to determine a fiber type based on a signature of the one or more of the nonlinear skirt and the center dip depth in the received spectrum of the one or more shaped ASE signals. The optical fiber can be characterized to determine chromatic dispersion, β₂, and fiber nonlinear coefficient, γ.

The chromatic dispersion, β₂, and the fiber nonlinear coefficient, γ, can be characterized by a measurement of a shape of the one or more of the nonlinear skirt and the center dip depth as a function of signal wavelength. The one or more shaped ASE signals can include a plurality of shaped ASE signals with a first set of shaped ASE signals utilized to determine launch power for every span in a section to yield an optimum center dip depth, and a second set of shaped ASE signals that sweep at different frequencies across a signal band to determine a corresponding center dip depth at the different frequencies. The optical fiber can be characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a separate differential delay measurement, to determine chromatic dispersion, β₂, and fiber nonlinear coefficient, γ. The optical fiber can be characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a separate Stimulated Raman Scattering (SRS) measurement, to determine chromatic dispersion, β₂, and fiber nonlinear coefficient, γ. The optical fiber can be characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a Least Mean Square (LMS) fit, to determine chromatic dispersion, β₂, and fiber nonlinear coefficient, γ.

In another embodiment, a method includes causing transmission of one or more shaped Amplifier Stimulated Emission (ASE) signals, from an ASE source, on an optical fiber; obtaining received spectrum of the one or more shaped ASE signals from the optical receiver connected to the optical fiber; and characterizing the optical fiber based in part on one or more of a nonlinear skirt and a center dip depth in the received spectrum of the one or more two-peak ASE signals. The one or more shaped ASE signals can be formed by the ASE source communicatively coupled to a Wavelength Selective Switch (WSS) that is configured to shape ASE from the ASE source to form the one or more shaped ASE signals with one or more peaks and with associated frequency. The one or more shaped ASE signals can have two distinct peaks at the transmission with a significant dip at a center frequency, and the received spectrum of the one or more two-peak ASE signals has much less of a dip that the center dip depth. The method can further include determining a fiber type based on a signature of the one or more of the nonlinear skirt and the center dip depth in the received spectrum of the one or more shaped ASE signals.

The optical fiber can be characterized to determine chromatic dispersion, β₂, and fiber nonlinear coefficient, γ. The one or more shaped ASE signals can include a plurality of shaped ASE signals with a first set of shaped ASE signals utilized to determine launch power for every span in a section to yield an optimum center dip depth, and a second set of shaped ASE signals that sweep at different frequencies across a signal band to determine a corresponding center dip depth at the different frequencies. The optical fiber can be characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a separate differential delay measurement, to determine chromatic dispersion, β₂, and fiber nonlinear coefficient, γ. The optical fiber can be characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a separate Stimulated Raman Scattering (SRS) measurement, to determine chromatic dispersion, β₂, and fiber nonlinear coefficient, γ. The optical fiber can be characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a Least Mean Square (LMS) fit, to determine chromatic dispersion, β₂, and fiber nonlinear coefficient, γ.

In a further embodiment, an Optical Add/Drop Multiplexing (OADM) node includes a Wavelength Selective Switch (WSS) system communicatively coupled to at least a first optical fiber and a second optical fiber; an Amplifier Stimulated Emission (ASE) source connected to the WSS system; a pre-amplifier connected to the WSS system and the first optical fiber; an Optical Channel Monitor (OCM) connected at least to an output of the pre-amplifier; and a processor configured to obtain received spectrum of one or more shaped ASE signals from the OCM, wherein the one or more shaped ASE signals are transmitted over the first optical fiber, and characterize the first optical fiber based in part on a nonlinear skirt shape and a center dip depth in the received spectrum of the one or more shaped ASE signals.

BRIEF DESCRIPTION OF THE DRAWINGS

The present disclosure is illustrated and described herein with reference to the various drawings, in which like reference numbers are used to denote like system components/method steps, as appropriate, and in which:

FIG. 1A is a graph of an optical signal and associated signal broadening over 100 km in Lambda Shifted (LS) fiber utilizing a one-peak signal, and FIG. 1B is a graph of another optical signal utilizing a two-peak signal;

FIG. 2A is a graph of measured results for received spectrum in LS fiber over 100 km, and FIG. 2B is a graph of measured results for received spectrum in Non-Dispersion Shifted Fiber (NDSF) over 40 km;

FIG. 3A is a graph of simulated results for received spectrum in Truewave (TW) Classic fiber, FIG. 3B is a graph of simulated results for received spectrum in LS fiber, FIG. 3C is a graph of simulated results for received spectrum in NDSF fiber, FIG. 3D is a graph of simulated results for received spectrum in Large Effective Area Fiber (LEAF), FIG. 3E is a graph of simulated results for received spectrum in enhanced LEAF (eLEAF), and FIG. 3F is a graph of simulated results for Dispersion Shifted Fiber (DSF);

FIG. 4A is a graph of measurement of center gap depth versus frequency versus β₂ for LS fiber, and FIG. 4B is a graph of a measurement of center gap depth versus frequency versus β₂ for NDSF fiber;

FIG. 5 is a graph of a simulation of center dip depth versus β₂ with different launching power;

FIG. 6 is a graph of a simulation of center dip depth versus β₂ with different gamma, γ;

FIG. 7 is a network diagram of an optical section with associated equipment for providing an in-situ nonlinear skirt measurement;

FIG. 8 is a flowchart of a measurement process which can be implemented in the optical section of FIG. 7;

FIG. 9A is a flowchart of a data analysis process which also uses a differential delay measurement, FIG. 9B is a flowchart of a data analysis process which uses an SRS measurement, and FIG. 9C is a flowchart of a data analysis process which sweeps dispersion, dispersion slope and γ;

FIG. 10 is a block diagram of a system for characterizing optical amplifiers utilizing the two-peak ASE signal;

FIG. 11A is a graph of optical spectrum in the system before subtracting ASE, FIG. 11B is a graph of optical spectrum in the system after subtracting ASE, and FIG. 11C is a graph of FWM level versus signal center frequency; and

FIG. 12 is a flowchart of a process for in-situ fiber characterization.

DETAILED DESCRIPTION OF THE DISCLOSURE

The present disclosure relates to systems and methods for in-situ fiber characterization using nonlinear skirt measurement. The present disclosure characterizes fiber dispersion and nonlinearity based on the spectral shape of an Amplified Stimulated Emission (ASE) signal. Of note, the ASE signal is more like a coherent, dual-polarization modulated signal than a single polarization CW source. The ASE signal can explore all states of polarization which averages these effects resulting in a more stable and accurate measurement. Also, the ASE signal is available from common infrastructure in a photonic line system without requiring additional equipment, and this can be utilized on a per section basis prior to system turn-up. For example, next-generation photonic line systems utilize ASE for channel holders and/or can generate ASE via amplifiers. An ASE source can be shaped by a Wavelength Selective Switch (WSS) at Tx. At a corresponding Rx, the nonlinear product can be characterized by the nonlinear skirt, i.e., spectral shape of the ASE signal. In an example, a two-peak Tx signal spectral shape is designed using an ASE source and transmitted over the fiber. Of note, other shapes are also contemplated such as a one-peak Tx signal. At a corresponding Rx, the nonlinear product can be characterized by center dip depth, i.e., the relative power difference between the signal peak and the valley of the overlap area of the nonlinear skirt between the two peaks. As a result, measurement depends on relative power between the peak and valley of the two-peak signal. Absolute power accuracy and power monitor Wavelength Dependent Loss (WDL) are not critical requirements for the measurement.

The present disclosure is characterized as in-situ due to utilizing existing common infrastructure in an optical networking system; no specialized hardware is required. The present disclosure measures dispersion and nonlinearity within the signal band (e.g., the C-band such as between about 1528 nm and 1565 nm, and/or the L-band such as between about 1565 nm and 1625 nm) rather than relying on out-of-band measurements and extrapolation. The present disclosure utilizes a relative power measurement, such that measurement can be easily carried out and post-processing is not complicated.

The accurate characterization of fiber dispersion and nonlinearity will help model the exact link performance, which is crucial for link budget and performance optimization. This is a key part of a “plug and play” approach for optical control. A fiber-type determination is currently a manual procedure and has been shown to be a source of system issues in many networks where there is often mis-provisioned fiber type. This disclosure removes the manual effort of fiber type identification and provisioning and the associated potential manual error. Furthermore, fiber-type does not accurately characterize the fiber. Even within each fiber type, there is a range of dispersion and nonlinear coefficient. It is also common in real systems to have mixed fiber types within a single span. This disclosure gives an appropriate value, averaged over the nonlinear-length, for the key performance parameters which can be used in optical control an optimization resulting in better performance and higher capacity.

When an optical signal propagates in a nonlinear medium such as optical transmission fiber, its spectrum will be broadened due to the combination of fiber nonlinearity and dispersion. The broadened spectral shape shows a distinct signature for fiber with different characteristics. Consequently, β₂ and γ can be backed out through measurement and then calculated. This disclosure proposes a process that characterizes β₂ and γ by measuring the broadened spectrum of a shaped ASE signal.

FIG. 1A is a graph 10A of an optical signal and associated signal broadening over 100 km in Lambda Shifted (LS) fiber illustrating a one-peak signal. FIG. 1B is a graph 10B similar to FIG. 1A illustrating a two-peak signal. A spectral shape of the Tx signal is illustrated by line 12 in the graph 10A, 10B, and the Rx signal after a 100 km LS span is illustrated by a line 14 in the graph 10A, 10B. Of note, the graphs 10A, 10B illustrate the Tx signal and the Rx signal on the same graph for illustration purposes to highlight the difference in the spectral shapes based on transmission over an optical fiber. The nonlinear broadening effect due to the transmission will result in a skirt shape on the Rx signal spectrum, a.k.a. nonlinear skirt 18. That is, when an optical signal is transmitted over fiber, the signal spectrum will be broaden due to the combination of fiber nonlinearity and dispersion, and the received signal shape is different after different type of fibers with different characteristics. Therefore, an LMS fit and the like can be performed to determine the various fiber parameters.

In FIG. 1B, it is observed that the gap between the two peaks, marked as a center dip depth 16 in the graph 10B, becomes shallower after transmission due to the broadening effect. It is discovered as outlined in this disclosure that β₂ and γ can be characterized by measuring center dip depth as a function of signal wavelength.

Of note, the present disclosure performs fiber characterization (e.g., fiber type determination, β₂ and γ measurements, etc. based on sending an ASE signal that is spectrally shaped and measuring the received spectral shape, in particular, the nonlinear skirt. In an example, a two-peak signal as shown in FIG. 1B is utilized where an ASE signal from an ASE source is shaped to form the two peaks. The two-peak signal simplifies measurement and post-processing by providing a relative measurement between the peak and valley of the two peaks. Those of ordinary skill in the art will appreciate it is possible to characterize fiber based on the nonlinear skirt shape of a one-peak signal as shown in FIG. 1A based on LMS fit of the shape of the nonlinear skirt. Further, the present disclosure contemplates other shapes including multiple peaks.

FIG. 2A is a graph of measured results for received spectrum in LS fiber over 100 km, and FIG. 2B is a graph of measured results for received spectrum in Non-Dispersion Shifted Fiber (NDSF) over 40 km. FIG. 3A is a graph of simulated results for received spectrum in Truewave (TW) Classic fiber, FIG. 3B is a graph of simulated results for received spectrum in LS fiber, FIG. 3C is a graph of simulated results for received spectrum in NDSF fiber; FIG. 3D is a graph of simulated results for received spectrum in Large Effective Area Fiber (LEAF), FIG. 3E is a graph of simulated results for received spectrum in enhanced LEAF (eLEAF), and FIG. 3F is a graph of simulated results for Dispersion Shifted Fiber (DSF).

Specifically, the measured/experimental results in FIG. 2 and the simulated results in FIG. 3 illustrate the change of the center dip depth 16 sweeping the signal across the C-band. Distinct signatures are observed for different fiber type with different characteristics. Of note, fiber type can be identified from the change of the center dip depth 16 over wavelength.

Furthermore, an important phenomenon is discovered that given the fiber to be transmitted, the normalized center dip depth over wavelength does not change with different signal launching power. This can be explained by a Gaussian Noise (GN) model, where the nonlinear product of fiber transmission is expressed as ^[1]

$\begin{array}{cc}{G}_{\mathrm{NLI}}\left(f,{\beta }_2,\alpha ,L_S,\gamma ,P_0\right)=\frac{16}{27}{\gamma }^2{P}_0^3{\int }_{-\infty }^{\infty }{\int }_{-\infty }^{\infty }{\frac{1-e^{-2\alpha L_s}{e}^{j4{\pi }^2{\beta }_2\left(f_1-f\right)\left(f_2-f\right)}}{2\alpha -j4{\pi }^2{\beta }_2\left(f_1-f\right)\left(f_2-f\right)}}^2·g_{Tx}\left(f_1\right)g_{Tx}\left(f_2\right)g_{Tx}\left(f_1+f_2-f\right)d{f}_{2}d{f}_1& \left(1\right)\end{array}$

where β₂ is the fiber dispersion, γ is the nonlinear coefficient, α is the fiber loss parameter, L_s is the span length, g_Tx(⋅) is the normalized power spectral density of the transmitted signal, P₀ is its launching power. This is taken from [1] A. Carena, V. Curri, G. Bosco, P. Poggiolini and F. Forghieri, “Modeling of the Impact of Nonlinear Propagation Effects in uncompensated Optical Coherent Transmission Links,” in Journal of Lightwave Technology, vol. 30, no. 10, pp. 1524-1539, May 15, 2012.

The normalized shape of G_NLI(f) is

$\begin{array}{cc}{g}_{\mathrm{NLI}}\left(f,{\beta }_2,\alpha ,L_S\right)={\int }_{-\infty }^{\infty }{\int }_{-\infty }^{\infty }{\frac{1-e^{-2\alpha L_s}{e}^{j4{\pi }^2{\beta }_2\left(f_1-f\right)\left(f_2-f\right)}}{2\alpha -j4{\pi }^2{\beta }_2\left(f_1-f\right)\left(f_2-f\right)}}^2·g_{Tx}\left(f_1\right)g_{Tx}\left(f_2\right)g_{Tx}\left(f_1+f_2-f\right)d{f}_{2}d{f}_1& \left(2\right)\end{array}$ which only depends on the term within the absolute operator when g_Tx(⋅) is fixed. The Tx signal spectrum with the two-peak spectral shape designed by this invention has no signal power between the two peaks. Therefore, the Power Spectral Density (PSD) in the center gap at the Rx side contains only nonlinear products. When PSD of the in band nonlinear product is negligible compare with signal PSD, the center dip depth, i.e., the delta of PSD between the peak and valley of the two-peak signal, can be written as

$\begin{array}{cc}\left({\beta }_2,P_0,\gamma ,L_S,\alpha \right)=10\log 10\left(\frac{g_{RX}\left(f_c\right)}{g_{\mathrm{NLI}}\left(f=0,{\beta }_2,L_s,\alpha \right)}\right)+10\log 10\left(\frac{16}{27}{\gamma }^2{P}_0^2\right)=\tilde{D}\mathrm{epth}\left({\beta }_2,L_S,\alpha \right)+A\left(\gamma ,P_0\right)+C& \left(3\right)\end{array}$ where f_c is the relative center frequency of the each of the two peaks, as marked in FIG. 1B. The first term in Eq (3) is the normalized center dip depth, denoted as {tilde over (D)}epth(β₂,L_s,α); the second term is the offset, denoted as A(γ,P₀), and C=10log10(16/27).

FIG. 4A is a graph of measurement of center gap depth versus frequency versus β₂ for LS fiber, and FIG. 4B is a graph of a measurement of center gap depth versus frequency versus β₂ for NDSF fiber. Given the fiber span under test, i.e., L_eff and γ are approximately constant over wavelength, the normalized center dip depth as a function of β₂, {tilde over (D)}epth|_Leff (β_s), remains unchanged with different P₀, while the offset, A(P₀), changes twice as fast as P₀. Note, the corresponding β₂ at each frequency on the top X-axis of FIGS. 4A and 4B is computed from measured dispersion and dispersion slope at 1550 nm.

FIG. 5 is a graph of a simulation of center dip depth versus β₂ with different launching power. FIG. 6 is a graph of a simulation of center dip depth versus β₂ with different gamma, γ. Note, the shape of the β₂ versus the center dip depth trace remain approximately unchanged when center dip depth is greater than about 15 dB. The center dip depth trace versus β₂ versus measure can provide the fiber type and dispersion slope. The shape of the center dip depth versus β₂ changes at different fiber lengths, but is unchanged with different gamma, γ at the same fiber length. For detection, the shape of the center dip depth versus β₂ can be stored at various different fiber lengths for simple comparisons. The simulation results in FIG. 5 demonstrate that, with fixed γ and L_eff, {tilde over (D)}epth|_Leff (β₂) remains approximately unchanged with different P₀, when the absolute value of Depth is larger than ˜15 dB. Depth gets distorted with high P₀, when the in band nonlinear product is no longer negligible. FIG. 6 demonstrates that with fixed P₀ and L_eff, {tilde over (D)}epth|_Leff (β₂) remains approximately unchanged, while absolute value of Depth changes twice as fast as the change of γ.

Accordingly, the relative center dip depth change over wavelength can be a distinct indicator of the type of fiber (e.g., NSDF, LS, TWc, LEAF, eLEAF, DSF, etc.). No absolute power information is required;

FIG. 7 is a network diagram of an optical section 50 with associated equipment for providing an in-situ nonlinear skirt measurement. The optical section 50 is a segment in an optical network between Optical Add/Drop Multiplexer (OADM) nodes 52, 54. The optical section 50 can be referred to as an Optical Multiplex Section (OMS), and one aspect of each optical section 50 is the spectral load is identical over the entire section. A real implementation of an optical network can include multiple optical sections 50 in a mesh, ring, linear, hub and spoke, etc. architecture. The in-situ nonlinear skirt measurement can be performed on a per optical section basis. The optical section 50 can also include intermediate optical line amplifier nodes 56. Further, a practical implementation of the optical section 50 includes two optical fibers 58, 60 for bidirectional communication. The in-situ nonlinear skirt measurement is performed on each optical fiber 58, 60 separately.

The OADM nodes 52, 54 include a Wavelength Selective Switch (WSS) 62 that faces the optical fibers 58, 60. The WSS 62 forms an optical degree that faces the optical fibers 58, 60. In this example, a single degree is illustrated at each of the OADM nodes 52, 54. Of course, practical implementations may include multiple degrees, each facing a different optical section 50. The WSS 62 is configured to add/drop spectrum to/from the degrees and/or locally. Each OADM node 52, 54 includes a post-amplifier 64 on the transmit side and a pre-amplifier 66 on the receive side. The amplifiers 64, 66 can be Erbium-Doped Fiber Amplifiers (EDFAs). Also, Raman amplifiers may be used as well in addition to EDFAs. The OADM nodes 52, 54 also include an Optical Channel Monitor (OCM) 68 (a.k.a. Optical Power Monitor (OPM), etc.) which is an optical receiver connected (e.g., by a tap) to an output of each of the amplifiers 64, 66. The OCM 68 can have two receivers to simultaneously monitor each of the optical fibers 58, 60 or a switch to allow a single receiver to monitor one of the optical fibers 58, 60 at a time.

The OADM nodes 52, 54 also can include an ASE source 70 coupled to the WSS 62. In newer optical line systems, the ASE source 70 can be used to fill unused spectrum to reduce power optimization time. Here, the ASE source 70 provides so-called channel holders used to fill the optical spectrum on the optical section 50 so that it always appears to have a full-fill configuration. Such an approach significantly reduces capacity change time.

Also, the OADM nodes 52, 54 can include other components 72 such as an OTDR, an OSC, a polarimeter, and a processor. Again, the components 72 are in-situ, i.e., part of the OADM nodes 52, 54. In an embodiment, the components 72 can provide a differential delay measurement, an SRS measurement, a fiber 58, 60 effective length (L_eff) measurement, etc. For example, these measurements are described in commonly-assigned U.S. patent application Ser. No. 15/986,396, filed May 22, 2018, and entitled “Optical fiber characterization measurement systems and methods,” the contents of which are incorporated by reference herein. Further, the processor can be used to obtain the measurement data and perform various data analyses described herein. Also, the intermediate optical line amplifier nodes 56 include in-line optical amplifiers 72.

As described herein, the in-situ fiber nonlinear skirt measurement can be performed with the ASE source 70 and the WSS 62 causing a two-peak signal to be transmitted on the optical fibers 58, 60 and received by the OCM 68. Referring back to FIGS. 1A and 1B, the graphs 10A, 10B illustrate the transmitted spectral shape in line 12. This spectral shape can be achieved through configuration of the ASE source 70 and the WSS 62. The received spectral shape in line 14 is received by the OCM 68 with the corresponding center dip depth 16 a function of the fiber 58, 60.

FIG. 8 is a flowchart of a measurement process 100. The measurement process 100 can be implemented in the optical section 50. The measurement process 100 starts with an initial scan at 1550 nm. At the beginning of a network section, e.g., at the upstream OADM, the measurement process 100 includes setting up the WSS 62 to shape the ASE source 70 to create the two-peak signal at 1550 nm (step 102).

For every span i=1˜N in the optical section, starting with i=1 (step 104), the measurement process 100 includes setting the optical amplifiers in a power mode for all spans (this power mode setting only needs to be done once, not necessarily for each iteration) and setting signal launching power at a reference power level, P0_ref for span i, for example P0_ref=15 dBm, and setting the rest of the spans at a much lower launching power, for example (P0_ref−15) (step 106). The measurement process 100 includes reading the OPM 68 at the downstream OADM, and recording the center dip depth of the received signal as Depth_ref|_span=i (step 108). The span count is incremented and steps 104-110 are repeated until the end of the section (step 110).

The signal broadening effect characterized by Depth_ref|_span=i is mainly generated by span i with high signal launching power. The purpose of steps 104-110 is to find the launching power for every span to yield the center dip depth around an optimum center dip depth, Depth_opt. Depth_opt is found when the PSD of the nonlinear product at the center gap is much higher than line EDFA ASE noise, while the corresponding in-band nonlinear product is still negligible compared to the signal. Depth_opt depends on the width of the signal and the gap of the two peaks. For example, when both the signal and gap width is 50 GHz, Depth_opt=15 dB. Since the absolute level of Depth changes twice as fast as P0, the launching power for span i=1˜N is computed byP0|_span=i=P0_ref+(Depth_ref|_span=i−Depth_opt)/2(dBm)  (4)

The next step is to sweep the signal through the signal band at X points with the optimum signal launching power P0|_span=i from Eq.(4). In the upstream OADM, for every wavelength to be tested (step 112), the measurement process 100 includes setting up the WSS/ASE source to create the two-peak signal at wavelength x, x=1˜X(step 114).

For every span i=1˜N in section (step 116), the measurement process 100 includes setting the optical amplifiers in power mode for all spans (this power mode setting only needs to be done once, not necessarily for each iteration), and setting launch power of span i to P0|_span=i, such that the center dip depth due to the signal broadening effect of span i is around the pre-defined Depth_opt. For the rest of the spans j=1˜N, j≠i, set launch power toP0|_span=j=P0_ref+(Depth_ref|_span=j=(Depth_opt+20))/2(dBm)  (5)such that the center dip depth due to the signal broadening effect of span=j is about 20 dB less than Depth_opt, and is deemed as negligible in the signal spectrum at the end of section (step 118).

For every span i and wavelength x, the measurement process 100 includes measuring the center dip depth of the received signal with the OCM at the downstream OADM and recorded as Depth_meas|_span=i,wvl=x (step 120). Steps 112-120 are repeated for all spans and wavelengths (steps 122, 124). Finally, the measurement process 100 has all required data measurements and can proceed to data analysis which is described in processes 200A, 200B, 200C.

FIG. 9A is a flowchart of a data analysis process 200A which also uses a differential delay measurement, FIG. 9B is a flowchart of a data analysis process 200B which uses an SRS measurement, and FIG. 9C is a flowchart of a data analysis process 200C which sweeps dispersion, dispersion slope and γ. Each of the processes 200A, 200B, 200C can be performed in a processor at the OADM nodes 52, 54, in a management system, or the like.

In FIG. 9A, the data analysis process 200A utilizes dispersion data at 1568 nm from a differential delay measurement such as from an OSC at 1510 nm and an OTDR at 1625 nm. This dispersion data is utilized in addition to data from the measurement process 100.

The data analysis process 200A, for each span i=1˜N at each measured wavelength, x=1˜X (step 202), includes loading measurement results from center dip depth measurement Depth_meas|_wvl=x,span=i, for each span i=1˜N at each measured wavelength, x=1˜X (step 204). The data analysis process 200A includes getting the measured dispersion at 1568 nm from the differential delay measurement, denoted as β₂|_{wvl=1568,span=i} (step 208); as well as the effective length, L_eff, from an OTDR measurement (step 206). Then, the data analysis process 200A includes loading the pre-computed normalized center dip depth as a function β₂ at the measured L_eff|_span=i, denoted as {tilde over (D)}epth|_Leff(β₂).

Next, the data analysis process 200A includes scaling {tilde over (D)}epth|_Leff(β₂) to Depth_meas|_wvl=x,span=i, by extrapolating Depth_meas|_wvl=x,span=i to Depth|_{wvl=1568,span=i} (step 210) and finding A to satisfy Eq. (6) (step 212):Depth_meas|_{wvi=1568,span=1}={tilde over (D)}epth|_Leff(β₂|_{wvi=1568,span=i})+A+c  (6)

Next, the data analysis process 200A, from Eq.3, includes determining γ is byγ=10^{0.1*(A/2″P0|span=i)}  (7)and β₂ of the fiber at each measured wavelength, β₂|_wvl=x,span=i, byβ₂|_wvi=x,span=i={tilde over (D)}_epth|_Leff⁻¹(Depth_meas|_wvl=x,span=i−A−c)  (8)Where {tilde over (D)}epth|_Leff⁻¹( ) is the inverse function of {tilde over (D)}epth|_Leff ( ) (step 214). The data analysis process 200A loops through every span in the section (step 216).

In FIG. 9B, the data analysis process 200B incorporates an SRS measurement. The data analysis process 200B, for each span i=1˜N at each measured wavelength x=1˜X (step 220); starts with loading results from center dip depth measurement, Depth_meas|_wvl=x,span=i (step 222), loading γ|_span=i from SRS measurement (step 226), and pre-computed {tilde over (D)}epth|_Leff (β₂) at OTDR measured L_eff|_span=i (step 224). The scaling factor A can be found by (step 228)a=2P0|_span=i+20log10(γ|_span=i)  (9)

Then, β₂ of the fiber at each measured wavelength, β₂|_wvi=x,span=i, is found by Eq (8) (step 230). The data analysis process 200B loops through every span in the section (step 232).

In FIG. 9C, the data analysis process 200C sweeps dispersion, dispersion slope and γ to scale {tilde over (D)}epth|_Leff (β₂) to a guessed Depth_guess(γ,β₂(λ)) based on Eq.(3). The data analysis process 200C, for each span i=1˜N at each measured wavelength x=1˜X (step 240), starts with loading results from center dip depth measurement, Depth_meas|_wvl=x,span=i (step 242), and pre-computed {tilde over (D)}epth|_Leff (β₂) at OTDR measured L_eff|_span=i (step 244). The data analysis 200C sweeps dispersion, dispersion slope and γ to scale {tilde over (D)}epth|_Leff (β₂) to a guessed Depth_guess(γ,β₂(λ)) (step 246). By comparing Depth_guess(γ,β₂(λ)) with the measured Depth_meas|_wvl=x, γ|_span=i and β₂|_wvl=x,span=i are found at Least Mean Square (LMS) fit (steps 248, 250). The data analysis process 200C loops through every span in the section (step 252).

In an embodiment, the measurement technique described herein with a two-peak ASE signal can be used to characterize optical amplifiers. For example, next-generation optical amplifiers with higher output power can have tighter Four-Wave Mixing (FWM) specifications. The various measurement techniques described herein work well to characterize this effect. For example, conventionally, the FWM specification of amplifiers was verified in the factory and lab using a specified measurement technique with two CW lasers (which also means single polarization) that had to be polarization controlled (scanned) to search for a peak interaction, with fixed spacing between the CW lasers which could not be widened because of polarization evolution, such that the walk-off could not be measured. This approach was only valid for FWM tones of 50 GHz narrow single polarization sources. This is not representative of the Nonlinear (NL) impact on wider, dual-polarization, coherent transceivers or modems.

The measurement technique employed here uses two-peaks of ASE and has the following benefits 1) ASE more accurately represents spectrally shaped dual-polarization coherent signals, 2) this eliminates the strong polarization hole burning that contaminates the CW approach, 3) the peak separation can be varied to get a direct measurement of walk-off, and 4) it is easy to create multiple peaks to get a full-fill representative G(length) FWM measurement.

FIG. 10 is a block diagram of a system 300 for characterizing optical amplifiers 302, 304 utilizing the two-peak ASE signal. The system 300 includes the ASE source 70 which is connected to the optical amplifiers 302, 304, such as via WSSs 306, 308 (or a single WSS). An Optical Spectrum Analyzer (OSA) 310 can be selectively connected to an output of the amplifier 302 and the amplifier 304.

A measurement process utilizing the system 300 can include creating a two-peak ASE signal with the ASE source 70 and the WSSs 306, 308, sending the two-peak ASE signal to the amplifiers 302, 304, and measuring the outputs of the amplifiers 302, 304 with the OSA 310. Here, and in the section 50, the WSSs can be configured to set attenuation to even the power of the two-peaks in the two-peak ASE signal.

First, the optical amplifier 302 can have its Total Output Power (TOP) adjusted to normalize signal power to a first frequency tested (e.g., 187000 GHz) and an OSA trace can be recorded. Next, the optical amplifier 304 can have its Total Output Power (TOP) adjusted to normalize signal power to a first frequency tested (e.g., 187000 GHz) and an OSA trace can be recorded after calibration. The OSA trace can be recorded at the output of the optical amplifier 304 with stepping signal frequent and TOP of either amplifier 302, 304.

FIG. 11A is a graph of optical spectrum in the system 300 before subtracting ASE, FIG. 11B is a graph of optical spectrum in the system 300 after subtracting ASE, and FIG. 11C is a graph of FWM level versus signal center frequency.

FIG. 12 is flowchart of a process 400 for in-situ fiber characterization. The process 400 includes causing transmission of one or more shaped Amplifier Stimulated Emission (ASE) signals, from an ASE source, on an optical fiber (step 402); obtaining received spectrum of the one or more shaped ASE signals from the optical receiver connected to the optical fiber (step 404); and characterizing the optical fiber based in part on one or more of a nonlinear skirt and a center dip depth in the received spectrum of the one or more two-peak ASE signals (step 406). The process 400 can also include determining a fiber type based on a signature of the one or more of the nonlinear skirt and the center dip depth in the received spectrum of the one or more shaped ASE signals (step 408). In another embodiment, the process 400 can have the step 402 be completed separately and the obtaining step 404 can be performed responsive to another device causing the transmission. For example, the process 400 here could be performed by a management system or the like that receives the spectrum of the one or more shaped ASE signals. The transmission or the causing of such transmission can be performed by an optical network element or node. The received spectrum by the optical network element or node can be provided to the processor or other apparatus implementing the process 400.

The one or more shaped ASE signals can be formed by the ASE source communicatively coupled to a Wavelength Selective Switch (WSS) that is configured to shape ASE from the ASE source to form the one or more shaped ASE signals with one or more peaks and with associated frequency. The one or more shaped ASE signals can have two distinct peaks at the transmission with a significant dip at a center frequency, and the received spectrum of the one or more two-peak ASE signals has much less of a dip that the center dip depth. The optical fiber can be characterized to determine chromatic dispersion, β₂, and fiber nonlinear coefficient, γ. The chromatic dispersion, γ₂, and the fiber nonlinear coefficient, γ can be characterized by a measurement of a shape of the one or more of the nonlinear skirt and the center dip depth as a function of signal wavelength.

The one or more shaped ASE signals can include a plurality of shaped ASE signals with a first set of shaped ASE signals utilized to determine launch power for every span in a section to yield an optimum center dip depth, and a second set of shaped ASE signals that sweep at different frequencies across a signal band to determine a corresponding center dip depth at the different frequencies. The optical fiber can be characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a separate differential delay measurement, to determine chromatic dispersion, β₂, and fiber nonlinear coefficient, γ. The optical fiber can be characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a separate Stimulated Raman Scattering (SRS) measurement, to determine chromatic dispersion, β₂, and fiber nonlinear coefficient, γ. The optical fiber can be characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a Least Mean Square (LMS) fit, to determine chromatic dispersion, β₂, and fiber nonlinear coefficient, γ.

It will be appreciated that some embodiments described herein may include one or more generic or specialized processors (“one or more processors”) such as microprocessors; Central Processing Units (CPUs); Digital Signal Processors (DSPs): customized processors such as Network Processors (NPs) or Network Processing Units (NPUs), Graphics Processing Units (GPUs), or the like; Field Programmable Gate Arrays (FPGAs); and the like along with unique stored program instructions (including both software and firmware) for control thereof to implement, in conjunction with certain non-processor circuits, some, most, or all of the functions of the methods and/or systems described herein. Alternatively, some or all functions may be implemented by a state machine that has no stored program instructions, or in one or more Application Specific Integrated Circuits (ASICs), in which each function or some combinations of certain of the functions are implemented as custom logic or circuitry. Of course, a combination of the aforementioned approaches may be used. For some of the embodiments described herein, a corresponding device in hardware and optionally with software, firmware, and a combination thereof can be referred to as “circuitry configured or adapted to,” “logic configured or adapted to,” etc. perform a set of operations, steps, methods, processes, algorithms, functions, techniques, etc. on digital and/or analog signals as described herein for the various embodiments.

Moreover, some embodiments may include a non-transitory computer-readable storage medium having computer readable code stored thereon for programming a computer, server, appliance, device, processor, circuit, etc. each of which may include a processor to perform functions as described and claimed herein. Examples of such computer-readable storage mediums include, but are not limited to, a hard disk, an optical storage device, a magnetic storage device, a ROM (Read Only Memory), a PROM (Programmable Read Only Memory), an EPROM (Erasable Programmable Read Only Memory), an EEPROM (Electrically Erasable Programmable Read Only Memory), Flash memory, and the like. When stored in the non-transitory computer-readable medium, software can include instructions executable by a processor or device (e.g., any type of programmable circuitry or logic) that, in response to such execution, cause a processor or the device to perform a set of operations, steps, methods, processes, algorithms, functions, techniques, etc. as described herein for the various embodiments.

Although the present disclosure has been illustrated and described herein with reference to preferred embodiments and specific examples thereof, it will be readily apparent to those of ordinary skill in the art that other embodiments and examples may perform similar functions and/or achieve like results. All such equivalent embodiments and examples are within the spirit and scope of the present disclosure, are contemplated thereby, and are intended to be covered by the following claims.

Claims

1. A system comprising: a processor communicatively coupled to an Amplifier Stimulated Emission (ASE) source and an optical receiver, wherein the processor is configured to cause transmission of one or more shaped ASE signals, from the ASE source, into an optical fiber of an optical section between Optical Add/Drop Multiplexer nodes,obtain a spectrum of the one or more shaped ASE signals from the optical receiver connected to the optical fiber, andcharacterize the optical fiber based in part on one or more of a nonlinear skirt and a center dip depth in the received spectrum of the one or more shaped ASE signals.

2. The system of claim 1, wherein the one or more shaped ASE signals are formed by the ASE source communicatively coupled to a Wavelength Selective Switch (WSS) that is configured to shape ASE from the ASE source to form the one or more shaped ASE signals with one or more peaks and with associated frequency.

3. The system of claim 1, wherein the one or more shaped ASE signals have two distinct peaks at the transmission with a significant dip at a center frequency, and the received spectrum of the one or more two-peak ASE signals has much less of a dip that the center dip depth.

4. The system of claim 1, wherein the processor is further configured to determine a fiber type based on a signature of the one or more of the nonlinear skirt and the center dip depth in the received spectrum of the one or more shaped ASE signals.

5. The system of claim 1, wherein the optical fiber is characterized to determine chromatic dispersion, β2, and fiber nonlinear coefficient, γ.

6. The system of claim 5, wherein the chromatic dispersion, β2, and the fiber nonlinear coefficient, γ, are characterized by a measurement of a shape of the one or more of the nonlinear skirt and the center dip depth as a function of signal wavelength.

7. The system of claim 1, wherein the one or more shaped ASE signals includes a plurality of shaped ASE signals with a first set of shaped ASE signals utilized to determine launch power for every span in a section to yield an optimum center dip depth, anda second set of shaped ASE signals that sweep at different frequencies across a signal band to determine a corresponding center dip depth at the different frequencies.

8. The system of claim 1, wherein the optical fiber is characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a separate differential delay measurement, to determine chromatic dispersion, β2, and fiber nonlinear coefficient, γ.

9. The system of claim 1, wherein the optical fiber is characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a separate Stimulated Raman Scattering (SRS) measurement, to determine chromatic dispersion, β2, and fiber nonlinear coefficient, γ.

10. The system of claim 1, wherein the optical fiber is characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a Least Mean Square (LMS) fit, to determine chromatic dispersion, β2, and fiber nonlinear coefficient, γ.

11. A method comprising: causing transmission of one or more shaped Amplifier Stimulated Emission (ASE) signals, from an ASE source, into an optical fiber of an optical section between Optical Add/Drop Multiplexer nodes;obtaining a spectrum of the one or more shaped ASE signals from the optical receiver connected to the optical fiber; andcharacterizing the optical fiber based in part on one or more of a nonlinear skirt and a center dip depth in the received spectrum of the one or more two-peak ASE signals.

12. The method of claim 11, wherein the one or more shaped ASE signals are formed by the ASE source communicatively coupled to a Wavelength Selective Switch (WSS) that is configured to shape ASE from the ASE source to form the one or more shaped ASE signals with one or more peaks and with associated frequency.

13. The method of claim 11, wherein the one or more shaped ASE signals have two distinct peaks at the transmission with a significant dip at a center frequency, and the received spectrum of the one or more two-peak ASE signals has much less of a dip that the center dip depth.

14. The method of claim 11, further comprising determining a fiber type based on a signature of the one or more of the nonlinear skirt and the center dip depth in the received spectrum of the one or more shaped ASE signals.

15. The method of claim 11, wherein the optical fiber is characterized to determine chromatic dispersion, β2, and fiber nonlinear coefficient, γ.

16. The method of claim 11, wherein the one or more shaped ASE signals includes a plurality of shaped ASE signals with a first set of shaped ASE signals utilized to determine launch power for every span in a section to yield an optimum center dip depth, anda second set of shaped ASE signals that sweep at different frequencies across a signal band to determine a corresponding center dip depth at the different frequencies.

17. The method of claim 11, wherein the optical fiber is characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a separate differential delay measurement, to determine chromatic dispersion, β2, and fiber nonlinear coefficient, γ.

18. The method of claim 11, wherein the optical fiber is characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a separate Stimulated Raman Scattering (SRS) measurement, to determine chromatic dispersion, β2, and fiber nonlinear coefficient, γ.

19. The method of claim 11, wherein the optical fiber is characterized based in part on a shape of the one or more of the nonlinear skirt and the center dip depth and a Least Mean Square (LMS) fit, to determine chromatic dispersion, β2, and fiber nonlinear coefficient, γ.

20. An Optical Add/Drop Multiplexing (OADM) node comprising: a Wavelength Selective Switch (WSS) system communicatively coupled to at least a first optical fiber and a second optical fiber;an Amplifier Stimulated Emission (ASE) source connected to the WSS system;a pre-amplifier connected to the WSS system and the first optical fiber;an Optical Channel Monitor (OCM) connected at least to an output of the pre-amplifier; anda processor configured to obtain a spectrum of one or more shaped ASE signals from the OCM, wherein the one or more shaped ASE signals are transmitted over the first optical fiber, andcharacterize the first optical fiber based in part on a one of a nonlinear skirt and a center dip depth in the received spectrum of the one or more shaped ASE signals.