Method and apparatus for evaluating nonlinear impairments in optical fiber links

Abstract

A method and apparatus for evaluating nonlinear impairments of an optical fiber link are provided. The method includes dividing a natural span of the optical fiber link into a plurality of subspans, each of the plurality of subspans being determined based on chromatic dispersion (CD) equivalence. The method also includes obtaining, for each of the plurality of subspans, a subspan function parameter and a subspan input power indicating an input power at a particular subspan, the subspan function parameter including a noise variance and a correlation between the particular subspan and other subspans in the plurality of subspans. The method also includes determining the nonlinear impairment of the optical fiber link based on the subspan input power and the subspan function parameter obtained for each of the plurality of subspans.

Description

[0001] Related applications

[0002] This application claims priority to previously filed U.S. non-provisional application No. 17 / 127,305, filed on December 18, 2020, entitled “METHOD AND APPARATUS FOR EVALUATING NONLINEAR IMPAIRMENT OF ANOPTICAL FIBER LINK,” the contents of which are incorporated by reference into this application in all jurisdictions. Technical Field

[0003] The present invention relates to the field of optical transmission technology, and in particular to a method and device for evaluating nonlinear damage of an optical fiber link. Background Art

[0004] In optical networks, a key issue related to transmission performance is nonlinear impairments on optical fibers, such as fiber nonlinear interference (NLI). In other words, fiber NLI, or nonlinear noise, is one of the most important factors limiting the performance of coherent optical communication systems. Because modern coherent transceivers typically have powerful digital signal processing capabilities, linear impairments such as chromatic dispersion can be fully compensated by the receiver's digital signal processor (DSP). Consequently, dispersion-uncompensated links can offer other advantages, such as cost savings (i.e., no need for dispersion-compensating fiber / modules), higher optical signal-to-noise ratio (OSNR) (i.e., no additional loss in dispersion-compensating fiber / modules), and lower fiber nonlinearity (i.e., NLI generated at different link locations is less coherent than NLI in dispersion-compensated links).

[0005] In dispersion-uncompensated links, NLI can be considered as additive noise despite the performance impairments, as it allows or makes modeling easier (e.g., by generating and including white noise). Consequently, many attempts have been made in both academia and industry to model NLI.

[0006] A well-known example is the so-called Gaussian noise (GN) model, which assumes that the channel is Gaussian (i.e., the Gaussianity assumption of the channel) and therefore ignores the effects of the modulation format. Although the GN model provides higher efficiency, it has lower accuracy (e.g., 5% to 15% arrival loss). The GN model has been improved to the enhanced GN model (eGN). Unlike the GN model, the eGN model includes the effects of the modulation format and can handle non-Gaussian channels. The eGN model is more accurate than the GN model. The eGN model has been extensively studied and validated by simulations and experiments. However, a disadvantage of the eGN model is that it is computationally intensive and therefore very time-consuming because it involves multiple integrals (e.g., triple integrals) and a large number of calculations. For example, for design tools or dynamic routing algorithms, the calculations can be performed tens of thousands of times. Therefore, faster models (e.g., on the order of milliseconds or faster) are highly desirable. Examples of eGN models are as follows:

[0007]

[0008] in:

[0009]

[0010]

[0011]

[0012]

[0013] Therefore, there is a need for a method for modeling nonlinear impairments (e.g., an NLI modeling method) that is faster than current methods. Since the evaluation and modeling of nonlinear impairments currently involve a large amount of computation, there is a need for a method and apparatus for evaluating nonlinear impairments in optical fiber links that does not suffer from one or more limitations of the prior art.

[0014] The purpose of this background information is to disclose information that the applicant believes may be relevant to the present invention. It is not necessary to admit, nor should it be construed, that any of the foregoing information constitutes prior art against the present invention. Summary of the Invention

[0015] The purpose of the embodiments of the present invention is to provide a method and apparatus for evaluating nonlinear impairments of an optical fiber link.

[0016] According to an embodiment of the present invention, a method for evaluating or estimating nonlinear impairments of an optical fiber link is provided. The method includes dividing a natural span of the optical fiber link into a plurality of subspans, each of the plurality of subspans being determined based on chromatic dispersion (CD) equivalence. For each of the plurality of subspans, the method also includes obtaining a subspan function parameter and a subspan input power indicating an input power at a particular subspan, the subspan function parameter including a noise variance and correlation between the particular subspan and other subspans in the plurality of subspans. The method also includes determining the nonlinear impairment of the optical fiber link based on the subspan input power and the subspan function parameter obtained for each of the plurality of subspans. A technical result of this embodiment may be that the NLI evaluation method can be used for any link with uneven fiber span lengths, mixed fiber types, and any signal format, and can be executed at a faster speed than current methods.

[0017] According to some embodiments, the step of determining the nonlinear impairment of the optical fiber link includes obtaining a natural span function parameter based on the subspan input power and the subspan function parameter obtained for each subspan of the multiple subspans, and determining the nonlinear impairment of the optical fiber link based on the natural span input power and the natural span function parameter.

[0018] According to some embodiments, the nonlinear impairments of the optical fiber link are evaluated or estimated independently of fiber type, fiber length, fiber attenuation, and stimulated Raman scattering (SRS) associated with the optical fiber link. According to some embodiments, the noise variance and correlation between the particular subspan and other subspans of the plurality of subspans depends on CD and signal modulation format.

[0019] According to some embodiments, the subspan function parameters obtained for each of the plurality of subspans are stored in a database. According to some embodiments, the natural span function parameters are obtained based on a link topology associated with the optical fiber link. According to some embodiments, the natural span function parameters are obtained based on a power ratio between the subspan input power and a natural span input power indicating an input power of the natural span. According to some embodiments, the natural span function parameters are obtained without the absolute values of the subspan input power and the natural span input power.

[0020] According to some embodiments, at least some of the plurality of subspans have different lengths. According to some embodiments, the input power of each of the plurality of subspans is obtained using the input power of the natural span and an optical fiber attenuation constant.

[0021] According to an embodiment of the present invention, a device for evaluating or estimating nonlinear impairments of an optical fiber link is provided. The device includes a processor and a machine-readable memory including machine-readable instructions. When executed by the processor, the machine-readable instructions cause the device to divide a natural span of the optical fiber link into a plurality of subspans, each of the plurality of subspans being determined based on chromatic dispersion (CD) equivalence. For each of the plurality of subspans, the machine-executable instructions, when executed by the processor, further cause the device to obtain a subspan function parameter and a subspan input power indicating an input power at the particular subspan, the subspan function parameter including a noise variance and a correlation between the particular subspan and other subspans in the plurality of subspans. When executed by the processor, the machine-readable instructions further cause the device to determine the nonlinear impairments of the optical fiber link based on the subspan input power and the subspan function parameter obtained for each of the plurality of subspans.

[0022] While embodiments have been described above in conjunction with aspects of the present invention, these embodiments may be implemented based on these aspects. It will be understood by those skilled in the art that embodiments may be implemented in conjunction with the aspects in which they are described, but may also be implemented together with other embodiments of that aspect. It will be apparent to those skilled in the art that embodiments may be mutually exclusive or otherwise incompatible with one another. Some embodiments may be described in conjunction with one aspect, but may also be applicable to other aspects, as will be apparent to those skilled in the art. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Further features and advantages of the present invention will become apparent from the following detailed description taken in conjunction with the accompanying drawings, in which:

[0024] Figure 1 A typical point-to-point optical link between optical transceivers is shown.

[0025] Figure 2A Various channels in an optical communication system are shown.

[0026] Figure 2B Phase modulation on the signal channel is shown.

[0027] Figure 2C The phase modulation by adjacent pump channels is shown.

[0028] Figure 3 shows the nonlinear distortion ε in a multi-span link i (t) simulation.

[0029] Figure 4 The nonlinear distortion ε induced by cross-phase modulation (XPM) is shown in Figure 2. i (t) The power of the signal channel and the pump channel.

[0030] Figure 5 An optical fiber link divided into subspans with small color spreads according to an embodiment of the present invention is shown.

[0031] Figure 6 The covariance matrix (ρ matrix) for calculating the self-phase modulation (SPM) of normalized nonlinear interference (NLI) noise according to an embodiment of the present invention is shown.

[0032] Figure 7 FIG. 4 shows a covariance matrix (ρ matrix) of an XPM for calculating normalized NLI noise with respect to a 50 GHz pump channel from a signal channel according to an embodiment of the present invention.

[0033] Figure 8 The conversion from a sub-span segment to a full span segment according to an embodiment of the present invention is shown.

[0034] Figure 9 The flowchart shows a process of calculating cumulative noise using sub-span to full-span conversion according to an embodiment of the present invention.

[0035] Figure 10 The electronic device according to the embodiment of the present invention is shown in a schematic diagram.

[0036] Figure 11 is a schematic diagram of an electronic device according to an embodiment of the present invention.

[0037] It should be noted that throughout the drawings, like features are identified by like reference numerals. DETAILED DESCRIPTION

[0038] Optical signals experience nonlinearities (e.g., nonlinear interference (NLI)) in optical fibers, which can lead to signal quality degradation. Therefore, modeling NLI is important for optical link design and quality of transmission (QoT) evaluation, and has been intensively studied.

[0039] The present invention provides an NLI modeling method that can be used for any link with non-uniform fiber span lengths (e.g., the fiber length between two amplifiers), mixed fiber types, and arbitrary signal formats, and can be performed at a faster speed than current methods. According to an embodiment, the NLI modeling method can be a subspan-based method. Using the subspan-based method, the fiber link is divided into subspans with smaller color step lengths. According to an embodiment, any fiber length and fiber type can be represented by this subspan-based method. In other words, the method provides versatility because it is applicable to various link types, including hybrid or non-uniform links. According to an embodiment, the subspan-based method can improve the computational speed of NLI evaluation (or estimation) and modeling.

[0040] The difference between the embodiments of the present invention may be that the existing technology (e.g., GN model or eGN model) uses complex analytical equations to calculate NLI, which may be computationally intensive because the calculation involves multiple integrals (e.g., triple integrals). In contrast, the embodiments of the present invention use offline simulation to generate parameter tables and use simple summation to calculate NLI (e.g., cumulative nonlinear noise). In addition, the subspan-based approach implemented in various embodiments of the present invention can improve the flexibility of the model. In various embodiments, full-span (or natural span) function parameters can be constructed based on subspan parameters, thereby taking advantage of both the subspan-based approach (e.g., flexibility - the parameter table can use this flexibility for the link) and the full-span-based approach (e.g., fast calculation - many orders of magnitude faster than the current GN or eGN model).

[0041] Figure 1A typical point-to-point optical link 100 between optical transceivers is shown. The point-to-point optical link 100 includes multiple transmitters (Tx) 115, each transmitter having a unique wavelength. The wavelengths are multiplexed by a wavelength division multiplexing (WDM) multiplexer 110, which can be configured as an arrayed waveguide (AWG) or a wavelength selective switch (WSS). The transmission medium is an optical fiber 130. An optical amplifier 140 is used to compensate for the losses associated with various optical components or optical fibers (spans). The span length (e.g., the length of the optical fiber between two amplifiers) can be as short as less than 1 km or as long as greater than 100 km. The number of spans can range from one to dozens or more. The most widely used optical fiber types are standard single mode fiber (SSMF) (G.652) and enhanced large effective area fiber (ELEAF) (G.655). At the receiving end, all wavelengths (eg, channels) are WDM demultiplexed by the WDM demultiplexer 120 and received by a receiver (Rx) 125 .

[0042] Point-to-point optical link 100 can be a point-to-point dense wavelength division multiplexing (DWDM) link. In this case, the typical maximum number of channels or wavelengths can be, for example, between 80 and 120, with 50 GHz spacing in the conventional frequency band (C-band). Channels are multiplexed by a DWDM multiplexer (i.e., WDM multiplexer 110 can be a DWDM multiplexer) and demultiplexed by a DWDM demultiplexer (i.e., WDM demultiplexer 120 can be a DWDM demultiplexer).

[0043] Furthermore, with respect to optical fibers in meshed optical networks with more advanced dynamic reconfigurable optical add / drop multiplexing (ROADM), channels can be routed to different fibers at the ROADM site. Furthermore, channel power can be equalized every few fiber spans, typically through WSS equalization within the ROADM site. Furthermore, chromatic dispersion (CD) may not typically be compensated by coherent detection systems.

[0044] In modern optical communication systems, the conventional band (C-band) is commonly used, with the long band (L-band) also being commercially available. The C-band, with a wavelength range of approximately 1530nm to 1565nm, is considered the wavelength band with the lowest loss. The L-band, with a wavelength range of approximately 1565nm to 1625nm, is considered the wavelength band with the second lowest loss. It is understood that the exact wavelength ranges for the C-band and L-band may vary depending on the manufacturer.

[0045] In traditional schemes, the transmission band is divided into 50GHz or 100GHz channels. However, as the transmission rate of the band exceeds 50Gbps, super channels emerge, in which the channel spacing becomes less fixed. Instead, more importantly, there are dozens to more than 100 channels in the transmission band, and various channels may exhibit Kerr nonlinearity in cases such as self-phase modulation (SPM), while other channels co-propagate in the same fiber in the case of cross-phase modulation (XPM). In other words, in the case of SPM, phase modulation can be induced by the signal channel itself, and in the case of XPM, phase modulation can be induced by channels other than the specific signal channel. It can also be noted that fiber Kerr nonlinearity can be a major impairment, so an efficient model for NLI is needed.

[0046] Nonlinear modeling can provide an estimate of the nonlinear noise from SPM and XPM. If the SPM and all XPM terms are known, the total nonlinear noise is the sum of SPM and XPM. Since all channels carry independent data, these channels are statistically independent.

[0047] Total nonlinear noise power It can be calculated as defined in Equation 1 below, where is the nonlinear noise power of the SPM, is the nonlinear noise power of the XPM-induced noise of channel ich (i.e., channel index ich), and Nch is the number of channels:

[0048]

[0049] Figure 2A Various channels in an optical communication system are shown. In an optical link (e.g., optical link 100), there may be multiple channels, and the channel powers may be different. Therefore, the sum of the nonlinear noise power can be obtained on all existing channels except the signal channel (e.g., signal channel 210) itself. It can be noted that for convenience, these noises are signal power normalized or signal-to-noise ratios, such as Figure 2AIt may also be noted that each noise term is generated in a fiber link (eg, fiber 130 ) depending on one or more factors, such as fiber type, fiber length, fiber attenuation constant, channel power, signal modulation format, wavelength, etc.

[0050] Fiber nonlinearity arises in a distributed manner throughout the transmission link. Fiber nonlinearity arises when the transmission medium's refractive index is intensity-dependent or when inelastic scattering occurs. For example, noise generated at different locations may not be independent. The challenge with fiber nonlinearity is accurately performing the accumulation of NLI noise.

[0051] Kerr nonlinearity is the phase modulation of the optical power in the optical fiber. With Kerr nonlinearity, the electric field becomes Among them E s (t) is the optical field of the signal without Kerr nonlinearity, φ NL (t) is the instantaneous nonlinear phase. It can be noted that φ NL (t) = γI(t), where γ is the nonlinear coefficient in rad / W / km and I(t) is the optical intensity / power in watts, including the signal channel itself and other channels on the fiber.

[0052] Figure 2B The phase modulation on the signal channel 220 itself is shown, which is called self-phase modulation (SPM). For SPM, is the average power of the signal channel 220.

[0053] Figure 2C The phase modulation performed by the adjacent (pump) channel 240 is shown, which is called cross-phase modulation (XPM). In XPM, each channel acts as a pump channel for the other channels on the fiber. For XPM, is the average power of the pump channel 240, excluding the signal channel 230 itself.

[0054] The nonlinear phase can be much smaller than 1, so where ε(t) is the normalized distortion of the signal due to Kerr nonlinearity or NLI. The normalized NLI power is related to |ε(t)| 2 or P 2 is proportional to , which can be considered an important scaling law for Kerr's NLI.

[0055] A semi-analytical model for accumulating NLI noise along a multi-span link (ie, a span-based approach) is defined below. It may be noted that this process can be applied to both SPM and XPM. Figure 3 shows the nonlinear distortion ε from the i-th span in a multi-span linki (t) simulation. It can be noted that the simulation can be performed offline. The multi-span link 300 includes Tx 310 and Rx 320, and the transmission medium between Tx 310 and Rx 320 is optical fiber 330. i is the accumulated chromatic dispersion (CD) after i spans. E0(t) is the initial optical electric field at Tx 310. For quadrature amplitude modulation (QAM) signals, this value can be a complex number. The normalized nonlinear distortion ε of the electric field induced by fiber span i is i (t) may be obtained through steps 301 to 303 as further defined below.

[0056] In step 301, the (initial) optical electric field is linearly propagated to the beginning of span i to obtain the waveform This is equivalent to adding a CD i–1 In step 302, the optical field propagates nonlinearly through the span i to obtain the waveform This waveform contains the NLI induced in span i. This can be done by performing a classical step-by-step approach. In step 303, by applying a negative CD i , the optical field propagates linearly back to Tx 310. This is equivalent to deleting CD i Step 303 is necessary because the link dispersion is completely removed in the coherent detection receiver (e.g., Rx 320) in order to obtain the signal waveform at Tx 310 (plus added noise, NLI, etc.). After step 303, the optical field includes the undistorted optical field E0(t), plus the nonlinear interference ε from span i. i (t).

[0057] The total NLI is the sum of all spans, The normalized NLI noise (power) is therefore the variance of ε(t), as defined in Equation 2.

[0058]

[0059] When simulating SPMε i When , the optical field only contains the signal channel. On the other hand, when simulating XPMε i When , the optical field includes a signal channel (eg, signal channel 410) and a pump channel (eg, pump channel 420), as shown in FIG. Figure 4 shown. Figure 4 The nonlinear distortion ε induced by the simulated XPM is shown in Figure 2. i (t) is the power of the signal channel 410 and the pump channel 420. Figure 4As shown, the pump channel 420 is separated from the signal channel 410 by a certain interval (in terms of wavelength). i At (t), the power of the signal channel 410 is significantly lower than the power of the pump channel 420. It should be noted that the signal power level is set to be lower than the pump power (for example, 30 dB lower) in order to obtain a pure XPM effect.

[0060] The total NLI is the sum of all spans, The normalized NLI noise (power) is therefore the variance of ε(t), as shown in Equation 3 below.

[0061]

[0062] Regarding Equation 3, is the normalized noise generated in span i, ρ i,j is the normalized covariance between segments i and j. It should also be noted that

[0063] Further regarding Equation 3, the first term is the sum of the noise generated in each span, and the second term The correlation between noises in different spans (eg, spans i and j, i≠j) is illustrated. It can be understood that the noises in different spans are partially coherent.

[0064] It should be noted that Among them, P i is the pump input power in span i (in the case of SPM, the pump channel is the same as the signal channel), κ i is the normalized noise generated by unit power in span i. Therefore, Equation 4 can be obtained.

[0065]

[0066] Equation 4 is used to calculate the cumulative noise in a multi-span link with arbitrary fiber input power. Equation 4 applies to SPM as well as each XPM term. Note that in the case of SPM, the pump channel is identical to the signal channel. Also note that κ i and ρ i,j It can be obtained by simulation and can be stored in, for example, a database for future use. Since the number of spans N can typically be considered to be less than 30, the calculation can be fast (less than a millisecond, so in the microsecond range). i and ρ i,jis a function of the accumulated dispersion before the i-th and j-th spans, and is a function of the fiber span parameters (e.g., dispersion coefficient, length, loss coefficient, effective area, etc.). It should be noted that this span-based approach may be very useful for uniform span links. However, for non-uniform span links (e.g., span lengths or fiber types may be different across the spans), the span-based approach is not practical because there are too many possible combinations. Although computationally fast, the span-based approach requires a very large parameter table. Therefore, it is desirable to find a simplified method to determine the NLI for non-uniform span links. According to an embodiment, the simplified method may have a parameter for κ i and ρ i,j The invention can also model various links (eg, uniform links and non-uniform span links) while retaining a limited table (eg, for a limited number of combinations).

[0067] The present invention provides a method for nonlinear noise accumulation that is independent of fiber type and fiber length. According to an embodiment, to achieve this independence, a natural fiber span is divided into smaller subspans, each of which is determined based on chromatic dispersion (CD) equivalence (e.g., the natural span of a fiber link is divided into smaller subspans with substantially equal chromatic dispersion steps) rather than fiber length. In other words, the fiber link is divided into small, equally sized (or equal amounts of) (accumulated) chromatic dispersion (CD), thereby eliminating fiber type dependencies associated with the link condition-related parameters κ and ρ.

[0068] According to an embodiment, the fiber link does not have a natural span (e.g., a full span, an undivided fiber span), but is divided into sub-spans with smaller color spreads, such as Figure 5 The equation for calculating the cumulative noise for the subspan-based approach is the same as that discussed above for the calculation of the cumulative noise for the span-based approach, but the index i is used to define the subspan, and the total number of subspans, Nss, is used. Equation 5 defines the cumulative NLI for an optical link when calculated using the method of an embodiment of the present invention.

[0069]

[0070] Among them, P i is the input power of the ith subspan (i.e., the input power at the ith subspan), κ i is the normalized noise variance or dispersion function in subspan i, ρ i,j is the correlation between subspans i and j. i and κ iThey may be collectively referred to as sub-segment function parameters. In some embodiments, the sub-segment function parameters may also include other factors that may affect the calculation of the cumulative NLI.

[0071] Figure 5 The optical fiber link 500 is divided into sub-spans with small color spreading according to an embodiment of the present invention. The multi-span link 500 includes a Tx 510 and an Rx 520 , and the transmission medium between the Tx 510 and the Rx 520 is an optical fiber 530 .

[0072] Depending on the embodiment, each subspan may have a different length. Generally speaking, fibers with lower dispersion coefficients may have larger subspan lengths. Further consideration Figure 5 For fiber link 500, span 502 has a lower dispersion coefficient than spans 501 and 503. Therefore, as shown, span 502 has a larger 'subspan' length than spans 501 and 503. With sufficiently small subspan sizes, the subspan-based approach discussed in this disclosure can be applied to a variety of non-uniform span lengths and any fiber type (e.g., any hybrid fiber type).

[0073] According to an embodiment, the power P at the input of the i-th subspan i The input power across the segment and the fiber attenuation constant can be used to calculate the power of all channels, even with the effects of stimulated Raman scattering (SRS).

[0074] In various embodiments, P is calculated by using the fiber loss coefficient i , different fiber loss coefficients can be considered. On the other hand, in the full-span based method, fiber loss affects the noise variance κ and the correlation ρ. According to embodiments, the fiber effective area can also be considered by appropriately scaling κ. Therefore, the method for determining (e.g., calculating) the NLI of an optical fiber of the present application can be completely independent of the fiber type.

[0075] In general, the subspan-based approach of the present application removes factors such as fiber type, fiber length, fiber attenuation, and SRS from parameters κ and ρ. In various embodiments, κ and ρ may depend only on chromatic dispersion (CD) and signal modulation format.

[0076] According to an embodiment, the covariance matrix can be calculated or the derivation of the covariance matrix can be performed in a simple manner. The covariance can be considered as the correlation between two CDs, except for small |CD i –CD j |, can essentially be zero. For XPM, the correlation length in CD is given by Approximate calculation, where T is the signal symbol pulse width, Δλ is the signal pump wavelength difference, Δλ s is the signal spectrum width. Given the correlation length in CD defined above, the correlation can be obtained by get.

[0077] According to an embodiment, the nonlinear distortion induced by XPM can be reduced as the spacing between the signal channel and the pump channel increases (e.g., the signal channel and the pump channel are more separated), and the relationship between the nonlinear distortion and the covariance can be inferred in the subspan method of the present application. In various embodiments, it may not be necessary to store the covariance matrix for each signal pump interval (i.e., the interval between the signal channel and the pump channel).

[0078] According to an embodiment, a subspan-based approach allows for flexible nonlinear noise accumulation. However, the number of subspans can be much larger than the number of natural spans (e.g., full spans, undivided fiber spans). For example, if the subspan step size is 10 ps / nm, the corresponding SSMF (with a dispersion coefficient of approximately 16.7 ps / (nm / km)) length is approximately 10 / 16.7 ≈ 0.6 km. In other words, in an 80 km SSMF span, there are more than 100 subspans. Therefore, the accumulation calculation may take more time.

[0079] However, in most cases, the time consumed by the cumulative calculations will not be a problem. According to an embodiment, most calculations are performed using double summations involving the ρ matrix. Since the ρ matrix is mostly zero except for the data points near the matrix diagonal (i.e., ρ i,j Only for small |CD i –CD j | is not zero), so only a smaller number of combinations can be double-summed, which greatly speeds up the calculation in most cases, thereby reducing the time required to perform the cumulative calculation. For example, for a 20-span SSMF link, the total NLI can be calculated in milliseconds (e.g., approximately 1-2 ms on a typical personal computer), which may be much faster than when using the GN / eGN model. In some embodiments, such as when the calculation is required by online routing and design tools, the total NLI can be calculated even faster because the model can be called hundreds of thousands of times.

[0080] Figure 6 The covariance matrix (ρ matrix) of 28Gb 100G quadrature phase shift keying (QPSK) for SPM according to an embodiment of the present invention is shown. The covariance matrix is used to calculate the normalized NLI noise power. Figure 6 It can be clearly seen that outside the very close range of the matrix diagonal, the values are essentially zero. Figure 7The covariance matrix (ρ matrix) of 28Gb 100G quadrature phase shift keying (QPSK) for calculating the XPM of normalized NLI noise power with respect to a 50GHz pump channel from a signal channel according to an embodiment of the present invention is shown. Figure 6 and 7 The units used in the axes are 1 km SSMF fiber (i.e., 1 km step size in SSMF) or 16.7 ps / nm dispersion. Figure 6 and 7 In the right graph, all curves are shifted to be centered around 200.

[0081] According to an embodiment, the full-span (natural span) function parameters κ and ρ can be constructed based on the subspan parameters κ and ρ, allowing the flexibility of the subspan-based method of the present invention and the faster calculation speed of the full-span-based method to be utilized. It should be noted that the parameter κ is a parameter indicating noise, and the parameter ρ is a parameter indicating correlation. It should also be noted that the subspan parameters κ and ρ can be obtained through simulation, and the obtained parameters can be stored in, for example, a database for use during calculation.

[0082] Figure 8 FIG shows the conversion from the sub-span method to the full-span method according to an embodiment of the present invention. Figure 8 As shown, according to an embodiment, each natural span (e.g., an undivided optical fiber span, such as span 1, span 2, span 3 ... span N) can be divided into sub-spans having the same chromatic spread (e.g., chromatic dispersion (CD) equivalence) but independent of the optical fiber length.

[0083] Taking into account Figure 8 and the following equations, using the following symbols, where: P is the input power; CD is the accumulated chromatic dispersion; N is the number of natural spans; nK is the number of subspans in the Kth natural span; ε K is the nonlinear distortion induced by the natural span K; ε iK is the nonlinear distortion induced by the i-th subspan of the natural span K. Further considering the notation, a single index (1, 2, 3...K...N) represents a natural span, and a double index (e.g., 11, 12...1K...1N) represents a subspan. Specifically, the first index of the double index represents a subspan, and the second index of the double index represents a natural span. For example, P K It may be the input power at the Kth natural span (i.e. the natural span input power of the Kth natural span), Pi K It may be the input power at the i-th subspan of the K-th natural span.

[0084] According to an embodiment, for the conversion from subspan segments to fullspan segments, the subspan segment accumulation equation is rewritten in a fullspan manner as shown below.

[0085]

[0086] Among them, N1, N2...Nn are the number of sub-spans in natural spans 1, 2...n respectively.

[0087] For example, for the Kth full span segment (ie, natural span segment) and the Lth full span segment, where K and L are not equal and are less than or equal to N, the cumulative noise of the two full span segments can be calculated as follows.

[0088]

[0089] Here, ρ iK,iL is the covariance of subspans iK and iL, which can be further expressed as the following equation.

[0090]

[0091] therefore,

[0092]

[0093] It can be noted that:

[0094]

[0095] is equivalent to:

[0096]

[0097] The full-span (natural-span) model can be derived from the sub-span model using the following parameters:

[0098]

[0099]

[0100] Among them, in calculating κ K and ρ K,L When , only the power ratio (loss) is required, such as and Therefore, the absolute power of the subspans is not required.

[0101] According to an embodiment, the full span (natural span) function parameter calculation does not require the channel input power into the optical fiber.

[0102] Based on the above, the cumulative noise of the entire span is given by the following equation:

[0103]

[0104] Among them, the highlighting parameters are derived based on the subspan parameters as shown below.

[0105]

[0106]

[0107] Figure 9 A flowchart is provided showing a process 900 for calculating cumulative noise using a subspan-to-full-span conversion according to an embodiment of the present invention. According to an embodiment, given link conditions, full-span (natural span) function parameters can be constructed based on subspan parameters, thereby leveraging the advantages of both the subspan-based approach (e.g., providing flexibility because the parameter table can be used for any link) and the full-span-based approach (e.g., providing fast calculation).

[0108] In step 910, once the link topology 901 is provided, the full span (natural span) function parameters (i.e., κ K , ρ K,L As described above, the sub-span segment parameters (i.e., κ) derived from the sub-span segment parameter database 902 may be used. iK , ρ iK,iL ) Dynamically calculate the full span (natural span) function parameter (i.e. κ K , ρ K,L In some embodiments, a finer subspan step can be used, since the conversion from a subspan to a full span can be performed (only) once. Full span (natural span) function parameter κ K and ρ K,L It can be obtained through the following methods:

[0109]

[0110]

[0111] In step 920, based on the full span (natural span) function parameter κ obtained in step 910, K and ρ K,L Calculate the cumulative noise. In step 920, the accumulation based on the full span can be expressed as follows:

[0112]

[0113] Figure 10A method for evaluating (or estimating) nonlinear impairments of an optical fiber link according to an embodiment of the present invention is shown. The method includes dividing a natural span 1010 of the optical fiber link into a plurality of subspans, each of the plurality of subspans being determined based on chromatic dispersion (CD) equivalence. For example, the natural span of the optical fiber link is divided into smaller subspans having substantially equal step sizes (or equal amounts) of (cumulative) chromatic dispersion (CD) rather than based on fiber length. The method also includes obtaining 1020, for each subspan in the plurality of subspans, subspan function parameters and a subspan input power indicating an input power at a particular subspan (e.g., an input power at an i-th subspan that is part of the natural span (full span)). The subspan function parameters include noise variance and correlation between the particular subspan (e.g., the i-th subspan divided from the natural span (full span)) and other subspans in the plurality of subspans. The method further includes determining 1020 nonlinear impairments of the optical fiber link based on the subspan input power and subspan function parameters obtained for each of the plurality of subspans. In various embodiments, the nonlinear impairments can be determined by calculating the cumulative NLI over the optical link according to Equation 5 above.

[0114] According to some embodiments, the step of determining the nonlinear impairment of the optical fiber link includes obtaining a natural span function parameter based on a subspan input power and a subspan function parameter obtained for each subspan of a plurality of subspans, and determining the nonlinear impairment of the optical fiber link based on the natural span function parameter and a natural span input power indicating an input power of the natural span.

[0115] According to some embodiments, nonlinear impairments of an optical fiber link are evaluated or estimated independently of fiber type, fiber length, fiber attenuation, and stimulated Raman scattering (SRS) associated with the optical fiber link. According to some embodiments, noise variance and correlation between a particular subspan and other subspans in a plurality of subspans depend on CD and signal modulation format.

[0116] According to some embodiments, the subspan function parameters obtained for each subspan in the plurality of subspans are stored in a database. According to some embodiments, at least some of the plurality of subspans have different lengths. According to some embodiments, the input power of each subspan in the plurality of subspans is obtained using the input power of the natural span and the fiber attenuation constant.

[0117] According to some embodiments, the natural span function parameter is obtained based on a link topology associated with the optical fiber link. According to some embodiments, the natural span function parameter is obtained based on a power ratio between a subspan input power and a natural span input power indicating an input power of the natural span. According to some embodiments, the natural span function parameter is obtained without the absolute values of the subspan input power and the natural span input power.

[0118] Figure 11 1 is a schematic diagram of an electronic device 1100 according to different embodiments of the present invention, which can perform any or all of the operations of the above-described methods and features, which are explicitly or implicitly described herein. For example, dedicated hardware capable of executing instructions for the above-described methods and features can be configured as the electronic device 1100. In addition, a network device that supports dense wavelength division multiplexing (DWDM) can be configured as the electronic device 1100. Furthermore, a computer equipped with network functions can be configured as the electronic device 1100. The electronic device can be a mobile device or a device forming part of a cell or base station, a wireless access node, a control function, an infrastructure, or other equipment in a wireless communication access network or a core network.

[0119] As shown, the device includes a processor 1110 (e.g., a central processing unit (CPU) or a special-purpose processor such as a graphics processing unit (GPU), or other such processor unit), memory 1120, non-transitory mass storage 1130, I / O interface 1140, network interface 1150, and transceiver 1160, all of which are communicatively coupled via a bidirectional bus 1170. According to certain embodiments, any or all of the depicted elements may be used, or only a subset of the elements may be used. In addition, device 1100 may include multiple instances of certain elements, such as multiple processors, multiple memories, or multiple transceivers. In addition, elements of the hardware device may be directly coupled to other elements without a bidirectional bus. In addition to processors and memories, other electronic devices such as integrated circuits may be used to perform the required logical operations.

[0120] The memory 1120 may include any type of non-transitory memory, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), read-only memory (ROM), or any combination thereof. The mass storage 1130 may include any type of non-transitory storage device, such as a solid-state drive, a hard disk drive, a magnetic disk drive, an optical disk drive, a USB drive, or any computer program product for storing data and machine-executable program code. According to certain embodiments, the memory 1120 or the mass storage 1130 may record thereon statements and instructions executable by the processor 1110 for performing any of the above-described method operations.

[0121] It should be understood that although specific embodiments of the technology have been described herein for illustrative purposes, various modifications may be made without departing from the scope of the technology. Accordingly, the specification and drawings are to be regarded only as illustrative of the invention as defined by the appended claims and are intended to cover any and all modifications, variations, combinations or equivalents that fall within the scope of the invention. Specifically, within the scope of the present technology, a computer program product or program element for storing machine-readable signals, or a program storage or memory device such as a magnetic wire or optical fiber, a magnetic tape or an optical disk, is provided for controlling the operation of a computer according to the method of the present technology and / or constructing part or all of the components of a computer according to the system of the present technology.

[0122] The actions associated with the methods described herein may be implemented as coded instructions in a computer program product. In other words, the computer program product is a computer-readable medium on which software codes are recorded to perform the methods when the computer program product is loaded into a memory and executed on a microprocessor of a wireless communication device.

[0123] Furthermore, each operation of the method may be executed on any computing device, such as a personal computer, server, PDA, etc., according to one or more or a portion of one or more program elements, modules, or objects generated from any programming language, such as C++, Java, etc. Furthermore, each operation or a file or object, etc., that implements each of the operations described may be executed by dedicated hardware or a circuit module designed for this purpose.

[0124] Through the description of the above embodiments, the present invention can be implemented only by hardware, or by software and a necessary general hardware platform. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product. The software product can be stored in a non-volatile or non-transient storage medium, which can be a compact disk read-only memory (CD-ROM), a USB flash drive, or a mobile hard disk. The software product includes many instructions that enable a computer device (a personal computer, a server, or a network device) to execute the method provided in the embodiment of the present invention. For example, such execution can correspond to a simulation of a logical operation as described herein. According to an embodiment of the present invention, the software product may additionally or alternatively include a plurality of instructions that enable a computer device to execute the operation of configuring or programming a digital logic device.

[0125] Although the present invention has been described with reference to specific features and embodiments thereof, it is apparent that various modifications and combinations of the present invention may be made without departing from the present invention. Accordingly, the specification and drawings are to be regarded only as illustrative of the present invention as defined by the appended claims and are intended to cover any and all modifications, variations, combinations or equivalents that fall within the scope of the present invention.

Claims

1. A method for evaluating nonlinear impairments of an optical fiber link, characterized in that: The method comprises: Dividing a natural span of the optical fiber link into a plurality of subspans, each of the plurality of subspans being determined based on chromatic dispersion (CD) equivalence; For each subspan of the plurality of subspans, obtaining a subspan function parameter and a subspan input power indicating an input power at the specific subspan, the subspan function parameter including a noise variance and a correlation between the specific subspan and other subspans of the plurality of subspans; The nonlinear impairment of the optical fiber link is determined based on the subspan input power and the subspan function parameter obtained for each subspan of the plurality of subspans.

2. The method according to claim 1, characterized in that The step of determining the nonlinear impairment of the optical fiber link comprises: Obtaining a natural span function parameter based on the subspan input power and the subspan function parameter obtained for each subspan of the plurality of subspans; The nonlinear impairment of the optical fiber link is determined based on the natural span function parameter and a natural span input power indicating an input power at the natural span.

3. The method according to any one of claims 1 to 2, characterized in that The nonlinear impairments of the optical fiber link are evaluated independently of the fiber type, fiber length, fiber attenuation, and stimulated Raman scattering (SRS) associated with the optical fiber link.

4. The method according to claim 3, characterized in that The noise variance and correlation between the particular subspan and other subspans of the plurality of subspans depends on CD and signal modulation format.

5. The method according to any one of claims 1 to 4, characterized in that The sub-span segment function parameter obtained for each sub-span segment of the multiple sub-span segments is stored in a database.

6. The method according to claim 2, characterized in that The natural span function parameters are obtained based on a link topology associated with the optical fiber link.

7. The method according to claim 2, characterized in that The natural span function parameter is obtained based on a power ratio between the sub-span input power and the natural span input power.

8. The method according to claim 7, characterized in that The natural span function parameter is obtained without the absolute values of the sub-span input power and the natural span input power.

9. The method according to any one of claims 1 to 8, characterized in that At least some of the plurality of subspans have different lengths.

10. The method according to any one of claims 1 to 9, characterized in that The input power of each subspan of the plurality of subspans is obtained using the input power of the natural span and an optical fiber attenuation constant.

11. A device for evaluating nonlinear impairments of an optical fiber link, characterized in that: The device comprises: processor; a machine-readable memory comprising machine-readable instructions that, when executed by the processor, cause the apparatus to: Dividing a natural span of the optical fiber link into a plurality of subspans, each of the plurality of subspans being determined based on chromatic dispersion (CD) equivalence; For each subspan of the plurality of subspans, obtaining a subspan function parameter and a subspan input power indicating an input power at the specific subspan, the subspan function parameter including a noise variance and a correlation between the specific subspan and other subspans of the plurality of subspans; The nonlinear impairment of the optical fiber link is determined based on the subspan input power and the subspan function parameter obtained for each subspan of the plurality of subspans.

12. The device according to claim 11, characterized in that The apparatus determines the nonlinear impairment of the optical fiber link by: Obtaining a natural span function parameter based on the subspan input power and the subspan function parameter obtained for each subspan of the plurality of subspans; The nonlinear impairment of the optical fiber link is determined based on the natural span function parameter and a natural span input power indicating an input power at the natural span.

13. The device according to any one of claims 11 to 12, characterized in that The nonlinear impairments of the optical fiber link are evaluated independently of the fiber type, fiber length, fiber attenuation, and stimulated Raman scattering (SRS) associated with the optical fiber link.

14. The device according to claim 13, characterized in that The noise variance and correlation between the particular subspan and other subspans of the plurality of subspans depends on CD and signal modulation format.

15. The device according to any one of claims 11 to 14, characterized in that The sub-span segment function parameter obtained for each sub-span segment of the multiple sub-span segments is stored in a database.

16. The device according to any one of claims 11 to 15, characterized in that The natural span function parameters are obtained based on a link topology associated with the optical fiber link.

17. The device according to claim 12, characterized in that The natural span function parameter is obtained based on a power ratio between the sub-span input power and the natural span input power.

18. The device according to claim 17, characterized in that The natural span function parameter is obtained without the absolute values of the sub-span input power and the natural span input power.

19. The device according to any one of claims 11 to 18, characterized in that At least some of the plurality of subspans have different lengths.

20. The device according to any one of claims 11 to 19, characterized in that The input power of each subspan of the plurality of subspans is obtained using the input power of the natural span and an optical fiber attenuation constant.

Images