C + L + S wave band multi-device configuration optimization method and C + L + S wave band multi-device configuration optimization device

Abstract

The invention provides a C + L + S wave band multi-device configuration optimization method and device, and the method comprises the steps: calculating the approximate optimal transmission power of an optical fiber entrance part through employing a local optimal approximate global optimal algorithm; taking the approximate optimal transmitting power as the receiving power of each span section, and recalculating the transmitting signal power at the starting position of the span section by reversely solving a stimulated Raman scattering effect power evolution equation; the gain and gain slope of the optical fiber amplifier are set, so that the signal power which begins to be input in each span segment approaches the transmitting signal power; constructing a loss function based on the generalized signal-to-noise ratio of each wave band; and the gain and the gain slope set after inversion of the stimulated Raman scattering effect are used as initial values, and the gain and the gain slope of the optical fiber amplifier and the channel loss value of the wavelength selection switch are iteratively optimized by using a gradient rising algorithm and taking the loss function improvement as a target. According to the method provided by the invention, only a small amount of iteration is needed, the operation speed is high, and multi-device second-level configuration optimization of the broadband optical transmission system can be realized.

Description

Technical Field

[0001] The present invention relates to the field of optical fiber communication technology, and in particular to a C+L+S band multi-device configuration optimization method and device. Background Art

[0002] With the ever-increasing demand for data traffic, the industry has proposed a variety of solutions to meet capacity growth trends through optical network upgrades. Multi-band transmission offers significant economic advantages due to its ability to effectively utilize existing fiber infrastructure, making it a more promising option compared to space-division multiplexing (SDM). Currently, C-band and L-band device technologies have matured and are gradually being deployed and promoted in existing networks. Meanwhile, the industry is shifting its R&D focus to the S-band, primarily based on two considerations: first, the S-band exhibits superior signal attenuation characteristics compared to the E-band and U-band; second, it can be amplified using the relatively mature thulium-doped fiber amplifier (TDFA) technology. Given the spectral proximity of the S-band and L-band and the maturity of TDFA technology, the S-band is a natural choice for expanding C+L transmission systems. However, it is worth noting that the introduction of the S-band, while expanding transmission bandwidth, also exacerbates stimulated Raman scattering (SRS), triggering power transfer from high-frequency channels to low-frequency channels. This not only creates a performance imbalance between the various bands but also significantly degrades S-band transmission quality. In this context, how to achieve performance balance and quality optimization of the C+L+S multi-band system has become a technical challenge that the industry urgently needs to overcome.

[0003] Generalized signal-to-noise ratio (GSNR) is a core metric for measuring quality of transmission (QoT). Optimizing GSNR in C+L+S band transmission systems, where nonlinear effects are significant and complex, is a current research hotspot. Currently, academic and industry research on GSNR optimization for multi-band transmission systems focuses on the following areas: first, maximizing link capacity while ensuring transmission stability, which requires precise configuration of power allocation across bands; second, improving the system's maximum GSNR value and optimizing its spectral flatness, which is particularly important for long-distance transmission; third, improving the GSNR performance of the system's worst channel, which directly determines the lower limit of the system's overall transmission quality; and finally, comprehensively considering the dynamic balance between amplified spontaneous emission (ASE) noise and nonlinear interference (NLI), which is crucial for achieving optimal transmission performance.

[0004] To achieve these optimization goals, numerous methods have been proposed in recent years, primarily through controlling transmit power or configuring amplifiers based on various algorithms. These include the local optimization global optimization (LOGO) algorithm, genetic algorithms (GAs), particle swarm optimization (PSO) algorithms, greedy search, and gradient descent algorithms. The LOGO algorithm is a hierarchical optimization strategy that decomposes the optimization problem into two levels: local and global, for collaborative solution. In the local optimization phase, the algorithm configures components for a single span or band. In the global optimization phase, it comprehensively considers the interactions between spans across the entire transmission link to achieve optimal overall performance. Genetic algorithms are heuristic search algorithms that gradually optimize solutions based on natural selection and genetic mechanisms. They simulate the evolutionary process, including selection, crossover, and mutation operations, to find the optimal solution in the search space. During the optimization process, the algorithm evaluates the fitness of each configuration (such as the average GSNR or flatness), retaining high-performing individuals and eliminating low-performing ones. After multiple generations of evolution, the algorithm converges to a near-optimal solution. The PSO algorithm is an optimization algorithm based on swarm intelligence. It seeks the optimal solution by simulating the collaborative search behavior of a swarm of particles in the solution space. Each particle represents a possible solution, and it adjusts its position based on its own experience and that of the swarm, gradually approaching the global optimum. The greedy search algorithm employs a stepwise optimization strategy, selecting the currently optimal local improvement at each step. Starting from an initial configuration, the algorithm compares the performance metrics of neighboring solutions, consistently selecting the direction that yields the greatest improvement in the objective function. This algorithm is computationally efficient and simple to implement, but can be trapped in local optima. It is suitable for quickly obtaining near-optimal configurations for each component in scenarios with a small parameter space. The gradient descent algorithm adjusts the configuration of each component to achieve optimal system performance within certain constraints, such as signal quality or noise limits. Its rapid convergence in the gradient direction makes it suitable for optimization problems with relatively smooth objective functions.

[0005] In the aforementioned prior art C+L+S band transmission system, the introduction of the S band results in a larger optimization space, more complex calculations, higher time costs, and more devices requiring configuration, such as transceivers, amplifiers, and wavelength selective switches (WSS). Existing technologies often only configure a single type of device, and each device type can only control a portion of the transmission performance. For example, an in-line amplifier can only improve the average performance of a single frequency band but lacks the ability to control the performance of a single channel. Therefore, single-objective optimization may produce suboptimal results. Furthermore, as the number of objects to be configured increases, the computational complexity of the aforementioned algorithm also increases, resulting in longer computation times for the complex iterative calculations during the optimization process. Summary of the Invention

[0006] In view of this, an embodiment of the present invention provides a C+L+S band multi-device configuration optimization method and apparatus to eliminate or improve one or more defects in the prior art.

[0007] In one aspect, the present invention provides a method for optimizing a multi-device configuration in the C+L+S band, the method comprising the following steps:

[0008] The local optimal approximation global optimal algorithm is used to calculate the approximate optimal transmission power at the fiber entry point;

[0009] The approximate optimal transmit power is used as the received power of each span, and the transmit signal power at the beginning of the span is recalculated by inversely solving the stimulated Raman scattering effect power evolution equation. The gain and gain slope of the optical fiber amplifier are set so that the input signal power at the beginning of each span approaches the transmit signal power.

[0010] The generalized signal-to-noise ratios of the C, L, and S bands are calculated to construct a loss function. The gain and gain slope of the optical fiber amplifier set after stimulated Raman scattering inversion are used as initial values. A gradient ascent algorithm is used to iteratively optimize the gain and gain slope of the optical fiber amplifier with the goal of improving the loss function to obtain the optimal transmission power.

[0011] Based on the initial loss value preset when the wavelength selective switch leaves the factory, the gradient ascent algorithm is used to iteratively fine-tune the channel loss value of the wavelength selective switch with the goal of improving the loss function.

[0012] In some embodiments of the present invention, a locally optimal approximate global optimal algorithm is used to calculate the approximate optimal transmission power at the fiber input point, including:

[0013] The local optimal approximation global optimal algorithm ignores the stimulated Raman scattering effect, and each span in the link can be optimized individually so that the optical performance of each span is optimized; by directly adding or superimposing the optimal optical performance of each span, the optimal optical performance of the entire multi-span link is approximately calculated.

[0014] In some embodiments of the present invention, the calculation process includes:

[0015] According to the local optimal approximation global optimal algorithm, the optical signal-to-noise ratio of any span is expressed as:

[0016]

[0017] Based on the local optimal approximation global optimal algorithm, the optical signal-to-noise ratio optimal nonlinear noise power spectrum can only obtain a suboptimal transmission power, which is expressed as:

[0018]

[0019] Among them, osnr n represents the optical signal-to-noise ratio of the nth span; G WDM,n Represents the nonlinear noise power spectrum of the optical signal-to-noise ratio; R s Indicates the symbol rate; a n Indicates the optical transmission attenuation of the nth span; B N represents the channel bandwidth; h represents the Planck constant; v represents the channel center frequency; F n represents the noise figure of the fiber amplifier in the nth span; ρ NLI,n represents the nonlinear crosstalk coefficient; P in represents the suboptimal transmit power; represents the optical signal-to-noise ratio optimal nonlinear noise power spectrum.

[0020] In some embodiments of the present invention, in the optical signal-to-noise ratio formula:

[0021] To optimize the optical signal-to-noise ratio (OSNR) nonlinear noise power spectrum, the calculation formula is:

[0022]

[0023] Where, ρ NLI,n represents the nonlinear crosstalk coefficient, which is calculated as follows:

[0024]

[0025] in, represents the optimal nonlinear noise power spectrum of the optical signal-to-noise ratio; h represents the Planck constant; v represents the channel center frequency; F n represents the noise figure of the fiber amplifier in the nth span; a n represents the light transmission attenuation of the nth span; α represents attenuation; γ represents the nonlinear coefficient; β represents the group velocity dispersion; L eff Indicates the effective length of optical fiber; B wDM Represents the entire transmission bandwidth.

[0026] In some embodiments of the present invention, the generalized signal-to-noise ratios of the C, L, and S bands are calculated separately to construct a loss function, including:

[0027] The loss function includes the negative of the mean and standard deviation of the generalized signal-to-noise ratio, and is calculated as follows:

[0028] J = mean(GSNR) - τstd(GSNR);

[0029] Wherein, J represents the loss function; meas(GSNR) represents the mean value of the generalized signal-to-noise ratio; std(GSNR) represents the standard deviation of the generalized signal-to-noise ratio; τ represents the weight coefficient.

[0030] In some embodiments of the present invention, a gradient ascent algorithm is used to iteratively optimize the gain and gain slope of the optical fiber amplifier with the goal of improving the loss function, including:

[0031] The gradient ascent algorithm is used to independently optimize the optical fiber amplifier of each span, or to jointly optimize the optical fiber amplifiers of multiple spans.

[0032] In some embodiments of the present invention, the method further comprises:

[0033] In an independent optimization scenario, it is assumed that the effects between the spans are not coupled; the gain and gain slope of the optical fiber amplifier set after the stimulated Raman scattering effect inversion are used as initial values; for each span, the current generalized signal-to-noise ratio is calculated, and the loss function is calculated based on the generalized signal-to-noise ratio; the gradient of the loss function relative to the gain and gain slope of the optical fiber amplifier is calculated; the gain and gain slope of each span are updated using the gradient ascent algorithm; when the loss function converges or reaches a preset number of iterations, the optimization is stopped to obtain the optimal gain and optimal gain slope of each span;

[0034] In a joint optimization scenario, it is assumed that the effects between the spans are coupled; the gain and gain slope of the optical fiber amplifier set after inversion of the stimulated Raman scattering effect are used as initial values; for all spans, the combined generalized signal-to-noise ratio is calculated, and the loss function is calculated based on the generalized signal-to-noise ratio; the gradient of the loss function relative to the gain and gain slope of the optical fiber amplifier is calculated; the gain and gain slope of all spans are updated using the gradient ascent algorithm; when the loss function converges or reaches a preset number of iterations, the optimization is stopped to obtain the optimal gain and optimal gain slope of all spans.

[0035] In some embodiments of the present invention, based on an initial loss value preset when the wavelength selective switch leaves the factory, the gradient ascent algorithm is used to iteratively fine-tune the channel loss value of the wavelength selective switch with the goal of improving the loss function, including:

[0036] The gradient ascent algorithm is used to fine-tune the loss value of each channel of the wavelength selective switch, or the super-channel optimization method is used to simultaneously adjust the loss values ​​of multiple adjacent channels.

[0037] In some embodiments of the present invention, the method further comprises:

[0038] In a channel-by-channel optimization scenario, the gradient of the loss function relative to the loss value of each channel of the wavelength selective switch is calculated; the loss value of each channel of the wavelength selective switch is updated using the gradient ascent algorithm; when the loss function converges or reaches a preset number of iterations, the optimization is stopped to obtain the optimal loss value of each channel of the wavelength selective switch;

[0039] In a super-channel optimization scenario, a preset number of adjacent channels in the wavelength selective switch are merged into a super-channel, and the gradient of the loss function relative to the loss value of the super-channel is calculated; the loss value of the super-channel is updated using the gradient ascent algorithm; when the loss function converges or reaches a preset number of iterations, the optimization is stopped to obtain the optimal loss value of the super-channel.

[0040] On the other hand, the present invention also provides a C+L+S band multi-device configuration optimization device, which is used to implement the steps of any of the C+L+S band multi-device configuration optimization methods mentioned above.

[0041] The present invention provides a method and apparatus for optimizing multi-device configurations in the C+L+S bands. The method comprises: using a local optimal approximation to a global optimal algorithm to calculate the approximate optimal transmit power at the fiber entry point; using the approximate optimal transmit power as the received power for each span, and recalculating the transmit signal power at the start of the span by inversely solving the stimulated Raman scattering power evolution equation; setting the gain and gain slope of the fiber amplifier so that the input signal power at the start of each span approximates the transmit signal power; constructing a loss function based on the generalized signal-to-noise ratio of each band; and using the gain and gain slope set after stimulated Raman scattering inversion as initial values, using a gradient ascent algorithm to iteratively optimize the gain and gain slope of the fiber amplifier and the channel loss value of the wavelength selective switch with the goal of improving the loss function. Furthermore, a super-channel optimization method is used to simultaneously adjust the loss values ​​of multiple adjacent channels of the wavelength selective switch, reducing parameter dimensionality, improving efficiency, and reducing time costs. The method provided by the present invention requires only a small number of iterations, operates rapidly, and can achieve multi-device configuration optimization for wide-band optical transmission systems. At the same time, it also takes into account the flat and stable output of the generalized signal-to-noise ratio, so that the operation of the entire system is stable and resources are used efficiently.

[0042] Additional advantages, objects, and features of the present invention will be set forth in part in the following description and will become apparent to those skilled in the art upon examination of the following or may be learned from practice of the present invention. The objects and other advantages of the present invention may be realized and obtained by the structures particularly pointed out in the description and drawings.

[0043] Those skilled in the art will understand that the purposes and advantages that can be achieved by the present invention are not limited to the above specific descriptions, and the above and other purposes that can be achieved by the present invention will be more clearly understood based on the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of this application, and do not constitute a limitation of the present invention. In the drawings:

[0045] Figure 1 Schematic diagram of the steps of a C+L+S band multi-device configuration optimization method according to an embodiment of the present invention.

[0046] Figure 2 Schematic diagram of the principle of a C+L+S band multi-device configuration optimization method in one embodiment of the present invention.

[0047] Figure 3 Schematic diagram of the flow of a C+L+S band multi-device configuration optimization method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0049] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, the accompanying drawings only show structures and / or processing steps closely related to the solutions according to the present invention, while other details that are not closely related to the present invention are omitted.

[0050] It should be emphasized that the term "include / comprises" when used herein refers to the existence of features, elements, steps or components, but does not exclude the existence or addition of one or more other features, elements, steps or components.

[0051] It should also be noted that, unless otherwise specified, the term "connection" herein may refer not only to a direct connection but also to an indirect connection involving an intermediate.

[0052] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent the same or similar components, or the same or similar steps.

[0053] It should be emphasized here that the step marks mentioned below do not limit the order of the steps, but it should be understood that the steps can be executed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be executed simultaneously.

[0054] In order to solve the problems in the existing optical transmission system device configuration optimization technology solutions, such as the need for a large number of complex iterations, being subject to local optimal misleading, slow operation speed, not being able to scale well with the number of spans, and thus not being suitable for real-time reconfiguration and control of dynamic networks, and not being suitable for rapid recovery in the event of a fault, as well as the larger optimization space and higher complexity as the band becomes wider, and the suboptimal performance of the single device configuration, the present invention proposes a C+L+S band multi-device configuration optimization method, which optimizes the final output performance by configuring multiple devices in the link, including transmitters, amplifiers, wavelength selection switches, etc. First, on the basis of the local optimal approximation global optimal algorithm, the stimulated Raman scattering effect is further considered, and finally, the gradient ascent algorithm is used to directly maximize the output generalized optical signal-to-noise ratio, adjust the optical fiber amplifier gain and gain slope settings of each span, and the configuration of each channel loss of the wavelength selection switch, and make the output generalized optical signal-to-noise ratio as flat as possible, so as to achieve fast and accurate optimization. Specifically, Figure 1 As shown, the method includes the following steps S101 to S104:

[0055] Step S101: using a local optimal approximation to global optimal algorithm to calculate the approximate optimal transmission power at the fiber input.

[0056] Step S102: Using the approximate optimal transmission power as the received power of each span; recalculating the transmission signal power at the beginning of the span by inversely solving the stimulated Raman scattering effect power evolution equation; and setting the gain and gain slope of the optical fiber amplifier so that the input signal power at the beginning of each span approaches the transmission signal power.

[0057] Step 103: Calculate the generalized signal-to-noise ratios of the C, L, and S bands respectively to construct a loss function; use the gain and gain slope of the fiber amplifier set after the stimulated Raman scattering effect inversion as initial values, and use a gradient ascent algorithm to iteratively optimize the gain and gain slope of the fiber amplifier with the goal of improving the loss function.

[0058] Step S104: Based on the initial loss value preset when the wavelength selective switch leaves the factory, a gradient ascent algorithm is used to iteratively fine-tune the channel loss value of the wavelength selective switch with the goal of improving the loss function.

[0059] like Figure 2 As shown in the figure, it is the principle diagram of the wide-band multi-device configuration optimization algorithm adapted to dynamic scenarios. Figure 3 Figure 2 shows a flowchart of a wide-band multi-device configuration optimization algorithm.

[0060] In step S101, the stimulated Raman scattering effect is ignored, and only the balance between the amplifier spontaneous emission and nonlinear crosstalk is considered. The local optimal approximate global optimal (LOGO) algorithm is used to calculate the approximate optimal transmission power at the fiber input of the optical fiber amplifier.

[0061] The LOGO (Lossless Optimization of Gain and Slope) algorithm is derived from a nonlinear Gaussian noise model in optical transmission. It ignores the effects of stimulated Raman scattering (SRS). Each span in a link (for example, the amplification section of each fiber amplifier) ​​can be individually optimized to achieve optimal optical performance (such as signal-to-noise ratio) for each span. By directly adding or superimposing the optimal optical performance of each span, the optimal optical performance of the entire multi-span link can be approximately calculated.

[0062] In some embodiments, the optical fiber amplifier uses an erbium doped fiber amplifier (EDFA) and a thulium doped fiber amplifier (TDFA).

[0063] In some embodiments, according to the LOGO algorithm, the optical signal-to-noise ratio of the nth span can be expressed as formula (1):

[0064]

[0065] Among them, osnr n represents the optical signal-to-noise ratio of the nth span; G WDM,n The nonlinear noise power spectrum representing the optical signal-to-noise ratio; R s Indicates the symbol rate; a n Indicates the optical transmission attenuation of the nth span; B N represents the channel bandwidth; h represents the Planck constant; v represents the channel center frequency; F n represents the noise figure of the n-th span fiber amplifier; ρ NLI,n represents the nonlinear crosstalk coefficient.

[0066] In formula (1), in order to make the optical signal-to-noise ratio osnr n The optimal nonlinear noise power spectrum is calculated as shown in formula (2):

[0067]

[0068] in, represents the optical signal-to-noise ratio osnr n The optimal nonlinear noise power spectrum and the explanation of other parameters refer to formula (1) and will not be repeated here.

[0069] In formula (2), ρ NLI,n It represents the nonlinear crosstalk coefficient, and its calculation formula is shown in formula (3):

[0070]

[0071] Where α represents the attenuation when the frequency (channel) correlation is not considered; γ represents the nonlinear coefficient; β represents the group velocity dispersion; L eff Indicates the effective length of optical fiber; B WDM Represents the entire transmission bandwidth.

[0072] Since the derivation of the LOGO algorithm is completed without considering the stimulated Raman scattering effect, the optical signal-to-noise ratio osnr is n Optimal nonlinear noise power spectrum Only a suboptimal transmission power can be obtained, and the calculation formula is shown in formula (4):

[0073]

[0074] Among them, P in Indicates suboptimal transmit power.

[0075] In step S102, the suboptimal transmission power P calculated in step S101 is in As the received power of each span. Since the stimulated Raman scattering effect was ignored in step S101, in step S102, the transmitted signal power at the beginning of the span is recalculated by inversely solving the stimulated Raman scattering effect power evolution equation. This process can obtain a more accurate signal power, thereby optimizing the gain and gain slope of the fiber amplifier. These gain and gain slope are used to adjust the fiber amplifier so that the signal power at the beginning of each span is closer to the ideal transmitted signal power calculated by inverse calculation of the stimulated Raman scattering effect.

[0076] Gain refers to the degree to which a fiber amplifier amplifies the input signal, typically expressed as the ratio of the signal's output power to its input power. It measures the change in signal intensity after passing through the amplifier. Gain is typically expressed in decibels (dB). The higher the gain, the stronger the signal is amplified. Conversely, a negative gain indicates signal attenuation. Gain slope refers to the variation in signal gain over frequency, or more specifically, the degree to which signal gain varies with wavelength (or frequency) in fiber optic communications. Tilt represents the variation in signal gain across different frequency bands (or wavelengths). In some fiber optic communication systems, since signals of different wavelengths experience different gains when passing through the amplifier, the gain slope describes this difference. Typically, the system desires to maintain a gain as flat as possible across different wavelengths to ensure that all signals are evenly amplified.

[0077] In step S103, considering the average signal-to-noise ratio of the channel, the present invention hopes to achieve a high and relatively flat generalized signal-to-noise ratio (GSNR) in a wide band, and therefore, a loss function is constructed based on the generalized signal-to-noise ratio.

[0078] In some embodiments, the loss function includes the negative of the mean and standard deviation of the generalized signal-to-noise ratio, and is calculated as shown in formula (5):

[0079] J=mean(GSNR)-τstd(GSNR); (5)

[0080] Where J represents the loss function; mean(GSNR) represents the mean of the generalized signal-to-noise ratio; std(GSNR) represents the standard deviation of the generalized signal-to-noise ratio; and τ represents the weight coefficient.

[0081] The gain and gain slope of the optical fiber amplifier set after the stimulated Raman scattering effect inversion in step S102 are used as initial values. The gradient ascent algorithm is further used to iteratively optimize the gain and gain slope values ​​with the goal of improving the loss function J to obtain a theoretically optimal transmission power.

[0082] In some embodiments, the gradient ascent algorithm can be used to independently optimize each span of the fiber amplifier, or to jointly optimize multiple spans of the fiber amplifier.

[0083] In the independent optimization scenario, it is assumed that the effects between spans are uncoupled. The gain and gain slope of the fiber amplifier, set after inversion of the stimulated Raman scattering effect, are used as initial values. For each span, the current generalized signal-to-noise ratio is calculated, and a loss function is calculated based on the generalized signal-to-noise ratio. The gradient of the loss function with respect to the gain and gain slope of the fiber amplifier is calculated, and the gain and gain slope of each span are updated using a gradient ascent algorithm. Optimization stops when the loss function converges or the preset number of iterations is reached, resulting in the optimal gain and gain slope for each span.

[0084] Here, assuming that the effects between the spans are not coupled, such optimization is performed span by span. For example, the optimization space of each span is two amplifiers, and each amplifier has two optimizable parameters, namely gain and gain slope. Then, the optimization space of each span is only 4 parameter dimensions, which can achieve faster search optimization. However, since the optimization is performed span by span, it may not be guaranteed to reach the global optimum.

[0085] In the joint optimization scenario, the effects between spans are assumed to be coupled. The gain and gain slope of the fiber amplifier, set after inversion of the stimulated Raman scattering effect, are used as initial values. For all spans, the combined generalized signal-to-noise ratio is calculated, and a loss function is calculated based on the generalized signal-to-noise ratio. The gradient of the loss function with respect to the gain and gain slope of the fiber amplifier is calculated, and the gain and gain slope of all spans are updated using a gradient ascent algorithm. Optimization stops when the loss function converges or when the preset number of iterations is reached, resulting in the optimal gain and gain slope for all spans.

[0086] Among them, the effects between each span are coupled, so the optimization is performed jointly on all spans. For example, the optimization space of 10 span links is 40 parameter dimensions. The space is relatively large and the running speed will be relatively slow, but it can ensure global optimization.

[0087] In step S104, similar to step S103, based on the initial loss value preset when the wavelength selective switch leaves the factory, with the goal of achieving a high and flat generalized signal-to-noise ratio, that is, improving the loss function, the gradient ascent algorithm is used to iteratively fine-tune the channel loss value of the wavelength selective switch.

[0088] In some embodiments, the wavelength selective switch has the same initial loss value preset for each channel when it leaves the factory. For example, the initial loss value ranges from 15 to 20 dB.

[0089] In some embodiments, a gradient ascent algorithm is used to fine-tune the loss value of each channel of the wavelength selective switch, or a super-channel optimization method is used to simultaneously adjust the loss values ​​of multiple adjacent channels.

[0090] Specifically, in a channel-by-channel optimization scenario, the gradient of the loss function relative to the loss value of each channel of the wavelength selective switch is calculated; the loss value of each channel of the wavelength selective switch is updated using a gradient ascent algorithm; when the loss function converges or reaches a preset number of iterations, the optimization is stopped to obtain the optimal loss value of each channel of the wavelength selective switch.

[0091] In a superchannel optimization scenario, a preset number of adjacent channels in a wavelength selective switch are merged into a superchannel. For example, four adjacent channels are merged into a superchannel, with the channels within the superchannel sharing the same loss value. The gradient of the loss function relative to the superchannel's loss value is calculated, and the superchannel's loss value is updated using a gradient ascent algorithm. When the loss function converges or reaches a preset number of iterations, the optimization is stopped to obtain the optimal loss value for the superchannel. This achieves the technical effect of optimizing transmission quality while reducing algorithm complexity and optimization time.

[0092] From a technical implementation perspective, this step-by-step multi-device configuration optimization algorithm is relatively low in complexity. By configuring different devices in each step without requiring numerous complex iterations, it achieves superior transmission performance in a shorter timeframe than methods that configure only a single device type. Furthermore, a hyperchannel optimization method is employed to simultaneously fine-tune the loss values ​​of adjacent channels, significantly reducing the algorithm's runtime.

[0093] In summary, the C+L+S band multi-device configuration optimization method provided by the present invention first uses numerical methods in steps S101 to S102 to quickly determine the approximate optimal transmission power and approach the optimal point of the amplifier configuration. In step S103, a small number of gradient ascent iterations are used to finally approximate the optimal value. Traditional search algorithms need to find the optimal value in a huge search space. For an eight-span link transmitting in the C+L+S band, each span contains three amplifiers, and each amplifier has two optimization parameters: gain and gain slope. In this way, the link contains a total of 48 optimizable parameters, which greatly increases the computational complexity. The step-by-step multi-device configuration optimization algorithm proposed by the present invention first considers linear noise and Kerr nonlinear effects in a step-by-step decoupling manner, and then considers SRS. Finally, through a small number of iterations, the coupling relationship between various effects is compensated, greatly improving the operation speed.

[0094] The generalized signal-to-noise ratio obtained after optimization in steps S101 to S103 of the present invention was compared with that of the CMA-GA algorithm. The results were basically similar. In addition, the optimization time of the present invention was only 47.43 seconds, while the optimization time of the CMA-GA algorithm was 95.55 seconds, which is about twice the time required for the optimization of the present invention.

[0095] In addition, the present invention also considers the configuration of multiple devices. In the C+L+S transmission band, the traditional method only configures a single device, and it is often difficult to obtain the optimal solution for the generalized signal-to-noise ratio in terms of high mean and high flatness. The proposed step-by-step multi-device configuration optimization algorithm further configures a wavelength selective switch in step S104, which can further improve the flatness of the generalized signal-to-noise ratio and achieve higher transmission quality compared to traditional methods.

[0096] The present invention also provides a C+L+S band multi-device configuration optimization device, which is used to implement the steps of the C+L+S band multi-device configuration optimization method.

[0097] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the C+L+S band multi-device configuration optimization method.

[0098] Corresponding to the above method, the present invention also provides a device, which includes a computer device, the computer device includes a processor and a memory, the memory stores computer instructions, the processor is used to execute the computer instructions stored in the memory, and when the computer instructions are executed by the processor, the device implements the steps of the method described above.

[0099] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the aforementioned edge computing server deployment method. The computer-readable storage medium may be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the art.

[0100] It should be understood by those skilled in the art that the various exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is specifically performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0101] It should be understood that the present invention is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted. In the above embodiments, several specific steps are described and illustrated as examples. However, the method of the present invention is not limited to the specific steps described and illustrated. Those skilled in the art may make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present invention.

[0102] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or replace features of other embodiments.

[0103] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations to the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A C+L+S band multi-device configuration optimization method, characterized in that: The method comprises the following steps: The local optimal approximation global optimal algorithm is used to calculate the approximate optimal transmission power at the fiber entry point; The approximate optimal transmit power is used as the received power of each span; the transmit signal power at the beginning of the span is recalculated by inversely solving the stimulated Raman scattering effect power evolution equation, and the gain and gain slope of the optical fiber amplifier are set so that the signal power input at the beginning of each span approaches the transmit signal power; The generalized signal-to-noise ratios of the C, L, and S bands are calculated respectively to construct a loss function; the gain and gain slope of the optical fiber amplifier set after stimulated Raman scattering effect inversion are used as initial values, and a gradient ascent algorithm is used to iteratively optimize the gain and gain slope of the optical fiber amplifier with the goal of improving the loss function; Based on the initial loss value preset when the wavelength selective switch leaves the factory, the gradient ascent algorithm is used to iteratively fine-tune the channel loss value of the wavelength selective switch with the goal of improving the loss function.

2. The C+L+S band multi-device configuration optimization method according to claim 1, characterized in that: The local optimal approximation to the global optimal algorithm is used to calculate the approximate optimal transmission power at the fiber entry point, including: The local optimal approximation global optimal algorithm ignores the stimulated Raman scattering effect, and each span in the link can be optimized individually so that the optical performance of each span is optimized; by directly adding or superimposing the optimal optical performance of each span, the optimal optical performance of the entire multi-span link is approximately calculated.

3. The C+L+S band multi-device configuration optimization method according to claim 2, characterized in that: The calculation process includes: According to the local optimal approximation global optimal algorithm, the optical signal-to-noise ratio of any span is expressed as: Based on the local optimal approximation global optimal algorithm, the optical signal-to-noise ratio optimal nonlinear noise power spectrum can only obtain a suboptimal transmission power, which is expressed as: Among them, osnr n represents the optical signal-to-noise ratio of the nth span; G wDM,n Represents the nonlinear noise power spectrum of the optical signal-to-noise ratio; R s Indicates the symbol rate; a n Indicates the optical transmission attenuation of the nth span; B N represents the channel bandwidth; h represents the Planck constant; v represents the channel center frequency; F n represents the noise figure of the fiber amplifier in the nth span; ρ NLI,n represents the nonlinear crosstalk coefficient; P in represents the suboptimal transmit power; represents the optical signal-to-noise ratio optimal nonlinear noise power spectrum.

4. The C+L+S band multi-device configuration optimization method according to claim 3, characterized in that: In the optical signal-to-noise ratio formula: To optimize the optical signal-to-noise ratio (OSNR) nonlinear noise power spectrum, the calculation formula is: Where, ρ NLI,n represents the nonlinear crosstalk coefficient, which is calculated as follows: in, represents the optimal nonlinear noise power spectrum of the optical signal-to-noise ratio; h represents the Planck constant; v represents the channel center frequency; F h represents the noise figure of the fiber amplifier in the nth span; a n represents the light transmission attenuation of the nth span; α represents attenuation; γ represents the nonlinear coefficient; β represents the group velocity dispersion; L eff Indicates the effective length of optical fiber; B WDM Represents the entire transmission bandwidth.

5. The C+L+S band multi-device configuration optimization method according to claim 1, characterized in that: Calculate the generalized signal-to-noise ratio of the C, L, and S bands respectively to construct the loss function, including: The loss function includes the negative of the mean and standard deviation of the generalized signal-to-noise ratio, and is calculated as follows: J = mean(GSNR) - τstd(GSNR); Wherein, J represents the loss function; mean(GSNR) represents the mean value of the generalized signal-to-noise ratio; std(GSNR) represents the standard deviation of the generalized signal-to-noise ratio; and τ represents the weight coefficient.

6. The C+L+S band multi-device configuration optimization method according to claim 1, characterized in that: Using a gradient ascent algorithm, with the goal of improving the loss function, iteratively optimizing the gain and gain slope of the optical fiber amplifier includes: The gradient ascent algorithm is used to independently optimize the optical fiber amplifier of each span, or to jointly optimize the optical fiber amplifiers of multiple spans.

7. The C+L+S band multi-device configuration optimization method according to claim 6, characterized in that: The method further comprises: In an independent optimization scenario, it is assumed that the effects between the spans are not coupled; the gain and gain slope of the optical fiber amplifier set after the stimulated Raman scattering effect inversion are used as initial values; the gradient of the loss function relative to the gain and gain slope of the optical fiber amplifier is calculated; the gain and gain slope of each span are updated using the gradient ascent algorithm; when the loss function converges or reaches a preset number of iterations, the optimization is stopped to obtain the optimal gain and optimal gain slope of each span; In a joint optimization scenario, it is assumed that the effects between the spans are coupled; the gain and gain slope of the optical fiber amplifier set after inversion of the stimulated Raman scattering effect are used as initial values; the gradient of the loss function relative to the gain and gain slope of the optical fiber amplifier is calculated; the gain and gain slope of all spans are updated using the gradient ascent algorithm; when the loss function converges or reaches a preset number of iterations, the optimization is stopped to obtain the optimal gain and optimal gain slope of all spans.

8. The C+L+S band multi-device configuration optimization method according to claim 1, characterized in that: Based on an initial loss value preset when the wavelength selective switch leaves the factory, using the gradient ascent algorithm, with the goal of improving the loss function, iteratively fine-tuning the channel loss value of the wavelength selective switch includes: The gradient ascent algorithm is used to fine-tune the loss value of each channel of the wavelength selective switch, or the super-channel optimization method is used to simultaneously adjust the loss values ​​of multiple adjacent channels.

9. The C+L+S band multi-device configuration optimization method according to claim 8, characterized in that: The method further comprises: In a channel-by-channel optimization scenario, the gradient of the loss function relative to the loss value of each channel of the wavelength selective switch is calculated; the loss value of each channel of the wavelength selective switch is updated using the gradient ascent algorithm; when the loss function converges or reaches a preset number of iterations, the optimization is stopped to obtain the optimal loss value of each channel of the wavelength selective switch; In a super-channel optimization scenario, a preset number of adjacent channels in the wavelength selective switch are merged into a super-channel, and the gradient of the loss function relative to the loss value of the super-channel is calculated; the loss value of the super-channel is updated using the gradient ascent algorithm; when the loss function converges or reaches a preset number of iterations, the optimization is stopped to obtain the optimal loss value of the super-channel.

10. A C+L+S band multi-device configuration optimization device, characterized in that: The device is used to implement the steps of the method according to any one of claims 1 to 9.

Images