Joint optimization method of wavelength selection and power allocation in wide spectrum domain optical communication system

Abstract

The present application relates to a kind of wavelength selection and power allocation in wide spectrum domain optical communication system combined optimization method, belong to communication technical field.The method includes the following steps: step 1: initialization particle swarm population size, particle swarm dimension, inertia weight, learning factor and maximum iteration number;Step 2: randomly initialize the position and speed of each particle.Step 3: obtain initial individual historical optimal position, group historical optimal position, individual historical optimal fitness, group historical optimal fitness;Step 4, update the speed and position of particle or update the individual optimal fitness value and position of each particle and group historical optimal fitness value and position.Step 5: reach maximum iteration number, output optimal channel combination and power allocation result;Step 6: measure the performance of wavelength selection and power allocation method;The present application allocates power for each channel separately, realizes the maximization of link capacity.

Description

Technical Field

[0001] This invention belongs to the field of communication technology and relates to a joint optimization method for wavelength selection and power allocation in a broadband optical communication system. Background Technology

[0002] With the widespread adoption of 5G / F5G services and internet video applications, data traffic in fiber optic communication networks is growing exponentially. To meet this rapidly increasing traffic demand, feasible measures must be taken to improve channel capacity. In the past, spectral efficiency was primarily improved through higher-order modulation and constellation shaping. However, the transmission capacity of single-mode fiber (SMF) channels is limited by the Shannon limit. Currently, various solutions have been discussed, including spatial division multiplexing (SDM) and utilizing multiple low-loss spectral bands (L, O, E, S, and U bands) of SMF. SDM can be implemented using multiple fibers (MFs) or multi-core fibers (MCF). MFs require the installation of new fibers or the activation of unused fibers, both of which are expensive and time-consuming processes. MCF technology requires the installation of new fibers and the deployment of potentially complex multiple-input multiple-output (MIMO) transceivers, which are currently rarely commercially available. Therefore, extending the operating band of existing SMFs beyond the C-band is considered one of the feasible solutions to address capacity bottlenecks in the near term. This solution maximizes the return on investment of deployed fiber optic infrastructure.

[0003] However, nonlinear effects present in fiber optic transmission limit the capacity of broadband communication systems. These include the optical Kerr effect and stimulated Raman scattering (SRS) between channels. The Kerr effect includes self-phase modulation (SPM), cross-phase modulation (XPM), and four-wave mixing (FWM). SRS causes power to shift from high frequencies to low frequencies, resulting in significant differences in performance between each WDM channel, which become more pronounced as fiber bandwidth increases. Therefore, the commonly used uniform transmit power distribution (i.e., all WDM channels have the same power) ultimately fails to provide optimal system performance. Non-uniform power allocation assuming all channels are occupied will also provide suboptimal system performance due to overestimating channel interference. Therefore, considering only the number of channels used, finding the channel combination and appropriate transmit power distribution that maximizes overall system performance is crucial for improving system throughput. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a joint optimization method for wavelength selection and power allocation in a broadband optical communication system.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A joint optimization method for wavelength selection and power allocation in broadband optical communication systems, comprising the following steps:

[0007] S1: Initialize the particle swarm population size N, particle swarm dimension D, inertia weight w, learning factors c1 and c2, and maximum number of iterations G;

[0008] S2: Randomly initialize the position and velocity of each particle; each particle k is associated with two vectors, namely the velocity vector. and position vector X k =[A k B k ]; where 2D represents the dimension of the solution space, This represents a binary discrete variable combination, indicating a feasible solution for the optimal channel combination. It is a combination of continuous variables, representing a feasible solution for optimal power allocation; the velocity and position of each particle are initialized by a random vector within the corresponding range;

[0009] S3: Calculate fitness values ​​using the initial positions, and obtain the initial individual historical best position, the group historical best position, the individual historical best fitness value, and the group historical best fitness value;

[0010] S4: If the number of iterations has not reached the maximum value, update the velocity and position of the particles; calculate the fitness value of each particle, update the individual best fitness value and position of each particle as well as the historical best fitness value and position of the population; then repeat S4; if the number of iterations has reached the maximum value, execute S5;

[0011] S5: Reach the maximum number of iterations and output the optimal channel combination and power allocation results.

[0012] Optionally, S3 specifically includes:

[0013] The transmit power of each channel and the selection of channel combinations are optimized by maximizing the capacity of the point-to-point fiber optic link, so as to maximize the total capacity of the selected channels; the optimization objective is shown in formula (1):

[0014]

[0015] The function C in formula (2) defines the Shannon information rate of the k-th channel, which defines the theoretical upper limit of the capacity of each channel;

[0016]

[0017] The constraints are as follows:

[0018]

[0019] P min <P(f i ) < P max (4)

[0020] Formula (3) represents the total number of channels used in the link, where δ i This is a binary decision variable; its value is 1 if the i-th channel is used, and 0 otherwise. The number of channels used is n, and the total number of channels in the link is N. ch Formula (4) represents the value space of the optimized power; SNR[P(f] in Formula (2) i [)] represents the signal-to-noise ratio of channel i;

[0021] To more accurately estimate the performance of broadband optical communication systems, the wavelength dependence of single-mode fiber parameters needs to be considered; specifically as follows:

[0022] Multi-band SMF spectral parameter modeling; Fiber loss is one of the most detrimental effects in fiber optic communication systems. After optical signals are transmitted through optical fibers, they are subjected to an OH loss. -1 The limitation of the ion peak is accurately fitted by the superposition of four Lorentz functions and one Gaussian function; the fiber attenuation coefficient α [dB / km] is approximately modeled as shown in formula (5):

[0023]

[0024] In the formula, A is the Rayleigh scattering loss, Λ is the Rayleigh scattering wavelength; C and D are absorption coefficients, representing the material properties; λ represents the wavelength; and peaks(λ) is the OH value. -1 absorption peak;

[0025] The SMF dispersive spectrum D and the dispersive slope S are effectively modeled using four Sellmeier fitting functions as shown in the following formulas (6)(7):

[0026]

[0027]

[0028] B1, B2, B3, and B4 are suitable sets of Sellmeier coefficients;

[0029] The fiber nonlinearity coefficient γ is the main indicator for measuring the Kerr nonlinearity of fiber, and its wavelength dependence is shown in formula (8):

[0030]

[0031] Where n2 is the nonlinear refractive index, a measure of the optical Kerr effect in silicon dioxide media, and A eff The effective fiber mode area; in the C-band, A at a center wavelength of 1550nm. eff ≈80μm 2 A eff The wavelength dependence of the effective mode area of ​​the fiber varies with different fiber designs. Assuming that the transverse electric field amplitude component has a Gaussian shape, the wavelength dependence of the effective mode area of ​​the fiber is given by the following empirical expression formula (9):

[0032]

[0033]

[0034] a is the fiber radius, NA is the numerical aperture, which is the maximum angle measured outside the fiber relative to the longitudinal direction of the incident beam, which allows the light to be confined within the fiber; the V value defines the number of modes supported by the fiber; to support single modes, the V value should not exceed 2.405; the calculation of V is shown in formula (10).

[0035] Optionally, the SNR[P(f i The signal-to-noise ratio (SNR) was calculated using a stimulated Raman scattering (SNR) model, as follows:

[0036] The performance of a broadband optical communication system is evaluated by introducing an effective received signal-to-noise ratio, which is decomposed into the contributions of linear spontaneous emission (ASE) noise and nonlinear interference (NLI) as shown in Equation (11):

[0037]

[0038] Where P i Let P be the transmit signal power of the i-th WDM channel. ASE,i ASE noise introduced by the optical amplifier; η(f i ) is in f i NLI coefficients at; η(f) i )≈η SPM (f i )+η XPM (f i )

[0039] The model for self-channel interference is constructed as Equation (12):

[0040]

[0041] in B i Channel bandwidth; β2 is the group velocity dispersion, and β3 is the group velocity dispersion slope;

[0042] The mutual channel interference model is constructed as shown in equation (13):

[0043]

[0044] Where φ i,m =2π(f m -f i )[β2+πβ3(f i +f m )],N ch This represents the total number of channels in the link.

[0045] The ASE noise model introduced by the amplifier is constructed as Equation (14):

[0046]

[0047] in, Here, f is the spontaneous emission factor, NF is the noise figure (NF = 4.5), h is Planck's constant, and f is the noise coefficient. c f is the reference frequency. i G represents the frequency interval of the i-th channel relative to the reference frequency. EDFA,iThe amplifier gain is calculated as shown in equation (11):

[0048]

[0049] Where L s G is the span length of the link. SRS (f i As shown in the following formula (12):

[0050]

[0051] g is the Raman gain slope, A e The effective area of ​​the optical fiber. The effective length of the span. To calculate the sum of the powers of the interacting channels using the triangular approximation within the SRS gain bandwidth of 15 THz; P m Let be the power of the m-th channel.

[0052] Optionally, S4 specifically includes:

[0053] If the number of iterations does not reach the maximum value, the power vector B k Update the particle's velocity and position according to formulas (17) and (18); channel combination vector A k Update the velocity and position of the particles according to formulas (17) and (19), calculate the fitness value of each particle, update the individual optimal fitness value and position of each particle as well as the group's historical optimal fitness value and position; then repeat S3; if the number of iterations reaches the maximum value, execute S4;

[0054]

[0055]

[0056]

[0057] In equation (17), ω is the inertial weight, and c1 and c2 are acceleration coefficients. and Let pBest be two uniformly distributed random numbers generated independently in the range [0,1] with dimension dth; in equation (17), pBest k It is the best fitness position found by the k-th particle so far, and gBest is the best position in the neighborhood; the next position of the k-th particle is obtained by equation (18). Formula (19) is a probability mapping method. The sigmoid function is used to map the velocity to the [0,1] interval as a probability, which is the probability that the particle will take a value of 1 in the next step. A random variable is generated; if the random variable is less than... but It is 1 if it is not 1, otherwise it is 0.

[0058] The beneficial effects of this invention are as follows: compared with traditional uniform power allocation schemes and non-uniform power allocation methods considering full load conditions, this method improves the overall link capacity by selecting channel combinations and allocating power. The innovation of this invention mainly lies in the parameter initialization and particle swarm position and velocity initialization completed in S1 and S2. S3 calculates the initial individual current optimal value and the current global optimal value. Through continuous iteration in S4, the position and velocity of the particles are updated to obtain the individual particle current optimal value and the current global optimal value. After S4 iterates to the maximum number of times, the optimal solution is output, obtaining the optimal channel combination and the corresponding optimal channel power. Ultimately, this maximizes the information throughput of the broadband optical fiber communication system.

[0059] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0060] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:

[0061] Figure 1 This is a flowchart of the algorithm for the joint optimization method of wavelength selection and power allocation in a broadband optical communication system proposed in this invention.

[0062] Figure 2 This is a schematic diagram of the multi-span link constructed in this invention;

[0063] Figure 3 A comparison chart of total link capacity constructed for this invention;

[0064] Figure 4 This is a comparison chart of the total signal-to-noise ratio of the links constructed in this invention;

[0065] Figure 5 This is a comparison chart of total capacity across link spans constructed in this invention;

[0066] Figure 6 An optimized transmit power distribution map constructed for this invention;

[0067] Figure 7 The channel signal-to-noise ratio distribution diagram constructed for this invention;

[0068] Figure 8 The channel capacity distribution map constructed for this invention. Detailed Implementation

[0069] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0070] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual pictures. They should not be construed as limiting the invention. To better illustrate the embodiments of the invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0071] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," "front," and "rear" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present invention. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0072] The flowchart of the joint optimization method for wavelength selection and power allocation in a broadband optical communication system is as follows: Figure 1 As shown, it includes the following steps:

[0073] S1: Initialize the particle swarm population size N, particle swarm dimension D, inertia weight w, learning factors c1 and c2, and maximum number of iterations G;

[0074] S2: Randomly initialize the position and velocity of each particle. Each particle k is associated with two vectors: a velocity vector and a velocity vector. and position vector X k =[A k B k ]. Where 2D represents the dimension of the solution space, This represents a binary discrete variable combination, indicating a feasible solution for the optimal channel combination. This is a combination of continuous variables, representing a feasible solution for optimal power allocation. The velocity and position of each particle are initialized by a random vector within a corresponding range.

[0075] S3: Calculate the fitness value using the initial position to obtain the initial individual historical best position, the group historical best position, the individual historical best fitness value, and the group historical best fitness value; the fitness calculation is related to the optimization objective of this invention. Here, the transmit power of each channel and the selection of channel combinations are optimized by maximizing the capacity of the point-to-point fiber optic link, so that the total capacity of the selected channels is maximized. The optimization objective is shown in formula (1):

[0076]

[0077] The function C in formula (2) defines the Shannon information rate of the k-th channel, which defines the theoretical upper limit of the capacity of each channel.

[0078]

[0079] The constraints are as follows:

[0080]

[0081] P min <P(f i ) < P max (4)

[0082] Formula (3) represents the total number of channels used in the link, where δ i This is a binary decision variable; its value is 1 if the i-th channel is used, and 0 otherwise. The number of channels used is n, and the total number of channels in the link is N. ch Formula (4) represents the value space of the optimized power. In formula (2), SNR[P(f i [] represents the signal-to-noise ratio (SNR) of channel i. This calculation requires the use of an SNR model that considers stimulated Raman scattering. Furthermore, to more accurately estimate the performance of a broadband optical communication system, the wavelength dependence of single-mode fiber parameters needs to be considered. The specific steps are as follows:

[0083] S3.1: Spectral parameter modeling of multi-band SMFs. Fiber loss is one of the most detrimental effects in fiber optic communication systems. After optical signals are transmitted through optical fibers, the optical power decreases due to Rayleigh scattering and infrared absorption. In addition, it is also affected by an OH... -1 The limitation of the ion peak can be accurately fitted by the superposition of four Lorentz functions and one Gaussian function. The fiber attenuation coefficient α [dB / km] can be approximately modeled as shown in formula (5):

[0084]

[0085] In the formula, A is the Rayleigh scattering loss, Λ is the Rayleigh scattering wavelength; C and D are absorption coefficients, representing the material properties; λ represents the wavelength; and peaks(λ) is the OH value. -1 absorption peak;

[0086] The SMF dispersive spectrum D and the dispersive slope S can be effectively modeled using four Sellmeier fitting functions as shown in the following formulas (6) and (7):

[0087]

[0088]

[0089] B1, B2, B3, and B4 are suitable sets of Sellmeier coefficients.

[0090] The fiber nonlinearity coefficient γ is the main indicator for measuring the Kerr nonlinearity of fiber, and its wavelength dependence is shown in formula (8):

[0091]

[0092] Where n2 is the nonlinear (Kerr) refractive index, a measure of the optical Kerr effect in silicon dioxide media, A eff This refers to the effective fiber mode area. Typically, in the C-band, the A-band is located at a center wavelength of 1550 nm. eff ≈80μm 2 A eff The wavelength dependence of the effective mode area of ​​the fiber will vary depending on the fiber design. Assuming that the transverse electric field amplitude component has a Gaussian shape, the wavelength dependence of the effective mode area of ​​the fiber can be given by the following empirical expression formula (9):

[0093]

[0094]

[0095] 'a' represents the fiber radius, and NA is the numerical aperture, which is the maximum angle measured outside the fiber relative to the longitudinal direction of the incident beam, allowing the light to be confined within the fiber. The V value defines the number of modes supported by the fiber. To support single-mode operation, the V value should not exceed 2.405. V is calculated as shown in formula (10). The constant parameter values ​​involved are shown in Table 1.

[0096] Table 1

[0097]

[0098] S3.2: Multi-band system performance modeling. The performance of a broadband optical communication system is evaluated by introducing the effective received signal-to-noise ratio (SNR), which can be decomposed into the contributions of linear amplified spontaneous emission (ASE) noise and nonlinear interference (NLI) as shown in Equation (11):

[0099]

[0100] Where P i Let P be the transmit signal power of the i-th WDM channel. ASE,i ASE noise introduced by the optical amplifier. η(f) i ) is in f i The NLI coefficient at η(f) i )≈η SPM (f i )+η XPM (f i ).

[0101] The model for self-channel interference is constructed as Equation (12):

[0102]

[0103] in B i This refers to the channel bandwidth. Under normal circumstances, it can be considered... β2 is the group velocity dispersion, and β3 is the group velocity dispersion slope.

[0104] The mutual channel interference model is constructed as shown in equation (13):

[0105]

[0106] Where φ i,m =2π(f m -f i )[β2+πβ3(f i +f m )],N ch This represents the total number of channels in the link.

[0107] The ASE noise model introduced by the amplifier is constructed as Equation (14):

[0108] P ASE,i =2(G EDFA,i -1)n sp h(f c +f i B i (14)

[0109] in, Here, f is the spontaneous emission factor, NF is the noise figure (NF = 4.5), h is Planck's constant, and f is the noise coefficient. c For reference frequency, f is used in the example. c =1576nm. f i G represents the frequency interval of the i-th channel relative to the reference frequency. EDFA,i The amplifier gain is calculated as shown in equation (11):

[0110]

[0111] Where L s This represents the span length of the link, set to L in the example. s =100km, G SRS (f i As shown in the following formula (12):

[0112]

[0113] g is the Raman gain slope, A e The effective area of ​​the optical fiber. The effective length of the span. To calculate the sum of the powers of the interacting channels using the triangular approximation within the SRS gain bandwidth (15THz). m Let be the power of the m-th channel.

[0114] S4: If the number of iterations has not reached the maximum value, the power vector B... k Update the particle's velocity and position according to formulas (17) and (18). Channel combination vector A k Update the particle velocity and position according to formulas (17) and (19), calculate the fitness value of each particle, and update the individual optimal fitness value and position of each particle, as well as the historical optimal fitness value and position of the population. Then repeat S3. If the number of iterations reaches the maximum value, execute S4;

[0115]

[0116]

[0117]

[0118] In equation (17), ω is the inertial weight, and c1 and c2 are acceleration coefficients. and Let pBest be two uniformly distributed random numbers generated independently in the range [0,1] with dimension dth. In equation (17), pBest kThe position with the best fitness found so far by the k-th particle is gBest, which is the best position in the neighborhood. The next position of the k-th particle is obtained through equation (18). Formula (19) is a probability mapping method, using the sigmoid function to map the velocity to the interval [0,1] as a probability. This probability is the probability that the particle will take a value of 1 in the next step. A random variable is generated; if the random variable is less than... but It is 1 if it is not 1, otherwise it is 0.

[0119] S5: Reach the maximum number of iterations and output the optimal channel combination and power allocation results;

[0120] S6: The performance of wavelength selection and power allocation methods is measured by metrics such as total throughput and total signal-to-noise ratio.

[0121]

[0122]

[0123] Formula (21) is used to calculate the total throughput of the entire link, R s Representing the symbol rate, formula (22) is used to solve for the total signal-to-noise ratio (SNR) of the entire link. i This represents the signal-to-noise ratio of the i-th channel.

[0124] The schematic diagram of the link used in the examples of this invention is as follows: Figure 2 As shown, within the framework of the analytical method of this invention, the spectral gaps between S, C, and L are omitted. Without loss of generality, this invention considers an ideal 100 GBd polarization-multiplexed Nyquist-spaced WDM fiber transmission system at a modulation bandwidth of up to 21 THz at the carrier wavelength corresponding to minimum fiber loss. It is assumed that all fiber spans and erbium-doped fiber amplifiers (EDFAs) are identical, with a length Ls = 100 km and an EDFA noise figure NF = 4.5 dB. The analytical method of this invention can be directly generalized by adding practical constraints and defects to real-world amplification schemes, such as considering residual transceiver distortion, segmented amplifier noise figure spectrum, amplifier gain spectrum slope, gain fluctuation, spectral gaps, etc. In the example scenario where the span remains constant, the span number is 12, the spectrum covers 210 channels in the S, C, and L bands, the particle swarm optimization algorithm is set to 100 particles, the number of iterations is 40,000, the initial power range is [-15dBm, 15dBm], the initial particle velocity is set to [-6dBm, 6dBm], w = 1.1, c1 = 2, c2 = 2.

[0125] Figure 3The graph shows the total throughput distribution of the link after numerical optimization using different numbers of channels. The horizontal axis represents the channel numbers of the 210 channels, with channel 0 corresponding to a wavelength of 1457.7 nm. This invention, the Joint method, compares different optimization strategies, including random power optimization, which randomly selects a channel combination and optimizes the power of the corresponding channels in the selected combination (the Random method in the figure); non-uniform power distribution, which adjusts the power of each individual channel in the entire link under the condition of full channel occupancy, and then selects channels from the channel combinations obtained by this method to obtain information such as the power, signal-to-noise ratio, and channel capacity of the corresponding channels (the Non-uniform method in the figure); and uniform flat power optimization, which assumes that each channel has the same power across the entire bandwidth and similarly selects channels from the channel combinations obtained by this method (the Uniform method in the figure). With 80 channels in use, the method proposed in this invention achieves approximately 18% higher total link throughput than the uniform power optimization method, 4.6% higher than the non-uniform power optimization method, and 1.3% higher than the random power optimization method. With 120 channels in use, the method proposed in this invention achieves approximately 19.8% higher total link throughput than the uniform power optimization method, 3.8% higher than the non-uniform power optimization method, and 1.26% higher than the random power optimization method. As the number of channels in use increases, the advantages of the method proposed in this invention diminish, but it still outperforms other comparative algorithms. Figure 4 The graph shows the total signal-to-noise ratio distribution of links after numerical optimization using different numbers of channels, and the trend is the same as that of the total throughput distribution.

[0126] Figure 5 This chart shows the total capacity of each method for different link spans when using 80 channels. When the span is 3, the proposed method achieves approximately 15.3% higher total throughput than the Uniform method, 3.8% higher than the Non-uniform method, and 1.8% higher than the Random method. When the span is 6, the proposed method achieves approximately 21.8% higher total throughput than the Uniform method, 7.6% higher than the Non-uniform method, and 1.9% higher than the Random method. When the span is 9, the proposed method achieves approximately 14.8% higher total throughput than the Uniform method, 4.6% higher than the Non-uniform method, and 2.2% higher than the Random method.

[0127] Figure 6 , Figure 7 and Figure 8The paper presents a series of numerically optimized transmit power, capacity, and signal-to-noise ratio (SNR) distributions per channel using different numbers of channels. These power distributions maximize system throughput. It is worth noting that the Uniform method significantly reduces the multidimensional optimization problem to a one-dimensional problem, greatly reducing computational costs and thus saving overall computation time. However, as bandwidth increases, the SRS effect causes this method to perform less effectively. The channel capacity and SNR achieved using the Non-uniform method are also lower than those proposed in this invention.

[0128] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A joint optimization method for wavelength selection and power allocation in a broadband optical communication system, characterized by: This method Includes the following steps: S1: Initialize particle swarm population size Particle swarm dimension Inertial weight Learning factors and Maximum number of iterations ; S2: Randomly initialize the position and velocity of each particle; each particle It is associated with two vectors, namely the velocity vector. and position vector ;in This represents the dimension of the solution space. This represents a binary discrete variable combination, indicating a feasible solution for the optimal channel combination. For a continuous combination of variables, it represents a feasible solution for optimal power allocation; The velocity and position of each particle are initialized by a random vector within a corresponding range; S3: Calculate fitness values ​​using the initialized positions to obtain the initial individual historical best position, the group's historical best position, the individual's optimal fitness value, and the group's historical best fitness value; S3 specifically includes: The transmit power of each channel is optimized and the channel combination is selected by maximizing the capacity of the point-to-point fiber optic link, so that the total capacity of the selected channels is maximized. The optimization objective is shown in formula (1): (1) The functional C of formula (2) is defined as follows: k The Shannon information rate of the channel, which defines the theoretical upper limit of the capacity of each channel; (2) The constraints are as follows: (3) (4) Formula (3) represents the total number of channels used in the link, where Belongs to binary decision variables, if the first i One channel is in use, with a value of 1 otherwise; the number of channels in use is... n The total number of channels in the link is Formula (4) represents the value space of the optimized power; in Formula (2) Indicates channel i Signal-to-noise ratio; To more accurately estimate the performance of broadband optical communication systems, the wavelength dependence of single-mode fiber parameters needs to be considered; specifically as follows: Multi-band SMF spectral parameter modeling; Fiber loss is one of the most detrimental effects in fiber optic communication systems. After optical signals are transmitted through optical fibers, they are subjected to a... The limitation of the ion peak was precisely fitted by the superposition of four Lorentz functions and one Gaussian function; fiber attenuation coefficient The approximate modeling is shown in Equation (5): (5) In the formula, A For Rayleigh scattering loss, This is the Rayleigh scattering wavelength; C and D The absorption coefficient is... This represents the wavelength. for absorption peak; SMF dispersive spectrum D and dispersion slope S Effective modeling using four Sellmeier fitting functions is achieved through the following formulas (6) and (7): (6) (7) in It is a suitable set of Sellmeier coefficients; The fiber nonlinearity coefficient γ is the main indicator for measuring the Kerr nonlinearity of fiber, and its wavelength dependence is shown in formula (8): (8) in It is a nonlinear refractive index, a measure of the optical Kerr effect in silicon dioxide media. The effective fiber mode area; in the C-band, at a center wavelength of 1550 nm. , The wavelength dependence of the effective mode area of ​​the fiber varies with different fiber designs. Assuming that the transverse electric field amplitude component has a Gaussian shape, the wavelength dependence of the effective mode area of ​​the fiber is given by the following empirical expression formula (9): (9) (10) For fiber radius, It is the numerical aperture; The value defines the number of modes supported by the optical fiber; to support single-mode, The value should not exceed 2.405; The calculation is shown in formula (10); S4: If the number of iterations has not reached the maximum value, update the velocity and position of the particles; calculate the fitness value of each particle, update the individual best fitness value and position of each particle as well as the historical best fitness value and position of the population; then repeat S4; if the number of iterations has reached the maximum value, execute S5; S5: Reach the maximum number of iterations and output the optimal channel combination and power allocation results.

2. The joint optimization method for wavelength selection and power allocation in a broadband optical communication system according to claim 1, characterized in that: The The signal-to-noise ratio (SNR) is calculated using a stimulated Raman scattering (SNR) model, as follows: The performance of a broadband optical communication system is evaluated by introducing an effective received signal-to-noise ratio, which is decomposed into linear self-emission noise and nonlinear noise interference contributions as shown in Equation (11): (11) in Let be the transmit signal power of the i-th WDM channel. ASE noise introduced into the optical amplifier; Is NLI coefficients at the location; The model for self-channel interference is constructed as Equation (12): (12) in , , , Channel bandwidth; , For group velocity dispersion, The group velocity dispersion slope; The mutual channel interference model is constructed as shown in equation (13): (13) in , This represents the total number of channels in the link. The ASE noise model introduced by the amplifier is constructed as Equation (14): (14) in, Spontaneous emission factor, Noise figure , h It is Planck's constant. For reference frequency; For the first i The frequency spacing of the channel relative to the reference frequency; The amplifier gain is calculated as shown in equation (15): (15) in This represents the span length of the link. As shown in the following formula (16): (16) The slope of the Raman gain. The effective area of ​​the optical fiber. The effective length of the span. To calculate the sum of the power of the interacting channels using the triangular approximation within the SRS gain bandwidth of 15 THz; For the first m Power of each channel.

3. The joint optimization method for wavelength selection and power allocation in a broadband optical communication system according to claim 1, characterized in that: S4 specifically includes: If the number of iterations has not reached the maximum value, the power vector Update the particle velocity and position according to formulas (17) and (18); channel combination vector Update the velocity and position of the particles according to formulas (17) and (19), calculate the fitness value of each particle, update the individual optimal fitness value and position of each particle as well as the group's historical optimal fitness value and position; then repeat S3; if the number of iterations reaches the maximum value, execute S4; (17) (18) (19) In equation (17), w For inertial weights, and For acceleration coefficient, and for d Two uniformly distributed random numbers are independently generated in the range [0,1]; in equation (17), It is the first time so far k The position with the best fitness found by each particle. It is the optimal position within the neighborhood; the first position is obtained through equation (18). The next position of each particle Formula (19) is a probability mapping method, using the sigmoid function to map the velocity to the [0,1] interval as a probability. This probability is the probability that the particle will take the value 1 in the next step. A random variable is generated. If the random variable is less than 1, the probability is calculated as follows: ,but It is 1 if it is not 1, otherwise it is 0.

Images