A genetic algorithm optimizes resource allocation of static service transmission in elastic optical network

Abstract

The application relates to a genetic algorithm optimization elastic optical network static service transmission resource allocation method, and belongs to the technical field of fiber communication. The method is used for solving the resource allocation problem of transmitting static services in an elastic optical network, and an optimization method for optimizing the resource allocation during service transmission is designed. First, a genetic algorithm is used to optimize the sorting of multiple static services, the service set is sorted in descending order according to the request rate of the services, the ancestor individual is generated by gene crossover of the ancestor individual, a variety of service sorting populations are generated, and the individuals in the population are sequentially subjected to resource allocation; in the resource allocation, the network spectrum index value occupied by the individual is designed as the fitness function; the core pressure value is designed to select the core and route of each service in the individual, and the bit loading mechanism is combined to select the spectrum block with the minimum frequency slot index value meeting the rate requirement for the service. The designed method can reduce the spectrum occupation degree of the network and improve the spectrum utilization rate.

Description

Technical Field

[0001] This invention belongs to the field of resource allocation technology for optical fiber communication and optical transmission, and relates to a resource allocation method for optimizing static service transmission in elastic optical networks using a genetic algorithm. Background Technology

[0002] With the rapid development of emerging technologies such as cloud computing, big data, the Internet of Things, and artificial intelligence, various new applications requiring high bandwidth and low latency are emerging in an endless stream, posing unprecedented challenges to the transmission capacity and service quality of communication networks. Traditional Wavelength Division Multiplexing (WDM) networks use a fixed spectrum grid allocation mechanism, resulting in coarse spectrum resource allocation that is difficult to adapt to the dynamic needs of diverse services, leading to low spectrum utilization and serious resource waste. Against this backdrop, Elastic Optical Networks (EONs) have emerged. Their fine-grained spectrum allocation mechanism can accurately allocate spectrum resources according to service needs, significantly improving network resource utilization efficiency, and has become the mainstream architecture and key technology for next-generation optical transmission networks.

[0003] To further break through the transmission capacity limit of a single optical fiber, multi-core fiber (MCF) technology integrates multiple independent fiber cores within a common cladding, enabling parallel transmission in the spatial dimension and opening up new avenues for the continuous growth of optical network capacity. Introducing multi-core fibers into elastic optical networks (ECNs) to form multi-core fiber elastic optical networks (MCF-EONs) can exponentially increase network transmission capacity, meeting the future demands for ultra-high-speed and ultra-high-capacity service transmission. However, the introduction of multi-core fibers also brings new technical challenges such as inter-core crosstalk and spectrum fragmentation, making network resource management more complex.

[0004] Bit loading is a key modulation technique in resilient optical networks. Based on the channel quality of each frequency slot (FS), different frequency slots are allocated spectrum blocks with different modulation levels for each service. This technique can further improve the utilization rate of spectrum resources. After introducing bit loading technology, the service capacity of resilient optical networks is significantly improved, achieving efficient utilization of spectrum resources while ensuring transmission reliability.

[0005] In recent years, intelligent optimization algorithms have been increasingly widely used in the field of communication networks. Inspired by mechanisms such as biological evolution and swarm intelligence in nature, these algorithms possess powerful global search capabilities and adaptability, effectively addressing optimization problems under complex constraints. Genetic Algorithms (GA) and Ant Colony Optimization (ACO) have shown promising application prospects in areas such as service routing calculation, spectrum allocation, and virtual network mapping. Introducing intelligent optimization algorithms into resource allocation methods for elastic optical networks (ECNs) using multi-core optical fibers not only overcomes the limitations of high complexity and low resource allocation efficiency of traditional methods, but also adaptively adjusts resource allocation strategies for batches of services within a time window based on changes in network resources and the dynamic state of numerous services arriving and departing within a defined time period, significantly improving the service performance of MCF-EONs. Summary of the Invention

[0006] In view of this, the purpose of this invention is to provide a resource allocation method for optimizing static service transmission in elastic optical networks using a genetic algorithm, which can improve the transmission performance of a large number of services within a given time window and reduce the service blocking rate.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] For multi-core fiber optic flexible optical network scenarios, a routing, modulation, spectrum, and core allocation (RMSCA) algorithm based on a genetic algorithm is proposed. In the static service request sorting stage arriving in batches within a given time window, the genetic algorithm optimizes the sorting of static services. Static services are initially sorted in descending order of request rate. Multiple individuals are generated from these individuals through random crossover of gene loci, forming a population. A core pressure function is designed to optimize the path and core selection process for each service within the individuals. A first-choice method is used to allocate idle spectrum blocks to services. The fitness function value of each individual is calculated, and the individual with the highest fitness value is selected for random crossover of gene loci to generate the population. After iteration, the individual with the highest fitness function value indicates that the service has successfully transmitted and occupied the fewest network slots, which is beneficial for improving resource efficiency and reducing the probability of service congestion. The method specifically includes the following steps:

[0009] S1: Set up service requests in the elastic optical network Request in Based on the rate of business requests The sorting is performed in descending order, which serves as the ancestor of the genetic algorithm. Each position of the individual is a gene position, and each gene corresponds to a business in the individual. Let x be the number of consecutive low-yield iterations, and set x=0; X is the upper limit of the number of consecutive low-yield iterations, Y is the threshold of the fitness function, I is the number of individuals in the population, Z is the upper limit of the number of iterations, and n is the number of iterations, which is set to n=0; where the source node, destination node, and rate information of each business request in R are known prior; an individual is a sorting result.

[0010] S2: By performing random crossover operations on the gene positions of ancestral individuals, new genes are generated. Each individual, including ancestral individuals and newly generated individuals A population of I individuals is formed by I individuals, where the j-th individual is denoted as . ;

[0011] S3: For each service in each individual in the population, execute the routing, fiber core, and spectrum resource allocation algorithms in turn;

[0012] S4: Calculate all individuals The fitness function value of I individuals is denoted as y. n ;like ,but Otherwise, let ;like or Then the output fitness function value is y. n The transmission resource allocation result of the service prioritization scheme corresponding to the individual; otherwise, the fitness function value is set to y. n The individual as the ancestral individual, let Proceed to step S2.

[0013] Furthermore, the specific process of step S1 is as follows:

[0014] S101: The elastic optical network is abstracted as an undirected weighted graph G(V, L, C, F), where V represents the set of nodes in the elastic optical network; L represents the set of optical fiber links between nodes in the elastic optical network, and all optical fiber links are multi-core fibers; C is the set of fiber cores of a single optical fiber link; F is the set of occupancy status representations of the frequency slots within the fiber core. If a frequency slot is occupied, the frequency slot status is 1; otherwise, the frequency slot status is 0.

[0015] S102: The business request set is defined as follows: Where N is the total number of business requests contained in the business request set. For the i-th business request, the business request is modeled as a triple. Where s and d are the source node and destination node of the service request, respectively, and b iThe request rate for the service is expressed in Gbps.

[0016] S103: The objective of this application is to optimize the individual's spectrum occupancy, which is defined as:

[0017] ;

[0018] In the formula, It is a binary variable, (e i , e j ) represents node e in the elastic optical network i To node e j The fiber optic link, if the fiber optic link (e i , e j If the k-th frequency slot of fiber core c is occupied, and its frequency slot index is the maximum value of the occupied frequency slot index in the current fiber core, then... If it is 1, otherwise, =0;

[0019] S104: For a set of business requests containing N business requests The j-th individual is defined as:

[0020] ;

[0021] In the formula, N is the total number of requests in the business request set. This is the index number of the sorted business, and Each Each individual in this application corresponds to a gene on a chromosome of an individual. Each individual in this application corresponds to a sorting scheme of a business request in R. An individual contains one chromosome, and each chromosome contains N gene positions.

[0022] S105: The fitness function of an individual is defined as:

[0023] ;

[0024] The design of this fitness function is related to the fitness function of the genetic algorithm. The fitness function value is negatively correlated with the spectrum occupancy degree SO. The lower the spectrum occupancy degree, the better the resource allocation performance of the sorting scheme corresponding to the individual. The smaller the maximum value of the frequency slot index of the MCF-EONs, the larger the fitness function value of the corresponding resource allocation scheme.

[0025] S106: The growth rate of the fitness function value among individuals is defined as follows:

[0026] ;

[0027] In the above formula, The fitness function value growth rate is given by n, where n is the number of iterations for each individual, and y is the number of iterations for each individual. n The fitness function value of an individual in the nth iteration is represented by the formula above. Calculate; y n-1 This represents the fitness function value of an individual in the (n-1)th iteration.

[0028] S107: The inter-core crosstalk intensity experienced by the f-th frequency slot in fiber core c of fiber link l. Defined as:

[0029] ;

[0030] In the formula, A c Let be the set of adjacent fiber cores of fiber core c; It is a Boolean variable, when the fiber core frequency gap When occupied, The value is 1, otherwise, =0; This indicates the load on fiber core c in fiber link l from adjacent fiber cores. The crosstalk value;

[0031] in, The calculation formula is:

[0032] ;

[0033] In the formula, This indicates the length of the fiber optic link l, with the unit of length being km; The power coupling coefficient is expressed as:

[0034] ;

[0035] In the formula, r, r0, , q represent the intercore coupling coefficient, the bending radius of the fiber core, the propagation constant, and the intercore spacing, respectively;

[0036] Furthermore, the specific method of step S3 is as follows:

[0037] S301: For each service within the individual network, Dijkstra's algorithm is used to find K candidate paths from the source node to the destination node in the elastic optical network, where K is the average degree value of the nodes in the elastic optical network rounded up; the candidate paths are sorted in ascending order according to their length values; and the core pressure value of each candidate path is calculated sequentially. P r This is a set of candidate paths for the business; the candidate fiber cores of the candidate paths are sorted in ascending order according to the fiber core pressure value.

[0038] Among them, the core pressure value of core c in candidate path p Defined as:

[0039] ;

[0040] In the formula, Represents a set of business requests The pressure value of the fiber core c in candidate path p, Represents a set of business requests The pressure value of fiber core c in candidate path p; service request set It is the set of services that have already been allocated resources in the service request set R. Combined, business request set If R is the set of business requests for which resources have not yet been allocated, then the following relation is satisfied: , ;

[0041] in, The calculation formula is:

[0042] ;

[0043] In the formula, This represents the number of frequency slots required for service request r to use BPSK (Binary Phase Shift Keying) modulation on fiber core c of candidate path p; |C| is the number of fiber cores in a single fiber link. Let p represent the set of candidate paths for business request r, and let p be the set of candidate paths for business request r. r A path in p is formed by sequentially connecting one or more fiber optic links, where l is a fiber optic link of p; this calculation formula represents: service request set All services passing through each fiber link l of candidate path p cause a fiber core stress value of c on candidate path p. The pressure;

[0044] in, The calculation formula is:

[0045] ;

[0046] In the above formula, The number of frequency slots occupied by service request r on fiber core c of candidate path p;

[0047] S302: For each individual service with unallocated resources, sequentially search the available modulation format set and idle spectrum block on the candidate path of the service from the candidate fiber core set of the candidate path set. Sort the available modulation format set in descending order of modulation level, and use the first-choice method to find the idle spectrum block. Calculate the total inter-core crosstalk intensity caused by allocating the idle spectrum block to the service. If an idle spectrum block with a total inter-core crosstalk intensity value not greater than the crosstalk threshold is found, allocate the idle spectrum block to the service, record the modulation format, candidate path, and fiber core information corresponding to the allocation of the idle spectrum block, and proceed to step S4; otherwise, proceed to step S303.

[0048] The method for determining the available modulation formats that meet the service rate requirements is as follows: Based on the length of the candidate path of the service request, determine the highest modulation level of the modulation format available for the service on that path. Then, the modulation formats not higher than that modulation level constitute the set of available modulation formats for the service on that path. Arrange the modulation formats in the set of available modulation formats in descending order of modulation level. Starting from the highest available modulation level, use the first-choice method to find idle spectrum blocks on the candidate path.

[0049] Among them, the total inter-core crosstalk intensity value of the spectral block of candidate path p The calculation formula is:

[0050] ;

[0051] In the above formula, FS B A set of frequency slot indices for free spectrum blocks; Let be the crosstalk threshold corresponding to the modulation level m of the service transmission on the candidate path. If the service uses modulation level m, the spectrum block index set FS on the candidate path p is... B To establish an optical transmission channel, the constraint condition of crosstalk threshold must be met:

[0052] ;

[0053] This constraint sets a hard threshold for the selection of modulation format, which is a key condition to ensure the transmission quality of service requests in MCF-EONs. Therefore, when selecting a modulation format for a service request, it is necessary to first calculate the total crosstalk strength value between cores based on the crosstalk model, and then select a modulation format that meets the crosstalk threshold constraint to achieve a balance between spectral efficiency and service transmission quality.

[0054] Among them, after selecting the modulation format, service r i The number of FSs contained in the consecutive idle spectrum blocks selected in the candidate path selection is:

[0055]

[0056] In the above formula, bi Indicates business r i The request rate, in Gbps; W s For unit bandwidth, in flexible optical networks, this application takes a value of 12.5 GHz; To protect the number of frequency slots, this application uses a value of 1 frequency slot (FS); m is the modulation level, which is determined by the length of the candidate optical path, as shown in Table 1;

[0057] Based on the length of the candidate path for the service request, the modulation format with the highest modulation level available for the service on that path is determined as shown in the table below:

[0058] Table 1: The relationship between path length, modulation level, and modulation format is as follows:

[0059] BPSK 1 9600 QPSK 2 4800 8QAM 3 2400 16QAM 4 1200 32QAM 5 600

[0060] In Table 1, BPSK (Binary Phase Shift Keying), QPSK (Quadrature Phase Shift Keying), 8QAM (8-Quadrature Amplitude Modulation), 16QAM (16-Quadrature Amplitude Modulation), and 32QAM (32-Quadrature Amplitude Modulation) are respectively two-phase phase shift keying, four-phase phase shift keying, 8th-order quadrature amplitude modulation, 16th-order quadrature amplitude modulation, and 32nd-order quadrature amplitude modulation.

[0061] S303: For services that fail to allocate spectrum blocks, select modulation formats sequentially from the set of available modulation formats for the service, and use a bit-loading-based spectrum allocation algorithm to select idle spectrum blocks for the service. If an idle spectrum block that meets the service rate requirements is found, calculate the total inter-core crosstalk intensity caused by allocating the idle spectrum block to the service. If an idle spectrum block with a total inter-core crosstalk intensity value not greater than the crosstalk threshold is found, allocate the idle spectrum block to the service, and record the modulation format, candidate path, and fiber core information corresponding to the allocation of the idle spectrum block to the service, and proceed to step S4; otherwise, mark the resource allocation of the service as failed, and proceed to step S2.

[0062] The process of the spectrum allocation algorithm based on bit loading is as follows: if the current modulation format selected by the service is the lowest modulation level, then mark the service resource allocation as failed and proceed to step S2; otherwise, sequentially search for idle spectrum blocks with reduced modulation level per frequency slot that meet the service rate requirements from each fiber core of each link of the candidate optical path. Attached Figure Description

[0063] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the following figures are provided for illustration:

[0064] Figure 1 Here is the flowchart for Algorithm 1: Overall Algorithm for Resource Allocation in Optimizing Static Service Transmission in Elastic Optical Networks Based on Genetic Algorithm;

[0065] Figure 2 Flowchart of Algorithm 2: Routing Fiber Core Selection and Spectrum Allocation Algorithm Based on Fiber Core Pressure Value;

[0066] Figure 3 Flowchart of Algorithm 3: Modulation Format Selection and Spectrum Allocation Based on Crosstalk Threshold;

[0067] Figure 4 Here is the flowchart for Algorithm 4: Bit-Loaded Spectrum Allocation Algorithm. Detailed Implementation

[0068] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0069] The algorithm based on genetic algorithms provided in this invention aims to solve the resource allocation problem in elastic optical networks by allocating appropriate spectrum resources for time-varying services, thereby improving network resource utilization and ensuring service transmission. This invention can also be implemented or applied through other specific embodiments, and the details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of this invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this invention; in the absence of conflict, the following embodiments and features can be combined with each other.

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

[0071] Appendix Figure 1Here is the flowchart for Algorithm 1: Optimizing Resource Allocation for Static Service Transmission in Elastic Optical Networks Based on Genetic Algorithm. The sorting of batch service requests arriving at the elastic optical network within a time window is a permutation optimization problem. Given a set of N service requests, there are N! possible sorting schemes. Different sorting schemes lead to different service transmission processing and resource allocation orders, resulting in completely different resource allocation results and thus significant differences in spectrum occupancy. Genetic algorithms, as a global optimization algorithm, can efficiently search for the optimal solution in a large solution space through iterative selection and crossover operations, perfectly suited to the optimization requirements of request sorting. (Appendix) Figure 1 The specific steps of the algorithm shown are as follows:

[0072] Input: Elastic Optical Network Resource status, set of business requests Each business request in R and business request rate b i , Inter-core coupling coefficient r, fiber bending radius r0, propagation constant Let the core spacing be q, the iteration counter be n=1, the number of low-yield iterations be x=1, the upper limit of continuous low-yield iterations be X, the growth rate threshold be Y, the iteration upper limit be Z, and the number of individuals in the population be I.

[0073] Output: The sorting results in R and the resource allocation results for each service in R;

[0074] 1: Arrange the business requests in the request set R according to the business request rate. The descending sorting business is in set R1, and R1 is used as the ancestor individual of the genetic algorithm. The business request for each sorting position of the ancestor individual is also a gene of the ancestor individual.

[0075] 2: For the ancestral individual, random gene crossover using a genetic algorithm generates a total of I-1 new individuals. Each individual R j Each chromosome consists of N ordered genes from business requests. Newly generated individuals and the current chromosome together constitute a population containing I individuals.

[0076] 3: for each individual R in population I j

[0077] 4: Execute Algorithm 2;

[0078] 5: Calculate individual R j Fitness value;

[0079] 6: end for

[0080] 7: From I individuals, select the individual with the highest fitness function value, save this individual as an ancestor individual, and set the highest fitness function value of this individual to y. n ; Calculate y n Growth rate of the fitness function value of the corresponding individual ;

[0081] 8: If If the x-axis is positive, then x = x + 1; otherwise, let x = 1.

[0082] 9: If or If the result is positive, output the resource allocation and sorting results of the business request set R, that is, the sorting and resource allocation results of the individual with the highest fitness value; otherwise, let n=n+1 and go to step 2.

[0083] Appendix Figure 2 Algorithm 2: The routing fiber core selection and spectrum allocation algorithm based on fiber core pressure value, the specific steps are as follows:

[0084] Input: Elastic Optical Network Resource status, individual business request ordering Inter-core coupling coefficient r, fiber bending radius r0, propagation constant Chip pitch q, highest modulation level R j The set of requests already processed Unprocessed request set Let K be the average degree of the nodes in the elastic optical network, rounded up to the nearest integer.

[0085] Output: Individual R j The results of resource allocation.

[0086] 1: Let R b =R j , for R b For the first business request, Dijkstra's algorithm is executed to find K candidate paths for the business, and the K shortest paths obtained are stored in the candidate path set P. r In the middle, and calculate R b The first service transmits the fiber core pressure value on the candidate path. Based on the fiber core pressure value, the candidate fiber cores for the candidate paths of this service are sorted in ascending order;

[0087] 2: Choose For the first service request, Algorithm 3 is executed to obtain the routing, modulation format, fiber core, and spectrum allocation results for that service request. Removed from, and in Add the request to it;

[0088] 3: If Otherwise, go back to step 1;

[0089] 4: Output R j The routing, modulation format, fiber core, and spectrum allocation results for all services.

[0090] Appendix Figure 3 Here is the flowchart for Algorithm 3: Modulation Format Selection and Spectrum Allocation Based on Crosstalk Threshold. The specific steps are as follows:

[0091] Input: Multi-core fiber optic flexible optical network Individual R j The set of business requests, and the set of candidate paths P for each business within an individual. r Coupling coefficient r, bending radius Propagation constant Core spacing q, core pressure value of service candidate path Crosstalk thresholds corresponding to each highest level m ;

[0092] Output: Individual R j The modulation format and spectrum block allocation results for each service request.

[0093] 1: According to R j The candidate path lengths of each service are determined, the set of available modulation formats for each service is determined, and the available modulation formats of the services are sorted in descending order according to the modulation level in the set of available modulation formats.

[0094] 2: According to R j The candidate path lengths for each service are sorted in ascending order in the candidate path set for each service; the fiber core pressure values ​​for each service are sorted in ascending order in the candidate fiber core set for each service.

[0095] 3: for R j Business with unallocated resources

[0096] 4: Calculate the total inter-core crosstalk strength of the idle spectrum block that meets the number of frequency slots required by the service from the set of optional modulation formats of the candidate fiber cores of the candidate path.

[0097] 5: If the total inter-core crosstalk strength value of the service meets the requirements... Required free spectrum block

[0098] 6: Assign the location information of idle spectrum blocks, modulation format, fiber core, and routing to services;

[0099] 7: else

[0100] 8: Add business logic to the collection In the middle, Algorithm 4 is executed to find the spectrum block for bit loading of the service;

[0101] 9: end if

[0102] 10: end for

[0103] 11: Output R j The routing, modulation format, fiber core, and spectrum allocation results for all services.

[0104] Appendix Figure 4 Here is the flowchart for Algorithm 4: a spectrum allocation algorithm based on bit loading. Algorithm 4 makes full use of the bit loading mechanism to find the free spectrum block and modulation format with the least FS usage under the current fiber core group and allocate it to the service.

[0105] Appendix Figure 4 The specific steps of Algorithm 4 shown are as follows:

[0106] Input: Elastic Optical Network Resource status, individual R j The resource allocation results for already allocated spectrum blocks are displayed; services without allocated spectrum resources are stored in [the relevant data]. In the middle, the inter-core coupling coefficient r, the bending radius r0, and the propagation constant are... Core spacing q; The candidate path set, candidate fiber core set, and available modulation format set for the medium-speed service are determined, and the modulation level of the highest modulation format in the available modulation format set is denoted as . ,set up , Each business request in the process is marked as Where s and d are the source and destination nodes of the service, and b i For business r i The request rate, let W s =12.5GHz;

[0107] Output: The modulation format and spectrum block allocation results for all services.

[0108] 1: Confirm n of the business requests min and n max n min n represents the number of frequency slots required for the highest modulation level of the available modulation format for the service selection. max The number of frequency slots required to select the lowest modulation level of the available modulation format for the service, denoted as m. max The lowest modulation level is denoted as m. min ;

[0109] 2: Among the selected fiber cores of the selected path of the service, find consecutive idle cores with a value greater than or equal to... The spectral blocks of each frequency slot are stored in set S;

[0110] 3: for each unchecked spectrum block in set S

[0111] 4: Order ,set up The spectrum block currently being queried is denoted as F. Block ;

[0112] 5: for each FS in F Block

[0113] 6: Let the modulation level of the FS be m, and calculate the total inter-core crosstalk strength value. ;

[0114] 7: Determine the crosstalk threshold based on the selected modulation level m. ;

[0115] 8: if

[0116] 9: Order ;

[0117] 10: if

[0118] 11: Save F Block and the modulation level m and corresponding modulation format of each FS in the spectrum block;

[0119] 12: break;

[0120] 13: else

[0121] 14: if modulation level m=m min

[0122] 15: Mark the spectral block F in S. Block Checked;

[0123] 16: if there are still unlabeled spectrum blocks in set S

[0124] 17: Proceed to step 3;

[0125] 18: else

[0126] 19: if ! =

[0127] 20: = m = m-1, proceed to step 2;

[0128] 21: else

[0129] 22: Output: Information indicating failure in spectrum resource allocation for the service; algorithm terminates.

[0130] 23: end if

[0131] 24: end if

[0132] 25: end if

[0133] 26: end if

[0134] 27: else

[0135] 28: m = m - 1, proceed to step 5;

[0136] 29: end if

[0137] 30: end for

[0138] 31:end for

[0139] The above steps fully describe the static resource allocation process for elastic optical networks based on genetic algorithms, and supplement possible optimization measures to ensure better practical application results of the algorithm.

[0140] Finally, it should be noted that the above preferred 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 through the above preferred embodiments, those skilled in the art should understand that various changes can be made to it in form and detail without departing from the scope defined by the claims of the present invention.

Claims

1. A genetic algorithm for optimizing resource allocation in static service transmission of a resilient optical network, characterized in that: First, a genetic algorithm is used to sort multiple static services, treating each sort as an individual. Then, a resource allocation algorithm is executed on each individual. For each individual, transmission resources are allocated sequentially to the sorted services within that individual. First, the pressure values ​​of fiber cores, links, and paths are calculated, and paths and fiber cores with lower pressure values ​​are selected for the services. In the spectrum allocation of services, a spectrum allocation strategy based on a bit loading mechanism and minimizing bandwidth occupancy is designed. After all resource allocation attempts for all services in each individual are completed, the fitness function value of the individual is evaluated, and the resource allocation result of the individual with the highest fitness function value is selected as the resource allocation for all static services. This method specifically includes the following steps: S1: Set up service requests in the elastic optical network Request in Based on the rate of business requests The sorting is performed in descending order, which serves as the ancestor of the genetic algorithm. Each position of the individual is a gene position, and each gene corresponds to a business in the individual. Let x be the number of consecutive low-yield iterations, and set x=0; X is the upper limit of the number of consecutive low-yield iterations, Y is the threshold of the fitness function, I is the number of individuals in the population, Z is the upper limit of the number of iterations, and n is the number of iterations, and let n=0. S2: By performing random crossover operations on the gene positions of ancestral individuals, new genes are generated. Each individual, including ancestral individuals and newly generated individuals A population of I individuals is formed by I individuals, where the j-th individual is denoted as . ; S3: For each service in each individual in the population, execute the routing, fiber core, and spectrum resource allocation algorithms in turn; S4: Calculate all individuals The fitness function value of I individuals is denoted as y. n ;like ,but Otherwise, let ;like or Then the output fitness function value is y. n The transmission resource allocation result of the service prioritization scheme corresponding to the individual; otherwise, the fitness function value is set to y. n The individual as the ancestral individual, let Proceed to step S2.

2. The resource allocation for optimizing static service transmission in a resilient optical network using a genetic algorithm according to claim 1, characterized in that: The specific method of S1 is as follows: S101: The elastic optical network is abstracted as an undirected weighted graph G(V, L, C, F), where V represents the set of nodes in the elastic optical network; L represents the set of optical fiber links between nodes in the elastic optical network, and all optical fiber links are multi-core fibers; C is the set of fiber cores of a single optical fiber link; F is the set of occupancy status representations of the frequency slots within the fiber core. If a frequency slot is occupied, the frequency slot status is 1; otherwise, the frequency slot status is 0. S102: The business request set is defined as follows: Where N is the total number of business requests contained in the business request set. For the i-th business request, the business request is modeled as a triple. Where s and d are the source node and destination node of the service request, respectively, and b i The request rate for the service is expressed in Gbps. S103: The objective of this application is to optimize the spectrum occupancy of an individual, where spectrum occupancy (SO) is defined as: ； In the formula, It is a binary variable, (e i , e j ) represents node e in the elastic optical network i To node e j The fiber optic link, if the fiber optic link (e i , e j If the k-th frequency slot of fiber core c is occupied, and its frequency slot index is the maximum value of the occupied frequency slot index in the current fiber core, then... If it is 1, otherwise, =0; S104: For a set of business requests containing N business requests The j-th individual is defined as: ； In the formula, N is the total number of requests in the business request set. This is the index number of the sorted business, and Each For each individual, there is a gene on the chromosome of an individual. Each individual in this application corresponds to a sorting scheme of a business request in R. An individual contains one chromosome, and each chromosome contains N gene positions. S105: The fitness function of an individual is defined as: ； S106: The growth rate of the fitness function value among individuals is defined as follows: ； In the above formula, The fitness function value growth rate is given by n, where n is the number of iterations for each individual, and y is the number of iterations for each individual. n y represents the fitness function value of an individual in the nth iteration. n-1 This represents the fitness function value of an individual in the (n-1)th iteration. S107: The inter-core crosstalk intensity experienced by the f-th frequency slot in fiber core c of fiber link l. Defined as: ； In the formula, A c Let be the set of adjacent fiber cores of fiber core c; It is a Boolean variable, when the fiber core frequency gap When occupied, The value is 1, otherwise, =0; This indicates the load on fiber core c in fiber link l from adjacent fiber cores. The crosstalk value.

3. The genetic algorithm for optimizing resource allocation in static service transmission of a resilient optical network according to claim 1, characterized in that: The specific method of S3 is as follows: S301: For each service within the individual network, Dijkstra's algorithm is used to find K candidate paths from the source node to the destination node in the elastic optical network, where K is the average degree value of the nodes in the elastic optical network rounded up; the candidate paths are sorted in ascending order according to their lengths; and the core pressure value of each candidate path is calculated sequentially. P r This is a set of candidate paths for the business; the candidate fiber cores of the candidate paths are sorted in ascending order according to the fiber core pressure value. Among them, the core pressure value of core c in candidate path p Defined as: ； In the formula, Represents a set of business requests The pressure value of the fiber core c in candidate path p, Represents a set of business requests The pressure value of fiber core c in candidate path p; service request set It is the set of services that have already been allocated resources in the service request set R. Combined, business request set If R is the set of business requests for which resources have not yet been allocated, then the following relation is satisfied: , ; in, The calculation formula is: ； In the formula, This represents the number of frequency slots required for service request r to use BPSK (Binary Phase Shift Keying) modulation on fiber core c of candidate path p; |C| is the number of fiber cores in a single fiber link, and p is the set of candidate paths P for service request r. r A path in the path is formed by sequentially connecting one or more fiber optic links, where l is a fiber optic link of p. in, The calculation formula is: ； In the above formula, The number of frequency slots occupied by service request r on fiber core c of candidate path p; S302: For each individual service with unallocated resources, sequentially search the available modulation format set and idle spectrum block on the candidate path of the service from the candidate fiber core set of the candidate path set. Sort the available modulation format set in descending order of modulation level, and use the first-choice method to find the idle spectrum block. Calculate the total inter-core crosstalk intensity caused by allocating the idle spectrum block to the service. If an idle spectrum block with a total inter-core crosstalk intensity value not greater than the crosstalk threshold is found, allocate the idle spectrum block to the service, record the modulation format, candidate path, and fiber core information corresponding to the allocation of the idle spectrum block, and proceed to step S4; otherwise, proceed to step S303. The method for determining the available modulation formats that meet the service rate requirements is as follows: Based on the length of the candidate path of the service request, determine the highest modulation level of the modulation format available for the service on that path. Then, the modulation formats not higher than that modulation level constitute the set of available modulation formats for the service on that path. Arrange the modulation formats in the set of available modulation formats in descending order of modulation level. Starting from the highest available modulation level, use the first-choice method to find idle spectrum blocks on the candidate path. Among them, the total inter-core crosstalk intensity value of the spectral block of candidate path p The calculation formula is: ； In the above formula, FS B A set of frequency slot indices for free spectrum blocks; S303: For services that fail to allocate spectrum blocks, select modulation formats sequentially from the set of available modulation formats for the service, and use a bit-loading-based spectrum allocation algorithm to select idle spectrum blocks for the service. If an idle spectrum block that meets the service rate requirements is found, calculate the total inter-core crosstalk intensity caused by allocating the idle spectrum block to the service. If an idle spectrum block with a total inter-core crosstalk intensity value not greater than the crosstalk threshold is found, allocate the idle spectrum block to the service, and record the modulation format, candidate path, and fiber core information corresponding to the allocation of the idle spectrum block to the service, and proceed to step S4; otherwise, mark the resource allocation of the service as failed, and proceed to step S2. The process of the spectrum allocation algorithm based on bit loading is as follows: if the current modulation format selected by the service is the lowest modulation level, then mark the service resource allocation as failed and proceed to step S2; otherwise, sequentially search for idle spectrum blocks with reduced modulation level per frequency slot that meet the service rate requirements from each fiber core of each link of the candidate optical path.

Images