Virtual network mapping method and device based on improved genetic algorithm

Abstract

The invention discloses a virtual network mapping method and device based on an improved genetic algorithm, wherein the method comprises: encoding each virtual network node mapping scheme into a chromosome to obtain a plurality of encoded chromosomes; subjected each of the plurality of chromosomes to chiasma, variation, feasibility check and selection operation; using a bandwidth resource consumedby a virtual link as the fitness function of a model; obtaining all the virtual link mapping schemes in the virtual network request according to the fitness function, and checking the feasibility ofall virtual link mapping schemes to achieve virtual network mapping. The method can further improve the quality of the virtual node mapping scheme through the improved mutation operation, shorten thepopulation initialization process time, increases the convergence speed of the iteration, and increases the utilization of physical network resources.

Description

technical field [0001] The invention relates to the technical field of genetic algorithms, in particular to a virtual network mapping method and device based on an improved genetic algorithm. Background technique [0002] The current genetic algorithm mainly uses such as figure 1 The flow shown is used to solve the related problems. The first step is to initialize the parameters of the algorithm, including the initial population size, crossover probability, mutation probability, etc.; the second step is to initialize the population, generally using a certain method to generate several solutions; the third step is to The particles in the population are crossed in pairs to generate particles with higher fitness; the fourth step is to mutate the particles in the population to generate particles with higher fitness; the fifth step is to check whether the generated particles meet the specified If yes, end the program, otherwise judge whether the maximum number of iterations is ...

AI Technical Summary

Problems solved by technology

[0003] However, the main disadvantages of the existing genetic algorithm are: the population initialization operation is too time-consuming, and the iterative convergence time is too long, requiring a lot of calculation and solution time, and the optimization operation is not performed after each iteration, resulting in particles generated in each iteration The quality is not high, increasing the number of iterations and iteration time

Images