Research On Multi Constraints QoS Unicastrouting Problem Based On Swarm Intelligence Optimization Algorithm

With the development of computer network technology,many network multimedia applications on the network quality of service(QoS)requirements are getting higher and higher,so to provide quality assurance is very important.The key to the service quality of the network lies in the performance of the routing algorithm.Therefore,it is very important to design a good algorithm for solving the problem of multi-constrained QoS.Because the multi-constrained QoS routing problem is NP-complete problem,if only the traditional routing algorithm can not be used to solve the polynomial time,and for the NP-complete problem,the use of the group of intelligent algorithms developed in recent years can often achieve better results.The network model for unicast routing problem is simplified and abstracted out of its mathematical model.The bandwidth constraint condition is filtered in the simplified model,and the constraint dimension dimension of the algorithm is reduced,which makes the problem simplified to some extent.In order to apply the genetic algorithm to the multi-constrained QoS routing problem,the corresponding coding method is adopted for the individual in the population,which greatly reduces the complexity of the conversion from coding to solution.Design a reasonable fitness function,reduce the computational complexity of the algorithm,and more truly reflect the characteristics of multi-constrained QoS routing problem;design a reasonable crossover operator,in the cross-operation through the "merit-based choice" approach makes The fitness value of the offspring is as large as possible,and the convergence speed of the algorithm is accelerated.After the cross operation,the appropriate algorithm is adopted to remove the maximum redundant ring and eliminate the phenomenon of the loop.The basic ant colony algorithm is studied and the ant colony algorithm is applied to the multi-constrained QoS routing problem.Aiming at the particularity of the problem,the appropriate algorithm is adopted to realize the initial operation of each ant in the ant population.Pheromone update strategy,through the optimal ants on the path of the pheromone on the second update,making the optimal solution on the corresponding pheromone concentration on the path to improve the ant colony algorithm to ensure positive feedback mechanism.The genetic algorithm and ant colony algorithm are combined to obtain the genetic-ant colony hybrid algorithm.The hybrid algorithm plays the advantages of genetic algorithm and ant colony algorithm.The local optimal solution obtained by the genetic algorithm is transformed into the initial value of the pheromone in the ant colony algorithm,and then the ant colony algorithm is used to solve the ant colony algorithm.In the ant colony algorithm,there is a good path selection orientation in the ant colony algorithm,Mechanism to work faster,to speed up the algorithm to find the global optimal solution speed.The simulation and testing of the Matplotlib graphics library based on Python are carried out for the specific network instance.The corresponding result is obtained by adjusting the parameter combination.The statistical results of each algorithm are analyzed and compared,and the feasibility and validity of the algorithm are verified.