Neural Network Multipath Routing in Software Defined Networks Based on Artificial Bee Colony Algorithm

Abstract

Software defined networks (SDN) are new network technology, developed with the aim of improving quality and varying network service requirements based on the benefits of a centralized network architecture. SDN not only makes it easy to configure network policies to control traffic, but it also provides a convenient way to get network status. Appropriate QoS policies need to be applied to meet the requirements of the different services, and there should also be a trade-off in link utilization. Multipath routing solutions are applied to solve this problem. Accordingly, instead of using one route, data can be transmitted simultaneously along many different routes. This paper proposes an intelligent multipath routing method based on artificial neural network that allows the controller to configure data transmission policies quickly and efficiently. Hyperparameters when designing the neural network model is optimized by artificial bee colony algorithm. The obtained results show that the proposed neural network model has the ability to predict routes with high accuracy in real time, which makes it possible to implement various load balancing schemes in order to increase the performance of the SDN.