Abstract
This paper presents a self-aware network approach with cognitive packets, with a routing engine based on random neural networks. The simulation study, performed using a custom simulator extension of OmNeT++, compares RNN routing with other routing methods. The performance results of RNN-based routing, combined with the distributed nature of its operation inaccessible to other presented methods, demonstrate the advantages of introducing neural networks as a decision-making mechanism in selecting network paths. This work also confirms the usefulness of the simulator for SDN networks with cognitive packets and various routing algorithms, including RNN-based routing engines.
Highlights
Modern computer networks are used to transmit increasing amounts of data
This paper is in turn a simulation evaluation of a method that is rather typical for machine learning (ML)
We must bear in mind that while the neural network method can be successfully applied in a distributed architecture, and software-defined networks (SDNs) is only a platform that facilitates its implementation, graph methods require a central point where data from the entire network are collected, and their distribution is impossible
Summary
Modern computer networks are used to transmit increasing amounts of data. Optimal routing in backbone networks is a constant challenge due to the tension between the increasingly ubiquitous IoT, with more and more devices being connected to the network via high-bandwidth 5G protocols, and users’ demands for higher bandwidth and low latency. The knowledge of the state of the entire network allows us to think about implementing new routing methods that use aggregated data in a way that was previously unavailable in order to search for optimal packet routes in the network.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have