Machine learning enhanced quality of service based routing with blockchain in multi-domain software defined networks

The main task of future networks is to create intelligent networking architectures for as much intellectualization, activation and customization as possible. Software Defined Networks (SDN) creation technology breaks the tight connection between the control plane and the data plane in traditional network architecture, making the controllability, security and economy of network resources a reality. Machine learning, one of the methods of artificial intelligence, when combined with SDN architecture, is effective in areas such as network resource management, end-to-end route planning, traffic programming, fault diagnosis or network security. The rapid increase in the number of devices in today's network architecture makes it difficult to carry out operations such as security, privacy, service provision and network management with a central control system. Therefore, the combined use of blockchain and machine learning for decentralized network management to be secure, smart and efficient is of interest in academic studies and industrial applications. SDN is easier to control as it is managed by software compared to traditional networks. In these networks, different machine learning techniques have been used in the literature on blockchain technology, such as increasing the blockchain transaction volume and reducing the average block generation time. The aim of this study is to create a wayfinding architecture in end-to-end multi-networks supported by blockchain with the help of Reinforcement Learning (RL) models.