Machine Learning for Detecting Security Attacks on Blockchain using Software Defined Networking

Abstract

As there are many portable devices such as mobile devices and Internet of Things (IoT) devices, and these devices are not secured due to mobility and other factors, attacks are the biggest threat on these devices, and security is the biggest concern. Different researchers used to research security and attacks, but they faced some latency, cost, and system failures. Over the years, blockchain has been used in various application areas to improve reliability, information protection, and frameworks. Even the blockchain structure is robust; even after this; blockchain is not resistant to cyber-attacks. As we have an idea about Ethereum is that it is a blockchain-based software-based platform with intelligent functionalities. One of the successful attacks on Ethereum has uncovered security propensity. As there are so many benefits as Software Defined Networking (SDN) is concerned, in Software Defined Networking (SDN) data packet routes via a single firewall which makes IDS more secure. This paper will detect security attacks on the blockchain using a machine learning approach and Software Defined Networks. This paper will detect security attacks on the blockchain using a machine learning approach and Software Defined Networks. This paper discusses an anomaly-based recognition method centered on an encoder-decoder prototype, and it is skilled via collective facts obtained by observing blockchain actions.