A Spatial-Temporal Attention-Driven Autoencoder Framework for Automated Blockchain Intrusion Response

Abstract

The attack surface of decentralized networks and blockchain ecosystems continues to grow, they are facing more frequent attacks. The growing attack surface of decentralized networks and blockchain ecosystems is creating more frequent attacks, as well as new cyber threats that are emerging. While the detection capability of the intrusion detection system based on deep learning has been enhanced, the incorporation of the spatial–temporal dependency is not yet sufficient in existing approaches and there is no seamless integration with automatic blockchain-based enforcement. Given these limitations, this paper introduces a peer-to-peer tri-tier architecture called Spatial