TianTong Technology Applies for Blockchain Attack Detection Patent Utilizing Graph Neural Networks to Identify Network Threats

TianTong Technology Applies for Blockchain Attack Detection Patent Utilizing Graph Neural Networks to Identify Network ThreatsFinancial World, October 18, 2024 - According to information from the State Intellectual Property Office, TianTong (Suzhou) Network Technology Co., Ltd

TianTong Technology Applies for Blockchain Attack Detection Patent Utilizing Graph Neural Networks to Identify Network Threats

Financial World, October 18, 2024 - According to information from the State Intellectual Property Office, TianTong (Suzhou) Network Technology Co., Ltd. has applied for a patent named "A Blockchain Attack Detection Method and Device Based on Graph Neural Networks", with publication number CN118784285A, filed in June 2024.

This patent aims to address the issue of cybersecurity in blockchain networks by utilizing Graph Neural Networks (GNNs) to identify and detect potential attack behaviors. The patent abstract describes the steps involved in the method, including collecting traffic data from the blockchain network, dividing the data into time slots and extracting features, constructing a graph structure for classification, and finally using the GNN model to classify the graph, thereby inferring the presence of attacks in the network, as well as the potential attack types and target entities.

The patent details the working principle of the blockchain attack detection algorithm based on graph neural networks. This algorithm first uses a pre-processing module to extract relevant information from traffic data and construct a graph structure, converting data and partial structural information of the network into node and edge features in the graph. Subsequently, the edge-GNN classifier within the algorithm classifies the graph constructed by the pre-processor based on these various graph features to identify the presence of anomalies in the traffic data and their types.

The blockchain attack detection method based on graph neural networks proposed in this patent provides a new solution for safeguarding blockchain network security. Leveraging the powerful capabilities of GNNs, this method can effectively identify attack behaviors in complex network environments, enhancing the security of blockchain networks.

The technology involved in this invention will have a significant impact on the field of blockchain network security. This patented technology is expected to be applied to various blockchain networks, such as Bitcoin, Ethereum, etc., effectively improving their defense capabilities, resisting various network attacks, and providing strong support for the healthy development of blockchain technology.