Alliance chain anomaly detection system based on machine learning and detection method thereof

Abstract

The invention discloses an alliance chain anomaly detection system based on machine learning and a detection method thereof. According to the detection method, the detection system is arranged in an alliance chain network based on a PBFT consensus algorithm; according to time interval data of a single node prepart stage and a commit stage in the PBFT consensus process, whether abnormal nodes existin the alliance chain network or not is determined firstly, if the abnormal nodes exist, the abnormal nodes are further determined, and whether the abnormal nodes exist in the alliance chain networkor not can be rapidly and accurately detected; two times of anomaly detection are adopted, it is ensured that under the condition that the abnormal nodes in the alliance chain are determined, resourceoccupation of the anomaly detection system is reduced, whether the abnormal nodes exist in the alliance chain network or not is detected through data of a single node in the first anomaly detection,and the abnormal nodes in the alliance chain are determined through the second anomaly detection; according to the system, on the premise of ensuring the detection effectiveness and reliability, the problem of resource occupation is fully considered, and the resource occupation of the detection system is reduced.

Description

technical field [0001] The invention relates to blockchain technology, in particular to a machine learning-based alliance chain anomaly detection system and a detection method thereof. Background technique [0002] Blockchain technology has received the attention of many experts and scholars since it was proposed. This technology has broad application prospects in many fields such as contract processing, data exchange, finance, credit reporting, Internet of Things, logistics, and economic settlement. This technology abandons the traditional trusted third party and is known for its decentralization. According to the degree of decentralization, it can be divided into public chain, alliance chain and private chain. Among them, the public chain is completely decentralized, allowing anyone to join or exit at any time, but it is difficult to correct if there is a problem after its operation, and its consensus mechanism has high overhead, high delay, and low throughput, which canno...

AI Technical Summary

Problems solved by technology

This method uses multi-source heterogeneous data when performing anomaly detection, and its data has a relatively large number of dimensions. Although data dimensionality reduction operations are performed during machine learning, the operation is more complicated

Especially when using complex deep learning, the cost of computing resources and time will be high when learning

Images