Research And Application Of Multi-source Information Fusion Method Of DDoS Attack In Internet Of Things

Distributed Denial of Service attack is a powerful and destructive attack method with simple operation.In recent years,with the rapid development of the Internet of Things,DDo S attack incidents on Internet of Things devices is more and more common,which has great social harm.At present,there is little research on DDo S attacks in the Internet of Things in academia.Aiming at the problem of low detection rate of DDo S attacks on Io T devices,this paper proposes a two-level multi-source information fusion method to identify DDo S attacks.The specific research work of this article is as follows:1.This article introduces the principles of DDo S attacks in the Internet of Things environment,discusses some classic DDo S attack methods,and finally analyzes and summarizes the reasons why Io T devices are vulnerable to DDo S attacks.2.This article introduces the related technology of multi-source information fusion.Aiming at solving the problem that it is difficult to efficiently identify DDo S attacks with a single feature,a two-level multi-source fusion method based on BP neural network and D-S evidence theory is proposed.First,according to the characteristics of DDo S attacks,we extract some kinds of feature from the original network flow data.Then,with the goal of improving the attack recognition rate of the model,the genetic algorithm is used to optimize the parameters of the BP neural network such as the initial weight,bias and learning rate to obtain the optimal BP neural network model.The optimized BP neural network model is used to perform feature-level fusion of DDo S attack data to obtain preliminary fusion results.Based on the preliminary fusion results,we use D-S evidence theory to carry out decision-level fusion.The experimental results show that the two-level fusion method which contains feature-level and decision-level can effectively improve the accuracy of DDo S attack judgment.3.This paper designs and implements a DDo S attack multi-source information fusion system based on the Internet of Things environment.First,a brief overview of the requirements of the system,and according to the requirements,design and implementation of network flow data acquisition module,feature extraction module,feature-level fusion module,decision-level fusion module,etc.The system test results show that the proposed two-level fusion method based on BP neural network and D-S evidence theory can improve the detection accuracy of DDo S attacks,and the system has certain scalability.