Research On The Key Technologies Of Detecting DDoS Attacks

Distributed Denial of Service(DDoS)is a kind of attack act that is aiming at exhausting the limited service capacity of the target host and making the valid users denied for requesting services.Once a DDoS attack occurs,the target host would be intruded by a number of coordinated attacks and be brought in with some catastrophic consequences.Therefore,it is important to detect,identify,process and prevent DDoS attacks effectively.Through studying on the principle,characteristics and methods of DDoS attacks,this thesis elaborates the behavioral patterns and inherent characteristics of DDoS attacks.Along with regarding the potential conditions where DDoS attack would occur as well as the network protocol architecture,a method of detecting DDoS attacks under high load communication scenario is proposed.The research in the thesis focuses on the research of multi-layer communication protocol,and classifies and analyzes the DDoS attacks.It focuses on large-data-intensive and high-load communication scenarios,and discusses feasible DDoS attack detection methods.In combination with demand-driven multi-Agent modeling technology to construct the DDoS attack detection method.Through analyzing the inherent characteristics of DDoS attack,the intelligent DDoS attack detection is realized.The simulation results show that the method proposed in this paper has better detection performance and less granularity dependence on the data unit,and can effectively detect target attack based on sample training.The detection scheme based on agent technology can correctly implement all kinds of preset behaviors and possess good expansibility,which can meet the technical requirement of subsequent design and implementation of software prototype.