The System Of DDoS Attack Collaborative Detection Based On DHT Design And Implementation

Distributed Denial of Service is a kind of attack which depletes the network’sresource and denies service to legitimate users. With the exposition of the DistributedDenial of Service, it‘s from become ever more diverse and the attack techniques areeasily to handle. Although the damage caused by DDoS are getting more and moreserious, but so far still no one thorough defense. This thesis regards real-time detectionof DDoS as target; deeply analysis features of DDoS, conducted research on thedetecting techniques of DDoS and its implementation analysis, here are the results:1. This thesis use detection node find out possible victim by traffic monitoring.Then confirmation aggressive behavior with collaborative fusion global informationbefore attack traffic arrived victim. Collaborative the local detection node could breaklimit of traditional DDoS detection. It can find out the attack information which hidingin normally flow and increasing accuracy rate that during the way which collaborativesingle node.2. To the problem of DDoS detection strength proportional to resourceconsumption，the system use Sketch date structure to compression storage networkflow. The Sketch date structure create Sketch matrix which store IP compressioninformation. Two Sketch matrixes make up sequence Sketch matrix. It is the basic oflocal detection. When warning appears, we use the Reverse Sketch to find out IPaddress of suspicious Sketch information. All of these technologies can make sureefficient local DDoS detection, save resource.3. System use DHT technology creates Aggregation Tree based on suspicious IPaddress of local detection. It insures collaborative system could monitor differentpossible victim, and avoid local detection information upload to finial decision nodewith repetitive information. Aggregation Tree produce shared path to fusion informationwithout extra resource, offer an efficient shared way, and prevent network congestion offinial decision node.4. Think about the visuality of DDoS detection system, we design demonstrateplatform of DDoS attacks detection system. The basic function of Demonstrate platformis simulating real network, then mark security of network on generating topologicalgraph. In addition, platform still response tidy and store attack evidence information. formake sure robustness and expandability of system, we design demonstrate window inMVC design mode, keep execute rate, reusability, expandability, maintainability.Structuring coupling platform, easy to update and control.