A DoS Attack Detection Method Based on Multi-source Data Fusion

Abstract

In this paper, a detection method for DoS attack is realized by combining neural network and D-S evidence theory. The basic probability assignment (BPA) value of various attacks can be acquired through BP neural network, and then final result will be obtained with the assistance of improved D-S evidence theory and acquired BPA. Neural network is adopted to improve the acquisition of BPA, and combinational rules of the D-S evidence theory are adjusted by means of Fuzzy Support Weighting to avoid unreasonable fusion result when there is a severe or complete conflict between different sources, reducing the error from conflict information and the method adopted on data fusion. Finally, the method is verified with data obtained in the experimental environment, proving that it is accurate in detecting common types of DoS attack (such as Neptune, Smurf attack, etc.).