Abstract: We applied artificial immune system algorithm to the real-time detection of botnet, and proposed an online detection model based on deterministic dendritic cell algorithm. Combining the characteristics of botnets, the behavior signals were defined, and the real\|time detection of the host of botnet based on the heuristic information was realized. The validity of the model was experimentally verified by using standard data sets. Experimental results show that  this model has the advantages of real-time, simple behavior definition and accepting the definition of multiple kinds of heuristic information. Moreover, the false negative rate and false positive rate of detection of botnet are low.

Key words: dendritic cell algorithm, artificial immune system, botnet detection, danger theory