PodBot: A New Botnet Detection Method by Host and Network-Based Analysis

The growing use of smart phones equipped with a variety of sensors makes them an ideal place for criminal activities and attacks such as bots. Unfortunately, the existing methods have made detection either by using network traffic or just using static analysis. In this paper, a method with a tool called PODBot has been introduced. The detection is done based on both application features and network traffic analysis. PODBot was evaluated on a set of botnets from well-known types and it could accurately detect over 87% in high risk and 96% in very high risk. In addition, in qualitative comparisons of similar tasks, due to the combination of the detection methods, it has features that improve the method relative to the previous methods.