Machine learning based malicious payload identification in software-defined networking

Qiumei Cheng, Chunming Wu, Haifeng Zhou, Dezhang Kong, Dong Zhang, Junchi Xing, Wei Ruan

Abstract: Highlights•A novel deep packet inspection method in software-defined networking is proposed.•Binary logistic regression is efficient in identifying unencrypted malicious payloads.•An adaptive packet window samples packets to ease the burden of the controller.•The decision tree is efficient in classifying malicious encrypted packets.•The throughput and overheads of the deep packet inspection method are acceptable.