Mf-Net: Multi-Frequency Intrusion Detection Network for Internet Traffic Data

Abstract

The rapid growth of Internet technology renders network cybersecurity an important research field. Considering traffic data relates to not only temporal information, but also attack frequency, this paper presents a novel deep learning framework termed the multi-frequency intrusion detection network (MF-Net). The core of MF-Net is the multi-frequency LSTM (MF-LSTM) and multi-frequency transformer (MF-Transformer) module, which regard Internet traffic as a superposition of time series data with a variety of frequencies. Experimental results with comparison to the state-of-the-art approaches demonstrate the superiority of MF-Net on network traffic intrusion detection tasks.

Keywords: Cyber Security, Intrusion Detection, deep learning, Multi-frequency Transformer, Multi-frequency LSTM