Research on malicious Program Traffic Detection Technology based on Deep Learning

Abstract

With the implementation of the strategy of "three types and two networks", there are network security risks in every link of the power system, which is mainly reflected in the fact that the power information system has a large user base and large exposure, and at the same time stores a large amount of electricity customer data. bearing control business, once the system is breached, it will have a great impact on the society and economy. However, the network attack on power information system has the characteristics of "strong professionalism, deep latency, long persistence and great destructiveness", so the security monitoring of power network and information system is faced with great challenges. Attackers generally use malicious programs to steal sensitive information, but with the progress of various intrusion detection technologies, malicious program anti-detection technology is also developing, which makes malicious program traffic very hidden and more difficult to be detected. Based on the shallow machine learning method, the ability to describe the characteristics of malicious programs is insufficient, and in the face of new malicious programs, such as inefficiency or failure, deep learning technology shows a strong learning ability in many application fields. this paper explores the research and application of malicious program traffic monitoring technology based on deep learning method. This paper studies the monitoring of abnormal traffic access behavior of malicious programs based on deep learning, and perceives abnormal traffic and security threats in real time.