Automatic penetration testing framework based on campus network
Xin GE, Minnan Yue, Jiangtao JIN · JOURNAL OF SHENZHEN UNIVERSITY SCIENCE AND ENGINEERING · 2020
The traditional penetration test is using the tools of scan and penetration by professional security personnels to simulate hacker attacks and find the vulnerable points in the network. It depends on the ability of personnels, especially their own experience, and may take a longtime. There are many systems in campus networks, and they are becoming more and more complex. One needs to take a long period and cost to carry out penetration tests for these services, so it is difficult to complete detections throughfully. Automatic penetration test is easy to operate, and can be carried out continuously. However, most penetration test platforms are single direction penetration or semi-automatic, lack comprehensive and flexible. In this paper, we propose a new automatic penetration testing framework (AAPF), which is loosely coupled and easily expanded. Using deep learning and artificial intelligence, AAPF can solve many key problems in the process of penetration automation. The tests in campus network show that the platform can improve the success rate of penetration test and significantly shorten the penetration time.