Polymorphic worm detection method based on frequency number CNN

• Main Authors: ZHOU HANXUN, YANG YANG, GUO WEI
• Format: Patent
• Language: Chinese; English
• Published: 01.01.2019
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC COMMUNICATION TECHNIQUE; ELECTRICITY; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS; TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHICCOMMUNICATION

A polymorphic worm detection method based on a frequency number CNN comprises the steps of: performing numeralization processing for effective load data, converting each character to a corresponding ASCII value and performing statistics for the number of appearance times of each character in the effective load data to process the effective load data to capture the character distribution features of the effective load data. The present invention provides a polymorphic worm detection method based on a frequency number CNN capable of improving the accuracy of the polymorphic worm detection, and facilitating learning of rich features by the CNN. 基于频数CNN的多态蠕虫检测方法，步骤为：首先，对有效载荷数据进行数值化处理，将每个字符转化为对应的ASCII值，其次通过统计有效载荷数据中每个字符出现的次数来处理有效载荷数据，可以捕获蠕虫有效载荷数据的字符分布特征。通过上述方法，本发明提供了种能够提高多态蠕虫检测的准确率，以便于CNN能够从中学习到丰富的特征的基于频数CNN的多态蠕虫检测方法。