Publication Date

Fall 2012

Degree Type

Master's Project


Computer Science


Metamorphic malware is capable of changing its internal structure without al- tering its functionality. A common signature is nonexistent in highly metamorphic malware. Consequently, such malware may remain undetected even under emulation and signature scanning combined. In this project, we use the concept of structural entropy to analyze variations in the complexity of data within a file. The process consists of two stages, namely, file segmentation and sequence comparison. In the file segmentation stage, we use entropy measurements and wavelet analysis to segment a file. The second stage measures the similarity of files by computing the edit distance between sequence segments. We apply this technique to the metamorphic detection problem and show that we can obtain strong results in certain challenging cases.