Forensic Data Properties of Digital Signature BDOC and ASiC-E Files on Classic Disk Drives
Name
Raul Nugis
Abstract
This thesis reviews the contents and observes certain properties of digitally signed documents of BDOC and ASiC-E container formats. After reviewing a set of sample containers, the author comes up with a header and footer combination (signature) significantly improving pinpointed carving-based recovery of those files from a deleted state on NTFS formatted uncompressed volumes in contiguous clusters, taking into account the geometry of classic disk drives. The author also describes forensically meaningful attributive data found in ZIP Headers and Central Directory, XML signatures as well as embedded ASN.1 encoded data of the sample files and suggests an algorithm for the extraction of such data. Based on these findings, the author creates scripts in Python and executes a series of tests for file carving and extraction of attributive data. These tests are run over the samples placed into unallocated clusters and the results are compared to several mainstream commercial forensic examination suites as well as some popular data recovery tools. Finally, the author web-scrapes a large number of real-life documents from a government agency’s public document registry. The carving signature and the data-extractive algorithm are thereafter applied on a larger scale and in an environment competitively supplemented with structurally similar containers.
Graduation Thesis language
English
Graduation Thesis type
Master - Cyber Security
Supervisor(s)
Pavel Laptev, Raimundas Matulevičius
Defence year
2018