Hiding Malicious Content in PDF Documents

Dan Sabin Popescu

Abstract


This paper is a proof-of-concept demonstration for a specific digital signatures vulnerability that shows the ineffectiveness of the WYSIWYS (What You See Is What You Sign) concept. The algorithm is fairly simple: the attacker generates a polymorphic file that has two different types of content (text, as a PDF document for example, and image: TIFF – two of the most widely used file formats). When the victim signs the dual content file, he/ she only sees a PDF document and is unaware of the hidden content inside the file. After obtaining the legally signed document from the victim, the attacker simply has to change the extension to the other file format. This will not invalidate the digital signature, as no bits were altered. The destructive potential of the attack is considerable, as the Portable Document Format (PDF) is widely used in e-government and in e-business contexts.

Keywords


Digital signatures, vulnerabilities, PDF

Full Text:

PDF

References


W. Diffie, M. Hellman, New Directions in Cryptography, IEEE Transactions on Information Theory, 1976.

European Union, Community Framework for Electronic Signatures, Directive 1999/93/EC of the European Parliament and of the Council, 1999.

A. Alsaid, C.J. Mitchell, Dynamic Content Attacks on Digital Signatures, Emerald Group Publishing, 2005.

F. Buccafurri, Digital Signature Trust Vulnerability: A New Attack on Digital Signatures, ISSA Journal, 2008/10.

F. Buccafurri, Fortifying the Dali Attack on Digital Signature, Security of Information and Networks, 2009.

ISO Standard 32000-1, Document management – Portable Document Format 1.7, Adobe Systems, 2008/01/07. [7] TIFF Revision 6.0, Adobe Systems,1992.

Digital Signatures in the PDF Language, Developer Technical Note, Adobe Systems, 2006


Refbacks

  • There are currently no refbacks.


Journal of Mobile, Embedded and Distributed Systems (JMEDS) ISSN: 2067 – 4074 (online)