NSF: CIF: Toward Trustworthy Information Forensics and Anti-Forensics

The path of technological evolution has naturally led to a critical issue to ensure content, devices, and intellectual property being used by authorized users for legitimate purposes, and to be able to forensically prove with high confidence when otherwise. When information goes through various devices and processing, there are inherent traces left from each processing step. These traces serve as intrinsic fingerprints and are essential for forensic analysis. This research involves developing a joint forensic and anti-forensic framework to leverage the universal intrinsic fingerprints to gather traces of evidence to answer who has done what, when, where, and how. The main thrusts of this research include:

(1) Investigation of the strategy space and adversarial dynamics on a broad range of forensic problems associated with content, device and emerging sensing paradigm;

(2) Development of a theoretical foundation to understand the fundamental performance limit and establish formal notions of forensicability in an information processing chain; and

(3) Development of an Information Forensic and Anti-Forensic Engine along with a benchmark set and an online portal for integrated research, education, and outreach.

This research establishes a comprehensive foundation for the field of trustworthy information forensics, with a broad range of important applications tackled from a new angle of thinking. The research program seamlessly integrates with the education and outreach program through the establishment of the Forensic Engine and the collaboration with law and policy professionals. Through establishing trust and provenance in digital domain, closely involving and training students, collaboration in the research community, and partnership with law professionals and industry, this research contributes to law enforcement and national security as well as promotes the emerging science of trustworthy forensics to a broad audience.