Computerized Forensics Research: The Good, the Bad, and the Unaddressed - PowerPoint PPT Presentation

digital forensics research the good the bad and the unaddressed l.
Skip this Video
Loading SlideShow in 5 Seconds..
Computerized Forensics Research: The Good, the Bad, and the Unaddressed PowerPoint Presentation
Computerized Forensics Research: The Good, the Bad, and the Unaddressed

play fullscreen
1 / 20
Download Presentation
hiram-spence
Views
Download Presentation

Computerized Forensics Research: The Good, the Bad, and the Unaddressed

Presentation Transcript

  1. Digital Forensics Research: The Good, the Bad, and the Unaddressed by Nicole L. Beebe, Ph.D. 5th Annual IFIP WG 11.9 January 27, 2009

  2. Discussion Topics • General Successes • General Failures • Research Needs

  3. Background (Bias) • My background • Ex-Law Enforcement (AFOSI, 1998-2007) • Private Sector Forensics (2001-Present) • Educational background • B.S. Electrical engineering • M.S. Criminal justice • Ph.D. Information systems • Assistant Professor in Information Systems Dept.

  4. Background – Contributors • Dr. Sujeet Shenoi – Univ. of Tulsa • Mark Pollitt – Ex-FBI, Univ. of Central Florida • Eoghan Casey – Stroz Friedberg LLC, Johns Hopkins Univ. • Dr. Simson Garfinkel – Naval Post Grad School (Harvard, MIT) • Eric Thompson – CEO, Access Data Inc. • Ovie Carroll – Ex-AFOSI, DoJ CCIPS Cybercrime Lab Dir. • Dave Baker – MITRE • John Garris – Ex-AFOSI, NASA OIG Computer Crimes SAiC • Randy Stone – Detective, Wichita Police Dept. • Dr. Marc Rogers – Purdue Univ. • Dr. Frank Adelstein – ATC-NY • Dr. Wietse Venema – IBM

  5. Background – Contributors • Gary King – AFOSI Computer Crime Investigations Program Mngr • Dr. Florian Bucholz – James Madison Univ. • Dr. VassilRoussev – Univ. of New Orleans • Jesse Kornblum – ManTech • Russell McWhorter – Bexar County Sherriff’s Office, Veridicus Inc. • DeWayne Duff – Ex-AFOSI, Stroz Friedberg LLC • Rod Gregg – Ex-FBI, Stroz Friedberg LLC • Drew Fahey – Ex-AFOSI, e-fense Inc. (developer of Helix) • … plus seven other researchers & practitioners • … and, of course, me.

  6. The Good • Unequivocal improvement in prominence & value of digital evidence in investigations • Becoming more scientific • Formalization/standardization of processes/approaches • Formulating DF problems into scientific research Q’s • DF research starting to enter mainstream research • Archeology of digital artifacts (Windows/Linux) • Cross-discipline knowledge sharing • Tackling the DF problem de jour (e.g. memory)

  7. More Kudos • HW write-blocking industry • Acquisition/collection phase in general • Live forensics • Contributions to qualification, certification, etc. discussion • Honorable mentions • The Sleuth Kit (high quality, open-source tool) • AFF (vendor neutral, compress-able imaging format)

  8. The Bad • Hyper-formalization of processes/approaches • Agencies getting dangerously close to checklists • Cross-discipline knowledge sharing incomplete • Lack of extension of information science research • Insufficient research into other OS & FS • HFS+, UFS, ZFS, proprietary systems, etc. • Data-centric not info/knowledge-centric • Researchers & practitioners are both guilty

  9. More Criticisms • Lacking a common body of knowledge • Accreditation “machine” has spun out of control • Bridging the gap between research & application • Still lacking rigor & relevance in research • Lack of a clear research agenda • Common CFP topics, but cover full-spectrum • Lack of federal funding of research (U.S. complaint) • Commercial industry shaping the agenda • Decisively toward e-discovery research questions

  10. The Unaddressed (or at least needing more attention)

  11. Volume & Scalability • Acquire & process more faster • Logical acquisitions – decision support systems • And/or non-“complete” physical acquisitions • Collaborative, distributed analysis • Collaboration management • Data storage/transfer (centralized, decentralized) • Lagging S/W development • H/W advances (multi-threading / massive parallelism) • Tools to handle large volumes of email • Data analytics, linkages & pattern analysis

  12. Intelligent Analytical Approaches • Need to extend artificial intelligence and other intelligent search/retrieval algorithms/approaches • Semantic vs. literal searching techniques • Improved data indexing / relational data optimization • Similarity matching mechanisms • “Fuzzy hashing” requires paradigm shift & scientific certainty research/support • Intelligent password recovery • Passphrase ID/extraction • Probabilistic approaches (length, location, signatures) • PW caching moving to CPU cache

  13. Fast-Paced Technological Landscape • Small device forensics (e.g. cell phones, PDAs, GPS, etc.) • “Homegrown” device forensics • Gaming devices • Virtual environments • Cloud computing environments

  14. “Ease of Use” • Tool – need to simplify ease of use for practitioner • Not too technical • Easy to use user interface • Protections against human error • but allow advanced mode for customizations • Information – reported findings must be usable • Data visualization • Cross-correlation, link-analysis (automated) • Reduce problem of info overload (need “zoom” capability) • Paradigm shift from hierarchical to temporal view

  15. S/W Development/Engineering • S/W must fully leverage H/W advances • Increased automation • Increased interoperability • Standardized, interoperable data formats (I/O) • Standardized APIs • Need OS independent DF platforms (e.g. Pyflag) • Need DF platforms that are all-in-one wrt data • Static media, volatile data, network dumps, etc.

  16. “Other” • Database forensics • Steganography • Live file systems • More work needed on volatile memory analysis • Knowledge of disturbance/distortion caused • Non-windows/linux file systems (HFS+, UFS, ZFS) • Solid state memory acquisition & analysis • Investigations involving multiple, distributed systems • New XML office document standards

  17. Issues of Science • A way to specify error rates like in traditional forensic sciences • Is this realistic? • Paradigm shift toward determining/quantifying certainty/confidence? • Formalization of hypotheses generation & testing • Repeatable experimentation & comparative eval. • Need for a common test corpora

  18. Philosophical Questions • Is DF field losing its “purity” to e-discovery field? • Are we immune to DMCA suits? • Can the research community influence technical specification documentation respecting DF needs? • Are we keeping pace with anti-forensics research?

  19. Research Agenda Summation • Volume • Non-device level acquisition • Intelligent searching, extraction & analysis • Technological changes • Move away from incremental knowledge contributions toward tougher challenges of significant contribution • Paradigm shifts • Non-binomial conclusions (scientifically derived) • Conclusion certainty vs. tool/process error rates • Need to study ease of use, HCI & adoption issues

  20. Questions/Comments? (210) 269-5647 Nicole.Beebe@utsa.edu http://faculty.business.utsa.edu/nbeebe