Date of Award
Engineering and Applied Science
Golden G. Richard III
Memory forensics (or memory analysis) is a relatively new approach to digital forensics that deals exclusively with the acquisition and analysis of volatile system memory. Because each function performed by an operating system must utilize system memory, analysis of this memory can often lead to a treasure trove of useful information for forensic analysts and incident responders. Today’s forensic investigators are often subject to large case backlogs, and incident responders must be able to quickly identify the source and cause of security breaches. In both these cases time is a critical factor. Unfortunately, today’s memory analysis tools can take many minutes or even hours to perform even simple analysis tasks. This problem will only become more prevalent as RAM prices continue to drop and systems with very large amounts of RAM become more common. Due to the volatile nature of data resident in system RAM it is also desirable for investigators to be able to access non-volatile copies of system RAM that may exist on a device’s hard drive. Such copies are often created by operating systems when a system is being suspended and placed into a power safe mode.
This dissertation presents work on improving the speed of memory analysis and the access to non-volatile copies of system RAM. Specifically, we propose a novel memory analysis framework that can provide access to valuable artifacts orders of magnitude faster than existing tools. We also propose two new analysis techniques that can provide faster and more resilient access to important forensic artifacts. Further, we present the first analysis of the hibernation file format used in modern versions of Windows. This work allows access to evidence in non-volatile copies of system RAM that were not previously able to be analyzed. Finally, we propose future enhancements to our memory analysis framework that should address limitations with the current design. Taken together, this dissertation represents substantial work towards advancing the field of memory forensics.
Sylve, Joseph T., "Towards Real-Time Volatile Memory Forensics: Frameworks, Methods, and Analysis" (2017). University of New Orleans Theses and Dissertations. 2359.
Available for download on Thursday, May 19, 2022