It may be one the oldest, most basic components of software code but the scientists at the Defense Advanced Research Projects Agency want to develop what they call revolutionary technologies for analyzing, identifying, and slicing binary executable components.
According to DARPA, the Department of Defense has critical applications that have been developed for older operating system versions and must be ported to future versions. In many cases, the application source code is no longer accessible requiring these applications to continue to run on insecure and out-dated configurations, impacting day-to-day operations. It is necessary to identify and extract functional components within this software for reuse in new applications.
More cool news: High-tech healthcare technology gone wild
DARPA said it defines binary executable components are defined as a fully encapsulated set of subroutines, data structures, imported APIs, objects, and global variables that accomplish a particular function.
DARPA says its Binary Executable Transforms (BET) program is focused on binary executables, not source code, and seeks to overcome the current limitations of existing binary analysis and program slicing techniques. Many program slicing techniques use source code, not machine code, to slice functional components due to the lack of correctness in binary analysis and disassembly.
That's where the BET program comes in. Specifically, BET is seeking innovative research in:
BET intends to generate novel research, publications, and prototype code to seed future programs requiring foundational technology in binary program analysis. The goal of this program is not to build systems or transitionable technology but to perform research that will eventually help the Department of Defense build such systems, DARPA stated.
"To bound the research problem, experimental binaries and approaches should be limited to x86 Windows or Linux operating systems. Approaches should be capable of processing either PE or ELF formats without artifacts or information dependencies particular to a specific binary format. Other formats may be considered but will require compelling rationale from the performer," DARPA said.
DARPA said it anticipates multiple awards not exceed $250,000 per phase/technical area for this research announcement.
Follow Michael Cooney on Twitter: nwwlayer8
Layer 8 Extra
Check out these other hot stories: