Biomolecular Computing for Automatic Target Recognition

The goal of this Phase I SBIR is to propose concepts for pattern recognition techniques for use in automatic target recognition (ATR) systems. The key approach will be the use of bacteriorhodopsin thin films as read/write holographic storage media in Fourier transform autoassociative pattern recognition architectures. We plan to test the feasibility of this approach by constructing a simple, unoptimized prototype ATR system and testing its performance. Briefly, the specific objectives of this Phase I proposal are to: a. Develop and test a prototype of a Fourier transform autoassociative automatic target recognition system in order to develop device concepts; b. Test the use of holographic thin films of bacteriorhodopsin in the above device; c. Develop concepts for optical, computer interface and materials improvements based upon test results from the optical bench prototype. During Phase II, we plan to build a second generation ATR system, using improvements in materials technology licensed from Syracuse University. In addition, the electronic and optical design of the ATR will be refined based upon Phase I results.