SBIR-STTR Award

DTI proposes to develop and demonstrate an algorithm for detecting wakes using dual polarization radar imagery. In Phase 1, we will exploit the information available in high-resolution Doppler spectra. Because of the nonlinear behavior of breaking waves that dominate the appearance of clutter and wakes in high-frequency radar data, wakes will be distinguished from clutter. Furthermore, the nonlinear behavior is different at HH and VV polarizations, increasing the information available for signal discrimination. DTI already has state-of-the-art radar modeling and simulation capability based on many years of experience in Navy R&D programs. Breaking wave physics is included in the Doppler spectrum model to ensure agreement with sea test data. We will simulate the appearance of wakes in Doppler spectra and quantify their strength with respect to ambient clutter as a function of sea state and viewing geometry. These results will guide the design of a detector algorithm optimized for dual polarization Doppler detection. The algorithm will be a variant of the maximum likelihood detector, called the eigen-image algorithm, which DTI has developed for similar applications in other programs. We will demonstrate successful performance of the algorithm on simulated and real data and outline the transition to a real-time implementation. Successful completion of the Phase 1 detector design and implementation as research code will serve as the basis for development of prototype code for real-time processing in Phase 2. The result will be improved capability for detecting wakes using dual polarization radar data. Operators on maritime patrol aircraft, who already are overburdened by multiple responsibilities, will have more time to give attention to other duties. The Navy is entering a period when forward operations in littoral environments are increasingly likely. With improved sensors and advanced platform technologies (such as unmanned aerial vehicles and the planned multi-mission aircraft) becoming available, and computer power that can make maximum use of available data, real-time detection of threats that can interfere with U.S. missions is both possible and necessary. These same tools can be used in the surveillance and defense of our own shores and for the evaluation of the threat to our submarines from enemy surveillance. The algorithms and software may be useful to nonmilitary users too. For example, locations of oil and gas seepage may be identified with variants of the software

Dynamics Technology, Inc. and its subcontractor, Lambda Science, Inc., will develop models and algorithms for maritime target detection in order to predict operational performance and to support requirements definition and engineering specifications for the BAMS UAV radar. Sea spikes from breaking waves are the biggest hindrance to radar detection of small surface targets, such as small boats, from high altitude and high grazing angle. We will extend existing clutter models, which have been developed to describe radar imagery, to MTI processing and validate them using program data. To reduce false alarms from the levels currently observed by the Global Hawk ISS and to achieve detection against targets with radar cross sections as small as 0.1 m2, we will develop algorithms for combining point target detection (small boats) with detection of their wakes. We will assess detection performance based on standard processing methods like STAP, and we will address the potential for better performance using alternative algorithms that may incorporate physical models of ocean clutter. We will incorporate engineering analysis of radar systems to ensure that predicted performance is consistent with a radar design that could be implemented on the BAMS-UAV