Rapid and Accurate High-Resolution Radar Signature Prediction of Sea Targets
Navy SBIR FY2013.1

Sol No.: Navy SBIR FY2013.1
Topic No.: N131-003
Topic Title: Rapid and Accurate High-Resolution Radar Signature Prediction of Sea Targets
Proposal No.: N131-003-0031
Firm: Delcross Technologies, LLC
3015 Village Office Place
Champaign, Illinois 61822
Contact: Robert Kipp
Phone: (312) 873-1101
Web Site: www.delcross.com
Abstract: Radar signatures of small boats and ships are inherently complex due to their myriad topside features: railings, antennas, masts, ladders, armaments, pilot stations, outboard engines, storage bins, etc. In boats, the use of fiberglass and other radar-penetrable materials for hulls and decks further complicates the radar signature by exposing interior geometry. Radar detection, identification, and tracking of watercraft depends on accurately characterizing radar signatures and their variability over anticipated operating conditions. However, at these wavelengths, watercraft can span hundreds to thousands of wavelengths, well beyond the capability of any full-wave electromagnetic solver. Asymptotic (ray-tracing) solvers do very well at quickly predicting the signatures of large, complex shapes, but they suffer in capturing the effects of detailed features measuring a few wavelengths or less. Yet, these detailed features can play an important role in the overall radar signature. We propose to develop a practical signature prediction capability for watercraft that hybridizes the solution of two mature electromagnetic modeling technologies based on ray tracing and method-of-moments. Phase I will focus on hybrid formulation development and proof of concept through numeric experiments. In Phase II, hybridization algorithms will be refined and implemented in a radar signature tool, including a commercial-grade GUI.
Benefits: Radar systems are fundamentally constrained by a variety of factors in their ability to detect, track, and identify targets, not the least of which is the radar response (signature) of the target, the response of its surrounding environment (i.e., clutter) and the electromagnetic interaction between the two. For this reason, the ability to predict target and clutter radar signatures is important in the development an assessment of radar systems for an intended application. Also for this reason, there is a long history of research and development in computational electromagnetic (CEM) techniques and tools for radar signature prediction of realistic targets described by high-fidelity CAD models and meshes. Such techniques are divided into two categories: full-wave and asymptotic. Full-wave techniques are the most accurate, being roughly direct numeric solutions of Maxwell's equations. However, they scale poorly with increasing size of the target measured in wavelengths. Asymptotic techniques (e.g., ray-tracing) are more approximate, but they scale well and simultaneously improve in accuracy with increasing target and feature size measured in wavelengths. Full-wave techniques and, to a lesser extent, ray-tracing techniques are represented among mature and commercially available tools for radar signature prediction. For radar operating frequencies ranging from S-band through Ku-band (3 - 18 GHz), most air, ground, sea, and space targets of interest are electrically large, yet containing exterior features that are electrically small to moderate. While this has motivated the development of acceleration techniques for full-wave codes that significantly improve their reach, their fundamental frequency-scaling problem remains. The need to simultaneously handle large-scale and small-scale features has also motivated over two decades of research into hybrid techniques that combines full-wave and asymptotic methods in a way that makes the best of both while overcoming their respective weaknesses. However, these remain boutique capabilities, more the realm of research papers than practical tools, because the full-wave/asymptotic interfacing requirements are too burdensome and error-prone for the typical end user to configure. From another perspective, we believe this reflects a lack of adequate effort in designing user interfaces that meet the unique and more difficult challenges presented by hybrid CEM techniques. The chief technical benefit of the proposed Phase I research is the development and demonstration of a practical full-wave/asymptotic hybridization algorithm that can handle realistic, multi-scale problem geometry at these frequencies. This development will anticipate the needs of Phase II development, as some formulations better lend themselves to intuitively deployed implementations than others. The main benefit of the Phase II project will be development of a commercial-grade tool where the necessary investment has been made in properly designing the hybridization feature so that it is intuitive and easy to deploy. It will thus offer the opportunity for improved accuracy relative to a purely ray-tracing approach in a feature that end-users will actually want to use. Such a tool has obvious commercial potential in defense engineering/analysis markets where radar signature assessment supported by high-fidelity simulation is an increasingly common activity. It is also highly relevant to the development and analysis of automotive and airborne anti-collision radar systems.