Long-range Target Identification

The objective of this research is to develop a target-recognition approach which is robust in challenging conditions that include long stand-off distances, low-visibility environments, and non-cooperative targets. The key element is an effective use of data from continuous surveillance, that is, redundant data that can in principle be used to improve resolution in selected target areas. Monopole Research and its consultants have extensive experience and expertise in (i) large scale numerical simulation of electromagnetic propagation phenomena,(i) array and distributed sensor imaging in complex environments, and (iii) optimization and statistical methodology for systematically improving resolution of images in clutter. These are areas that are directly relevant to long-range target imaging in complex environments. The focus of the Phase I research will be to assess in depth the use of multiple frames over relatively long time intervals in order to improve resolution in imaging three dimensional moving targets under a variety of conditions. Both zero offset (SAR) and extended offset, numerically generated array data will be used to assess the scope and effectiveness of the imaging algorithms. We will address in detail the following issues, which form the scope of the research effort (i) How robust multi-frame image resolution improvement is to estimation errors in target and/or array motion. (ii) How well we have to estimate residual environmental effects, between the array and the moving target, so loss of image resolution can be adaptively reduced, especially when using multiple frames. (iii) Do computationally expensive interferometric array imaging methods enhance resolution sufficiently in multi-frame imaging so that the development of fast algorithms for their implementation can be undertaken. BENEFIT: The developed algorithms and the corresponding numerical simulations will present a significant step towards identifying a robust target recognition approach which would make an efficient use of data from continuous surveillance and which would lead to extract, with improved resolution, characteristic three-dim features of non-cooperative targets. We believe, the algorithms developed under this contract will provide the basis for a commercial package of interest to numerous Government agencies, as well as to the private sector.