Advanced Radar Data Fusion - Multiple Input Multiple Output Sensor Acquisition (MIMOSA) System

Radar targets provide a rich scattering environment that produces from 5 to 25 dB target fluctuations. The spatial dependence of target scattering has been understood for some time and it has been shown that targets produce essentially independent scattering returns when radiated from sufficiently different directions. Thus, if receivers are distributed over a wide enough area, which is the premise of MIMO radar, we can exploit the angular spread of the scattering to obtain improved performance by combining these independent looks at the target. This is essentially what is known as diversity gain in MIMO communications and it is exploited extensively in space-time coding and related MIMO techniques. Effectively, we exploit what is normally considered to be a deficiency, target fluctuations, to improve performance. In fact, other effects that are also normally considered to hurt performance, such as reflections of the returns off the ground or other objects, can also be exploited to improve performance in a similar way. The proposed MIMOSA concept will exploit MIMO technologies leading to the development of a new capability to autonomously collect, process, and fuse information from a variety of radars (either at the same or different frequencies and multiple perspectives) and from other sensors to form a single integrated picture of the battlespace. The solution will exploit MIMO technologies originally designed for RF communications applications. This approach will be used to more accurately and reliably support acquisition, track, discrimination, and engagement objectives of threatening objects across a spectrum of threat classes and environments. This novel and synergistic MIMO-based approach shows promise of providing a more accurate picture of the adversary threat cloud than any single sensor or group of sensors operating independently can offer. In order to achieve this, the fusion of data at several levels will be necessary. This will include methods for fusing multi-sensor data for 3-D imaging to discriminate targets based on multiple radar returns as well as data collected by airborne or space borne IR sensors and active LADAR devices. Further fusion of data with other radar or a-priori data will also be supported. Software algorithms and hardware that enable this synergistic fusion and interpretation of data from disparate ground-based and overhead sensors (air or space-based IR and LADAR) are expected to enhance system acquisition, tracking and discrimination of threat objects in a cluttered environment and provide enhanced situational awareness. The proposed R&D will investigate: (i) the deployment of MIMO-based radars that take advantage of antenna/spatial/frequency diversity for target detection/tracking and feature exploitation, (ii) use of robust feature aided tracking (FAT) algorithms; (iii) application of multi-modality sensor data registration/fusion schemes to support real-time performance; (iv) (vi) spatio-temporal sensor calibration to reduce bias errors; and (v) the development of 3-D imaging techniques for visualizing multiple targets and fused target track data.