Processing of multi-aperture SAR to produce fine-resolution images of arbitrarily large extent

The spatial extent that can be accurately imaged by a synthetic aperture radar (SAR) is unfortunately limited. This restriction results from the fact that the number of independent pixel estimates produced by a SAR is constrained by the number of independent measurements that it collects. The solution is to increase the number of independent sensor measurements, without modifying sensor resolution. This can be accomplished by subdividing the SAR aperture into an array of receiver elements, each with a coherent receiver. This sensor collects information over both time and space, and thus space-time processing must be implemented. Three SAR processors are evaluated: a correlation processor, a maximum likelihood estimator, and a minimum mean-squared error estimator. Simulations demonstrate that all three processors can create fine resolution SAR images over an arbitrarily large area, provided that the number of array elements is sufficiently large. Additionally, it is shown that the ML and MMSE processors provide quality SAR images from a sparse receiver array, such as those produced by a “formation-flying” collection of radar satellites