Reduced-Dimension Reiterative Superresolution

Adaptive filtering via reiterative superresolution (RISR) has been shown to be an effective technique for mitigating the impact of target masking due to sidelobe leakage; however, the associated high computational cost may prevent application when compute resources are constrained. This paper presents a reduced-dimension RISR technique that first performs static filtering on subsets of data to reduce the adaptive degrees of freedom, then adaptively combines these outputs to achieve results approaching that of full-dimension RISR filtering. By reducing the adaptive degrees of freedom, the computational cost of the algorithm is reduced significantly. The reduced-dimension RISR filter formulation described is agnostic to the domain in which it is being applied, allowing for application to any sampled domain.