Application of compressive sensing to refractivity retrieval with a network of radars

The emerging technology of compressive sensing (CS) has been recently applied to many fields to effectively reconstruct high quality images using much fewer number of samples than those in conventional image reconstruction. In this work, the CS framework is applied for the first time to retrieve refractivity field from a network of weather radars. It has been shown that radar-derived refractivity field can be used as a proxy for near-surface moisture and has the potential to improve the prediction of convective initialization and understanding of other weather phenomena. The problem of refractivity retrieval is formulated for networked radars and is solved using CS with the goal of robust reconstruction. Moreover, the feasibility of compressing sensing for refractivity retrieval is demonstrated and verified using simulation. Preliminary results have shown qualitatively and quantitatively that CS technique performs better than the conventional constrained least square (CLS) approach and is less susceptible to noise contamination.