Compressive sensing of up-sampled model and atomic norm for super-resolution radar

Compressive sensing (CS) for radar signal processing is known to be capable of various applications. This signal processing technique shows excellent performance for detecting objects. However, the grid problem of CS is an obstacle to more precise performance. In this paper, we introduce two methods to overcome this grid problem and evaluate the performance of the methods. The first method is an up-sampled model, which is a method of dividing the grids into smaller pieces. The second method is an atomic norm minimization, which is a detectable method for continuous parameters.