Thinned Knowledge-Aided STAP by Exploiting Structural Covariance Matrix

What is it about?

Covariance matrix estimation is vital in space-time adaptive processing. Exploiting the structural information of the covariance matrix can achieve performance enhancement, as compared to the well known sample covariance matrix (SCM). This paper presents a unified framework to analyze the performance (SCNR loss) with linearly structured covariance estimates, and a novel asymptotically efficient estimate.

Featured Image

Why is it important?

1) Previous works validate the merits of exploiting the structured covariance matrix by the simulations. This paper presents a method to analyze the SCR loss, which can determine the required training size. This method can be used for popular structures, such as Toepltiz, Toeplitz-Block-Toeplitz, Kronecker, Persymmetric. ciculant structured CM. 2) We propose a novel asymptotically efficient covariance matrix estimate.

Perspectives