Convergence rate of the dominant mode rejection beamformer for a single interferer

The dominant mode rejection (DMR) adaptive beamformer (ABF) is a reduced-rank version of the standard minimum variance distortionless response (MVDR) ABF. DMR uses a structured estimate of the covariance matrix derived from an eigendecomposition of the sample covariance matrix. This paper exploits results from random matrix theory to fit a distribution for the SINR loss of the DMR ABF in the single interferer case. Monte Carlo simulations confirm the accuracy of these expressions. SINR loss quantifies the rate at which the performance of the DMR ABF converges to the performance of the optimal processor. For the single interferer case, the mean SINR loss only depends on the number of snapshots.