Wage, Kathleen EKamaraju, Sai Kasyap2017-12-072017-12-07https://hdl.handle.net/1920/10813In sonar array signal processing, one of the essential objectives is mitigating the effect of loud non-stationary interfering sources. The Dominant Mode Rejection adaptive beam- former is a beamformer that is often used to detect low powered signals in the presence of high power interferers. The DMR beamformer constructs its weights using a structured co- variance estimate which is obtained by doing the principal component analysis of the sample covariance matrix (SCM). Notch depth is a parameter which quantifies how well a beam- former can attenuate an interferer. In prior work, Buck and Wage developed an analytical model that predicts average DMR notch depth as a function of interferer strength and the number of snapshots available to estimate the eigendecomposition of the SCM [IEEE Stat. Sig. Proc. Workshop, 2012]. The DMR notch depth model is based on results from random matrix theory on the accuracy of the eigenvectors of the SCM. This thesis assesses the validity of the notch depth model using data recorded with a vertical line array during the 2010 Philippine Sea Experiment. In the PhilSea10 analysis the DMR beamformer removes loud mechanical vibration noise (array strum) from the data set. Results show good agreement with theory when the experimental data satisfies the assumptions of the model.enRandom matrix theoryNotch depthArray signal processingDMR BeamformerPhilSea10BeamformingThesis