High Resolution Direction of Arrival Estimation for Uniform Linear Arrays in

• Main Authors: Gustav Almquist, Presence Of Radar Clutter, Saab Bofors, Dynamics Ab
• Other Authors: The Pennsylvania State University CiteSeerX Archives
• Format: Text
• Language: English
• Published: 2004
• Subjects: DML
• Online Access: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.623.5599; http://www.ee.kth.se/php/modules/publications/reports/2004/IR-S3-EX-0433.pdf

1 This Master of Science Thesis investigates the impact of radar clutter on direction of arrival estimators based on uniform linear arrays. The studied estimators are the subspace methods MUltiple SIgnal Classification (MUSIC) and Estimation of Signal Parameters using Rotational Invariance Techniques (ESPRIT) and the maximum likelihood method Deterministic ML (DML). The clutter is modelled with various distributions based on the Gaussian and Weibull probability density functions. Five different filters are constructed to reduce the effects of the clutter in the receiver, one filter with fixed coefficients and four adaptive filters. One of the adaptive filters is a whitening filter and the others are Wiener filters. The Thesis also proposes different methods based on the distributions of the received signals to classify an incoming signal as either a reflection from a target or from clutter. These methods use estimated information about the distributions of the signals to choose the most probable signal. Two of these methods are based on Gaussian Mixture Models and the Expectation Maximization algorithm. Analytical analysis and simulations show that the direction of arrival estimation isn’t dependent on the distribution of the clutter. For the direction of arrival estimation it is sufficient to describe the clutter in terms of