Comparison of superresolution algorithms with different array geometries for radio direction finding
Jenn, David C.
Pace, Phillip E.
MetadataShow full item record
The objective of this thesis is to investigate and evaluate the effectiveness of modern estimation methods with different array geometries as they apply to the problem of bearing estimation. The algorithms were selected from those that apply to multidimensional direction finding, and include MUSIC, PHD, minimum norm, and Capon's beam-former. These four techniques based on their high resolution capability, and their ability to deal with three-dimensional non-uniform arrays and to estimate both azimuth and elevation angle of arrival (AQA). Computer simulations were run linear arrays, circular array, and combinations of the two. The test conditions included: (1) two closely spaced emitters, and (2) various levels of additive white Gaussian noise.
Approved for public release, distribution is unlimited