Robust direction-of-arrival estimation in the presence of faulty elements in a uniform linear antenna array

Abstract

Many direction-of-arrival (DOA) estimation algorithms require accurate measurements from all sensing elements on an antenna array. However, it would be practical to perform DOA estimation even in the presence of faulty elements, as well as to identify the locations of these faults in the array. In this work, we are interested in understanding how the redundancy of a uniform linear array can be used to robustly perform DOA estimation in the presence of sensor failures. Particularly, we look at new formulations for antenna array measurements to account for the presence of faulty elements, new algorithms for fault detection and robust DOA estimation, and regimes for which robust DOA estimation can be achieved. In this thesis, we provide three different approaches to the problem (based on sparse array interpolation, sparse signal recovery and penalized-weight alternating methods), each applicable to different regimes for the problem. The performance of the respective approaches for robust DOA estimation are demonstrated through computational simulations.