| The estimation of the Direction-of-Arrival (DOA), also known as direction finding problems, has been an active research area for the last couple of decades. While one DOA estimation method may be better than the other depending on the nature of the applications, all methods can be categorized into either subspace-based methods or beamformer-based methods. Subspace-based methods are known to provide higher resolution but most of them work efficiently with relatively high signal to noise ratio (SNR). For low Array-Signal-to-Noise-Ratio (ASNR), however, their performance degenerates in a similar way as the conventional beamformer-based methods do. In this dissertation, we introduce a new method for the DOA estimation for extremely noisy environment (ASNR = {dollar}-{dollar}10 dB), which we refer to as 'MaxMax' method. MaxMax method is partially based on the eigenstructure of the temporal data covariance matrix constructed at the end of the beamformer while it scans the entire directions, and since it does not entirely rely on the eigenstructure of the covariance matrix like subspace-based methods, its performance becomes more distinct especially for extremely low ASNR environment with small number of sensors. MaxMax method also provides the accurate estimates of the number of signals far beyond the limitation of the conventional methods. Simulation results show that MaxMax method yields superior DOA estimates to those of the subspace-based methods and the conventional beamformer-based methods while providing the outstanding estimates of the number of signals via a novel sampling technique 'spatial sampling.'... |
