Phase-only and amplitude-only adaptive algorithms based on a D(3)LS method for smart antenna systems and airborne radar applications

Adaptive array signal processing has been used in many applications. One principle advantage of an adaptive array is the ability to recover the desired signal while also automatically placing deep pattern nulls along the direction of the interferences. Adaptive antenna array systems perform by changing amplitudes and phases of the voltages induced at each of the antenna elements. Forming such a weighted sum of the complex voltages could be slow and ineffective for large arrays. Some existing antenna systems possess the capability of using the weights as the phase only at each of the antenna elements to mitigate the undesired interference while preserving simultaneously the desired signal. For these reasons, in this dissertation we propose an adaptive methodology where the weights can be phase-only or amplitude-only. The use of these algorithms results on a faster response and simpler software and hardware design. These algorithms are based on a Direct Data Domain Least Squares (D 3 LS) approach, in which we utilize only a single snapshot of the data for adaptive processing. In most adaptive algorithms it is necessary to know something about the signal that we are trying to extract in a noisy environment. The D 3 LS approach is also available where the direction of arrival (DOA) of the signal of interest (SOI) is known a priori . Further, for Space-Time Adaptive Processing (STAP) applications, the algorithm requires knowledge of the signal Doppler frequency. Hence we need to guarantee a good estimate for the DOA and the Doppler frequency of the SOI. One of the open problems is how to provide a good estimate for the DOA of the SOI when there is uncertainty associated with the assumed DOA due to the atmospheric diffraction or measurement errors. It will be shown that the norm of the adapted weights provide a refined estimate for the actual DOA of the SOI when there are uncertainties associated with their initial estimates. Specifically, the existence of a minimum in the sum of the weights can be used further to estimate the target return angle. It could also be used to perform the detection process as well. So this thesis represents the minimum norm properties of the optimum weights. This could lead to a more accurate estimation of the direction of arrival of the signal or on a detection process, when a good estimate for the direction of arrival information is not available a priori