Sparse Direction-of-Arrival Estimation in Antenna Arrays with Practical Limitations

The process of identifying the direction from which a radio frequency (RF) wave or other types of waves is coming from is called Direction of Arrival (DOA) estimation. DOA estimation applications are making it an area of continuous interest. There are many applications where DOA is already used such as in smart antennas, mobile terminal localization, and interference cancellation among many more. There are different DOA estimation techniques and each has its own advantages and disadvantages as well as applications. When DOA algorithms are implemented, several factors that limit its performance should be investigated, especially when considering actual antenna types and their use in future mobile terminals and wireless access points that will be limited in terms of physical dimensions, storage and processing. This research work address these physical limitations like the limitation on the number of antenna elements within the array, spacing between them and most importantly the effect of real antennas compared to isotropic radiators which the current literature only focus on. We will also use the sparse framework representation of the problem to reduce the processing required to estimate the DOA. This field has attracted more attention with newly published works utilizing Compressive Sensing (CS) in DOA estimation. Moreover, the use of coprime and nested arrays is attracting more researchers as viable antenna array solutions for such DOA methods. We are investigating the use of these sparse processing algorithms in the presence of the physical limitations. The main objective is to strike an optimal balance between complexity, actual implementation limitations and accuracy. The proposal is to evaluate the performance of different DOA algorithms namely: Beamforming, Capon, MUSIC, l_1- Singular Value Decomposition (l_1-SVD) and compare their performance. All algorithms will be examined in sparse framework for practical cases where we have limited number of antennas and limited spacing in between. Actual antenna elements will be used taking into account their radiation properties and behavior. The results will be based on theoretical simulation models and augmented with practical prototyping of the proposed antenna array for the purpose of DOA estimation optimization.