This paper is presented for performance evaluation of different adaptive beamforming algorithms in smart antenna system (SAS). These algorithms can be divided into Blind، Non-Blind and Constrained Minimum Variance Algorithms. This work introduces analytical detailed study of adaptive beamforming algorithms، which are classified Least- Mean Squares (LMS)، Normalized LMS (NLMS)، Variable Step Size LMS (VSSLMS)، Recursive Least-Squares (RLS) and Conjugate Gradient Method (CGM) as Non-blind algorithms. Blind algorithms are classified Constant Modulus Algorithms (CMA) and Recursive Least-Squares Constant Modulus Algorithm (RLSCMA). In addition to Constrained Adaptive Algorithms or Constrained Minimum Variance Algorithms، which are classified Linearly Constrained Minimum Variance (LCMV)، Multiple Constrained MV (MCMV) and Complementally Transformed Minimum Variance (CTMV). Comparison between these algorithms is performed in terms of the interference rejection capability and number of iterations required to get to the steady state or rate of convergence. Depending on results of comparison، suitable or optimal algorithms are selected in smart antenna system according to radio situation and purpose of this SAS. Results of simulations in MATLAB are presented for performance illustrations.

This paper indicates importance of algorithms of estimation direction of arrival (DOA) of incoherent signals in smart antenna system (SAS). DOA algorithms can be divided into three basic categories, namely, conventional (classical), subspaces methods, and parametric methods like maximum likelihood (ML) techniques. The paper introduces, using Uniform Linear Array, brieflly study of conventional methods, which include Delay-and-Sum and Capon's Minimum Variance, and analytical detailed study of subspaces methods, which include MUSIC, Root-MUSIC and ESPRIT. This paper compares between subspaces algorithms and determines advantages and disadvantages in terms of DOA accuracy, resolution and execution time or computation efficiency. Results indicate that subspaces methods outperform conventional methods. In addition, ESPRIT and Root-MUSIC outperform MUSIC according to DOA estimation accuracy and resolution, but MUSIC outperforms Root-MUSIC and ESPRIT when DOAs are placed on grid of angular scanning. MUSIC scans all the angles and step of angular scanning in MUSIC decides the estimation accuracy and resolution. A small step of scanning can improve the accuracy and resolution ability, but increase execution time. As well as, ESPRIT is the best due to estimation execution time. Depending on results of comparison, we can choose which of these algorithms (suitable or optimal algorithm) in smart antenna system according to radio situation and purpose of this SAS. All mentioned algorithms are simulated using MATLAB. Results of simulations support the theoretical expressions.