Analysis of methods and algorithms for controlling the direction pattern of smart antennas on moving objects

The paper presents a systematic analysis of modern methods and algorithms for controlling the directivity pattern of Smart antennas used on moving objects in dynamic conditions. The theoretical foundations of adaptive directivity pattern formation, the principles of digital beamforming (Digital Beamforming), as well as the features of the functioning of the LMS, NLMS, RLS, MUSIC, ESPRIT, PSO and GA algorithms are considered. Their advantages, disadvantages and conditions for effective application are analyzed, taking into account the speed of adaptation, computational complexity, accuracy of estimating the direction of arrival of signals (DOA) and resistance to interference. The study found that LMS/NLMS class algorithms provide simple implementation and high speed, but are limited in accuracy under conditions of intense interference. RLS methods are characterized by improved convergence, but require significant computational resources. The MUSIC algorithm demonstrates high resolution in determining the direction of arrival of signals, but is not suitable for real-time use due to the complexity of spectral analysis. Stochastic and evolutionary algorithms (PSO, GA) have shown potential for optimizing the phase and amplitude parameters of antenna elements, but have a low convergence rate under variable conditions. The results of the analytical review indicate the relevance of the task of creating combined and adaptive approaches that combine high accuracy of assessment with operational response to environmental changes. The conclusions obtained can be used as a theoretical basis for further research in the direction of developing new methods for controlling the directivity pattern of Smart antennas on moving objects, in particular using machine learning and intelligent optimization.