Joint Mechanical-Electrical Beam Steering for Maritime-LEO Communications: Optimizing Performance Through ESPRIT-Based DOA Estimation

Non-terrestrial networks, particularly satellite communications, provide ubiquitous connectivity in next-generation wireless systems. As signals from terrestrial networks cannot reach vessels far from shore, satellite communication is critical in maritime environments. Accurate beam alignment is essential for maintaining high throughput in maritime-satellite communications, especially under continuous vessel rotation. Variations in the direction of arrival (DOA) and estimation’s computational complexity significantly affect estimation accuracy and beam-pointing performance. This paper proposes a joint mechanical-electrical beam steering (JBS) approach that integrates electrical beam steering based on estimation of signal parameters via rotational invariant techniques (ESPRIT) with mechanical beam steering. The combined method reduces DOA variation and enhances estimation accuracy. An analytical model for two-dimensional ESPRIT is presented, incorporating the two-dimensional Cramér-Rao bound to derive the optimal satellite direction and doublet spacing for electrical steering. The proposed optimization function accounts for ESPRIT estimation accuracy, antenna gain, and vessel rotational motions. Analysis results show that the analytical model of ESPRIT closely matches simulation results. Furthermore, the proposed JBS algorithm significantly enhances the signal-to-interference-plus-noise ratio (SINR), DOA estimation accuracy, and beam-pointing accuracy compared to standalone mechanical or electrical beam steering approaches. The beam-pointing root-mean-square error for the elevation angle is reduced from 1.9684 to 0.1276 degrees, and for the azimuth angle from 7.3975 to 0.6761 degrees, when comparing the proposed JBS against standalone mechanical beam steering. This substantial improvement in alignment accuracy results in a significant improvement of the post-processing SINR, maintaining values above 0 dB with the proposed JBS, whereas the standalone mechanical approach experiences a severe drop to below −60 dB.