Side Angle Information Assisted Near-Field Beam Training for XL-Array Communications

In this letter, we propose efficient near-field beam training design for extremely large-scale array (XL-array) communication systems to establish high-quality communication links before data transmissions. Specifically, different from the existing near-field beam search methods where wireless sensing has not been well exploited, we propose in this letter two new beam training schemes by effectively utilizing side angle information obtained from anchor nodes/sensors around the user. Specifically, Scheme 1 searches for the user angle in a reduced angular domain within the neighborhood of the side angle. In contrast, Scheme 2 first applies an adaptive beam search to progressively determine the candidate user angles, and then uses a customized polar-domain codebook to determine the user range. Finally, numerical results show that 1) our proposed two schemes significantly reduce the training overhead of various benchmark schemes, while achieving comparable beamforming performance; and 2) Scheme 2 is generally more robust to noise than Scheme 1.