A Robust Two-Stage Uplink Equalizer Under Hardware Impairments in Non-Terrestrial Networks

Abstract

Applying extremely large-scale antenna arrays (ELAA) to non-terrestrial networks (NTNs) can significantly enhance uplink communication performance while supporting direct communication using lightweight user equipments (UEs) with low transmit power. However, ELAA poses challenges to the consistency calibrations of the receive amplitudes and phases between different antennas. Antenna calibration failure invalidates position-based beamforming weights for uplink reception. Perantenna channel-estimation-based equalization is possible, but it is computationally expensive in ELAA and vulnerable to low received SNR. Therefore, this paper proposes a robust two-stage uplink equalizer for NTNs under hardware impairment. The first stage estimates the actual end-to-end channel differences relative to a reference antenna using data and pilots with low complexity to improve the received SNR. The second stage estimates the combined channel using the SNR-improved pilots to construct the final normalized weight. Simulation results verify that the proposed method outperforms a complexity-reduced channel estimation based equalizer in terms of bit error rate.