SNR Estimation for 5G Satellite-to-Ground Communication in Ultra-Low SNR Environment Based on Channel Frequency Response Reconstruction

Accurate estimation of transmission parameters is indispensable in 5G satellite-to-ground communication systems. However, the tricky problem in multipath fading channels is to accurately estimate the average signal-to-noise ratio (SNR) within the operating frequency band. In this work, a novel SNR estimation method is proposed based on the channel frequency response (CFR) reconstruction. The noise variance is obtained based on the variance of CFR difference before and after the least squares discrete Fourier transform (LS-DFT) algorithm, while the signal power is calculated based on the power difference of CFR and the noise variance. In addition, a numerical simulation is performed to evaluate the effectiveness of the proposed algorithm in real satellite-to-ground communication scenarios. The analysis reveals that the proposed algorithm achieves the normalized mean square errors (MSE) of −13.32 dB observed under the ultra-low SNR condition of −15 dB for the non-terrestrial network (NTN) tapped delay line (TDL-B) channels. These results indicate a significant enhancement of 4 dB compared to the maximum likelihood (ML), Boumard’s and Zhang’s methods.