Attenuation Modelling and Machine Learning Based SNR Estimation for 5G Indoor Link

Abstract

A significant number of propagation models have been proposed for 5G communication systems in recent years. Among the different environments studied, indoor propagation environments have emerged in importance. The present paper reviews current research in indoor propagation modelling at different frequencies relevant to 5G signal propagation. The paper goes on to present a model for attenuation of 5G signals in the X band. The model takes into account the variation of signal attenuation due to varying number of human bodies and other obstacles present in the indoor environment at different times of a day, leading to time-dependent difference in signal to noise ratio (SNR). A non-linear polynomial based machine learning technique is then used to obtain a least-squares (LS) estimate of SNR from the model.