On the performance of ECF-based multi-threshold receiver in NOMA systems for vehicular communications with unknown impulsive noise

Abstract

In this paper, we derive an analytical expression for the bit error probability of a multi-threshold (MTh) detector in the downlink three-user non-orthogonal multiple access (NOMA) based vehicular communication system with unknown noise. The MTh detector directly detects the signal of vehicle user with the highest channel gain without detecting the signals of weaker users and removing their effects. It is shown that the bit error probability performance of the MTh detector is superior to that of the successive interference cancellation (SIC) detector in a special case. Then, we propose a blind empirical characteristic function (ECF) based method to estimate the signal levels of vehicle users, required to implement the MTh detector, in unknown noise. Next, we derive an analytical expression for the variance of the proposed ECF-based estimator which is validated via computer simulations. It is shown that the ECF-based estimator is asymptotically unbiased and consistent. Furthermore, we obtain the normalized Cramér-Rao lower bound (NCRLB) for the estimators of signal levels which shows that the ECF-based estimator is almost efficient in small generalized signal-to-noise ratio (GSNR) values. Numerical results show that the ECF-based MTh detector outperforms the absolute median-based SIC detector in the mixture of Gaussian and α-stable noise. © 2020 Elsevier Inc.