Bit Error Rate Evaluation of SCMA in Rician Fading Channels Considering Presence of Co-Cell Interference

Abstract

Non-orthogonal multiple access (NOMA) techniques have been proposed for fifth generation (5G) mobile systems, among them stands out sparse code multiple access (SCMA), which is a code domain NOMA (C-NOMA) technique. With SCMA, the system can be overloaded, that is, the number of active terminal devices can be greater than the number of frequency channels available in the system. Several works evaluate its performance in single-cell scenarios. However, this technique will operate in cellular environments where there is interference from co-cells. Therefore, an overloaded system can represent high levels of co-cell interference (CCI). Hence, this work evaluates the SCMA performance, in terms of the bit error rate (BER), considering the uplink of a cellular system that uses power control and operates in presence of CCI in a Rician fading channel. A 16-point SCMA configuration with 4 subcarriers and 6 users is used as part of the system model. Besides, the channel model assumes presence of additive white Gaussian noise (AWGN) and exponential path-loss. Monte-Carlo simulations are carried out to evaluate the mean BER of a group of users in a cell of interest. Results show that SCMA receiver is highly affected by CCI. Thus, if the channel reuse factor is small, the mean BER is extremely high since CCI produces BER floors which cannot be reduced if the signal-to-noise ratio is increased. However, if the line-of-sight (LOS) component power is increased for the users of interest, the system performance improves.