Multiple Access Interference Cancellation in Downlink NOMA Under Imperfect CSI Based on Single-Channel Blind Separation

Non-orthogonal multiple access (NOMA) is one of the most potential multiple access techniques in the next generation wireless communications system. Compared to orthogonal multiple access(OMA), NOMA shows much better performance in spectrum efficiency and system capacity. However, there is a critical problem that lie in putting NOMA into practice, that is multiple access interference cancelation(MAIC). The most popular MAIC technique used in NOMA is successive interference cancellation (SIC). However, this MAIC method has two disadvantages. One is that interference cancellation performance depends on channel state information (CSI). Due to the influence of factors such as channel estimation errors, time-varying channel transmission, and feedback channel transmission delay, it is difficult for the receiver to obtain accurate channel state information. Second, the MAIC performance is very poor while the power of different user signals is similar. This paper proposes a multiple access interference cancellation method based on single-channel signal separation, which solves the above two problems in the downlink signal reception of the NOMA system. The simulation results show the effectiveness of the proposed algorithm.

I. Introduction

The next generation wireless communication system(5G) is expected to support unprecedented requirements for the wireless access connection, such as targeting user-experienced rate of 0.1-1Gbps, peak rate of tens of Gbps, connection density of 1 million per square kilometer, and roundtrip latency of about 1 msec, and so on. It is obvious from the development vision of the 5G system that the 5G communication system has the characteristics of high system capacity, high spectrum efficiency, and a large number of terminal connections. However, spectrum resources is becoming more and more tight, and it is difficult to realize this vision just by increasing bandwidth. In order to meet the application requirements of the 5G, the improvement of the spectrum efficiency is a problem that needs to be solved urgently. Multiple access technique is one of the most important aspects in improving the spectrum efficiency.