Non-Orthogonal Multiple Access (NOMA) with asynchronous interference cancellation

Non-orthogonal multiple access (NOMA) allows allocating one carrier to more than one user at the same time in one cell. It is a promising technology to provide high throughput due to carrier reuse within a cell. In this thesis, a novel interference cancellation (IC) technique is proposed for asynchronous NOMA systems, which uses multiple symbols from each interfering user to carry out IC. With the multiple symbol information from each interfering user the IC performance can be improved substantially. The proposed technique creates and processes so called "IC Triangles". That is, the order of symbol detection is based on detecting all the overlapping symbols of a stonger user before detecting a symbol of a weak user. Also, successive IC (SIC) is employed in the proposed technique. Employing IC Triangles together with the SIC suppresses co-channel interference from strong (earlier detected) signals for relatively weak (yet to be detected) signals and make it possible to achieve low bit error rate (BER) for all users. Further, iterative signal processing is used to improve the system performance. Employing multiple iterations of symbol detection which is based on exploiting a priori estimate obtained from the previous iteration can improve the detection and IC performances. The BER and capacity performance analyses of an uplink NOMA system with the proposed IC technique are presented, along with the comparison to orthogonal frequency division multiple access (OFDMA) systems. Performance analyses validate the requirement for a novel IC technique that addresses asynchronism at NOMA uplink transmissions. Also, numerical and simulation results show that NOMA with the proposed IC technique outperforms OFDMA for uplink transmissions. It is also concluded from the research that, in the NOMA system, users are required to have large received power ratio to satisfy BER requirements and the required received power ratio increases with increasing the modulation level. Also, employing iterative IC provides significant performance gain in NOMA and the number of required iterations depend on the modulation level and detection method. Further, at uplink transmissions, users' BER and capacity performances strongly depend on the relative time offset between interfering users, besides the received power ratio.