Simulation And Analysis Of NOMA And SCMA Non-Orthogonal Multiple Access Systems

With the commercial deployment of the 4th generation (4G) mobile communication networks, investigations on the 5th generation (5G) mobile communication technology are carrying out in full swing. In general, it appears that multiple access technology is updated in each generation of mobile communication networks. Power domain non-orthogonal multiple access(NOMA) and sparse code multiple access(SCMA) has currently attracted much attention to industry for their excellent throughput and capability of supporting overloaded users.For power domain NOMA,where multiple users are multiplexed in power domain at the transmitter, the separation of multi-user signals is conducted at the receiver based on successive interference cancellation (SIC). Based on low density signature (LDS), SCMA combines spreading and modulation together, where the incoming bits are mapped to a multidimensional codeword directly. At the receiver, the message passing algorithm (MPA) is used to eliminate the interference from other users with near-optimal performance and acceptable complexity. In this thesis, key technologies of NOMA and SCMA are studied by simulation approach.For power domain NOMA, a downlink system level simulation platform is designed and implemented based on IT++. In order to modify and extend conveniently, the modular design criterion is employed. Then user pairing schemes and power allocation algorithms are researched based on the simulation platform. For user pairing schemes, the exhaustive search scheme and group pairing scheme are considered. It is shown by simulation results that the latter is a tradeoff between throughput and complexity. Two sub-optimal power allocation algorithms are studied, i.e. fixed power allocation (FPA) and fractional transmit power allocation (FTPA). Simulation results indicate that the latter is superior to the former under various power allocation factor.For SCMA, a link level downlink simulation platform is builted, and the new multi-dimensional codebooks designed by our project team are implemented based on the simulation platform. It is demonstrated by simulation results and performance analyses that the bit error rate (BER) of the designed muti-dimensional codebooks improves significantly, compared with the existing codebooks. In addition, based on equal power allocation (EPA) and dynamic power allocation (DPA) algorithms,the BER performance of average and worst user are investigated, and it is shown that the overall average BER performance is improved when allocate more power to the user with poor channel condition.