The Study Of Interference Cancellation In MC-CDMA Systems

Multi-carrier code division multiple access (MC-CDMA) system which combines the technologies of code division multiple access (CDMA) and orthogonal frequency division multiplexing (OFDM) has advantages of high spectrum efficiency, high-speed data transmission and robustness against frequency selective fading. It is a promising candidate in the future multi-carrier transmission, but it also inherits the disadvantage of these two technologies.Multiple access interference (MAI) is the key factor which deteriorates the capacity of CDMA systems. The key disadvantage of OFDM is that the system is sensitive to the carrier frequency offset. Even slight one will destroy the orthogonality among sub-carriers, cause inter-carrier interference (ICI), and degrade the system performance. MAI and ICI are two main interferences in MC-CDMA systems, and the study of interference cancellation in MC-CDMA systems is an academia research hotpot.This paper focuses on the cancellation of MAI and ICI in MC-CDMA systems. Firstly, a signal model of MC-CDMA systems in the case of carrier synchronization is derived. Using the idea of multi-user detection (MUD) in CDMA systems, the MUD in MC-CDMA systems is considered as a combinational optimization and discrete particle swarm optimization(DPSO)algorithm is applied to acquire the optimal value.The DPSO algorithm is easy to carry out, and has few parameters to set, but it is also easy to fall into the local convergence and its rate of the convergence is not fast enough. In order to solve these problems, the Hopfield neural network (HNN) is applied to DPSO to increase the diversity of particles and improve the ability of optimization and the rate of convergence. So a method of MUD for MC-CDMA systems based on PSO with HNN is proposed. Simulation results show that the performance of the proposed MUD method is not only better than that of the MUD method based on PSO, but also better than that of the MUD method based on HNN.Secondly, the received signal model of MC-CDMA systems in the case of existing carrier frequency offset is analyzed. The signal model is different from that in OFDM because of the multi-user symbols. Therefore, the frequency offset estimation algorithm in OFDM systems is not fit for MC-CDMA systems, it needs some improvements. The maximum likelihood estimation of frequency offset based on the maximum likelihood principle is derived, it can be viewed as a consecutive optimization, and PSO algorithm is applied to solve this problem. The frequency offset estimation method for MC-CDMA systems based on PSO and pilot symbols is presented. Simulation results show that the performance of the proposed scheme is better than that of the frequency offset estimation based on virtual carriers (VC).Finally, in order to simultaneously combat both of the MAI and ICI in MC-CDMA systems, a joint scheme which estimates frequency offset and multi-user symbols is proposed. The estimation expressions of frequency offset and multi-user symbols are derived based on the maximum likelihood principle, and then the PSO algorithm with a compression factor is applied to gain optimal value. Simulation results show that the performance of bit error rate (BER) of the proposed MUD is lower than that of MUD which doesn't take the frequency offset compensation into account. The mean square error of the proposed frequency offset estimation nearly reaches that of frequency offset estimation based on PSO and pilot symbols, however, the proposed scheme doesn't require any training sequences and pilot symbols, it has high spectrum efficiency.