【摘要】 本文研究了块衰落信道下非正交多址系统(Non-orthogonal Multiple Access,NOMA)中基于Raptor码的两用户下行传输系统.基站将要发送给两用户的信息先分别进行Raptor码编码及调制,再将其按一定功率分配因子进行叠加后广播给两个用户.各用户选择采用串行干扰消除(Successive Interference Cancellation,SIC)或直接利用置信传播算法(Belief Propagation,BP)恢复所需信息.为了进一步提升系统性能,本文针对基站仅具有信道统计状态信息下,以最大化系统平均下行速率为目标,利用外信息(Extrinsic Information Transfer,EXIT)分析,联合优化了两用户的Raptor码度数分布以及基站叠加编码功率分配因子.本文所提出的优化方法不需要实时信道状态信息,因此可以减少系统反馈开销.仿真结果显示,本文提出的优化方法能够降低系统误码率(Bit Error Rate,BER)以及提升系统平均下行速率.更多还原

【Abstract】 This paper studies two-user dow nlink transmission based on Raptor codes in a non-orthogonal multiple access system(NOMA) under block-fading channel. At the base station,the message intended for each user is encoded by Raptor code and then modulated. The signals for both users are superimposed with a certain power allocation factor and the resulted signals are broadcast to both users. Each user either uses Successive Interference Cancellation(SIC) or directly applies the Belief Propagation Algorithm(BP) to recover the required message. In order to enhance the system performance further,under the assumption of solely statistical channel state information(CSI) at the base station,we aims at maximize the average dow nlink system throughput over all possible channel states.We resort to external information(EXIT) analysis to jointly optimize the Raptor code degree profiles for both users and power allocation factor. Instant CSI is not required by the proposed optimization scheme,therefore,the system feedback overhead can be reduced.It can be show n by simulation results that the proposed scheme can reduce the system bit error rate(BER) and improve the average system dow nlink throughput.更多还原