Separation of co-frequency mixed signals in single channel is a challenging problem, especially for high-order MPSK modulation signals. A separation approach of mixed signals in single channel is proposed for co-frequency high-order MPSK signals, based on the cyclo-stationarity of digital communication signals and low-rank semi-definite relaxation (SDR). Firstly, the problem of mixed signals separation in single channel is formulated as a rank-1 semi-definite programing (SDP), which can be effectively solved by alternative variable algorithm. Then, by introducing decision feedback equalization (DFE), a separation method of mixed MPSK signals is proposed, which is based on the combination of DFE and low rank SDR. The complexity of the proposed method is polynomial with the length of channel and has linear relation to the modulation order. Simulation results show the effectiveness of the proposed method for high order MPSK signals under the condition of the medium or high SNR.%单通道同频混合信号的分离是一个具有挑战性的问题,尤其是对于高阶MPSK调制信号.利用数字通信信号的循环平稳性和低秩半定松弛的原理,提出一种单通道同频混合的高阶MPSK信号的分离方法.首先,将单通道同频混合信号分离问题转化为秩为1的半定规划问题,并通过交替变量法对其进行求解;然后,引入判决反馈均衡,提出一种判决反馈均衡和低秩半定松弛优化相融合的同频混合信号的分离方法.方法的复杂度与等效信道长度呈多项式关系,且与信号的调制阶数线性相关.仿真结果表明,在中高信噪比条件下,所提方法能够有效分离同频的高阶MPSK混合信号.
展开▼
