盲接收条件下单信道时频混叠信号的调制识别

Modulation Recognition for Time-Frequency Overlapped Signals in Single-Channel under Blind Reception Conditions

针对现有调制识别算法对单信道时频混叠信号失效问题,从信号循环平稳域出发,通过分析不同调制信号自身具有以及做非线性变换处理衍生出的二阶循环平稳结构特征,结合循环平稳性检测,提出了一种新的适用于单信道时频混叠信号的调制识别算法。该算法无需完成任何参数估计以及同步等预处理过程。对一些典型常用调制信号(如BPSK、QPSK、16QAM、OQPSK等)的随机混合,能够同时有效辨识混叠信号中所包含的信号个数以及每个信号的调制类型,理论推导及仿真结果表明,该算法可在较低信噪比下实现对单信道时频混叠信号的调制识别,有效提高了盲接收条件下对QPSK/16QAM时频混叠信号的分类能力。

Considering the current modulation recognition algorithms losing effectiveness for time-frequency overlapped signals received by single channel,through analyzing the second order cyclostationary structural characteristics owned by different modulated signals or derived from the nonlinearities change,a novel method based on cyclostationary test is proposed for time-frequency overlapped signals in single-channel. The method does not need any preprocessing tasks,such as estimation of any parameters and synchronization. It also can effectively identify the source number and modulation type of each signal when the received signal is the random mixture of several typical common modulation signals（ BPSK、QPSK、16QAM、OQPSK etc.） simultaneously. Theoretical analysis and simulation result show that the method can realize the modulation identification for time-frequency overlapped signals in single-channel with low SNR,and effectively promote the classification capability for QPSK /16 QAM time-frequency overlapped signals under blind reception conditions.