对线性调频-二进制相位编码(LFM-BPSK)复合调制信号的参数估计,现有方法在低信噪比(SNR)环境下估计精度较低。针对该问题,文中推导出了LFM-BPSK 信号的Born-Jordan(BJD)时频分布,分析了LFM-BPSK 信号的时频特性,并提出了基于BJD 时频分布的参数估计算法。首先,该算法在BJD 时频变换的基础上使用霍夫变换对LFM-BPSK 信号初始频率和调频斜率进行估计,并将二维的BJD 时频变换转化为一维时域信号,通过降维后的时域信号求解码元宽度;然后,为了探寻参数估计的理论下界,文中还推导了LFM-BPSK 信号参数估计的修正克拉美罗下界(MCRLB);最后,利用蒙特卡洛实验对三个参数进行了测试。仿真实验表明,文中算法在低SNR 下的参数估计精度明显优于现有的算法,其性能接近MCRLB。

The existing methods for parameters estimation of hybrid modulated signal combining linear frequency modulation and binary phase shift keying (LFM-BPSK) have low estimation accuracy in low signal-to-noise ratio (SNR) environments. In order to solve this problem, the Born-Jordan (BJD) time-frequency distribution of LFM-BPSK signal is derived in this paper, and the time-frequency characteristics of LFM-BPSK signal are analyzed, furthermore a parameter estimation algorithm based on the BJD time-frequency distribution is proposed. First, the initial frequency and frequency modulation slope of the LFM-BPSK signal are estimated using the Hough transform based on the BJD time-frequency transform, and the two-dimensional BJD time-frequency transform is converted into a one-dimensional time-domain signal, and the coder width is then calculated using the time-domain signal after dimension reduction. Then, in order to explore the theoretical lower bound of parameters estimation, the modified Cramer-Rao lower bound (MCRLB) for LFM-BPSK signal parameter estimation is also derived. Finally, three parameters are tested using Monte Carlo experiments. Simulation results show that the parameters estimation accuracy of the proposed algorithm is significantly better than existing algorithms at low SNR, and the performance is close to that of MCRLB.

TN971.+1