基于单矢量水听器时频声强法的非合作水声跳频信号多维参数估计

Multi-dimensional Parameter Estimation of Non-cooperative Underwater Acoustic Frequency-hopping Signal Based on Time-frequency Acoustic Intensity Method of Single Vector Hydrophone

针对传统方法对水下非合作跳频信号的多维参数估计效果不佳的问题,提出一种基于单矢量水听器时频声强法的非合作水声跳频信号空时频参数估计方法。利用矢量水听器声压通道与振速通道的接收信号建立时频域的短时互谱分布,分析噪声与信号在时频域分布的差异,通过迭代去噪法与连通域方法,对噪声进行抑制并平滑每跳的能量分布;采用复声强法对短时互谱矩阵进行处理,得到清晰稳定的跳频信号空时频分布;提取空时频分布的脊线并利用Haar小波变换对脊线进行奇异点检测,估计出信号的多维参数。仿真和水池试验结果表明,当信噪比大于-5 dB时,所提方法得到的参数估计归一化均方根误差小于0.05。

Aiming at the problem that the traditional methods are not effective in estimating the multi-dimensional parameters of underwater non-cooperative frequency-hopping(FH)signals,an estimation method of space-time-frequency parameters of non-cooperative underwater acoustic FH signal based on time-frequency intensity method of single vector hydrophone is proposed.The short-time cross-spectrum distribution in time-frequency domain is established by using the received signals of sound pressure channel and vibration velocity channel of vector hydrophone.The difference of noise and signal distributions in time-frequency domain is analysed,and an iterative denoising method and a connected domain method are used to suppress noise and smooth the energy distribution of each hop.A clear and stable space-time-frequency distribution of FH signals is obtained by processing the short-time cross-spectrum matrix using the complex acoustic intensity method.The ridge of space-time-frequency distribution is extracted,and the singularity detection is performed on the ridge using Haar wavelet transform to estimate the multi-dimensional parameters of the signal.The simulated and pool test results show that the normalized root mean square error of the obtained parameters is less than 0.05 when SNR is greater than-5 dB.