基于对称小波降噪及非对称高斯拟合的激光目标定位

Laser Target Localization Based on Symmetric Wavelet Denoising and Asymmetric Gauss Fitting

针对数字体制机载多脉冲激光测距机信号处理算法引入测距误差的问题,提出减少脉冲激光回波波形失真的降噪算法以及改善激光测距精度的校正措施,可用于脉冲激光目标的精确定位。结合数字体制多脉冲激光目标检测过程分析数字信号处理算法引入的测距误差,指出信号采样率限制了目标定位的距离分辨率,低通滤波环节的非线性相频特性是造成波形失真、峰值点偏移的主要因素。然后,针对这些问题给出了改善多脉冲激光目标定位精度的具体算法:基于对称小波基实施回波信号小波分解,利用改进Donoho阈值处理小波系数,用以减小重构目标波形峰值点位置的偏移;采用差分组合函数搜索目标回波的峰值点,基于非对称高斯脉冲模型进行高采样率拟合,对峰值点位置加以校正,改善定位的精确性。通过仿真选取合适的小波基,对比不同算法信噪比改善及峰值点定位的性能。在室外高塔开展消光比激光测距实验,结果显示信噪比为1.5的激光目标定位精度小于2m,验证了多脉冲激光目标精确定位算法的性能。

Aiming at the problem of range error caused by signal processing algorithm of digital airborne multi-pulse laser rangefinder, a noise reduction algorithm to reduce the distortion of pulsed laser echo waveform and correction measures to improve the accuracy of laser ranging are proposed, which can be used for precise positioning of pulsed laser targets. First of all, the range error introduced by the digital signal processing algorithm is analyzed by combining target detection process of digital multi-pulse laser. It is pointed out that the signal sampling rate limits the range resolution of the target location, and the nonlinear phase frequency characteristic of the low-pass filtering is the main factor that causes the distortion of the waveform and the peak offset. Then, aiming at these problems, a specific algorithm for improving the accuracy of multi-pulse laser target localization is given. The wavelet decomposition of the eeho signal is implemented based on the symmetric wavelet basis, and the wavelet coefficients are processed by the improved Donoho threshold to reduce the offset of the peak position of the reconstructed target waveform. The peak point of the target echo is searched by the differential combinatorial function, and based on the asymmetric Gauss pulse model, the high sampling rate fitting is performed to correct the position of the peak point and improve the accuracy of localization. Finally, the suitable wavelet bases are selected through simulation, the improvement of signal-to-noise ratio （SNR） and the performance of peak point localization with different algorithms are compared. In the outdoor tower, the extinction ratio laser ranging experiment is carried out. The results show that the laser target localization accuracy is less than 2 m when the SNR is 1.5 and the performance of the multi- pulse laser target precise positioning algorithm is verified.