星载光子计数激光测距雷达的实时去噪方法

Real time denoising method for spaceborne photon counting laser ranging radar

星载光子计数体制激光测距雷达系统具有高重频、高精度等显著优势,但也面临原始数据量大且噪声数据占比过高的问题。为适应星上数据通道的传输能力,需压缩原始数据量并保障信号光子的查全率,因此必须发展以硬件为主体的实时去噪算法。本文提出一种粗精结合的快速去噪算法,首先基于激光器发射脉宽、系统噪声率、目标特性以及接收光子事件的局部密度信息进行粗去噪,剔除部分噪声光子;再利用直方图统计,对保留的光子事件进行精去噪,确定信号光子区间及最终的信号光子及其时间信息。通过蒙特卡洛仿真和ICESat-2实测数据对算法进行验证,测试结果表明,本算法查全率大于94%、查准率大于93%、调和平均值大于94%,运行效率提高了10%。算法可以实现光子事件的快速实时去噪,为星上硬件实时去噪处理提供了理论基础。

The spaceborne photon counting laser ranging radar system has significant advantages such as high repetition rate and high precision,but it also faces the problem of large amount of original data and high proportion of noise data.In order to adapt to the transmission capacity of the satellite data channel,it is necessary to compress the original data volume and ensure the recall rate of signal photons,so it is necessary to develop a hardware based real-time denoising algorithm.This article proposes a fast denoising algorithm that combines coarse and fine processing.Firstly,based on the laser emission pulse width,system noise rate,target characteristics,and local density information of received photon events,coarse denoising is performed to remove some noisy photons;then,using histogram statistics,fine denoising is performed on the retained photon events to determine the signal photon interval and the final signal photon and its time information.The algorithm is verified by Monte Carlo simulation and ICESat-2 measured data.The test results show that the recall ratio,precision ratio and harmonic average of the algorithm are more than 94%,93%and 94%respectively,and the operational efficiency is improved by 10%.The algorithm can achieve fast and real-time denoising of photon events,providing a theoretical basis for real-time denoising processing of onboard hardware.