基于单光子计数激光雷达的Bayesian检测研究

Research on Bayesian detection for single photon counting lidar

在单光子计数激光雷达检测领域,目前的检测方法在低信噪比情况下虚警概率会增加,同时也无法适应噪声变化的问题。针对这些问题,提出了一种基于Bayesian的检测方法,该方法首先通过雷达方程估计回波信号光子数的范围,将其作为先验信息,而后结合二项分布建立了累计概率模型,基于Bayesian判决准则计算得到检测阈值,此阈值能够在检测概率与虚警概率中间择其平衡。这种方法不仅克服了低信噪比检测困难的情况,还减少了先验信息的获取难度。实验结果表明,对比固定阈值其虚警概率降低了10倍。对比“恒虚警”其检测概率提高了约20%。验证了方法具有良好的检测效果,具备一定的可操作性。

In the field of single photon counting lidar detection,the current detection methods have an increased probability of false alarms at low signal-to-noise ratios,as well as the inability to adapt to noise variations.In response to these problems,a Bayesian-based detection method is proposed,which first estimates the range of the photon number of the echo signal by the radar equation and uses it as a priori information,and then combines the binomial distribution to establish a cumulative probability model,and calculates the detection threshold based on the Bayesian judgment criterion,which is a balanced choice between the detection probability and the false alarm probability.This method not only overcomes the difficulty of low signal-to-noise ratio detection,but also reduces the difficulty of obtaining prior information.The experimental results show that the false alarm probability is reduced by a factor of 10 compared to a fixed threshold.The detection probability is increased by about 20%compared with"constant false alarm".It is verified that the method has good detection effect and has certain operability.